(Lippincott Manual Series) Canan Avunduk - Manual of Gastroenterology - Diagnosis and Therapy-LWW (2008)
(Lippincott Manual Series) Canan Avunduk - Manual of Gastroenterology - Diagnosis and Therapy-LWW (2008)
MANUAL OF GASTROENTEROLOGY:
DIAGNOSIS AND THERAPY
Fourth Edition
79466_FM.QXD 1/8/08 4:09 PM Page ii
79466_FM.QXD 1/8/08 4:09 PM Page iii
MANUAL OF GASTROENTEROLOGY:
DIAGNOSIS AND THERAPY
Fourth Edition
Care has been taken to confirm the accuracy of the information presented and to
describe generally accepted practices. However, the authors, editors, and publisher are not
responsible for errors or omissions or for any consequences from application of the
information in this book and make no warranty, expressed or implied, with respect to the
currency, completeness, or accuracy of the contents of the publication. Application of the
information in a particular situation remains the professional responsibility of the practitioner.
The authors, editors, and publisher have exerted every effort to ensure that drug
selection and dosage set forth in this text are in accordance with current recommendations
and practice at the time of publication. However, in view of ongoing research, changes in
government regulations, and the constant flow of information relating to drug therapy and
drug reactions, the reader is urged to check the package insert for each drug for any change
in indications and dosage and for added warnings and precautions. This is particularly
important when the recommended agent is a new or infrequently employed drug.
Some drugs and medical devices presented in this publication have Food and Drug
Administration (FDA) clearance for limited use in restricted research settings. It is the
responsibility of the health care provider to ascertain the FDA status of each drug or
device planned for use in their clinical practice.
To purchase additional copies of this book, call our customer service department at (800)
638-3030 or fax orders to (301) 223-2320. International customers should call (301) 223-2300.
Visit Lippincott Williams & Wilkins on the Internet: at LWW.com. Lippincott Williams &
Wilkins customer service representatives are available from 8:30 am to 6 pm. EST.
10 9 8 7 6 5 4 3 2 1
79466_FM.QXD 1/8/08 4:09 PM Page v
CONTENTS
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
5. Endoscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
8. Manometric Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
9. Imaging Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
VII
79466_FM.QXD 1/8/08 4:09 PM Page viii
VIII Contents
Contents IX
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484
79466_FM.QXD 1/8/08 4:09 PM Page x
79466_FM.QXD 1/8/08 4:09 PM Page xi
PREFACE
XI
79466_FM.QXD 1/8/08 4:09 PM Page xii
79466_CH01.QXD 1/2/08 11:58 AM Page 1
SOCIAL IMPACT OF 1
DIGESTIVE DISEASES
N early everyone has experienced a digestive illness. Perhaps it was a self-limiting bout
of “intestinal flu” or heartburn or a more serious condition such as chronic ulcerative col-
itis, Crohn’s disease, or carcinoma of the colon. Nevertheless, the social impact of digestive
diseases is probably underappreciated.
Cardiovascular disease and cancer are dramatic illnesses that command a great deal
of public attention. These disorders are associated with a high mortality and generally
affect the older segment of the population. On the other hand, digestive disorders afflict a
large number of people in all age groups and are associated with considerable pain, dis-
ability, and time lost from school, work, and other activities and considerable health care
costs.
In a typical year, digestive disorders affect about 40 million people in the United
States. The illnesses are responsible for the long-term inability to pursue ordinary activities,
such as school or work, in about a million people, and exact an enormous toll in dimin-
ished productivity: more than 150 million days with some restriction in activity, nearly 70
million bed-days, and more than 20 million days lost from work per year. These conditions
also result in 40 million annual physician contacts outside the hospital. The illnesses pre-
dominantly affect the young and middle-aged, with 70% occurring between the ages of 15
and 64 years. In addition to the losses caused by these predominantly chronic conditions,
acute digestive disorders annually affect more than 11 million people and result in 48 mil-
lion days of restricted activity, 27 million bed-days, and 9 million days lost from work.
Considering all digestive diseases, there are more than 4 million hospitalizations per
year, resulting in 28 million hospital days and accounting for 13% of all hospitalizations.
Moreover, many digestive diseases are fatal, causing about 9% of all deaths in the United
States, or about 220,000 deaths per year, of which 61% are caused by malignancies and
16% by chronic liver disease and cirrhosis. A conservative estimate of the economic bur-
den from digestive diseases is in excess of $60 billion per year, or roughly 10% of the total
costs of all illnesses.
Selected Readings
Lembo AJ. The clinical and economic burden of irritable bowel syndrome. Pract.
Gastroenterol. 2007;Sept:3–9.
Longstreth GF, et al. Functional bowel disorders. Gastroenterology. 2006;130:1480–1491.
Martin BC, et al. The annual cost of constipation in the U.S. ambulatory and inpatients
settings [abstract]. Gastroenterology. 2005;128(suppl 2):A-283. Abstract M974.
Singh G, et al. Use of healthcare resources and cost of care for adults with constipation.
Clin. Gastroentrol Hepatol. 2007;5:1053–1059.
Spiller K. Clinical update: irritable bowel syndrome. Lancet. 2007;369:1586–1588.
3
79466_CH02.QXD 1/2/08 11:58 AM Page 4
S urvival is the most important drive for human existence. Thus, eating, digesting,
and eliminating are very important to human beings. Any disruption of any of these func-
tions causes much concern to the individual.
Patient complaints may be directly related to gastrointestinal (GI) dysfunction, such
as difficulty or painful swallowing, keeping foods down or difficulty in eliminating
(i.e., having diarrhea or constipation). The complaints may be indirectly related to the GI
system, such as pain in the chest from gastroesophageal reflux (GER), or pain in the
abdomen or pelvis or from an intestinal disorder. Change in color, consistency, or shape of
one’s stool, unexplained weight loss, jaundice, or abdominal swelling may bring the patient
to the health care professional.
A patient and a physician may differ in their perspectives on the patient’s complaints.
The physician’s orientation generally is in terms of disease categories. The physician wants
to make as accurate a diagnosis as possible and treat accordingly. On the other hand, the
patient comes to the physician with one or several complaints that usually describe signs
or symptoms perceived as “abnormal.” A patient may complain of food sticking on swal-
lowing; the physician thinks of an esophageal disease. A patient complains of yellow eyes
or jaundice; the physician wonders if the patient has hemolysis or liver disease or biliary
obstruction.
This divergence of orientation is sometimes a help and sometimes a hindrance. It is
a help when the health care professional understands that diagnostic categories are
merely aids to understanding and dealing with disease. It is a hindrance when the health
care provider relies unquestioningly on an established or an apparent diagnosis and
ignores other possibilities. For example, a 55-year-old woman with years of irritable
bowel symptoms who develops pencil-thin stools should not simply be reassured and
sent home; she needs an evaluation for carcinoma of the rectum or colon. On the other
hand, the 45-year-old man with chest pain does not necessarily have coronary heart
disease. His symptoms may be related to GER.
In the past 50 years, because of the diagnostic and therapeutic advances in medicine,
and the wide availability of the Internet, the expectations of both patients and physicians
have changed. Diagnostic tests have become much more sophisticated and accurate and
available in most medical facilities. Special expertise in noninvasive or minimally invasive
interventions has replaced invasive surgery as well as exploratory operations. Drugs
specially designed to act on a specific target have replaced drugs with poorly understood
and general effects. Many patients, before they consult their health care provider, go to the
World Wide Web first and gather information on their complaints. This sophistication
generates a milieu for preventive and early diagnostic and therapeutic measures.
Astute physicians and health care professionals are aware of their orientation and their
shortcomings and listen carefully to patients’ complaints. Sometimes the patient has a hid-
den agenda. A 42-year-old woman complains of recent abdominal pain and constipation.
On additional inquiry, it is learned that her mother died recently of colon cancer. This
patient is afraid that she also has cancer but may be reluctant to disclose that fear unless the
physician asks her directly. In another patient, a 32-year-old man with a positive family
history of early onset heart disease, the recent onset of epigastric or chest pain may be
merely the overt manifestation of nearly overwhelming anxiety, which seems to be related
to difficulties he is having at home and at work or exacerbation of gastroesophageal reflux
disease (GERD) or recent onset of angina. The physician should not only listen carefully but
4
79466_CH02.QXD 1/2/08 11:58 AM Page 5
also should be purposely redundant. It is helpful to review the same information several
times, sometimes at different visits: “I know you told me about your abdominal pain last
time, but I would like to go over the information again. When did you first notice the pain
and what was it like?”
Selected Readings
Aziz Q, et al. Brain-gut axis in health and disease. Gastroenterology. 1998;114:599.
Bloomer JR, et al. Intermittent unexplained abdominal pain—is it porphyria? Clin.
Gastroentrol. Hepatol. 2007;5(11):1255–1259.
Drossman DA, et al. Psychosocial factors in the case of patients with GI disorders.
In: Yamada T, ed. Textbook of Gastroenterology. Philadelphia: Lippincott–Raven
Publishers; 2003:636–654.
Knoll BM, et al. 56-year-old man with rash, abdominal pain and orthralgias. Mayo Clin.
Prac. 2007;82(6):745–748.
Zinn W. The emphatic physician. Arch Intern Med. 1994;153:306–312.
79466_CH03.QXD 1/2/08 11:59 AM Page 6
B ecause you are familiar with examining patients, this chapter is not intended to be
comprehensive, but rather to orient you to the essential components of examining a patient
with digestive complaints.
The physician should be sensitive to the patient’s physical and emotional comfort.
Is the patient in as comfortable a position as possible? Is the patient warm enough? Have
you ensured the patient’s privacy by closing doors and adjusting drapes? How does the
patient feel about others in the room—colleagues, residents, students? Saying “This may be
a little awkward or embarrassing for you” may reassure and relax the patient. As you con-
duct the examination, you should inform the patient of what you intend to do, particularly
with regard to aspects of the examination that may be sensitive and embarrassing.
II. THE ABDOMINAL EXAMINATION. The abdomen conventionally is divided into four
quadrants: right upper, left upper, right lower, and left lower. It is also common,
however, to refer to areas of the abdomen by more specific terms, such as epigastric,
periumbilical, suprapubic, and right or left flank (Fig. 3-1).
A. Patient position. The patient should lie in a supine position, although the head
may be elevated slightly for comfort. Some patients lift their arms over their heads,
which tightens the abdominal wall and makes palpation and interpretation of signs
of peritoneal irritation difficult. The arms should remain at the patient’s side.
Flexion of the knees also may relieve abdominal tightness.
B. Inspection
1. Skin of the abdomen. Are there any scars, dilated veins, rashes, or other
marks?
2. Is the umbilicus normal? Is there an umbilical or abdominal wall hernia?
3. Contour of the abdomen. Is the abdomen protuberant or concave? Are there
any bulges?
4. Can you see peristaltic movements or pulsations?
C. Auscultation. In examining the abdomen, it is probably wise to listen before
performing percussion and palpation because these maneuvers may alter the
quality of bowel sounds.
1. Bowel sounds. What is the character of the bowel sounds? In healthy people,
the character and frequency of bowel sounds vary widely. In people who are
hungry, bowel sounds may be active, whereas after a meal the abdomen typi-
cally becomes rather quiet. Bowel sounds may be increased in frequency and
intensity in diarrheal conditions. Intestinal obstruction is characterized initially
by increased bowel sounds that progress to high-pitched tinkling sounds and
6
79466_CH03.QXD 1/2/08 11:59 AM Page 7
rushes. On the other hand, bowel sounds become decreased or absent in condi-
tions that cause paralytic ileus, such as peritonitis or electrolyte imbalance.
2. Other abdominal sounds. The physician should listen carefully also for
sounds in the abdomen other than bowel sounds.
a. Arterial bruits may indicate narrowing or turbulence in the aorta or renal,
iliac, or femoral arteries, depending on location. In some patients with renal
artery stenosis, the bruit is heard best over the back in the lumbar area.
b. A venous hum is a soft sound in both systole and diastole that indicates
increased collateral circulation between the portal and systemic venous sys-
tems. This rare finding usually is associated with cirrhosis of the liver.
c. A friction rub may be heard with respiration over the ribs that overlie the liver
or spleen. The grating sound of a friction rub is caused by inflammation of the
peritoneal surface, which may be caused by tumor, infection, or infarction.
D. Percussion and palpation. Most experienced physicians combine percussion and
palpation in the examination of the abdomen. Because patients typically are some-
what tense during the abdominal examination, the examiner should avoid sudden
movements. Much more information can be elicited by proceeding slowly and gen-
tly and sometimes firmly.
First, place one hand gently on the abdomen. This will tell you whether the
abdomen is tense, firm, or soft and whether the patient is made uncomfortable by
light pressure. Percuss the abdomen lightly in all four quadrants to determine tym-
pany and dullness.
1. Liver. The size, shape, and consistency of the liver are best estimated by first
percussing the upper and lower limits of liver dullness and then palpating the
lower edge in both phases of respiration. Remember to feel for the liver not
only in the right upper quadrant but also in the epigastrium. Some patients
with hepatomegaly have predominantly left lobe enlargement, which may be
evident only in the epigastrium. Occasionally, a markedly enlarged left lobe is
mistaken for the spleen.
2. The spleen can be identified as an area of dullness just above the left costal
margin posterior to the midaxillary line. In adults, the normal spleen usually
cannot be palpated, although about 10% of teenagers and young adults have
normal, palpable spleens. When the spleen enlarges, it expands anteriorly and
79466_CH03.QXD 1/2/08 11:59 AM Page 8
III. THE ANORECTAL EXAMINATION is an important part of the general physical exam-
ination. It may be uncomfortable for the patient and cause embarrassment, but the
discomfort and embarrassment can be ameliorated greatly by an understanding,
unhurried, gentle attitude on the part of the physician.
A. Patient position. Most physicians ask the patient to lie on the left side with knees
and hips flexed. Others prefer the patient to stand, leaning over an examining
table. The former position is preferred and recommended.
B. Perianal area. The buttocks should be spread apart to allow inspection of the
anus and perianal skin. Protuberant hemorrhoids, anal tags, fissures, or abscesses
may be seen. The perianal area is palpated with a gloved finger. Tenderness and
masses are noted.
79466_CH03.QXD 1/2/08 11:59 AM Page 9
C. Insertion. Before inserting a gloved index finger into the rectum, the physician
lubricates both the finger and the anus well. Sometimes it is helpful to ask the
patient to strain down gently, as if having a bowel movement. This maneuver
relaxes the anal sphincter and allows easier penetration of the finger. The longitu-
dinal axis of the anal canal is aimed roughly to the umbilicus; thus, the examiner
gently presses the finger into the anal canal in the direction of the umbilicus. The
rectum is found to turn posteriorly and open out into the hollow area formed by
the sacrum and coccyx.
D. Examination of the rectum should be gentle. A tender anal canal may result from
proctitis, inflamed or thrombosed hemorrhoids, a stricture, or a fissure. Note the
tone of the anal sphincter. Palpate the lateral, posterior, and anterior walls of the
rectum, noting any tenderness or nodularity. Anteriorly, in male patients, the
prostate is felt as a bilobed structure with a median sulcus. What are the size,
shape, and consistency? Is the prostate tender? Are there any nodules? In female
patients, the cervix (and sometimes a vaginal tampon) can be felt as a firm mass
through the anterior rectal wall.
Although most of the rectal wall that is accessible to the examining finger is
below the peritoneal reflection, most examiners can reach above the peritoneal
reflection anteriorly. Thus, tenderness of peritoneal inflammation or metastatic
nodules may be identified.
E. Stool. On withdrawal of the finger, note the color and consistency of the stool.
Is there any mucus or gross blood? If not, test the stool for occult blood.
Selected Reading
Bates B. The abdomen. In: Bates B, Hoeckelman RA, eds. Guide to Physical Examination
and History Taking. 9th ed. Philadelphia: Lippincott Williams & Wilkins; 2007:339–368.
79466_CH04.QXD 1/2/08 12:00 PM Page 10
T he mind and body are not independent but parallel. “For there are not two
processes, and there are not two entities; there is but one process . . . one entity, seen now
inwardly as mind, now outwardly as matter, but in reality an inextricable mixture and
unity of both. Mind and body do not act upon each other, because they are not other, they
are one.” Baruch Spinoza, as paraphrased by Will Durant, The Story of Philosophy, 1954.
An extensive review of the psychological and emotional aspects of digestive disor-
ders is beyond the scope of this book. Because psychological factors do play an important
role in gastroenterologic disorders, however, and patients certainly react emotionally to
illness, we need to consider at least briefly the psychosomatic component of gastrointesti-
nal disease.
Physicians tend to think of disorders as affecting organ systems and resulting in
observable pathophysiologic changes. Most astute physicians and health care providers,
however, either consciously or intuitively understand that a disorder is a complex interac-
tion among pathophysiologic processes, the patient’s emotional makeup, and the patient’s
perception of the disorder. A large number of factors may affect this interaction, including
the patient’s age, sex, socioeconomic and marital status, other “medical” disorders, emo-
tional stress, family history, and society’s view of the disorder. Additional complicating
influences are the physician’s view of the patient, the patient’s disorders, and the options of
managing the patient’s problems.
There is more than a mere relation between the “psyche” and the “soma.” Although
there is a tendency to separate the psychological and somatic aspects of a disease, attach
labels, and categorize, disorders exist as inseparable entities of psyche and soma. In
some patients the psychological and emotional aspects seem to predominate and exert
profound somatic effects, such as in bulimia, anorexia nervosa, and irritable bowel
syndrome. Also, it is believed that exacerbations of some chronic diseases, such as Crohn’s
disease and ulcerative colitis, may be triggered by emotional factors, yet have their patho-
genesis firmly rooted in the soma. In addition, in these chronic, sometimes devastating ill-
nesses, it is easy to understand that patients also experience psychological trauma due to
their illness.
Finally, no matter what the illness is and no matter how trivial it appears, illness
of any degree represents a threat to the patient’s integrity. An illness is never purely
somatic—there is always a person who must experience the illness.
Selected Readings
DiBaise JK. Psychotherapy and functional dyspepsias: Brain-gut interactions. Am J
Gastroenterol. 2001;96:278.
Drossman DA, et al. Psychosocial factors in the care of patients with GI disorders.
In: Yamada T, ed. Textbook of Gastroenterology. Philadelphia: Lippincott–Raven
Publishers; 2003:636–654.
Drossman DA. The physician-patient relationship. In: Corazziari E, ed. Approach to the
Patient with Chronic Gastrointestinal Disorders. Milan: Messaggi; 1999:133–139.
Longstreth GF, et al. Severe irritable bowel and functional abdominal pain syndromes:
Managing the patient and health care costs. Clin Gastroenterol Hepatol. 2005;3:
397–400.
Talley NJ, et al. Practice Parameters Committee of the American College of Gastroenterology.
Guidelines for the management of dyspepsia. Am J Gastroenterol. 2005;100:2324–2337.
10
79466_CH05.QXD 1/2/08 12:01 PM Page 11
ENDOSCOPY 5
S ince Hirschowitz and colleagues described the first flexible fiberoptic endoscopes,
the technical explosion in fiberoptic endoscopy has produced instruments that are durable,
safe, relatively comfortable, and capable of visualizing the gastrointestinal tract with great
diagnostic accuracy.
The typical modern fiberoptic endoscope is a highly sophisticated instrument
(Fig. 5-1). The shaft is 8 to 12 mm in diameter. The fiberoptic bundles pass through the
shaft to transmit light to the tip and the image to the endoscopist. In the handle of the
endoscope are controls for maneuvering the tip and buttons to regulate irrigation water, air
insufflation, and suction for removing air, secretions, and blood. An instrument channel
allows the passage of biopsy forceps, small brushes for obtaining cytology samples, snares
for removing polyps and foreign bodies, and devices to control bleeding.
The adaptation of endoscopes to video monitoring systems has been developed using
computer chip technology and now is in routine use in all endoscopy suites. The endo-
scopist conducts the examination by viewing the video screen, not by looking directly
through the fiberoptic system of the endoscope. The video screen also allows a variable
number of people, including the patient if desired, to observe what the primary endoscopist
is seeing, and the procedure can be videotaped for clinical and educational purposes.
Figure 5-1. Flexible fiberoptic endoscope for examination of the upper gastrointestinal tract. An
open biopsy forceps is shown at the tip of the endoscope, having been passed through the biopsy
channel in the handle. This instrument also has been adapted to perform endoscopic ultrasonogra-
phy by the addition of an ultrasound transducer at the tip. (GF-UE160-AL5, reprinted with Courtesy
of Olympus®.)
13
79466_CH05.QXD 1/2/08 12:01 PM Page 14
Figure 5-2. Endoscopic ultrasound system. The flexible endoscope on the right of the table has
ultrasound transducers at the tip of imaging and fiberoptics that allow direct visualization of lesions.
(SSD-ALPHA5, reprinted with Courtesy of Olympus®.)
During routine upper gastrointestinal endoscopy, the entire esophagus, stomach, and
proximal duodenum are examined. Several therapeutic endoscopic techniques have been
developed that allow endoscopists to treat bleeding lesions (see Chapter 14) and relieve
esophageal obstruction caused by benign or malignant processes. Endoscopic placement of
gastric feeding tubes—percutaneous endoscopic gastrostomy (PEG)—has largely replaced
surgical gastrostomy.
A variation of upper gastrointestinal endoscopy, endoscopic retrograde cholan-
giopancreatography (ERCP), combines endoscopic and radiologic techniques to visualize
the biliary and pancreatic ductal systems. ERCP methods also have been used therapeuti-
cally to perform sphincterotomy of the sphincter of Oddi, to remove common bile duct
stones, and to place stents that bypass obstructing lesions.
During examination of the lower gastrointestinal tract by colonoscopy, a skilled
endoscopist can reach the cecum in more than 95% of patients, and in many instances the
terminal ileum can also be visualized. The major therapeutic capability of colonoscopy,
popularized in the United States in the mid-1980s by President Ronald Reagan’s experience
with a cancerous polyp, is polypectomy, usually by electrocautery.
An important development in endoscopy is endoscopic ultrasonography (EUS)
(Figs. 5-1–5-3). High-frequency, high-resolution ultrasound transducers that are built into
the tip of the endoscope can be passed to sites that are relatively close to the target organ.
Applications include evaluation of lesions involving the esophagus, mediastinum, stomach,
duodenum, pancreas, colon, and rectum. EUS compares favorably with computerized
tomography in tumor staging and determination of lymph node involvement. Indications
for EUS are summarized in Table 5-1. Also, lymph nodes in the thorax and abdomen may
be biopsied and pancreatic pseudocysts may be drained using EUS.
79466_CH05.QXD 1/2/08 12:01 PM Page 15
Chapter 5: Endoscopy 15
Figure 5-3. Endosonographic view of a villous adenoma (arrows). The tumor is invading the sec-
ond hyperechoic layer ( T2 lesion).
79466_CH05.QXD 1/2/08 12:01 PM Page 16
Chapter 5: Endoscopy 17
Sigmoidoscopy
Condition EGD ERCP or colonoscopy
Dysphagia X
Caustic or foreign body ingestion X
Dyspepsia X
Persistent nausea and vomiting X
Need to obtain small-intestinal biopsy X
Acute or chronic gastrointestinal bleeding X X
Inflammatory bowel disease X X
Chronic abdominal pain X X X
Suspected polyp or cancer X X X
Obstructive jaundice X
Change in bowel habits X
Diarrhea longer than 1 wk X
EGD, esophagogastroduodenoscopy (upper gastrointestinal endoscopy); ERCP, endoscopic retro-
grade cholangiopancreatography.
an hour or more after the procedure, it is required that the outpatients arrange for
transportation home with an accompanying person after EGD.
Deep sedation, using Propofol IV, may be used in selective patients with med-
ical or psychiatric comorbid conditions.
D. Therapeutic EGD. A number of methods have been used to treat actively bleeding
lesions of the upper gastrointestinal tract. Endoscopic band ligation and/or
injection sclerotherapy of esophageal varices is the most widely practiced thera-
peutic endoscopic method for the treatment of bleeding esophageal varices.
Sclerotherapy or banding reduces both mortality and the frequency of rebleeding
from esophageal varices. Other methods of controlling actively bleeding ulcers and
erosions include local injection of the bleeding site with epinephrine or hyper-
tonic saline, electrocoagulation, and/or placement of clips on the bleeding
lesion and laser photocoagulation. These methods are described in more detail
in Chapter 14. Endoscopic treatment of gastroesophageal reflux disease (GERD)
will be discussed in Chapter 20.
E. Complications of upper gastrointestinal endoscopy. Endoscopy of the gas-
trointestinal tract is generally regarded as safe, but adverse events do occur (Table 5-4).
Major complications of EGD—perforation of the esophagus or stomach, genera-
tion of new hemorrhage, pulmonary aspiration, serious cardiac arrhythmia—occur
with a frequency of from 1 in 1,000 to 1 in 3,000 instances. Mortality ranges from
1 in 3,000 to 1 in 16,000 endoscopies. Sedation and inhibition of the gag reflex
contribute to the risk of aspiration of blood, secretions, and regurgitated gastric
contents.
79466_CH05.QXD 1/2/08 12:01 PM Page 18
Perforation of viscus X X
Bleeding X X
Cardiac arrhythmias X X
Medication reactions X X
Vasovagal reactions X X
Pulmonary aspiration X
Cardiac failure (due to overhydration during X
bowel preparation)
Hypotension (due to underhydration during X
bowel preparation)
EGD, esophagogastroduodenoscopy.
Chapter 5: Endoscopy 19
C. Stent insertion. Other techniques may also be used with ERCP to insert short
tubes, or stents, through obstructing lesions, such as pancreatic or biliary cancer or
strictures of the distal bile or pancreatic duct. The stents allow relief of the obstruc-
tion and may be used as definitive therapy in patients with inoperable disease or
temporarily until the obstructive lesion is treated by irradiation or surgery.
reaching the descending colon and may even reach the splenic flexure. Further-
more, patients tolerate flexible sigmoidoscopy better than rigid sigmoidoscopy.
2. Preparation of the patient. Tap water or commercial enemas (i.e., Fleet
enemas) usually are sufficient preparation for either rigid or flexible sigmoi-
doscopy. However, the sigmoidoscopic examination should be performed
without enema preparation when patients have watery diarrhea or suspected
colitis. Most patients do not require conscious sedation. For flexible or rigid sig-
moidoscopy, the patients are usually placed in the left lateral decubitus position.
3. Biopsy and polypectomy. As in EGD, the mucosa and lesions within the rec-
tum or colon can be biopsied through the rigid or flexible instrument. Most
endoscopists avoid biopsy immediately before barium enema because of the
small chance of introducing air and barium into the mucosal defect caused by
biopsy. Also, although polyps can be removed through either the rigid or flexi-
ble sigmoidoscope by standard electrocautery methods, this should not be done
unless the entire colon has been properly prepared to avoid the possibility of
gas ignition or explosion.
C. Colonoscopy
1. Instruments. Modern fiberoptic colonoscopes are similar in design to upper gas-
trointestinal endoscopes but are longer, ranging in length from 120 to 180 cm.
2. Preparation of the patient. Standard bowel preparation for colonoscopy is
similar to the preparation for barium enema x-ray examination: 1 to 2 days of
clear-liquid diet followed by a strong cathartic.
Liquid preparations include a balanced electrolyte polyethylene glycol
(PEG) lavage, exemplified by GoLYTELY or Colyte. Typically about 3.8 L of the
solution must be consumed either orally or through a nasogastric tube over
about 2 to 4 hours. The lavage solution is consumed 6 to 12 hours before the
procedure. The advantages of this preparation are that it is quick, gentle and
very effective in cleansing the bowel. The PEG solution also appears to be safe
to use in anyone in whom it is safe to perform colonoscopy. One drawback is
the difficulty some patients have in consuming a gallon of fluid over a few
hours. Alternatively, 2 L of HalfLytely may be given after 4 Dulcolax tablets.
MoviPrep, a newer, better tasting 2-L PEG solution, may be used instead with-
out additional need for Dulcolax tablets.
An easier colon preparation is Fleet phospho-soda, 1.5 oz. in 4 oz. of
water taken twice, 6–8 hours apart, along with a clear-liquid diet. However, the
safety of this preparation has been questioned due to the possibility of causing
dehydration and renal insufficiency in some patients.
Colon preparation has recently been further simplified by giving patients
28 to 32 tablets containing Fleet phospho-soda (Visicol or Osmo-prep) 6 to 12
hours prior to the procedure, along with a clear-liquid diet.
Because colonoscopy generally takes longer and is more uncomfortable
than flexible sigmoidoscopy, it is customary to administer intravenous medica-
tions such as midazolam ( Versed ), meperidine hydrochloride ( Demerol), or
fentanyl to promote relaxation and diminish discomfort. In some cases, intra-
venous Propofol may be administered by an anesthesiologist for deeper sedation.
3. Biopsy and therapeutic procedures. Colonoscopic biopsies of the mucosa
and colonic lesions are obtained routinely for diagnostic purposes. The most
common colonoscopic therapeutic procedure is polypectomy. Pedunculated
polyps may be removed with a wire snare, which is maneuvered to encircle the
stalk. After the snare is tightened, the stalk is severed directly or by passing
electric current through the snare. Sessile polyps may be removed by polypec-
tomy in several pieces or biopsied through the colonoscope. Raising a bleb with
normal saline injection under a sessile polyp allows more complete removal as
routine polypectomy. Some sessile lesions may require surgical removal.
Bleeding lesions of the colon may be treated during colonoscopy in the same
manner as upper gastrointestinal lesions.
D. Complications of colonoscopy. The two major complications of colonoscopy,
perforation and hemorrhage, occur in less than 1% of instances. These are more
79466_CH05.QXD 1/2/08 12:01 PM Page 21
Chapter 5: Endoscopy 21
Figure 5-5. Photomicrograph of a normal human small-intestinal biopsy taken from the distal
duodenum. (From Eastwood GL. Core Textbook of Gastroenterology. Philadelphia: Lippincott
Williams & Wilkins; 1984:105. Reprinted with permission.)
Chapter 5: Endoscopy 23
Selected Readings
Avunduk C, et al. Endoscopic sonography of the duodenum, pancreas, liver and the biliary
tract: Findings in benign and malignant conditions. Appl Radiol. 1997;25–33.
Avunduk C, et al. Endoscopic Sonography of the stomach: Findings in benign and
malignant lesions. Amer J Roentgenol. 1994;163:591–595.
79466_CH05.QXD 1/2/08 12:01 PM Page 24
E xamination of the abdominal cavity and its organs by means of a laparoscope has
been available for nearly a century. Until recently, laparoscopy was largely a diagnostic
procedure; the instruments were used primarily to visualize and biopsy abdominal organs
and other structures, although some treatment was possible in the form of aspiration of
cysts and abscesses, lysis of adhesions, ligation of the fallopian tubes, and ablation of
endometriosis or cancer by laser. In recent years, rapid and dramatic developments in
operative laparoscopy have made laparoscopic cholecystectomy and appendectomy com-
monplace; other, more complicated operative procedures, such as partial gastrectomy and
partial colectomy, have been described using laparoscopic methods.
I. LAPAROSCOPY
A. Indications. The reliance on diagnostic laparoscopy varies from one medical center
to another. In some centers, laparoscopy is used routinely in evaluation of the
abdominal conditions discussed in this chapter, whereas in other centers it is used
rarely. This variability in the use of diagnostic laparoscopy can be attributed to the
advances in computed tomography (CT) and other imaging techniques of the past
decade, which have provided alternatives to laparoscopy that are either less invasive
or more readily available. In hospitals and medical centers where diagnostic
laparoscopy is available, it is usually performed for the following indications.
1. Biopsy of the liver. This procedure may be done to evaluate a diffuse condi-
tion of the liver, such as cirrhosis or an infiltrating disease, or to biopsy a focal
defect of the liver that has been identified by CT or ultrasound examination.
During laparoscopy, the appearance of the liver also can be assessed; for exam-
ple, the collateral vessels of portal hypertension or nodularity of cirrhosis or
neoplasm may be evident.
2. Determination of cause of ascites. When the cause of ascites is unknown,
laparoscopic examination of the abdominal organs, the omentum, and the peri-
toneum may provide an answer. The most common causes are disseminated
cancer, usually ovarian, and cirrhosis.
3. Staging of Hodgkin’s disease and non-Hodgkin’s lymphoma.
4. Evaluation of patients with fever of unknown origin.
5. Evaluation of patients with chronic or intermittent abdominal pain.
Diagnoses include abdominal adhesions, Crohn’s disease, appendicitis, and
endometriosis.
B. Contraindications include a perforated viscus, abdominal wall infections, diffuse
peritonitis, and clinically significant coagulopathy. Chronic lung disease and con-
gestive heart failure are relative contraindications. If laparoscopy is necessary in
those instances, sedation should be minimized and the patient should have pulse
oximetry and cardiovascular monitoring. Tense ascites interferes with adequate
visualization and should be treated before attempting laparoscopy.
C. Technique. Most diagnostic laparoscopic procedures are performed electively in
patients who have fasted and are under sedation and local anesthesia. A small skin
incision is made, usually above and to the left of the umbilicus, avoiding surgical
scars and abdominal masses. Nitrous oxide or carbon dioxide gas is introduced by
a needle to distend the abdomen, and a trocar and cannula are passed through the
incision into the peritoneal cavity. The laparoscope is inserted into the abdomen
25
79466_CH06.QXD 1/2/08 12:02 PM Page 26
and, by maneuvering the instrument or positioning the patient, most of the abdom-
inal contents can be examined. Tissue samples can be obtained by brushes, nee-
dles, or forceps that are passed through the laparoscope. At the conclusion of the
examination, the gas is withdrawn, the instrument is removed, and the small inci-
sion is closed with sutures or clips.
D. Therapeutic laparoscopy. Some conditions can be treated by the laparoscopic
techniques described. These treatments include aspiration of cysts and abscesses,
lysis of adhesions, ligation of the fallopian tubes, and ablation of endometriosis or
cancer by laser. However, the most dramatic advances in therapeutic laparoscopy
have been in the area of operative laparoscopy, described in the following sections.
II. LAPAROSCOPIC SURGERY. The recent advances in laparoscopic surgery have been
the result of two important contemporary factors: (a) the remarkable continuing devel-
opments in fiberoptic technology and (b) the strong economic incentives to minimize
the duration of hospitalization and use of inpatient hospital facilities.
A. Laparoscopic cholecystectomy
1. Indications and contraindications for laparoscopic cholecystectomy are the
same as for traditional operative cholecystectomy (Table 6-1).
2. Technique. The equipment and instruments required to perform laparoscopic
surgery are reviewed in the selected reading by Gadacz and associates. Briefly,
the equipment includes a powerful (xenon) light source, a carbon dioxide insuf-
flator, a high-resolution end-viewing camera with a high-resolution video
monitor, an irrigation device that instills fluid at a high flow rate, and an elec-
trocautery or laser device. A variety of instruments also are needed, including a
Veress needle for insufflation, cannulas with trocars, endoscopes, retractors,
graspers, dissectors, a clip applier, irrigation and aspiration catheters, coagula-
tors, and catheters for performing cholangiography.
Laparoscopic cholangiography and cholecystectomy can be performed
under general or epidural anesthesia. Preoperative antibiotics generally are used
at the discretion of the operator but are indicated in patients with recent chole-
cystitis, heart valve prostheses, and other medical risk factors. Before beginning
the procedure, the urinary bladder is drained with a Foley catheter, and the
stomach is decompressed with a nasogastric tube. Two video monitors, one on
each side of the operating table, allow all members of the operating team a view
of the procedure. Several cannulas are inserted through the abdominal wall for
insufflation and surgical manipulation; the surgical laparoscope is inserted just
above the umbilicus. A detailed description of the laparoscopic cholecystectomy
procedure is provided by Gadacz, et al.
3. Results. As experience with laparoscopic cholecystectomy increases, the proce-
dure is recognized to be safe and effective. Operative time is less than 2 hours.
Most patients are able to leave the hospital in fewer than 2 days and are able to
return to work more quickly than after standard operative cholecystectomy.
Thus, there are economic savings both in decreased hospital costs and in
reduced time away from work. Fewer than 5% of patients require standard
laparotomy because of a complication of the laparoscopic procedure, such as
bleeding, bile duct leak, bile duct injury, or technical difficulties.
B. Other laparoscopic operations. Laparoscopic appendectomy and inguinal
herniorrhaphy are performed routinely, and other more extensive abdominal oper-
ations, such as gastrectomy, colectomy, esophageal fundoplication, gastric stapling,
and intestinal bypass are also performed, but require more advanced expertise.
Laparoscopic techniques are also used to treat pulmonary and pericardial lesions
in the chest. Intraluminal laparoscopic surgery via the lumena of the GI tract, e.g.,
stomach, is the promising new era of surgical management of intraabdominal
diseases.
Selected Readings
Anvari M, et al. Five-year comprehensive outcomes evaluation in 181 patients after
laparoscopic Nissen fundoplication. J Am Coll Surg. 2003;196:51–57.
Doherty GM, et al. Current Surgical Diagnosis and Treatment. 12th ed. New York, NY:
McGraw-Hill; 2006:662–667.
Giger U, et al. Laparoscopic cholecystectomy in acute cholecystitis: indication, technique,
risk and outcome. Langenbecks Arch Surg. 2005;390:373–80.
Gurusamy KS, et al. Early versus delayed laparoscopic cholecystectomy for acute cholecystitis.
Cochrane Database Syst Rev. 2006;4:CD005440.
Iannelli A, et al. Therapeutic laparoscopy for blunt abdominal trauma with bowel injuries.
J Laparoendosc Adv Surg Tech A. 2003;13:189–191.
Kalloo An, et al. Flexible transgastric peritoneoscopy: A novel approach to diagnostic and
therapeutic interventions in the peritoneal cavity. Gastrointest Endosc. 2004;60:
114–117.
Kamoz T, Granderath F, Pontner R. Laparoscopic antireflux surgery. Surg Endosc.
2003;17:880–885.
Madan A. Laparoscopic bariatric surgery. US Gastroenterol Rev. 2000;1:29–33.
McQuay N, et al. Laparoscopy in the evaluation of penetrating thoracoabdominal trauma.
Am Surg. 2003;69:788–791.
Polanivelu C, et al. Laparoscopic lateral pancreaticojejunostomy: A new remedy for an old
ailment. Surg Endosc. 2006;20:458–461.
Roggin KK, et al. What is the long term safety and efficacy of laparoscopic resection for
gastric and gastrointestinal stroma tumors? Nat Clin Pract Gastroenterol Hepatol.
2007;4:76–78.
Soper NJ, et al. Laparoscopic general surgery. N Engl J Med. 1994;330:409.
Tom J, et al. Laparoscopic surgery for Chron’s disease: a meta analysis. Dis Colon Rectum.
2007;50(5):576–585.
79466_CH07.QXD 1/2/08 12:04 PM Page 28
Uncooperative patient
Bleeding disorder
Infection in skin, pleura, right lower lung, or peritoneum overlying the liver
Suspected liver abscess or vascular lesion
Difficulty in determining liver location, as with ascites
Severe extrahepatic obstruction
28
79466_CH07.QXD 1/2/08 12:04 PM Page 29
maximal liver dullness between the anterior and midaxillary lines. If the biopsy is
being done under ultrasound, the site is chosen by the radiologist.
C. Preparation of the biopsy site. Using sterile gloves, cleanse the biopsy site and
surrounding skin with acetone-alcohol and iodine solutions, and arrange sterile
drapes around the biopsy site. Infiltrate the skin with local anesthetic. The anes-
thetic also may be infiltrated beneath the skin to the liver capsule, with care taken
to keep the needle at the upper edge of the rib to avoid the artery that runs below
the rib.
D. Taking the biopsy. Make a 4-mm skin incision at the biopsy site with a no. 11
knife blade. Fill a 20-mL syringe with about 10 mL of sterile saline (avoid bacte-
riostatic preparations if you plan to culture the biopsy tissue), and attach the
biopsy needle. (Generally, this is a Menghini or modified Menghini needle.) Insert
the biopsy needle through the skin incision parallel to the surface of the bed, aim-
ing toward the xiphoid. Advance the needle into the intercostal muscles, and flush
the needle with 0.2 to 0.5 mL of saline.
Have the patient hold the breath at full expiration for the intercostal
approach; at full inspiration for the subcostal approach. It is wise to have the
patient practice this maneuver several times before the biopsy is obtained. Apply
constant suction on the syringe and, in a rapid, smooth motion, advance and
Figure 7-1. Photomicrograph of the liver, showing the central area of a classic lobule with a cen-
tral vein. (From Snell RS. Clinical and Functional Histology for Medical Students. Boston: Little,
Brown; 1984:479. Reprinted with permission.)
79466_CH07.QXD 1/2/08 12:04 PM Page 30
Figure 7-2. Photomicrograph of the liver, showing the peripheral area of a classic lobule with a
branch of the hepatic artery (A), a branch of the portal vein (V), and a small bile duct (D). (From Snell
RS. Clinical and Functional Histology for Medical Students. Boston: Little, Brown; 1984:479.
Reprinted with permission.)
withdraw the needle 4 to 5 cm. The total duration of this movement should not
exceed 1 second. Ask the patient to resume breathing. A second biopsy through
the same incision at a slightly different angle will increase the diagnostic yield in
patients with suspected cancer of the liver. The biopsy may be expelled from the
needle temporarily into saline or directly into 10% formalin. Apply an adhesive
bandage over the biopsy wound.
Percutaneous liver biopsies may also be obtained using automatic biopsy
needles. These automatic needle “guns” are usually preferred by radiologists and
Hemorrhage Hemothorax
Bile peritonitis Penetration of abdominal viscera
Pneumothorax
79466_CH07.QXD 1/2/08 12:04 PM Page 31
III. COMPLICATIONS FROM LIVER BIOPSY are infrequent but may be dramatic
(Table 7-2 on page 30). Deaths have been reported in about 1 in 1,000 biopsies and
major complications in about 3 in 1,000. Complications can be minimized by using a
needle 1.2 mm in diameter or less, avoiding biopsy in high-risk patients, and adhering
strictly to protocol for obtaining the biopsy.
Selected Readings
Bravo AA, et al. Liver biopsy. N Engl J Med. 2001;344:495.
Brunt EM, et al. Liver biopsy interpretation for the gastroenterologist. Curr Gastroenterol
Rep. 2000;2:27.
Cjaza AJ, et al. Optimizing diagnosis from the medical liver biopsy. Clin Gastroenterol
Hepatol. 2007;5(8):898–907.
Edmison JM, et al. How good is transjugular liver biopsy for the histologic evaluation of
liver disease. Nat Clin Pract Gastroenterol Hepatol. 2007;4:306–308.
LefKowitch JH. Hepatobiliary pathology. Curr Opin. Gastroenterol. 2006;22:198–208.
Schiaro TD, et al. Importance of specimen size in accurate needle biopsy evaluation of
patients with chronic hepatitis. C. Clin Gastroenterol Hepatol. 2005;3:930–935.
Sherman KE, et al. Liver biopsy in cirrhotic patients. AJ Gastroenterol. 2007;102:789–793.
79466_CH08.QXD 1/2/08 12:04 PM Page 32
8 MANOMETRIC STUDIES
32
79466_CH08.QXD 1/2/08 12:04 PM Page 33
TABLE 8-1
Disorders in Which Esophageal Manometry May Be
Useful in Making a Diagnosis
are within the stomach. As the tube is withdrawn gradually, each probe passes
through the LES. As the probes move up into the body of the esophagus, peristaltic
contractions are recorded sequentially as they pass down the esophagus, first with
the proximal probe, next the middle probe, and last the distal one. When the mid-
dle or distal probe lies within the LES, the relation between esophageal peristaltic
contractions and sphincteric relaxation can be observed (Fig. 8-2). Similarly, when
the upper probe lies in the pharynx and the middle probe is within the UES, the
relation between pharyngeal contractions and UES relaxation can be recorded.
Figure 8-1. The esophageal manometry tube is coiled on a chart recorder. Note the three small
transducers arranged serially, 5 cm apart, beginning 5 cm from the distal end of the tube.
79466_CH08.QXD 1/2/08 12:04 PM Page 34
Figure 8-2. Manometric appearance of normal peristaltic contractions and relaxation of the lower
esophageal sphincter (LES). The upper, middle, and lower probes are each separated by 5 cm.
(From Eastwood GL. Core Textbook of Gastroenterology. Philadelphia: Lippincott Williams & Wilkins;
1984:27. Reprinted with permission.)
C. Interpretation of manometric findings and the role manometry plays in the diag-
nosis of specific esophageal disorders are discussed in subsequent chapters. It is
sufficient at this point to say that some conditions may have quite specific diag-
nostic findings by manometry, such as the hypercontracting LES of achalasia,
which fails to relax completely on swallowing, whereas other conditions, such as
esophageal motor disorders causing chest pain, may be accompanied by nonspe-
cific or normal findings at the time of manometry; in these instances, manometry
is of little diagnostic value.
under conditions that patients experience during their everyday lives. This proce-
dure is particularly useful in documenting nocturnal reflux symptoms or symptoms
that occur in relation to other experiences or activities. Patients are asked to keep
a record of their symptoms. The best correlations with symptoms appear to be
with total time that pH is less than 4.0 and with the number of times intra-
esophageal pH below 4.0 exceeds 5 minutes.
C. A wireless pH monitoring (Bravo) device has been developed to improve patient
comfort and increase pH test sensitivity. It is placed endoscopically in the distal esoph-
agus and allows for prolonged monitoring (48 hours vs. 24 hours). The extended
monitoring period helps to better understand GER physiology by studying the day-
to-day and moment-by-moment variability of GER. It may also be used to assess the
effectiveness of acid suppressive therapy without the need for repeat testing.
D. Multichannel intraluminal impedance (MII) is a new technique for the evalua-
tion of gastroesophageal reflux disease (GERD) and for esophageal motor function
testing. It can assess esophageal bolus transit similar to barium swallow and the
proximal extent of the reflux event. MII can be used in combination with
esophageal manometry for concurrent assessment of esophageal motor function
and motility.
E. High-resolution manometry (HRM) is solid-state manometry with 36 circumferen-
tial sensors spaced at 1-cm intervals (4.2-mm outer diameter). Each of these sensors
contains 12 circumferentially isolated sensors that detect pressure over a 2.5-mm
length of the esophagus.
The procedure involves placing the catheter in the esophagus and recording
10 swallows, then withdrawing. Pressures detected by each sensor are averaged to
obtain a mean pressure measurement for each sensor, making each of the 36 sen-
sors a circumferential pressure detector. The data obtained are then processed by a
computerized program (ManoScan) to create high-resolution plots or conventual
line traces.
HRM is simpler, faster, and more precise. It provides a complete observation
of the esophageal motor function from the pharynx to the LES without the need
for catheter reposition and gives accurate sphincter pressures and assessment of
peristaltic performance.
III. ANORECTAL MANOMETRY may be useful in the evaluation of patients with consti-
pation or stool incontinence. Measurements include documentation of baseline pres-
sures of the internal and external anal sphincters and the relaxation response of the
internal sphincter to distention of the rectal vault by stool or an inflated balloon. For
example, in Hirschsprung’s disease, the internal anal sphincter does not relax ade-
quately in response to rectal distention.
Incontinence from internal anal sphincteric dysfunction can be caused by hemor-
rhoid surgery or neuromuscular disorders affecting smooth muscle. The internal anal
sphincter is under autonomic control, whereas the external anal sphincter is under vol-
untary control. External sphincteric dysfunction may be related to sacral nerve disease
and disorders of striated muscle.
Selected Readings
Adler DG, et al. Primary esophageal motility disorders. Mayo Clin Proc. 2001;76:195.
Fox M, et al. High-resolution manometry predicts the success of oesophageal bolus
transport and identifies clinically important abnormalities not detected by conventional
manometry. Neurogastroenterol Motil. 2004;16(5):533–542.
79466_CH08.QXD 1/2/08 12:04 PM Page 36
Kahrilas PJ. Esophageal motility disorders: current concepts of pathogenesis and treatment.
Can J Gastroenterol. 2000;14:221–230.
Pandolfino J. Esophageal monitoring devices. US Gastroenterol Rev. 2007;(1)23–27.
Pandolfino J, et al. AGA medical position statement? Clinical use of esophageal manometry.
Gastroenterology. 2005;128:107–108.
Park W, et al. Esophageal impedance recording: Clinical utility and limitations. Curr
Gastroenterol Rep. 2005;7:182–189.
79466_CH09.QXD 1/2/08 12:05 PM Page 37
IMAGING STUDIES 9
I t is an unusual patient who, complaining of some digestive disorder, does not have a
radiologic or ultrasound study sometime during the course of the evaluation. In the pre-
endoscopic era, many gastroenterologists performed plain and barium-contrast radiographic
studies in their own offices. Now the term gastroenterologist is virtually synonymous with
endoscopist, and roentgenographic studies are performed (appropriately) by a radiologist.
Nevertheless, all physicians who see patients with digestive complaints should be familiar
with the radiologic, ultrasound, and radionuclide studies that are commonly available.
37
79466_CH09.QXD 1/2/08 12:05 PM Page 38
(see Section I.J) has largely replaced oral cholecystography for identifying gall-
stones, although the two studies may be complementary in some patients.
During oral cholecystography, radiopaque iodine-containing dye is ingested in
the form of tablets, absorbed from the intestine, extracted by the liver, and
excreted into bile, where it is concentrated in the gallbladder, which is then visu-
alized radiographically. Usually the biliary ductal system is not opacified. Failure
to see the gallbladder after a double-dose oral cholecystogram is a strong indi-
cator of gallbladder disease. However, failure to opacify the gallbladder may
also result from poor absorption of the dye, intrinsic liver disease, or a serum
bilirubin level in excess of 3 mg/dL. Today, this technique is rarely observed.
2. During intravenous cholangiography, a radiopaque dye is injected by vein
and, like the oral dye, is extracted by the liver and excreted in the bile.
However, unlike oral cholecystography, both the major bile ducts and the gall-
bladder opacify during intravenous cholangiography. The study is severely lim-
ited when the serum bilirubin level is greater than 3 mg/dL. Furthermore, the
extent and clarity of opacification of the bile duct system is considerably less
than that achieved with ERCP or percutaneous cholangiography. Thus, cur-
rently there is little clinical use for intravenous cholangiography.
3. Percutaneous transhepatic cholangiography (PTC) is performed by passing
a long, thin needle (about 23-gauge) through the skin into the liver. As the nee-
dle is withdrawn slowly, small amounts of radiopaque dye are injected until the
bile duct is visualized. Sufficient dye is subsequently injected to opacify the bil-
iary system. This test is more than 90% successful in patients with extrahepatic
obstruction and dilated bile ducts. It is only 50% to 60% successful in patients
with normal-size ducts. The ready availability of ERCP has decreased the fre-
quency with which PTC is used. Often, when one of the two studies is unsuc-
cessful, the other is successful in visualizing an obstructed biliary system.
G. Intravenous pyelogram. Although an intravenous pyelogram (IVP) is most com-
monly used in the evaluation of the genitourinary tract, it can be of help as an
adjunct to the evaluation of some GI and other abdominal disorders. Retro-
peritoneal tumors and the inflammatory masses that occur in Crohn’s disease may
cause deviation or obstruction of the ureters. Similarly, masses may impinge on the
bladder. Rarely, an enterovesical fistula can be visualized by IVP or by retrograde
cystography.
H. Computed tomography scan. The nearly universal availability of computed
tomography (CT) scanning has altered the diagnostic approach to many patients
with abdominal complaints. The CT scan often aids in making a diagnosis quickly,
abbreviating the diagnostic evaluation, shortening hospital stay, and avoiding
unnecessary studies. For example, lesions in locations and in organs within the
abdomen that formerly was obscure, such as the retroperitoneum and pancreas,
now often can be readily visualized by CT scanning.
The CT scan has largely replaced the radionuclide liver–spleen scan for
detecting masses of the liver because of its ability to resolve smaller lesions. In
addition, the CT scan shows the size and shape of the liver more accurately and
provides information about other abdominal organs that is not available by
liver–spleen scan. Furthermore, the relative densities of mass lesions can be esti-
mated by CT scanning to determine whether such lesions are solid or cystic.
The diagnostic accuracy of the CT scan is enhanced by the injection of con-
trast material intravenously to visualize the vascular system and by oral adminis-
tration of contrast to delineate the GI tract.
The ability to obtain tissue samples by fine-needle aspiration has enhanced
the diagnostic capability of CT scanning even more. In this manner, malignant
lesions, abscesses, and benign cysts can be histologically and cytologically con-
firmed. Abscesses and infected cysts also can be drained and treated by CT-guided
placement of drainage catheters.
CT colography has been developed for examination of the colon, especially
for the detection of colonic polyps and neoplasms. The patients are prepared sim-
ilarly as for colonoscopy. At the time of the scanning, air is introduced into the
79466_CH09.QXD 1/2/08 12:05 PM Page 40
colon, and the colon is kept distended during the CT scanning. Patients do not
receive any sedation and thus the procedure may be painful. Polyps larger than
7 to 10 mm are seen with acceptable accuracy; however, smaller polyps may be
missed by this technique. If a polyp is found, patients are then referred for
colonoscopy.
CT enterography is a newer technique gaining popularity used in the exam-
ination of the small intestine, especially for the detection of Crohn’s disease and its
complications. It is able to detect early lesions (i.e., aphthous ulcers, bowel wall
inflammation and thickening, and early stricture and fistula formation).
I. Angiography. Selective arteriography of the celiac axis, superior mesenteric artery,
inferior mesenteric artery, or their branches may reveal vascular tumors, document
the effects of other tumors on abdominal organs, and identify sites of bleeding. For
extravasation of dye to be seen at a bleeding site, it is estimated that the rate of
bleeding must be at least 0.5 to 1.0 mL per minute.
Selective arteriography can also be therapeutic. Autologous clots or
vasopressin can be injected into appropriate vessels. In patients with cancer,
chemotherapeutic drugs can be delivered directly to the tumor. This procedure has
been used particularly in the treatment of malignant lesions in the liver.
J. Ultrasonography, which depends on the reflection of sound waves to create an
image rather than ionizing radiation, can visualize the organs of the abdomen with
high accuracy by means of modern high-resolution equipment. Ultrasonography
usually is less expensive than CT and is better in detecting some conditions, such as
gallstones. Stomach or bowel gas may obscure visualization of the organs beneath.
The inclusion of an ultrasound device in the tip of an endoscope (endo-
scopic ultrasound) has produced a sophisticated technique for the examination
of the GI tract and adjacent structures (see Chapter 5).
Selected Readings
Beebe TJ, et al. Assessing attitudes roward laxative preparation in colorectal cancer
screening and effects on future testing: potential receptivity to computed tomographic
colonography. Mayo Clin Proc. 2007;82:666–671.
Beets-tan R, et al. Preoperative MR imaging of anal fistulas: Does it really help the
surgeon? Radiology. 2001;218:75.
Gourtsoyiannis NC, et al. Magnetic resonance imaging evaluation of small intestinal
Crohn’s disease. Best Pract Res Clin Gastroenterol. 2006;20:137–156.
Kim DH, et al. CT colonography versus colonoscopy for the detection of advanced neoplasia.
N Surg J Med. 2007;357:1403–1412.
79466_CH09.QXD 1/2/08 12:05 PM Page 42
NUTRITIONAL REQUIREMENTS 10
AND ASSESSMENT
I. METABOLISM
A. Normal energy metabolism. Every day normal adults need roughly 25 to 30 kcal
of fuel/kg of body weight. For a 70-kg person, this is about 2,100 kcal per day. A
typical American derives 40% to 60% of daily calories from carbohydrates, 20%
to 45% from lipids, and 10% to 20% from protein. Glucose is the body’s pre-
ferred source of immediate energy. The healthy body manufactures glucose from
carbohydrates, protein, and the glycerol backbone of triglycerides (TGs). Since
very little carbohydrate can be stored in the body, most of the body’s reserve energy
stores are made up of lipids. There is no storage form of protein. Even though pro-
tein can be metabolized for energy, this results in wasting of lean body mass and
negative nitrogen balance.
B. Metabolism of carbohydrates
1. Glucose. Most of the ingested carbohydrate is broken down to glucose, which
enters the circulation. Glucose is taken up by all cells of the body and burned for
immediate energy. Under normal, well-fed circumstances, the cells of the central
nervous system depend on glucose for energy. However, neurons of the brain
cortex and blood cells can use only glucose as fuel under all circumstances.
2. Glycogen. Glucose, fructose, and galactose can be converted to glycogen, a
polymer of glucose, which is stored mainly in the liver (200 mg) and muscle
(300 mg) as a readily available energy reserve. Liver glycogen can be converted
directly to glucose for systemic distribution. However, muscle glycogen is
burned by muscle itself. Total glycogen stores of the body can meet the body’s
energy needs for 36 to 48 hours.
The body protects its glycogen stores for emergencies. At times of tempo-
rary glucose insufficiency, the body manufactures glucose by gluconeogenesis,
from protein and glycerol of lipids. Also free fatty acids (FFAs) and amino acids
are burned for direct energy. Excess dietary carbohydrate is converted to TGs
for storage in the adipose tissue.
C. Metabolism of lipids. Lipids constitute the body’s main energy reserve. A
nonobese, 70-kg man has 12 to 18 kg of fat stores. This figure is somewhat higher
for women. Fat supplies 9 kcal/g compared to 4 kcal/g from glucose or protein.
Ingested lipids are hydrolyzed in the intestinal lumen, and then absorbed into
enterocytes of the small intestine, where they are resynthesized into TGs. In the
enterocytes, TGs made up of long-chain FFAs form chylomicrons by the addition
of apoproteins. Chylomicrons are secreted into the intercellular space and are
absorbed into the lymphatics. Short- and medium-chain FFAs are directly absorbed
into the portal vein. Lipoprotein lipase of endothelial cells release FFAs from TGs,
and FFAs enter cells of various tissues (e.g., heart and muscle), where they are oxi-
dized for energy, and cells of adipose tissue to form TGs for storage. Adipose tis-
sue can also convert carbohydrates and proteins into TGs by lipogenesis.
Mobilization of TGs for energy by lipolysis begins with the hydrolysis of TGs
into FFAs and glycerol. Glycerol may be either converted into glucose by gluco-
neogenesis or directly oxidized further. FFAs enter some tissue cells, are broken
down to acetyl coenzyme A, and oxidized through the Krebs cycle. At times of
starvation and lack of glucose as fuel, large quantities of FFAs enter tissue cells.
The Krebs cycle may become overloaded, and FFAs may be incompletely oxidized.
45
79466_CH10.QXD 1/2/08 12:06 PM Page 46
Intermediate products in the form of acids and ketone bodies accumulate in the
blood, leading to ketosis and acidosis.
D. Metabolism of protein. The body of a 70-kg man contains 10 to 14 kg of protein.
Because there is no storage form of protein in the body, the protein compartment
must be maintained by daily intake. A typical adult requires about 0.8 to 1.0 g of pro-
tein/kg of body weight per day. For a 70-kg man, this is about 65 to 70 g.
Ingested protein may be used for protein synthesis or fuel, especially if the
body requires more energy than is supplied by carbohydrate and lipid intake. In
this event, body protein may also be catabolized for energy. One third of the body’s
total protein is potentially available as an energy source in case of dire need.
Further protein catabolism, however, severely jeopardizes health.
If protein is ingested in amounts greater than needed for protein synthesis and
energy production, it is stripped of its nitrogen and converted to glucose, glycogen,
and TGs for storage. Protein, besides being needed for structural purposes, is also
needed for replacement, repair, and growth of tissue (cell components) and main-
tenance of circulating proteins (e.g., albumin, transferrin, coagulation proteins,
enzymes, and antibodies).
E. Nitrogen balance. In a healthy adult, ingested protein must supply enough amino
acids to maintain a constant level of body protein. Thus, the intake of protein
must equal or exceed the breakdown of body protein. The effect of diet on the
body’s protein compartment may be approximated by nitrogen balance. Nitrogen
balance is the difference between nitrogen intake and output.
1. When protein synthesis is equal to protein degradation, one has a neutral
nitrogen balance. In adults, this is a sign of health.
2. Positive nitrogen balance occurs when protein synthesis exceeds protein degra-
dation. This suggests tissue growth. This state is normal and expected in chil-
dren. In adults, it may mean rebuilding of wasted tissue.
3. In negative nitrogen balance, the protein breakdown is in excess of protein
synthesis. This catabolic state usually occurs in sepsis, trauma, and burns, and
when the carbohydrate and lipid intake is less than the body’s energy needs,
necessitating use of the body’s own protein for fuel.
4. Calculation of nitrogen balance. Nitrogen balance can be calculated with
reasonable accuracy. Nitrogen makes up 16% of ingested protein. The division
of protein intake in grams by 6.25 (the reciprocal of 0.16) will give nitrogen
intake. Most of the nitrogen output from the body is into the urine as urea,
which can be measured. Other excreted nitrogen is in feces and urine as
nonurea nitrogen, amounting to about 4 g per day. The addition of 4 g to the
urine nitrogen measured will give the daily nitrogen output. Thus, nitrogen bal-
ance can be calculated by subtracting nitrogen output from nitrogen input.
Nitrogen balance nitrogen in nitrogen out
protein intake [daily urinary nitrogen 4 g 6.25
(for nonurea nitrogen)]
For example, if protein intake is 75 g, urine urea nitrogen is 500% or 5 g/L
with 2,000-mL urine output per 24 hours.
Nitrogen balance 75 g protein
[(5 g/L 2 L) 4]6.25 g protein/g nitrogen
12 g nitrogen intake 14 g nitrogen output
22 g per day
F. Energy metabolism in starvation. During periods of starvation, when ingested nutri-
ents are unavailable, the body goes through different stages of metabolic adaptation.
Energy requirements are met by metabolism of substrates from the energy reserves,
which are drawn on simultaneously, but not equally, following a careful sequence.
1. For immediate use, the glycogen in the liver is depolymerized to glucose for
systemic use. Muscle glycogen is oxidized locally. The lactate produced may be
converted to glucose in the liver for systemic use. If used up entirely, the glyco-
gen reserves are depleted in 36 to 48 hours.
79466_CH10.QXD 1/2/08 12:06 PM Page 47
Men
Height (cm) Weight (kg) Height (cm) Weight (kg) Height (cm) Weight (kg)
Women
From Blackburn GL, Bistrian BR, Maini BJ. Nutritional and metabolic assessment of the hospitalized
patient. J Parenter Enteral Nutr. 1977;1:14. Reprinted with permission from the American Society for
Parenteral and Enteral Nutrition.
Mena Womenb
Ideal Ideal
From Blackburn GL, Bistrian BR, Maini BJ. Nutritional and metabolic assessment of the hospitalized
patient. J Parenter Enteral Nutr. 1977;1:15. Reprinted with permission of the American Society for
Parenteral and Enteral Nutrition.
number of ICU days, with the number of days on a ventilator, and with the num-
ber of hospital days. It was the only measurement that correlated with the devel-
opment of both a new infection and ventilator dependency.
C. Rate of nutritional depletion. Not only is the severity of the protein-intake
depletion important, but also its rate of progression. Ingestion of less than 30 g
per day of protein and less than 1,000 kcal per day means that protein–calorie
undernutrition will progress rapidly. Associated fever, infection, trauma, or mal-
absorption will further accelerate the rate of nutritional depletion. It is important
to note that the stress of sepsis induces an obligatory protein loss regardless of the
provision of nutrients. Thus, nutritional status in these patients needs to be closely
monitored.
III. NUTRITIONAL REQUIREMENTS. The body needs seven groups of nutrients: carbo-
hydrates, lipids, proteins, vitamins, electrolytes, trace elements, and water.
The calorie and energy requirements of healthy individuals vary depending on age, body
size, and level of physical activity. Calorie intake should balance energy expenditure to
maintain body weight. The energy requirements of seriously ill or injured patients are
different than in health and depend on the nature and severity of the illness.
A satisfactory estimate of basal calorie requirement or basic energy expenditure
(BEE) can be calculated from the Harris–Benedict equation, which takes into account
sex, height, age, and weight (Table 10-5). For most patients, the Harris–Benedict
equation provides an adequate estimate of caloric needs. However, in some clinical
situations (e.g., impending respiratory failure, active weaning from mechanical venti-
lation), a more individualized and accurate estimate is needed. Indirect calorimetry is
useful in such instances.
Indirect calorimetry is a method of determining caloric expenditure in different
metabolic situations. The carbon dioxide and oxygen levels in inspired and expired air
are measured and compared. The average of several measurements taken over a 10- to
20-minute period gives a close approximation of caloric expenditure.
Indirect calorimetry also provides the respiratory quotient (RQ), which reflects
the number of carbon dioxide molecules produced per molecule of oxygen consumed.
The RQ varies, depending on the metabolic substrate. For example, the RQ is 1.0 for
pure carbohydrate metabolism, 0.8 for protein, and 0.7 for fat. Because respiratory
effort may decrease as carbon dioxide production drops, patients with respiratory
distress may benefit from a reduction in carbon dioxide production achieved by low-
ering the RQ.
Lowering the RQ involves two principles: avoidance of excess calories and sub-
stitution of fat for carbohydrate. A common error is to assume that providing addi-
tional fat calories will reduce the RQ. On the contrary, excess calories may cause the
RQ to increase above 1.0, possibly to as high as 4.0, depending on the amount of
excess calories.
From Blackburn GL, Bistrian BR, Maini BJ. Nutritional and metabolic assessment of the hospitalized
patient. J Parenter Enteral Nutr. 1977;1:14. Reprinted with permission of the American Society for
Parenteral and Enteral Nutrition.
79466_CH10.QXD 1/2/08 12:06 PM Page 51
A. Proper use of indirect calorimetry and the RQ requires recognition of the fol-
lowing factors:
1. Excess calories tend to greatly increase the RQ and production of carbon dioxide.
2. When the RQ is greater than 1.0, total calories should be decreased initially
to approximate the caloric requirement; indirect calorimetry may then be
repeated.
3. When the RQ is less than 1.0 but greater than 0.85 and further reduction of the
RQ is deemed beneficial, fat calories may be substituted for carbohydrate calo-
ries; however, simply adding fat calories is counterproductive.
4. Lowering the RQ is beneficial only when respiratory failure is impending or
when weaning from mechanical ventilation is difficult. Patients undergoing
long-term ventilatory support or who will not undergo weaning for several
days do not benefit from high-fat feedings or a reduction in RQ.
B. If lipid substitution is desired, a dietitian can be helpful in making appropriate
changes in enteral feeding or in adjusting glucose calories and substituting lipid in
parenteral feedings. Although lipid substitution can be beneficial, studies have
shown that intravenous administration of lipid has potentially adverse effects.
Despite these reports, intravenous lipid feedings of 50 g per day or less have no
definitive adverse effects. This amount is present in a standard 500-mL unit of
10% intravenous lipid and provides 450 kcal.
Water 1 mL/cal
Energy 1,800–2,500 kcal
Protein 45–56 g
Linoleic acid 4–6 g
Vitamin A 2,640–3,000 IU
Vitamin D 200 IU
Vitamin E 8–10 IU
Vitamin K 0.5–1.0 mg/kg
Ascorbic acid 60 mg
Folic acid 400 µg
Niacin 6.6 mg/1,000 kcal
Riboflavin 1.2–1.6 mg
Thiamine 0.5 mg/1,000 kcal
Vitamin B6 (pyridoxine) 2.2 mg
Vitamin B12 3.0 mg
Pantothenic acid 5–10 mg
Biotin 100–200 mg
Calcium 800 mg
Phosphorus 800 mg
Iodine 100–130 µg
Iron 10–18 mg
Magnesium 300–350 mg
Zinc 15 mg
Copper 2 mg
Potassium 2,500 mga
Sodium 2,500 mga
Chloride 2,000 mga
Chromium 50–120 µga
Manganese 6–8 mga
Molybdenum 400 µga
Selenium 50–100 µga
aRecommended daily allowance not established. Values are those provided in a normal diet.
Adapted from Food and Nutrition Board, Recommended Daily Allowances. Washington, D.C., 1974;
National Academy of Sciences, National Research Council.
79466_CH10.QXD 1/2/08 12:06 PM Page 53
from muscle, taken up by the splanchnic bed, and delivered to the gut and liver.
In the gut, glutamine is converted to glutamate and enters the Krebs cycle. The
liver metabolizes it to glucose, ketones, and ammonia. Glutamine supplementa-
tion of external and parenteral solutions seems to improve gut mass function and
limit bacterial translocation from the gut to the circulation in critically ill patients.
B. Fluid and electrolytes. The fluid and electrolyte requirements of individuals vary
depending on the disease process. Therapy should be tailored to the patient’s indi-
vidual needs and requirements based on appropriate clinical and laboratory eval-
uations. In general, the patient should receive maintenance requirements of water
and electrolytes. Previous and ongoing losses should be replaced with attention to
the patient’s cardiopulmonary and renal status. A convenient method for calcu-
lating maintenance fluid requirements is as follows:
1. First 10 kg of body weight 1,000 mL.
2. Next 10 kg of body weight 500 mL.
3. Each kilogram of body weight thereafter 20 mL.
C. Vitamins and trace elements serve as parts of enzymes or coenzymes and are
needed by the body in small quantities each day. These should be included in the
patient’s daily diet (Table 10-6).
Selected Readings
Banh L. Serum Proteins as Markers of Nutrition: What are we treating? Practical
Gastroenterol. 2006;xxx(10):46.
Heller AR, et al. Omega-3 fatty acids improve the diagnosis–related clinical outcome. Crit
Care Med. 2006;34:972–979.
Heyland DK, et al. Antioxidant nutrients: a systematic review of trace elements and
vitamins in the critically ill patient. Intensive Care Med. 2005;31:327–337.
Holick MF, et al. Vitamin D deficiency. N Eng. J Med. 2007;357:266–281.
Krystofiak C. Gastrointestinal Disease. In The A.S.P.E.N. Nutrition Support Core
Curriculum: A Case-Based Approach—The Adult Patient. Edited by Gottschlich MM.
Silver Spring, MD: American Society for Parenteral and Enteral Nutrition; 2007:163–186.
Ockenga J, et al. Glutamine-enriched total parenteral nutrition in patients with
inflammatory bowel disease. Eur J Clin Nutr. 2005;59:1302–1309.
Wooley JA, et al. In Nutrition Support Core Curriculum, edn 2. Edited by Gottschlich MM.
Silver Spring, MD: ASPEN; 2007:19–32.
79466_CH11.QXD 1/2/08 12:08 PM Page 54
11 ENTERAL NUTRITION
54
TABLE 11-2 Selected Enteral Nutritional Supplements
79466_CH11.QXD
Volume to
meet 100%
1/2/08
Lactose-free
12:08 PM
formulas with
intact nutrients
Isocal 1/mL 300 34 (13) 5 133 (50) 44.3 (37) 20 1,877 Mead
Johnson
Precision 1/mL 300 7.5 5 37.5 (60) 7.8 (28) 28 1,800 Doyle
Page 55
Isotonic
Osmolite 1/mL 300 37 6 145 (55) 38.4 (31.5) 50 2,000 Ross
Enrich 1.1/mL 300 40 — 162 (52) 37.1 (30) — — Ross
(with fiber)
Jevity 1/mL 310 44.3 — 151.4 36.7 — — Ross
(with fiber)
Precision 1.1/mL 530 26 (9.5) 4 248 (89) 1.6 (1.3) — 1,710 Sandoz
(low residue)
Meritine 1/mL 600 (24) 10 (46) (30) — —
Ensure 1/mL 450 37 (14) 6 145 (55) 37 (31.5) — 1,391 Ross
Ensure 1.5/mL 600 (17) 9 (53) (30) — — Ross
Plus HN
Sustacal 1/mL 625 (24) 10 (55) (21) — — Mead
Johnson
Magnacal 2/mL 590 (27) 11 (50) (23) — — Organon
(continued)
55
56
TABLE 11-2 Continued
79466_CH11.QXD
Volume to
meet 100%
Nitrogen for USRDA
Product Protein content CHO Fat % vitamins &
1/2/08
description Calories mOsm/kg g/L (%) g/L (%) g/L (%) g/L (%) MCT minerals Source
Elemental
Vivonex TEN 1/mL 630 38 (15) 7 206 (82) 2.8 (2.5) — 2,000 Norwich
(30% Eaton
12:08 PM
branched-
chain AA)
Travasorb HN 1/mL 560 45 (18) 7 175 (70) 13.5 (12) 60 2,000 Travenol
Criticare HN 1/mL 650 38 (14) 6 222 (83) 3.4 (3) — 1,900 Mead
Johnson
Page 56
CHO, carbohydrates; MCT, medium-chain triglycerides; USRDA, United States Recommended Daily Allowance; AA, amino acids; HN, high nitrogen;
TEN, total enteral nutrition.
79466_CH11.QXD 1/2/08 12:08 PM Page 57
The same nutrient formulas may be used with PEG tubes as with nasogastric
or nasoduodenal tubes.
B. Types of infusion
1. Continuous drip infusion is the method of choice in initiating tube feedings;
a defined amount is given continuously every hour with the use of an infusion
pump. Although large volumes may be administered over 24 hours, the amount
entering the GI tract at any given time is quite small. This method minimizes
the potential for aspiration, abdominal distention, and diarrhea.
a. For most patients, tube feedings are started at 50 mL/hour using a lactose-
free, 1 kcal/mL formula with intact nutrients. Thereafter, the rate is increased
by increments of 25 mL/hour daily until the desired rate is achieved.
b. If a nutrient-dense or elemental formula is used, the initial starting dilu-
tion should be at least isotonic to plasma. Hypotonic and isotonic solutions
behave almost identically in terms of small intestinal flux of fluid, and thus
there is no advantage to overdiluting a feeding product.
c. When feeding into the small bowel, isotonic solutions (3,000 mOsm) are
started at a continuous rate of 25 to 50 mL/hour every 8 hours until the
needed volume is reached. Osmolarity is then increased until the patient’s
nutrient requirements are reached.
d. Patient position. The patient’s head and shoulders must be kept at a 30- to
45-degree elevation to minimize the risk of aspiration.
2. The “cyclic” method of drip infusion may be used once a patient has been
stabilized on maintenance therapy to provide the patient “flat-in-bed” time by
increasing the rate of infusion per hour during the day and stopping the infu-
sion at night, thus still delivering the same volume of nutrients to the patient
over the 24 hours. The patient’s head and shoulders must be elevated during
feeding and 1 hour after the feeding has been stopped to ensure that gastric
emptying has occurred. Gastric residuals must be measured 2 to 3 hours after
the feeding is stopped.
V. COMPLICATIONS
A. Mechanical complications
1. Tube clogging. Viscous preparations may obstruct the lumen of the feeding
tube. This can be prevented by flushing the tube every 4 to 8 hours with 20 mL
of water or cranberry juice.
2. Pharyngeal irritation and esophageal erosion are rare with the softer tubes.
3. Tracheoesophageal fistula may occur in patients with endotracheal or tra-
cheostomy tubes receiving mechanical ventilation.
4. Aspiration is the most serious problem with tube feedings. It may be mini-
mized by placement of the tube well beyond the pylorus into the duodenum, by
keeping the gastric volumes less than 100 mL, and by elevation of the patient’s
head and shoulders at 30 to 45 degrees while the patient is being fed.
B. Gastrointestinal complications. Patients may have nausea, vomiting, crampy
abdominal pain, distention, flatulence, bloating, and diarrhea.
It is not unusual for tube-fed patients to have no bowel movements for 3 to 5
days because most commercial formulas are low in residue. Likewise, it is not
unusual for patients to have loose stools. As long as the volume of the stools is not
large, this may be tolerated. Diarrhea can result from gut atrophy, osmotic over-
load, malabsorption, lactose intolerance, concurrent medications (e.g., antibiotics,
nonsteroidal antiinflammatory drugs, and magnesium-containing antacids), addi-
tives, or vehicles added to medications and fecal impaction with the liquid feces
escaping around the impaction.
Starting patients on slow, continuous infusion of tube feeding without lac-
tose, with gradual increase in concentration and rate of delivery, or the use of for-
mulas with fiber, minimizes this problem. If diarrhea continues, DTO (deodorized
tincture of opium) may be added to the tube feedings with care that ileus does not
develop.
79466_CH11.QXD 1/2/08 12:08 PM Page 60
1. Confirm placement of feeding tube by abdominal x-ray films before administration of feeding.
2. Elevate head of bed 30 to 45 degrees when feeding into the stomach.
3. Name of formula _________________________ ; volume of formula ____________ mL at
____________ strength to be given over ____________ hour(s); rate of feeding ____________
mL/h.
4. Change administration tubing and feeding bag daily.
5. Irrigate feeding tube with 20 mL water at completion of each intermittent feeding when
tube is disconnected.
6. Check for gastric residual every ____________ hour(s) in patients receiving gastric feedings;
withhold feedings for ____________ hour(s) if residual is 50% greater than ordered volume.
7. Make sure formula does not hang for more than 8 hours.
8. Record intake and output daily; chart volume of formula separately from water or other
oral intake for every shift.
9. Weigh the patient daily; chart on graph.
10. Obtain calorie counts from dietitian daily for 5 d, then once weekly if patient is taking food
by mouth.
11. Order a complete blood count with red blood cell indices, serum chemistries, total iron
binding capacity, and serum iron and magnesium levels weekly.
12. Order serum glucose, electrolytes, BUN, and creatinine.
13. Obtain 24-hour urine collection for volume, creatinine, and urea to start at 8 A.M. weekly.
Selected Readings
Bruder EA, et al. Colonmetric carbon dioxide detection of enteral feeding tube placement.
US Gastroenterol Rev. 2007;1:17–19.
DeLegge MH, et al. Randomized prospective comparison of direct percutaneous
endoscopic jejunostomy (DPEJ) vs. percutaneous endoscopic gastrostomy with jejunal
extension (PEG-J) feeding tube placement for enteral feeding. Gastrointest Endosc.
2004;59:I58(A).
DeLegge MH. Endoscopic options for enteral feeding. Gastroenterol Hepatol. 2007;3(9):
690–692.
Guidelines for the use of parenteral and enteral nutrition in adult and pediatric patients.
J Parenter Enteral Nutr. 2002;26:1–138.
Ho KM, et al. A comparison of early gastric and post pyloric feeding in critically
ill patients: a meta-analysis. Intensive Care Med. 2006;32:639–649.
Javid PJ, et al. The role of enteral nutrition in the reversal of parenteral nutrition-associated
liver dysfunction in infants. J Pediatr Surg. 2005;40:1015–1018.
79466_CH11.QXD 1/2/08 12:08 PM Page 61
Marin M, et al. Enteral formulations. In: Merritt R, ed. The ASPEN Nutrition Support
Practice Manual. 2nd ed. Silver Spring, Md.: American Society for Parenteral & Enteral
Nutrition; 2005:63–75.
Mizock BA. Risk of aspiration in patients on enteral nutrition: frequency, relevance, relation
of pneumonia, risk factors, and strategies for risk reduction. Curr Gastroenterol Rep.
2007;9(4):338–344.
Uklej A. Gastric versus post-pyloric feeding: relationship to tolerance, pneumonia risk, and
successful delivery of enteral nutrition. Curr Gastroenterol Rep. 2007;9(4):309–316.
Zhou M, et al. Immune-modulating enteral formulations: optimum components,
appropriate patients, and controversial use of arginine in sepsis. Curr Gastroenterol
Rev. 2007;9(4):329–337.
ABLE 11.2
TAB
79466_CH12.QXD 1/2/08 12:08 PM Page 62
12 PARENTERAL NUTRITION
II. INDICATIONS. PN should be considered in any patient who cannot ingest or absorb
sufficient calories through the gastrointestinal tract. Table 12-1 lists some of the
disease categories with indications for PN.
62
79466_CH12.QXD 1/2/08 12:08 PM Page 63
Gastrointestinal disease
Inflammatory bowel disease
Radiation enteritis
Short-bowel syndrome
Severe malabsorption state
Intestinal fistula
Pancreatitis
Diverticulitis
Intestinal obstruction
Preoperative preparation of malnourished patients
Carcinoma of the head and neck
Esophageal stricture or carcinoma
Tracheoseophageal fistula
Gastric outlet obstruction
Severe peptic ulcer disease
Inflammatory bowel disease
Postoperative surgical complications
Paralytic ileus
Gastrointestinal tract fistula
Enterocutaneous fistula
Pancreatic or biliary fistula
Miscellaneous
Extensive burns or trauma
Cancer patients receiving radiation or chemotherapy
Anorexia nervosa
Some forms of liver disease
Renal failure
Most pathogens responsible for infected catheters originate from superficial sites such
as tracheostomies or abdominal wounds. The most common organisms associated
with catheter infections are: Staphylococcus epidermidis, Staphylococcus aureus,
Klebsiella pneumoniae, and Candida albicans.
A specific procedure using aseptic technique must be followed in the care of
the catheter and dressing. The catheter should be used exclusively for TPN and not
for any other purpose (e.g., blood drawing; central venous pressure measurement;
administration of drugs, antibiotics, or blood products).
E. Caloric requirements. Nutritional support regimens are usually based on estimates
of energy expenditure. These estimates were thought to be increased substantially in
patients with severe trauma or sepsis because of a presumed hypermetabolic state.
However, using actual energy expenditure measurements, a large increase in the
metabolic rate has not been seen in stressed patients. Excessive caloric intake can
produce complications such as hepatomegaly and liver dysfunction due to fatty infil-
tration of the liver, respiratory insufficiency due to excessive carbon dioxide produc-
tion during increased lipogenesis, and hyperglycemia with osmotic diuresis due to
glucose intolerance.
1. A patient’s energy requirements depend on a number of factors, including
age, sex, height, and degree of hypermetabolism. Resting energy expenditure
(REE) may be measured using the principles of indirect calorimetry from mea-
surements of carbon dioxide production and oxygen consumption. If metabolic
nutritional analysis is not available, it is possible to estimate the basal energy
expenditure (BEE) using the Harris–Benedict equation.
79466_CH12.QXD 1/2/08 12:08 PM Page 64
the favored energy source in sepsis, in which glucose utilization is depressed with
increased insulin resistance. It is recommended that in the septic hypermetabolic
patient, glucose intake be restricted to one half or less of the REE.
VII. EXAMPLE. Patient is a 30-year-old woman with Crohn’s disease who has an ileocolic
fistula and requires TPN prior to surgery: weight (W) 50 kg (110 lbs), height (H)
166.4 cm (5 ft 1/2 in.)
A. Energy requirements per day
1. BEE 655 (9.6 W) (1.7 H) (4.7 age)
For this patient:
BEE 655 (9.6 50) (1.7 166.4) (4.7 30) 1,300 kcal per day
2. Parenteral anabolic requirements 1.8 BEE
1.8 1,300 kcal per day
2,300 kcal per day
B. Protein requirements per day
1. Ideal kcal per nitrogen ratio 150:1
2. 1 g nitrogen 6.25 g of protein
3. Protein (g) 6.25 energy requirement per day
150
4. This patient needs
Protein (g) 6.25 2,300 kcal per day
150
100 g
5. To provide 100 g of protein per day, 1,000 mL of 10% aminosyn solution is needed.
C. Fluid requirements per day
1,000 500 20 (30) 2,100 mL per day
79466_CH12.QXD 1/2/08 12:08 PM Page 67
VIII. ADDITIVES. The basic TPN solution does not contain electrolytes, trace elements, or
vitamins. The electrolyte additives to TPN must be individualized to prevent fluid
and electrolyte abnormalities.
A. Electrolytes (Table 12-2)
1. Sodium (Na), the principal extracellular cation, must be administered in suffi-
cient quantities to provide for maintenance needs and to replace any existing
deficits or ongoing losses. The quantity of Na added to TPN is determined by the
patient’s extracellular fluid volume status and serum Na concentration. Patients
Sodium chloride
Sodium acetate 60–200 mEq Na
Potassium chloride
Potassium acetate 50–150 mEq K
Sodium phosphate 50–150 mM PO4
Potassium phosphate
Magnesium sulfate 8–24 mEq Mg
Calcium gluconate 8–32 mEq Ca
malnutrition with a total lymphocyte count of 800 and a serum albumin level
of 1.8 g/dL. His serum chemistries are:
Na: 133 mEq/L Cl: 96 mEq/L Creatinine: 1.1 mg/dL
K: 3.4 mEq/L HCO3: 19 mEq/L BUN: 13
W: 80 kg H: 180 cm Age (A): 52 years
BEE 66 (13.7 W) (5 H) – (6.8 A)
66 (13.7 80) (5 180) – (6.8 52) years
1,708 kcal
This patient has sepsis and requires mechanical ventilation. REE should exceed
BEE by 30%.
REE BEE 1.3
1,708 1.3 2,221 kcal
Minimal protein requirement 1.0 to 1.5 kg per day
TPN Solution
1 L AA 10% 100 g protein (16 g nitrogen)
1 L 50% D/ W 500 g dextrose 175 kcal
500 mL 10% Intralipid 550 kcal
Total kcal per day 2,300 kcal
This formula provides 44.2 kcal/kg (76% glucose, 24% lipid)
Energy–nitrogen ratio 144 kcal/g nitrogen
23 kcal/g, protein
Electrolytes to be added to each liter of TPN:
Sodium chloride (NaCl): 50 mEq
Potassium chloride (KCl): 20 mEq
Magnesium sulfate (MgSO4): 5 mEq
Na acetate: 25 mEq KPO4: 10 mEq
Administration Orders
Day 1 (infuse over 24 hours)
AA 10% 500 mL
50% D/ W 500 mL
Electrolytes (as above)
Multivitamins 5 mL
Trace elements 5 mL
Lipid emulsion 10% 500 mL
Day 2 (infuse over 24 hours)
AA 10% 1,000 mL
50% D/ W 1,000 mL
Electrolytes (as above)
Multivitamins 5 mL
Trace elements 5 mL
Lipid emulsion 10% 500 mL
After the TPN is initiated, the concentrations of serum electrolytes and phosphorus
should be determined twice a day until daily requirements are ascertained. Serum
glucose concentration should be measured every 4 to 6 hours. Serum Ca, Mg,
creatinine, and blood urea nitrogen (BUN) should be determined daily. When the
patient is stabilized on TPN, serum electrolytes, creatinine, and BUN should be
measured every other day. Ca and Mg levels may be measured twice weekly. Serum
transaminase, alkaline phosphate, and bilirubin levels should be measured weekly
to facilitate early diagnosis of hepatic steatosis.
To determine the efficacy of TPN therapy, the total lymphocyte count and
serum albumin and transferrin levels should be measured on a weekly basis. The
optimal method for assessing the adequacy of TPN is through nitrogen balance
studies. A 24-hour urine sample is collected for the urea and creatinine determina-
tions. To calculate a patient’s nitrogen balance see section III.F.
If lipids are given in addition to the TPN on a daily basis, serum triglyceride lev-
els should be measured daily for several days to document that the lipid load is not
exceeding the patient’s metabolic capabilities. Serum triglyceride concentrations may
be measured weekly when patients are on a stable TPN regimen.
vein tolerance for osmolality is about 800 mOsm/L. The osmolality of a solution
can be calculated as follows:
Osm (mOsm/L) 10 protein (g) 6 CHO (g)
(0.3 mL 20% Intralipid) total fluid (liter)
E. Example. For the patient receiving the TPN solution described in section VII.E
1. The osmolality of the TPN solution can be calculated as follows:
Protein 100 g, CHO 1,700 kcal 500 g3
Osm (mOsm/L)
10(100) 6(353) 0.3(550) 1,427
2.3
This solution is too hyperosmolar to be used in PPN.
2. The osmolality can be decreased by increasing the total volume of the solu-
tion. Because 800 mOsm/L is the desired osmolality, we can solve for total
volume needed to give osmolality of 800 mOsm/L.
800 10(100) 6(353) 0.3(550)
total fluid volume in liter
Total volume 4 L
3. Most patients cannot tolerate such high volumes of IV fluids; thus PPN can
only deliver maintenance amounts of calories.
4. For this patient the maintenance energy requirement is
BEE 1.5 1.5 1,300 1,900 kcal per day
5. The maintenance protein requirement is
Protein 6.25 1,900 80 g per day
150
6. Maximum fat allowance 1,125 kcal per day
7. Final PPN solution per day:
Solution Volume (mL) Kcal
10% AA 800 —
50% dextrose 4703 800
20% Intralipid 550 1,100
Total 1,820 1,900
8. Total volume to allow 800 mOsm/L
10(80) 6(235) 0.3(550)
3L
800
9. To bring the total volume of PPN solution to 3 L, we need to add 1,180 mL
of distilled water.
10. Adequate electrolytes, vitamins, and trace elements as in TPN are added to the
final solution if the patient is not receiving these nutrients enterally.
In many instances, if the PPN is to be administered for a short time a
more simplified PPN solution may be made from:
Solution Volume (mL) Kcal Amount (g) Osmolality (mOsm/L)
8.5 % AA 1,000 340 85 850
10% Dextrose 1,000 340 100 505
10% Lipid 500 550 50 260
Total 2,500 1,230 235 595
F. Method of administration. The protein and CHO solution may be given simulta-
neously with the lipid emulsion. The infusion sets are connected by a Y-connector
3 CHO (g)
kcal (see section V.A).
3.4
79466_CH12.QXD 1/2/08 12:08 PM Page 74
that delivers the lipid emulsion to the vein in a piggybacked fashion. This method
seems to “soothe” the vein and decrease the incidence of phlebitis. In most medical
centers, the PPN solutions contain the lipid, CHO, and protein solutions mixed
together into an emulsion. This eliminates the need for piggybacking of the lipid
emulsion. The management and monitoring of patients receiving PPN are the same
as for those receiving TPN, and the same meticulous care and team approach are
necessary to give the best results.
Selected Readings
Bistrian BR, et al. Nutritional and metabolic support in the adult intensive care unit: Key
controversies. Crit Care Med. 2006;34:1525–1531.
Cheung NW, et al. Hyperglycemia is associated with adverse outcomes in patients receiving
total parenteral nutrition. Diabetes Care. 2005;28:2367–2371.
Heller AR, Rossler S, Litz RJ, et al. Omega-3 fatty acids improve the diagnosis-related
clinical outcome. Crit Care Med. 2006;34:972–979.
Huschak G, et al. Olive oil based nutrition in multiple trauma patients: a pilot study. Intensive
Care Med. 2005;31:1202–1208.
Krein SL, et al. Use of central venous catheter-related bloodstream infection prevention
practices by US hospitals. Mayo Clin Proc. 2007;82:672–678.
Kudsk KA. Immunonutrition in surgery and critical care. Annu Rev Nutr. 2006;26:463–479.
Marik PE. Maximizing efficacy from parenteral nutrition in critical care: appropriate
patient population, supplemental parentral nutrition, glucose control, parenteral
glutamine, and alternative fat sources. Curr Gastroenterol Rep. 2007;9(4):345–353.
Sacks GS, et al. Parenteral nutrition implementation and management. In: Merritt R, ed.
The ASPEN Nutrition Support Practice Manual. 2nd ed. Silver Spring, Md.: American
Society for Parenteral & Enteral Nutrition; 2005:108–117.
van der Voort PH, et al. Intravenous glucose intake independently related to intensive care
unit and hospital mortality: an argument for glucose toxicity in critically ill patients.
Clin Endocrinol. 2006;64:141–145.
Waitzberg DL, et al. Postsurgical infections are reduced with specialized nutrition support.
World J Surg. 2006;30:1592–1604.
79466_CH13.QXD 1/2/08 12:09 PM Page 75
Gastroenterologic Emergencies IV
79466_CH13.QXD 1/2/08 12:09 PM Page 76
79466_CH13.QXD 1/2/08 12:09 PM Page 77
T he term acute abdomen evokes an image of a patient suffering from sudden, severe
abdominal pain, perhaps accompanied by vomiting, attentively surrounded by physicians
and surgeons who are earnestly deciding whether to take the patient to the operating room.
Indeed, many instances of acute abdomen are surgical emergencies. However, the differen-
tial diagnosis is extensive, and the management of an acute abdomen varies according to
the diagnosis.
77
79466_CH13.QXD 1/2/08 12:09 PM Page 78
* Some diagnoses may fall into more than one category. For example, cholecystitis is an inflammatory
condition of the gallbladder, but it usually develops as a result of obstruction of the cystic duct by a
gallstone.
79466_CH13.QXD 1/2/08 12:09 PM Page 79
Blood studies
Complete blood count
Electrolytes
Calcium and magnesium
Amylase
Bilirubin, SGOT, SGPT, alkaline phosphatase
Arterial blood gases
Urinalysis
Electrocardiogram
Chest x-ray films
Abdominal x-ray films (flat and upright or decubitus)
Additional studies that may be indicated
Ultrasound scan
CT scan
HIDA scan
Intravenous pyelogram
Diagnostic paracentesis
Selected Readings
Al-Salamah SM, et al. Role of ultrasonography, computed tomography and diagnostic
peritoneal lavage in abdominal blunt trauma. Saudi Med J. 2002;23:1350–1355.
Brown TA, et al. Acute appendicitis in the setting of clostridium difficile colitis. Clin
Gastroenterol Hepatol. 2007;5(8):969–971.
Fry LC, et al. The yield of capsule endoscopy in patients with abdominal pain or diarrhea.
Endoscopy. 2006;38:498–502.
Iancelli A, et al. Therapeutic laparoscopy for blunt abdominal trauma with bowel injuries.
J Laparoendosc Adv Surg Tech A. 2003;13:189–191.
Khawaja FJ, et al. 86-year-old woman with abdominal pain and diarrhea. Mayo Clin Proc.
2007;82(4):487–489.
Knoll BM, et al. 56-year-old man with rash, abdominal pain and orthralgias. Mayo Clin
Proc. 2007;82(6):745–748.
Maglinte DD, et al. Current concepts in imaging on small bowel obstruction. Radiol Clin
North Am. 2003;41:263–283.
Ng B, et al. 49 year old woman with acute abdominal pain and nausea. Mayo Clin Proc.
2001;76:649.
Poulin EC, et al. Early laparoscopy to help diagnose acute non-specific abdominal pain.
Lancet. 2000;355:861.
Wolfe JM, et al. Analgesic administration to patients with acute abdomen: A survey of
emergency medicine physicians. Am J Emerg Med. 2000;18:250.
79466_CH14.QXD 1/2/08 12:10 PM Page 82
I. INITIAL MANAGEMENT
A. History
1. Vomiting or passage of blood per rectum. The action of gastric acid on
blood quickly forms dark particles that resemble coffee grounds. Vomiting of
red blood (hematemesis) or of coffee-ground-appearing material usually signi-
fies a source of bleeding in the esophagus, stomach, or duodenum, but it can
result from swallowed blood from the respiratory tract. On the other hand,
passage of red- or maroon-colored stool per rectum (hematochezia) usually
indicates that the source is in the rectum, colon, or terminal ileum.
82
79466_CH14.QXD 1/2/08 12:10 PM Page 83
B. Physical examination
1. The physical examination is unlikely to indicate a precise cause of bleeding.
However, coolness of the extremities, palmar creases, and pallor of the con-
junctivae, mucous membranes, and nail beds may be evident as a result of blood
loss and peripheral vasoconstriction. The signs of chronic liver disease or
abdominal tenderness may provide relevant information.
2. The rectal examination is important and should not be omitted, even
in seemingly obvious upper gastrointestinal bleeding. The anus, perianal
area, and lower rectum can be assessed, as can the character and color of the
stool.
Occult blood in the stool can be detected with as little as 15 mL of blood
loss per day. Stools may remain positive for occult blood for nearly 2 weeks
after an acute blood loss of 1,000 mL or more from an upper gastrointestinal
source.
3. Postural signs. As the patient loses intravascular volume due to blood loss,
cardiac output and blood pressure fall, and pulse rate increases. Under condi-
tions of severe volume loss, postural compensation of blood pressure and pulse
are inadequate. Thus, so-called postural signs are present if, when the patient
sits from a supine position, the pulse rate increases more than 20 beats per
minute and the systolic blood pressure drops more than 10 mmHg. Under these
circumstances, it is likely that blood loss has exceeded 1 L. However, age, car-
diovascular status, and rate of blood loss all influence the development of pos-
tural signs.
C. Fluid, electrolyte, and blood replacement
1. A large-bore intravenous catheter should be inserted promptly into a peripheral
vein. Blood can be drawn at this time for laboratory studies (see section I.D). In
a profusely bleeding patient, a single peripheral intravenous catheter may not
be sufficient to provide adequate blood replacement; two or more intravenous
catheters may be required. In an acute emergency in which a peripheral vein is
not available, venous access should be established via a jugular, subclavian, or
femoral vein.
2. Infusion of fluids and blood. Normal saline is infused rapidly until blood for
transfusion is available. In patients who have excess body sodium, such as
those with ascites and peripheral edema, the physician may be reluctant to
infuse large amounts of saline. In those instances, the restoration of hemody-
namic stability should take precedence over other considerations. In other
words, if the patient is bleeding profusely and blood for transfusion is not yet
available, saline should be infused without regard for the patient’s sodium bal-
ance. If bleeding is less severe, hypotonic sodium solutions may be infused
until blood for transfusion arrives. Appropriate treatment of acute gastroin-
testinal bleeding includes not only replacement of blood, usually in the form of
packed red cells, but also infusion of supplemental electrolyte solutions and,
when necessary, clotting factors.
3. A central venous pressure catheter or Swan-Ganz catheter may be nec-
essary to evaluate the effects of volume replacement and the need for continued
infusion of blood, particularly in elderly patients or patients with cardio-
vascular disease.
4. Monitoring of urine output provides a reasonable indication of vital organ
perfusion. In severely ill patients, a urinary catheter may be necessary.
D. Laboratory studies
1. Initial blood studies should include a complete blood count (CBC) and levels
of electrolytes, blood urea nitrogen (BUN), creatinine, glucose, calcium, phos-
phate, and magnesium; blood should be drawn for typing. Hemoglobin and
hematocrit levels usually are low, and the level of these measures may have
some relation to the amount of blood loss. However, some patients bleed so
rapidly that there is insufficient time for the blood volume to equilibrate, and
the hemoglobin and hematocrit levels are normal or only slightly reduced. In
acutely bleeding patients, changes in blood pressure and pulse and the direct
79466_CH14.QXD 1/2/08 12:10 PM Page 85
evidence of continued bleeding via the nasogastric tube or per rectum are bet-
ter indicators than the hemoglobin and hematocrit levels for determining the
administration of electrolyte solutions and the replacement of blood. Clotting
status should be assessed with platelet count, prothrombin time, and partial
thromboplastin time. Depending on the clinical presentation, other blood stud-
ies may be important, such as serum amylase, liver tests, and cardiac enzymes.
In severely ill patients, arterial blood gases should be monitored.
2. Leukocytosis, usually not in excess of 15,000/L, can accompany acute gas-
trointestinal bleeding. However, an elevated white blood cell count should not
be attributed to acute blood loss without a search for sources of infection.
3. Elevated BUN in a patient whose BUN has recently been normal or whose
serum creatinine level is normal suggests an upper gastrointestinal bleeding
source. The rise in BUN results from the hypovolemia of acute blood loss, but
digestion of blood proteins in the small intestine and the absorption of
nitrogenous products can also contribute. In patients with impaired liver
function, the increased protein load may be sufficient to induce or aggravate
hepatic encephalopathy. The magnitude of the protein load can be calculated
roughly by multiplying the grams of hemoglobin and serum protein per
deciliter of blood by the estimated volume of blood lost. For example, if a
patient has an initial hemoglobin level of 13 g/dL and serum protein of 7 g /dL
and loses 1,000 mL (or 10 dL) of blood, approximately 200 g of protein is
presented to the small intestine ([13 g 7 g] 10 200 g). Thus, gastric
lavage and control of bleeding are additionally important in patients with
liver disease.
4. Later studies. Because of rapid fluid shifts during gastrointestinal bleeding
and the infusion of blood, blood products, and other fluids, frequent assess-
ment of serum electrolytes, calcium, phosphate, and magnesium levels is neces-
sary. Extensive transfusion dilutes platelets and clotting factors, particularly
factors V and VII. This condition can be treated by infusion of fresh-frozen
plasma and platelets as necessary. Also, a high proportion of patients who bleed
while taking therapeutic anticoagulants do so from a clinically significant
lesion. Thus, it is important to evaluate these patients for gastrointestinal
pathology in addition to correcting their clotting status. Most patients who
receive blood transfusions do not need calcium supplements, although hypocal-
cemia as a result of binding of calcium by anticoagulants in banked blood may
occur after massive transfusions. Patients who receive more than 100 mL of
blood per minute may be given 0.2 g calcium chloride (CaCl2) via another
intravenous line during the time the blood is infusing. Measurement of the ion-
ized calcium and monitoring of the electrocardiographic QT interval are rec-
ommended during the rapid infusion of anticoagulated blood.
E. Nasogastric intubation and gastric lavage. A nasogastric (NG) tube should be
passed in all patients with acute gastrointestinal bleeding unless the source is obvi-
ously the lower gastrointestinal tract. Blood from an esophageal or gastric source
pools in the stomach, and in more than 90% of bleeding duodenal ulcers the blood
refluxes back across the pyloric channel into the stomach. If the aspirate is clear or
clears readily with lavage, the NG tube may be removed. If there is fresh blood or
a large amount of old blood or retained material, the stomach should be lavaged
by means of a large-bore sump tube (20–24 French) or an Ewald tube. Removal of
as much of the gastric contents as possible facilitates subsequent endoscopy and
may contribute to hemostasis by allowing the walls of the stomach to collapse.
Because of its adverse effects, the NG tube should be removed promptly when it
no longer fulfills a useful purpose.
1. Benefits of NG intubation
a. To document the presence of blood
b. To monitor the rate of bleeding
c. To identify recurrence of bleeding after initial control
d. To lavage and decompress the stomach
e. To remove gastric acid
79466_CH14.QXD 1/2/08 12:10 PM Page 86
II. DIAGNOSTIC AND THERAPEUTIC STUDIES. Schemes for the diagnostic evaluation
of acute upper and lower gastrointestinal bleeding are shown in Figs. 14-1 and 14-2,
respectively.
A. Endoscopy. Gastrointestinal endoscopy, of both the upper gastrointestinal tract
and the colon and rectum, is discussed in Chapter 5 with regard to diagnostic
capability and methods of performing the procedures.
1. Upper gastrointestinal bleeding
a. Diagnostic endoscopy. Endoscopy usually is recommended as the initial
diagnostic procedure in acute upper gastrointestinal bleeding because
Figure 14-1. Scheme for the diagnostic evaluation of acute upper gastrointestinal bleeding.
(*Stigmata of recent hemorrhage are a visible vessel, fresh blood clot, black eschar, or active bleeding.)
79466_CH14.QXD 1/2/08 12:10 PM Page 87
Figure 14-2. Scheme for the diagnostic evaluation of acute lower gastrointestinal bleeding.
Laser Injection
Characteristic Electrocoagulation photocoagulation therapy
From Eastwood GL. Endoscopy in gastrointestinal bleeding: Are we beginning to realize the dream?
J Clin Gastroenterol. 1992;14:187. Reprinted with permission.
79466_CH14.QXD 1/2/08 12:10 PM Page 89
III. GASTRIC AND ANTISECRETORY AGENTS. Because gastric acid plays a pathogenic
role in many causes of upper gastrointestinal bleeding, it is reasonable to inhibit acid
secretion with antisecretory agents such as histamine-2 (H2) antagonists, or proton-
pump inhibitors (PPIs). The intent is to maintain the intraluminal gastric pH above
4.0. This is difficult to achieve with the use of H2 antagonists. PPIs inhibit gastric
acid secretion much more effectively and at adequate doses increase the gastric pH to
above 4. In the United States, pantoprazole sodium (Protonix), esomeprazole (Nexium),
and lansoprazole (Prevacid) are available in intravenous form as well as in an oral
formulation.
The reduction of intraluminal acid may be effective in two ways. First, the direct,
harmful effects of acid and pepsin on the bleeding lesion are diminished. Second, a less
acid environment allows platelets to aggregate and thus promotes clotting. After the
NG tube is removed, the patient is treated with an antisecretory agent, preferably with
a PPI. A bolus of 80 mg of intravenous PPI is administered, followed by a continuous
infusion at 80 mg an hour up to 72 hours.
patients with severe liver disease and for those who may require liver trans-
plantation. Patients with compensated cirrhosis may be better treated with distal
splenorenal shunt-type portosystemic shunt surgery.
4. Long-term management. After acute variceal bleeding has been controlled by
endoscopic variceal banding or injection sclerosis, octreotide infusion, or bal-
loon tamponade, a decision regarding long-term management of the varices
must be made. Some physicians elect no specific therapy, hoping that bleeding
will not recur. Because the risk of rebleeding is high, however, some form of
treatment usually is indicated. The daily oral administration of beta-blockers
(e.g., nadolol) and nitrates (e.g., isosorbide dinitrate) has been shown to reduce
portal hypertension in patients and decrease the risk of rebleeding. Because
long-term survival appears to be increased after repeated endoscopic band lig-
ation or injection sclerosis to eradicate the varices or portosystemic shunt
surgery, most patients should be considered for one of these treatments. Total
health care costs are probably less and the risk of encephalopathy is lower in
patients who undergo band ligation or injection sclerosis. Thus, repeated
endoscopic band ligation or injection sclerosis, consisting of repeated injections
separated by 1 to 4 weeks for several months, is a reasonable first choice.
Complications of endoscopic band ligation and injection sclerosis
include acute perforation of the esophagus, mucosal ulceration and necrosis,
and stricture formation. Patients who have recurrent variceal bleeding despite
band ligation and injection sclerotherapy and whose risk of surgery is acceptable
may benefit from portosystemic shunt surgery.
B. Mallory-Weiss gastroesophageal tear. Bleeding from a Mallory-Weiss mucosal
tear at the esophagogastric junction traditionally has been associated with repeated
vomiting or retching before the appearance of hematemesis. However, in a large
review of patients with documented Mallory-Weiss lesions, such a history was
obtained in less than one third of the patients. The diagnosis of Mallory-Weiss
mucosal tear must be made by endoscopic visualization of the lesion. Most patients
stop bleeding spontaneously. In those with active bleeding, endoscopic placement of
clips, bipolar electrocautery or injection of epinephrine may be used successfully.
C. Aortoenteric fistula. Patients with an aortoenteric fistula typically have massive
hematemesis or hematochezia. The bleeding may stop abruptly; if it recurs, it often
is fatal. This is a surgical emergency and immediate surgical intervention is necessary.
Selected Readings
Angtuaes TL, et al. The utility of urgent colonoscopy in the evaluation of acute lower
gastrointestinal tract bleeding: A 2-year experience from a single center. Am J Gastroenterol.
2001;96:1782.
Bardou M, et al. Meta-analysis proton pump inhibition in high-risk patients with acute
peptic ulcer bleeding. Aliment Pharmacol Ther. 2005;21:677–686.
Bianco MA, et al. Combined epinephrine and bipolar coagulation vs. bipolar coagulation
alone for bleeding peptic ulcer: A randomized, controlled trial. Gastrointest Endosc.
2004:60:910–915.
Chan FK, et al. Preventing recurrent upper gastrointestinal bleeding in patients with
Helicobacter pylori infection who are taking low-dose aspirin or naproxen. N Engl J Med.
2001;344:967.
Chan FKL, et al. Combination of a cyclo-oxygenase 2 inhibitor and a proton-pump
inhibitor for prevention of recurrent ulcer bleeding in patients at very high risk: a
double-blind, randomised trial. Lancet. 2007 May 12;369:1621–6.
Cipolletta L, et al. Endoclips versus heater probe in preventing early recurrent bleeding
from peptic ulcer: A prospective and randomized trial. Gastrointest Endosc. 2001;
53:147–151.
Corely DA, et al. Octreotide for acute esophageal variceal bleeding: A meta analysis.
Gastroenterology. 2001;120:946.
A 40-year-old woman with epistaxis, hematemesis, and altered mental status. Case records
of the Massachusetts General Hospital. N Engl J Med. 2007;356:174–182.
Geier A, et al. Profuse rectal bleeding of no visible cause. Lancet. 2007;369:1664.
79466_CH14.QXD 1/2/08 12:10 PM Page 93
Gerson LB, et al. Endoscopic band ligation for actively bleeding Dieulafoy’s lesions.
Gastrointest Endosc. 1999;50:454.
Howarth DM. The role of nuclear medicine in the detection of acute gastrointestinal
bleeding. Semin Nucl Med. 2006;36:133–46.
Lau JY, et al. Oweprazole before endoscopy in patients with gastrointestinal bleeding.
N Eng J Med. 2007;356:1631–1640.
Lau JUW, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic
treatment of bleeding ulcers. N Engl J Med. 2000;343:310.
Leontiadis GI. Proton pump inhibitor treatment for acute peptic ulcer bleeding. Cochrane
Database Syst Rev. 2006;1:CD002094.
Park JJ. Meckel diverticulum: the Mayo Clinic experience with 1476 patients (1950–2002).
Ann Surg. 2005;241:529–33.
Sanyal AJ. Octreotide and its effects on portal circulation. Gastroenterology. 2001;120:303.
Schmulewitz N, et al. Dieulafoy lesions: A review of 6 years of experience at a tertiary care
center. Am J Gastroenterol. 2001;96:1688.
Sharara AI, et al. Gastroesophageal variceal hemorrhage. N Engl J Med. 2001;345:669,
2001.
Sung JJY, et al. The effect of endoscopic therapy in patients receiving omeprazole for
bleeding ulcers with nonbleeding visible vessels or adherent clots. A randomized
comparison. Ann Intern Med. 2003;139:237–243.
Villanueva C, et al. Endoscopic ligation compared with combined treatment with nadolol
and isosorbide mononitrate to prevent recurrent variceal bleeding. N Engl J Med. 2001;
345:647.
79466_CH15.QXD 1/2/08 12:11 PM Page 94
I. ESOPHAGUS. Foreign bodies may cause obstruction above the upper esophageal
sphincter and may compromise the airway. These patients should be urgently handled
by ear, nose, and throat specialists.
Most obstructions from foreign body ingestions involve the esophagus; many
occur above a benign or malignant stricture, web, or ring. The four areas of physio-
logic narrowing in the esophagus—the cricopharyngeal muscle, the aortic arch, the
left main-stem bronchus, and the gastroesophageal junction—are also common sites
for obstruction. Sharp objects such as fish or poultry bones, pins, or toothpicks
may perforate the esophagus, resulting in sepsis or hemorrhage. Button (miniature,
7.9–11.6 mm) battery ingestions are not uncommon in children. Most of these spon-
taneously pass; however, those with larger diameters (15.6–23.0 mm) may impact in
the esophagus, causing tissue necrosis, perforation, or hemorrhage.
A. Clinical findings
1. Signs and symptoms. Acute esophageal obstruction may result in substernal pain
at the level of obstruction or be referred to the sternal notch. The pain may be mild
or severe or may mimic a myocardial infarction. There may be profuse salivation
and regurgitation. In patients who have ingested a sharp object like a fish bone,
odynophagia and a sensation of the object lodged in the esophagus may be present.
2. Physical examination is usually unrewarding. When perforation is suspected,
subcutaneous air in the soft tissues should be sought by looking for crepitus by
palpation of the upper thorax and neck.
B. Diagnostic tests
1. Radiographic techniques
a. Photographic densities. Plain x-rays are frequently used in the detection of
foreign bodies. However, not all foreign bodies are radiopaque due to differ-
ences in their densities.
i. Foreign bodies of high density are highly radiopaque and have low
photographic density on a radiograph. If the object is of adequate size,
it is easily differentiated from the surrounding tissues. Common exam-
ples include objects made of iron, steel, and some alloys, as in nails,
screws, chips, bullets, and many coins.
ii. Foreign bodies with physical densities somewhat higher than body
tissues (e.g., glass, aluminum, chicken bones, plastics) have photographic
densities slightly less than body tissues and form more subtle images.
94
79466_CH15.QXD 1/2/08 12:11 PM Page 95
II. STOMACH
A. Clinical findings. Foreign bodies in the stomach usually do not cause any symp-
toms. The presence of nausea and vomiting may indicate pyloric obstruction; pain,
bleeding, and fever may suggest mucosal injury or perforation.
79466_CH15.QXD 1/2/08 12:11 PM Page 96
B. Diagnosis
1. Chest and abdominal radiographs help in determining the location and
nature of the swallowed object. Contrast studies may be needed if the object
cannot be visualized by plain films or if there is a question of perforation. The
presence of air under the diaphragm also should be ascertained.
2. Endoscopy visualizes the foreign object, but this may be difficult when the
stomach is full of food.
C. Management. Most ingested objects that have made their way into the stomach
pass through the pylorus and the rest of the gastrointestinal tract. Most coins, such
as dimes and nickels, do not cause obstruction, but quarters may not pass in chil-
dren. With rounded objects, it is thought to be safe to wait several days for the
spontaneous passage of the foreign body. The progress is followed by daily x-rays.
Because objects may become embedded in the wall of the stomach after several
days, endoscopic removal is advised. However, bones, denture fragments, pencils,
toothpicks, needles, razor blades, and other sharp objects should be removed
immediately, endoscopically or surgically. Induced vomiting, especially in intoxi-
cated persons and in children, carries the risk of aspiration of the gastric contents
and the foreign body and is not recommended.
Selected Readings
Athanassiadi K, et al. Management of esophageal foreign bodies: A retrospective review of
400 cases. Eur J Cardio Thorac Surg. 2002;21:653–656.
Focht D, Kaul A. Food impaction and eosinophilic esophagitis. J Ped. 2005;147:540.
Janik JE, et al. Forceps extraction of upper esophageal coins. J Pediatr Surg. 2003;38:
227–229.
Jeen YT, et al. Endoscopic removal of sharp foreign bodies impacted in the esophagus.
Endoscopy. 2001;33:518–522.
Lam HC, et al. Esophageal perforation and neck abscess from ingested foreign bodies:
Treatment outcomes. Ear Nose Throat J. 2003;82:786–794.
Mosca S, et al. Endoscopic management of foreign bodies in the upper gastrointestinal
tract: Report on a series of 414 adult patients. Endoscopy. 2001;33:692–696.
Panda NK, et al. Management of sharp esophageal foreign bodies in young children:
A cause for worry. Int J Pediatr Otorhinolaryngol. 2002;64:243–246.
Soprano JV, et al. Four strategies for the management of esophageal coins in children.
Pediatrics. 2000;105:e5.
Vanelli PM, et al. Exploring the link between eosivephilic esophagitis and esophageal
foreign bodies in the pediatric population. Pract Gastroenterol. 2007;xxx1(10):81–85.
79466_CH16.QXD 1/2/08 12:11 PM Page 97
CAUSTIC INGESTION 16
T he ingestion of caustic chemicals may cause tissue injury on contact with the
oropharynx, esophagus, stomach, and duodenum. Accidental ingestion of these caustic or
corrosive substances is most frequent in small children and inebriated individuals.
Intentional ingestion occurs more commonly in persons who are suicidal.
Caustic substances can be divided into acids and alkalis. Many household and indus-
trial products contain acids or alkalis in liquid or solid form (Tables 16-1 and 16-2).
97
79466_CH16.QXD 1/2/08 12:11 PM Page 98
A. The severity of the tissue injury after caustic ingestion depends on the nature,
concentration, and quantity of the caustic substance and the duration of tissue
contact. The higher the pH, the more destructive is the alkaline agent. The critical
pH that causes esophageal ulceration is 12.5. Most cases of deep ulceration that
progress to stricture formation involve lye solutions of pH 14.0.
B. Areas of injury. The oropharynx and esophagus are most commonly injured in
alkali ingestions. Twenty to thirty percent of the patients who have esophageal
injury also have gastric injury. The severity of the injury varies from inflammation
and ulceration to necrosis and perforation of the viscera.
1. Solid alkali ingestion. Alkali swallowed in solid form adheres on contact to
mucous membranes of the oropharynx or esophagus, usually sparing the
stomach. If solid alkali is swallowed with water, it is carried further down the
digestive tract.
2. Liquid alkali ingestion, especially of lye, exposes most of the mucosal sur-
face of the oropharynx and upper gastrointestinal tract to the caustic sub-
stance. The esophagus is the most commonly injured organ. The stomach is
also usually injured. In addition to ulceration and necrosis of the affected tis-
sues, bleeding may occur after liquid-lye ingestion. Respiratory distress
secondary to soft-tissue swelling of the epiglottis, larynx, vocal cords, or tra-
chea may occur due to aspiration of lye. Tracheoesophageal fistulas may
develop after severe injury. The most common delayed complication of lye
ingestion is esophageal stricture formation. Strictures may involve a variable
length of the esophagus, ranging from a focal narrowing to complete steno-
sis of the entire esophagus. Antral scarring and pyloric stenosis also have
been reported.
79466_CH16.QXD 1/2/08 12:11 PM Page 99
performed to localize the site of perforation. After 1 to 2 months, the extent of fibro-
sis and stricture formation may be assessed periodically by radiocontrast studies.
B. Endoscopy of the upper gastrointestinal tract within the first 24 hours of inges-
tion is recommended to establish the extent and severity of the tissue damage and
to remove by suction any remaining caustic material from the stomach. Rigid
scopes should not be used. Endoscopes with diameters less than 1 cm are safe in
experienced hands. If esophageal or gastric perforation is suspected clinically or
from plain films, endoscopy should not be performed.
C. The severity of tissue injury is classified by endoscopic appearance.
Grade 0: Normal endoscopic examination
Grade I: Edema and hyperemia of the mucosa
Grade IIa: Friability, hemorrhages, erosions, blisters, whitish membranes or exu-
dates, superficial ulcerations
Grade IIb: Grade IIa plus deep discrete or circumferential ulcerations
Grade IIIa: Grade II plus multiple ulcerations and small scattered areas of necrosis
(areas of brown-black or grayish discolorations)
Grade IIIb: Grade II plus extensive necrosis
VI. THERAPY
A. Patients with grade 0 and grade I injuries are expected to heal with no specific ther-
apy. They should receive nothing by mouth until they can eat without discomfort.
A barium swallow and upper gastrointestinal series may be done after 1 to 2
months to look for scarring and strictures.
B. Patients with grade IIa and some patients with grade IIb injuries recover rapidly
and may be discharged from the hospital within 5 to 12 days. Healing usually
occurs by the third or fourth week without sequelae. Some grade II and all grade
III injuries heal with scarring. Complications such as hemorrhage and perforation
should be carefully looked for in patients with grades IIb and III injuries. Most
patients with grade IIIb injury require immediate surgical attention.
C. Patients with more extensive tissue injury should be treated in the intensive care
unit with special attention to the possibility of viscous perforation. To protect the
injured tissues, the upper gastrointestinal tract should not be used. Total parenteral
nutrition should be started immediately. Intravenous histamine-2 (H2)-blocker therapy
may be used to reduce gastroduodenal acidity to prevent further acid–peptic injury.
D. Steroids. In pharmacologic doses, corticosteroids impair wound healing, depress
immune defense mechanisms, and mask the signs of infection and viscous perfo-
ration. They cannot salvage an already injured organ. Because the risks outweigh
the possible benefits, their use is not recommended.
E. Antibiotics. Serious infections are infrequent with caustic injury. In first- or
second-degree burns, antibiotics should be withheld until evidence of infection is
present. In patients in whom the possibility of perforation exists, antibiotics are
recommended to reduce the risk of infectious mediastinitis or peritonitis.
F. Surgery. The coordinated intensive efforts of both medical and surgical teams are
necessary for the best outcome in patients severely injured with caustic ingestion.
In the event of extensive tissue necrosis, perforation, peritonitis, or severe hemor-
rhage, emergency surgery may be necessary.
In severe gastric burns, antral and pyloric stenosis may require partial or
total gastrectomy. Total gastrectomy and esophageal replacement may be neces-
sary in cases of simultaneous severe esophageal and gastric injury.
G. Stricture formation is usually noted 2 to 8 weeks after a second- or third-degree
caustic injury with circumferential lesions.
1. Dilatation. Single and minor strictures usually respond to repeated dilatation.
However, because early dilatation increases the risk of perforation, it should be
delayed until after the acute injury has healed.
2. Surgery. Extensive esophageal strictures may require surgical resection with
transposition of a segment of jejunum or colon. Placement of an intraluminal
silicone stent under endoscopic guidance for 4 to 6 weeks has been used to pre-
vent severe stricture formation.
79466_CH16.QXD 1/2/08 12:11 PM Page 101
H. Long-term follow-up. Esophageal and gastric cancer may develop in the scarred
mucosa 10 to 15 years after the initial injury. These patients require ongoing clinical
and endoscopic surveillance and follow-up.
Selected Readings
Bernhardt J, et al. Caustic acid burn of the upper gastrointestinal tract: First use of
endosonography to evaluate the severity of injury. Surg Endosc. 2002;16:1004.
Boukhir S, et al. High doses of steroids in the management of caustic esophageal strictures
in children. Arch Pediatr. 2004;11:13–17.
Erdogan E, et al. Esophageal replacement using the colon: A 15-year review. Pediatr Surg.
2000;16:546–549.
Hamza AF, et al. Caustic esophageal strictures in children: 30-year experience. J Pediatr
Surg. 2003;38:828–833.
Katzka DA, et al. Caustic injury to the esophagus: Current treatment options. Gastroen-
terology. 2001;4:59–66.
Kukkody A, et al. Long-term dilation of caustic strictures of the esophagus. Pediatr Surg Int.
2002;18:486–490.
Nunes AC, et al. Risk factors for stricture development after caustic ingestion. Hepato-
gastroenterology. 2002;49:1563–1566.
Zwischenberger JB, et al. Surgical aspects of esophageal disease. Perforation and caustic
injury. Am J Repir Crit Care Med. 2001;164:1037–1040.
79466_CH17.QXD 1/2/08 12:13 PM Page 102
F ulminant hepatic failure (FHF), acute hepatic failure, and fulminant hepatitis
all refer to acute severe impairment of liver function accompanied by coagulopathy,
advanced stages encephalopathy, and coma in patients who have had liver disease for less
than 8 weeks. FHF, in most instances, is complicated by multiorgan failure and cerebral
edema, lasts 1 to 4 weeks, and ends fatally in 60% to 95% of patients. FHF is a rare con-
dition with an incidence of 2,000 cases per year in the United States.
In a subgroup of patients, the duration of illness before the onset of encephalopathy
is more prolonged (subacute FHF) but as in FHF, there is no evidence of previous liver dis-
ease. In patients with late-onset hepatic failure, hepatic encephalopathy and other evidence
of hepatic decompensation appear between 8 and 24 weeks after the first symptoms.
Patients with late-onset disease are significantly older than those who have FHF; median
ages of onset are 44.5 years and 25.5 years, respectively.
Liver transplantation may be the ultimate solution in FHF.
New terminology has been introduced and is based on the interval from the onset of
jaundice to the development of encephalopathy.
Hyperacute liver failure, with an interval of 7 days
Acute liver failure, with an interval of 8–28 days
Subacute liver failure, with an interval of 4–12 weeks.
102
79466_CH17.QXD 1/2/08 12:13 PM Page 103
Viral agents
Hepatitis A virus Herpes viruses
Hepatitis B virus Adenovirus
Hepatitis C virus Cytomegalovirus
Hepatitis D virus Paramyxovirus
Hepatitis E virus Epstein-Barr virus
Toxic substances
Acetaminophen Valproic acid
Halothane Disulfiram
Isoniazid Nortriptyline
Rifampicin White or yellow phosphorus
Amine oxidase inhibitors Emetic toxin of Bacillus cereus
Hydrocarbons Mushroom poisoning (Amanita phalloides)
Carbon tetrachloride Some herbal medicines
Nonsteroidal antiinflammatory drugs Dideoxyinosine
Ischemic liver necrosis
Wilson’s disease with intravascular hemolysis Shock (hypotension hypoxemia)
Acute Budd-Chiari syndrome Autoimmune hepatitis
Congestive heart failure Heat stroke
Acute steatosis syndromes
Reye’s syndrome
Acute fatty liver of pregnancy
Tetracycline
Massive blastic infiltration of the liver Hodgkin’s lymphoma
Lymphoreticular malignancies Burkitt-type lymphoma
Malignant histiocytosis
Non-Hodgkin’s lymphoma
Acute leukemia
Acute phase of chronic myelogenous
leukemia
Acute monoblastic leukemia
Metastatic liver disease from primary lung or breast cancer and melanoma
E. Rarely, FHF may develop in patients who have one of the hematolymphoid malig-
nancies, such as malignant histiocytosis, Burkitt’s lymphoma, the acute phase of
chronic myelogenous leukemia, acute monoblastic leukemia, and Hodgkin’s and
non-Hodgkin’s lymphomas. Massive infiltration of hepatic parenchyma with malig-
nant cells results in infarction and necrosis, leading to FHF.
of the disease. A mutant form of hepatitis B and rarely hepatitis C is implicated in most
instances of late-onset hepatic failure. Determination of viral DNA or RNA titers of
hepatitis B and C viruses by polymerase chain reaction respectively may give informa-
tion that is more accurate in the cases of viral hepatitis B or C.
Liver biopsy may be helpful in establishing a diagnosis but may be difficult to
perform because of the severe coagulopathy, which is not correctable with replace-
ment of clotting factors.
III. PROGNOSIS. Survival from FHF depends on the ability of the liver to regenerate with
restitution of the normal hepatic function. Prothrombin time greater than 100 seconds,
regardless of the stage of encephalopathy or the presence of any three of the following
findings, indicates a poor prognosis in FHF caused by viral hepatitis or drug toxicity
excluding acetaminophen toxicity:
Arterial pH 7.3
Age 10 or 40 years
Jaundice 7 days before the onset of encephalopathy
Prothrombin time 50 seconds
Serum bilirubin 18 mg/dL
Prognosis in FHF depends on the age of the patient, cause of the acute liver
failure, clinical course, occurrence of secondary complications, and duration and
severity of the coma.
A. Causes of death in FHF are neurologic complications (67%), gastrointestinal
hemorrhage (13%), bacterial and/or fungal infection and sepsis (13%), hemody-
namic complications (8%), and progressive respiratory and renal failure.
Verbal response
Oculocephalic–oculovestibular reflexes
None
No reaction
Incomprehensible
Partial or dysconjugate
Confused
Full
Normal
Normal
Eye opening
Best motor response
None
None
Noxious stimuli only
Abnormal extensor
Verbal stimuli
Abnormal flexor
Spontaneous
Withdraws or localizes
Pupils Obeys commands
Nonreactive
Respiration
Sluggish
Nil or ventilator
Brisk
Irregular
Regular 22 breaths/min
Regular 22 breaths/min
0 No abnormality detected
1 Trivial lack of awareness, shortened attention span, impairment noted on
arithmetic testing
2 Lethargy, disorientation in time, clear personality change, inappropriate
behavior
3 Very drowsy, semicomatose but responsive to stimuli, confused, gross
disorientation in time or space, bizarre behavior
4 Comatose, unresponsive to painful stimuli with or without abnormal
movements (e.g., decorticate or decerebrate posturing)
79466_CH17.QXD 1/2/08 12:13 PM Page 106
F. Gastrointestinal and other bleeding. More than half of the patients with FHF
have severe gastrointestinal bleeding from acute erosions in the stomach and
esophagus. Coagulopathy, thrombocytopenia, abnormal platelet function, and
DIC contribute to the bleeding propensity of these patients. Because gastric acid
plays a major role in the formation of erosions and ulcers, prophylactic use of
intravenous proton pump inhibitors or histamine-2 (H2)-receptor antagonists has
been shown to decrease the frequency of upper gastrointestinal bleeding.
Retroperitoneal hemorrhage, epistaxis, or bleeding into the lungs may also
occur. Prophylactic use of fresh-frozen plasma is not of proven benefit. In bleeding
patients, however, maintenance of the blood volume by the use of blood products
and correction of clotting factor deficiencies should be tried.
G. Hypoglycemia is common in patients with FHF and may lead to abrupt deepen-
ing of coma. Blood sugar needs to be closely monitored.
H. Sepsis usually complicates FHF due to leukocyte and macrophage dysfunction,
bacterial gut translocation, decreased opsonin function and complement, release of
endotoxin and cytokines, as well as iatrogenic causes such as placement of naso-
gastric tubes, catheters, and central lines. Prophylactic antibiotics are not generally
recommended, but if sepsis is suspected, cultures should be obtained and sepsis
promptly treated.
V. MANAGEMENT
A. General measures. If hepatic regeneration occurs, complete recovery from FHF
is theoretically possible. The major factors that affect progression of liver injury
are intercurrent infections and respiratory and hemodynamic instability, resulting
in alterations of cerebral and hepatic perfusion. Vigorous maintenance of vital
functions and prompt identification and treatment of all anticipated complica-
tions, particularly of cerebral edema to prevent brain damage, are essential. The
patient should be treated in an intensive care unit, where monitoring personnel are
in constant attendance and prepared for emergencies.
The patient’s vital signs and cardiac rhythm should be monitored continu-
ously. A nasogastric tube, passed to decompress the stomach and monitor for
gastrointestinal bleeding, may be used as an access to give oral medicines. Close
attention should be given to the patient’s fluid balance. A urinary catheter and a
central venous catheter should be placed. It may be necessary to monitor the
plasma glucose as frequently as hourly. Hemoglobin level, blood urea nitrogen
(BUN), and electrolytes should be checked every 12 hours. Measurement of arter-
ial blood gases is essential in determining the acid–base status and oxygenation.
The usual indications for endotracheal intubation and assisted ventilation should
be observed. Frequent surveillance for infection is necessary. The prophylactic use
of antibiotics is not recommended.
Patients should not be given prophylactic infusions of fresh-frozen plasma or
concentrates of clotting factors to treat the coagulopathy without any evidence of
bleeding. Daily administration of parenteral vitamin K is appropriate. Suppression
of gastric acid secretion, maintaining the intragastric pH above 5.0, has been shown
to be effective in preventing upper gastrointestinal bleeding and reducing the
requirements for blood transfusions in patients with FHF. Hypothermia should be
avoided. In patients with delirium or convulsions, intravenous diazepam may
be used cautiously.
Acetylcysteine may be given intravenously in a dose of 150 mg /kg of
body weight in 250 mL of 5% dextrose over a period of 15 minutes and then in
a dose of 50 mg/kg in 500 mL of 5% dextrose over a period of 4 hours. This
regimen has been shown to increase survival of patients with established liver dam-
age induced by acetaminophen even when administered more than 15 hours after
the acetaminophen overdose. This beneficial effect seems to result from an increase
in tissue oxygen transport (delivery and consumption) in response to acetylcys-
teine. The beneficial effect was also seen in eight persons with FHF from other
causes.
79466_CH17.QXD 1/2/08 12:13 PM Page 108
Selected Readings
Albataineh H, et al. Acute liver failure secondary to Clarithromyah: A case report and
literature review. Pract Gastroenterol. 2007; xxx1(7):87–89.
Keefe EB. Acute liver failure. In: McQuaid KR, Friedman SL, Grendell JH, eds. Current
Diagnosis and Treatment in Gastroenterology. 2nd ed. New York: Lange Medical
Books/McGraw-Hill; 2003:536–545.
79466_CH17.QXD 1/2/08 12:13 PM Page 110
Khashab M, et al. Epidemiology of acute liver failure. Curr Gastroenterol Rep. 2007;
9:66–73.
Montalti R, et al. Liver transplantation in fulminant hepatic failure: experience with
40 adult patients over a 17-year period. Transplant Proc. 2005;37:1085–1087.
O’Grady JG. Acute liver failure. Post Grad Wed J. 2005;81:148–154.
Ostapowicz G, et al. Results of a prospective study of acute liver failure at 17 tertiary care
centers in the United States. Ann Intern Med. 2002;137:947–954.
Polson JL. AASL position paper: the management of acute liver failure. Hepatology.
2005;41:1179–1197.
Riordan SM, et al. Fulminant hepatic failure. Clin Liver Dis. 2000;4:24–45.
Riordan SM, et al. Use and validation of selection criteria for liver transplantation in acute
liver failure. Liver Transplant. 2000;6:170–173.
Schiodt FV, et al. Etiology and outcome for 295 patients with acute liver failure in the
United States. Liver Transplant Surg. 1999;5:29–34.
Voquero J, et al. Mild Hypothermia for the treatment of acute liver failure. Nat Clin Pract
Gastroenterol Hepatol. 2007;4(10):528–529.
79466_CH18.QXD 1/2/08 12:14 PM Page 111
P regnant women may present with specific diseases that occur exclusively during
pregnancy as well as those that are present at the time of pregnancy and those that occur
coincidentally with pregnancy. In this chapter, only the diseases that occur exclusively in
pregnancy will be discussed.
I. HYPEREMESIS GRAVIDARUM
A. Definition and epidemiology. Hyperemesis gravidarum involves intractable vom-
iting during pregnancy that may lead to electrolyte abnormalities, dehydration,
and malnutrition. It is more common in the first trimester, in women younger than
25 years of age, and with multiple gestations. Incidence in the United States may
be as high as 6 in 1,000 deliveries.
B. Clinical findings. Patients usually present with nausea, vomiting, dysphagia,
odynophagia, epigastric pain, and dehydration. Fifty percent of patients may
have elevations in the serum aminotransaminases (alanine aminotransferase and
aspartate aminotransferase) as well as, occasionally, also of bilirubin and alkaline
phosphatase levels. There may be concomitant gastroesophageal reflux disease
(GERD) and in the severely malnourished patient, esophageal candidiasis, or her-
pes simplex virus (HSV) infection. Some patients may have hyperthyroidism.
C. Treatment. Most patients do well with intravenous hydration and antiemetics.
Some patients may require gastric acid suppressive drugs such as proton-pump
inhibitors (PPIs) or histamine receptor blockers to control GERD, and esophagi-
tis. While no formal studies of these drugs have been performed on pregnant
women, there have been no reports of adverse effects to the mothers or fetuses
during clinical use. Prokinetic drugs such as metoclopramide hydrochloride
(Reglan) 10 to 20 mg intravenously or by mouth four times daily may be used
concomitantly with PPIs. Antiemetics (i.e., ondansetron [Zofran] 4–5 mg) may
be given by mouth three to four times a day or 8 mg intravenously every 4 to
8 hours.
In refractory cases, upper gastrointestinal (UGI) endoscopy may be required
to document UGI mucosal injury due to peptic and/or infectious (candida or HSV)
concomitant disease. A minority of patients may require total or peripheral
parenteral nutrition.
111
79466_CH18.QXD 1/2/08 12:14 PM Page 112
levels may be present. For early diagnosis, the specific biochemical marker of the
change of 3-hydroxysteroid-sulfate ratio of progesterone metabolites may be used.
C. Clinical outcome. Mothers do well in nearly in all cases; however, fetal outcome
may be complicated by prematurity, prenatal death, fetal distress, and meconium
staining of the amniotic fluid.
D. Treatment. Symptomatic treatment of pruritus addresses increased serum bile salt
concentration.
1. Cholestyramine resin (Questran) (4 mg, 1–4 times daily before meals) works
by intraluminal binding of bile salts.
2. Ursodeoxycholic acid (Actigall) (300 mg, 3–4 times daily) works by modifi-
cation of the serum bile acid concentrations by inhibiting the absorption of
more hydrophobic bile acids that are thought to be more pruritic.
3. Dexamethasone (2 mg per day) may be used in refractory cases. Its mode of
action is not well delineated.
.
III ACUTE FATTY LIVER OF PREGNANCY
A. Definition and epidemiology. In acute fatty liver of pregnancy (AFLP),
microvascular fatty infiltration of the hepatocytes leads to progressive liver
failure. Incidence in the United States is 1 in 7,000 to 15,000 deliveries. AFLP
is more common in cases of multiple gestations, male fetuses, and in first
pregnancies. It generally occurs in the third trimester, as early as 26 weeks of
gestation.
B. Clinical findings. Presenting symptoms include nausea, vomiting, headache,
malaise, and abdominal pain (diffuse, right upper quadrant, or epigastric). Jaundice
may follow initial symptoms. Serum alkaline phosphatase, bilirubin, and transami-
nase levels are mildly to moderately elevated. Hyperuricemia may occur in 80% of
patients. Hypoglycemia may be present. Preeclampsia may occur in 20% to 40% of
patients. Progressive liver failure may occur with coagulopathy, encephalopathy,
and renal failure.
C. Diagnosis. Diagnosis should be prompt with high clinical awareness. Abdominal
ultrasound shows diffusely increased echogenicity of the liver. Computed tomog-
raphy scan of the abdomen may be more sensitive than abdominal ultrasound.
D. Treatment and outcome. Treatment is rapid delivery of the fetus. If fulminant
hepatic failure develops, liver transplantation may be lifesaving. Maternal mortality
is reported to be 8% to 18% and fetal morality 18% to 23%.
Selected Readings
Bor S, et al. Association of heartburn during pregnancy and the rise of gastroesophageal
reflux disease. Clin Gastroenterol Hepatol. 2007;5:1035–1039.
Floreami A, et al. Intrahepatic cholestatis of pregnancy: three novel MDR3 gene mutations.
Ailment Pharmacul Ther. 2006;23:1649–1653.
79466_CH18.QXD 1/2/08 12:14 PM Page 114
Specific Complaints V
and Disorders
79466_CH19.QXD 1/2/08 12:14 PM Page 116
79466_CH19.QXD 1/2/08 12:14 PM Page 117
ESOPHAGEAL INFECTIONS 19
117
79466_CH19.QXD 1/2/08 12:14 PM Page 118
Figure 19-1. Algorithm depicting diagnostic approach in patients with suspected fungal esophagitis.
This may be a focal narrowing in the upper esophagus or may involve the
entire length.
C. Treatment
1. Before the initiation of any specific therapy for esophageal candidiasis, the
underlying predisposing factor should be identified.
2. Patients with AIDS and mild-to-moderate symptoms who have oral thrush may
be treated initially with one of the topical agents such as nystatin (Mycostatin)
suspension, clotrimazole troches, or one of the “azoles” may be an alterna-
tive form of therapy before undergoing a diagnostic procedure (Table 19-1).
3. Patients with leukopenia or with severe symptoms or systemic signs should
undergo endoscopy to obtain biopsy and brushing specimens for fungi, viruses,
and bacteria.
4. Orally administered systemic antifungal agents include the azoles and polyene
antibiotics. The most commonly used azoles include ketoconazole (Nizoral),
fluconazole (Diflucan), and itraconazole (Sporanox). All are effective in treat-
ing esophageal candidiasis.
Ketoconazole may be used at high doses. However, its gastrointestinal
side effects; decreased absorption in achlorhydric patients or patients on acid
antisecretory drugs such as histamine-2 (H2) blockers, proton-pump inhibitors,
and prostaglandin analogs; and hepatotoxicity may limit its usefulness.
Fluconazole is extremely efficacious in treating oral candidiasis and Candida
and other fungal esophagitis with or without tissue invasion. Fluconazole may be
administered via oral or parenteral routes. Its oral absorption is efficient and does
not require the presence of gastric acid. It is minimally metabolized and excreted
by the urine. Drug interactions have been demonstrated between fluconazole
79466_CH19.QXD 1/2/08 12:14 PM Page 120
Topical
Nystatin 1–3 million units (10–30 mL) 4–5 times daily; swish
and swallow for 5–10 days
Clotrimazole 1 troche (10 mg) 5 times daily for 5–10 days
Systemic
Amphotericin B 0.3–0.6 mg/kg/d intravenously
Ketoconazole 200–400 mg once daily, orally for 5–10 days
Fluconazole 50–200 mg once daily, orally for 5–10 days
100–200 mg twice daily intravenously for 5–10 days
Itraconazole 200 mg once daily for 5–10 days
(and the other azoles) and other medications, including phenytoin, oral antico-
agulants, sulfonylureas, cyclosporin A, rifampin, and barbiturates. Fluconazole
augments the effects of warfarin, necessitating careful observation of patients
receiving both of these agents. Fluconazole may increase serum levels of
cyclosporin A. Serum cyclosporin A and creatinine levels must be monitored
carefully in patients receiving both of these drugs. Fluconazole and the other
azoles also appear to increase the serum levels of phenytoin and oral hypo-
glycemic agents. Fluconazole appears to have minimal antisteroidogenic effects
in humans at currently recommended doses. This results from fluconazole highly
specific affinity for fungal cytochrome P-450 enzymes, with virtually no affinity
for the mammalian system. This is an important advantage of fluconazole over
ketoconazole in patients with AIDS, who often have adrenal insufficiency sec-
ondary to adrenal cytomegalovirus (CMV) infection.
Itraconazole has a longer half-life than ketoconazole, but its absorption
is also reduced by hypochlorhydria. These two azoles are metabolized by the
liver and excreted in the bile. Dose adjustments are not required in the patient
with renal failure. Total treatment dosage is about 100 to 200 mg.
5. Intravenous (IR) amphotericin B should be considered in symptomatic
patients who fail to respond to the above regimen and in those in whom sys-
temic involvement is suspected. In the absence of systemic Candida infection,
a low-dose regimen of 10 to 20 mg per day for 10 days may be given. The
dosage may be increased if the patient does not respond favorably. If systemic
infection is present, the dosage should be increased gradually to 0.5 mg/kg per
day. Most patients are treated for 6 weeks. The major serious side effect of
amphotericin is renal toxicity, which is usually reversible.
6. A major problem in the treatment of esophageal candidiasis is relapse after
therapy is discontinued, especially in patients with AIDS, in whom immuno-
suppression is unrelenting. If the underlying predisposing factors persist, the
chances for permanent cure are low, and maintenance therapy may be neces-
sary (e.g., fluconazole 100 mg /day). However, in patients with reversible
predisposing conditions (e.g., radiotherapy, steroid use), a single course of
therapy for 10 to 21 days should be successful.
7. Nutritional support of the patient is very important. If the patient can swal-
low, the diet should be supplemented with liquid enteral formulas to ensure
that adequate calories are received. If the patient is unable to swallow, par-
enteral nutrition should be given until the patient can receive enteral feedings.
viruses are the most common etiologic agents of viral esophagitis, with HSV and CMV
predominating. In patients with AIDS, Epstein-Barr virus (EBV), varicella zoster
virus, and HIV-1 also have been implicated in esophageal ulcerations.
A. HSV esophagitis. Herpetic esophagitis may be seen in healthy, normal individu-
als, especially after strenuous physical exertion and stress. The infection in such
individuals is self-limited and resolves in approximately 7 to 10 days. Recovery
generally indicates intact humoral and cellular immunity. Persistence or recurrence
of the disease may be a sign of acquired immunodeficiency, requiring further
workup of the patient.
1. Predisposing factors
a. Immunocompromise
i. Malignancy, mainly of the hematopoietic and lymphoreticular systems
ii. Transplantation
iii. Immunosuppressive drugs, steroids
iv. Antineoplastic chemotherapy
v. AIDS
b. Severe debilitation
i. Elderly
ii. Burns
c. Antecedent trauma
i. Nasogastric tubes
ii. Tracheal intubation
iii. Gastroesophageal reflux disease and peptic esophagitis
2. Diagnosis
a. Clinical presentation. Herpetic infection of the esophagus usually presents
as a triad of fever, odynophagia, and substernal pain that increases with
feeding. The pain may radiate to the back. Patients may complain of pain
on palpation of the xiphoid process. Gingival stomatitis may be present.
Disseminated HSV infection is seen in 30% of the patients. Multiple organ
involvement with the virus (e.g., gastric and respiratory infection) is usually
life-threatening. Simultaneous infection with other organisms (bacteria,
fungi, and other viruses) is common. Severely debilitated patients may not
complain of pain; therefore, a high index of suspicion should be present if
these patients have dysphagia and decreased oral intake.
b. Diagnostic studies
i. Endoscopy. Endoscopic examination of the esophagus is the preferred
diagnostic approach in these patients. Biopsies and brushings of the
affected mucosa should be obtained. Biopsies should be obtained from
the edge of the ulcers.
a) The endoscopic appearance varies according to the stage of the viral
infection.
1) Early—vesicles of various sizes.
2) Mid—small, punched-out superficial ulcers covered with yellow,
fibrinous exudates.
3) Late—coalescing ulcers forming a diffuse, erosive esophagitis
with large areas of shallow ulceration.
b) The lesions often become overgrown with Candida and bacteria.
The virus may be cultured from ulcer margins and vesicles. A smear
of the ulcer base should be processed for Candida. Histologically,
the epithelial cells at the border of the ulcers contain inclusion bod-
ies. The chromatin of the infected nuclei is displaced toward the
periphery of the nucleus, giving a “rim” appearance. Multinucleated
giant cells are often present.
ii. Barium swallow. Double-contrast radiography may show ulcers or
plaques. The picture may be indistinguishable from that of Candida
esophagitis.
c. Diagnosis of herpes esophagitis is often difficult because the characteristic
nuclear inclusions or multinucleate giant cells of HSV infection may be
79466_CH19.QXD 1/2/08 12:14 PM Page 122
III. IDIOPATHIC ESOPHAGEAL ULCER (IEU) related to primary HIV infection. Large,
deep ulcers resembling CMV or HSV ulcers may occur in patients with AIDS and,
when other etiologies are excluded are attributed to HIV. Endoscopic cure may be
obtained with treatments with oral prednisone or thalidomide.
Selected Readings
Bini EJ, et al. Natural history of HIV-associated esophageal disease in the era of protease
inhibitor therapy. Dig Dis Sci. 2000;45:1301–1306.
Denning DW. Echinocardin anti fungal drugs. Lancet. 2003;362:1142–1452.
Kearney DJ, et al. Esophageal disorders caused by infection, systemic illness, medications,
radiation and trauma. In: Feldman M, Freedman LS, Schlessinger MH, eds. Philadelphia:
WB Saunders; 2002:623–646.
Keate RF, et al. Lichen planus: Report of three patients with oral tacrolimus or
intraesophageal corticosteroid injection or both. Dis Esophagus. 2003;16:47–53.
Pappas PG. et al. Guidelines for treatment of candidias. Clin Infect Dis. 2004;38:161–189.
Ramanthan J, et al. Herpes simplex esophagitis in the immunocompetent host: An overview.
Am J Gastroenterol. 2000; 95:2171–2176.
Rebolic AC, et al. Anidula fungia versus flucarazole for invasive candidiasis. N Eng J Med.
2007;356:2472–2482.
Walsh TJ. Echinocardins—an advance in the primary treatment of invasive candidiasis.
N Eng J Med. 2006;354:1215–1256.
79466_CH20.QXD 1/2/08 12:15 PM Page 124
I. DEFINITIONS
A. Gastroesophageal reflux (GER) occurs when gastric contents escape into the
esophagus. This process may or may not produce symptoms. The most common
symptoms are heartburn, regurgitation, chest pain, and dysphagia.
B. Reflux esophagitis can be defined as esophageal inflammation caused by refluxed
material.
C. Gastroesophageal reflux disease (GERD) includes the constellation of symp-
toms and consequences to the esophagus from reflux damage.
II. PATHOGENESIS. The extent and severity of esophageal injury due to GER depend on
the frequency and the duration of esophageal exposure to the refluxed material, the
volume and potency of gastric juice available for reflux, and the ability of the
esophageal mucosa to withstand injury and to repair itself.
The pathogenesis of reflux esophagitis or GERD is a multifactorial process. The
following factors all contribute to the development of GERD:
A. Antireflux mechanisms. A positive pressure gradient exists between the abdomen
and the thorax. If there were no physiologic barrier at the area of the gastro-
esophageal junction, GER would occur continuously, especially with increases in
intraabdominal pressure or changes in gravitational position and during events
associated with abdominal muscle contraction, such as coughing, sneezing, strain-
ing, bending, turning in bed, and exercise. The antireflux barrier can be divided
into two categories.
1. Anatomic factors extrinsic to the lower esophageal sphincter (LES) that
augment the LES to prevent GER include a distal esophageal mucosal flap,
the acute esophagogastric angle, compression of the esophagogastric junction
by gastric sling fibers, the diaphragmatic crus acting as pinchcock, a hiatal tun-
nel, the sling action of the right diaphragmatic crus, and the intraabdominal
junction of the esophagus. The longer the intraabdominal segment, the less
likely reflux is to occur.
The presence of hiatal hernia with loss of the abdominal esophageal seg-
ment supported by the diaphragm and the normal acute esophagogastric angle
may lead to GER. However, a direct causal relationship has not been found
between hiatal hernia and GER. Nevertheless, a hiatal hernia generally (90%)
accompanies reflux esophagitis. It is possible that hiatal hernia enhances the
likelihood of LES dysfunction due to the loss of angulation at the esopha-
gogastric junction and the direct transmission of intragastric pressure to the
infrathoracic LES. Also, the hiatal hernia may act as a reservoir of refluxate
and impair esophageal clearance in the recumbent position, thus promoting
esophageal injury.
2. The closure strength and efficacy of LES
a. LES corresponds to the 2- to 4-cm zone of asymmetrically thickened smooth
muscle at the esophagogastric junction.
b. LES maintains a high-pressure tone during resting conditions and relaxes
with swallowing, esophageal distention, and vagal stimulation. These prop-
erties are independent of the diaphragm and persist even when the LES is
in the thorax, as in patients with hiatal hernia.
124
79466_CH20.QXD 1/2/08 12:15 PM Page 125
Gastrin Secretin
Pitressin Cholecystokinin
Angiotensin II Glucagon
Cholinergics (e.g., bethanechol) Vasoactive intestinal polypeptide
Gastric alkalinization Progesterone (birth control pills)
Metoclopramide Theophylline
Anticholinesterases Caffeine
Protein meal Gastric acidification
Prostaglandin F2 Fatty meals
Chocolate (xanthines)
Carminatives (spearmint, peppermint)
Smoking
Ethanol
-Adrenergic antagonists
-Adrenergic agonists
Anticholinergics
Calcium channel blockers
Nitrates
Prostaglandin E2, prostaglandin A2
Morphine, meperidine
Diazepam, other benzodiazepines
this layer appears to be necessary for the development of esophageal ulcers, stric-
tures, and Barrett’s epithelium. There is evidence that epithelial cell turnover and
replication is increased after hydrogen (H+) injury. Basal cell hyperplasia seen in
mucosal biopsies of patients with reflux esophagitis lends further support to this
finding. Normal turnover rate for esophageal epithelium is 5 to 8 days. This rate
seems to be increased to 2 to 4 days with injury. This will allow for epithelial
renewal and repair in a short time if further injury is prevented.
F. Summary. Patients with reflux esophagitis have heterogeneous abnormalities that
contribute to the development of esophagitis. Because patients have different
underlying abnormalities responsible for their reflux esophagitis, correct diagnosis
of the specific abnormalities involved allows for designing and selecting appropri-
ate therapy. Thus, therapy may be individualized and directed toward increasing
the LES pressure, enhancing esophageal clearance, promoting salivary output,
improving gastric emptying, suppressing gastric acidity, binding bile salts and pro-
teolytic enzymes, and promoting intrinsic epithelial defenses. Nighttime GER is
most deleterious to the esophageal mucosa and it should be considered with each
patient and addressed therapeutically.
III. DIAGNOSIS
A. Clinical presentation. The prevalence of heartburn, the most common clinical
manifestation of GER, is difficult to determine. Most people consider this sensa-
tion normal and do not seek medical attention. It is estimated that at least one
third to one half of the U.S. population experience heartburn at least once a month
and up to 20% of the population experience heartburn daily. The most common
symptoms of GERD are as follows:
1. Heartburn (pyrosis). A substernal burning pain, radiating upward. Ingestion
of antacids usually relieves this symptom within 5 minutes.
2. Regurgitation. Reflux of sour or bitter material into the mouth usually at
night, while lying down, or when bending over. It suggests severe reflux.
3. Dysphagia. Difficulty in swallowing. Dysphagia usually indicates a narrowing
or stricture of the esophagus; however, it may occur due to inflammation and
edema, which may resolve with aggressive medical therapy of the GERD.
4. Odynophagia. Pain on swallowing, which sometimes accompanies severe
esophagitis.
5. Water brash. Filling of the mouth suddenly with a clear, slightly salty fluid,
which comes in large quantities. The fluid is not refluxed from the stomach but
is secreted by the salivary glands in response to GER.
6. Chest pain. Resembling angina of cardiac origin, chest pain is an atypical pre-
sentation of GERD. This pain may result from acid-induced irritation of the
nerve endings in the elongated rete pegs protruding into the surface epithelium
or from GER-induced esophageal spasm or GER-induced angina pectoris.
In a study of the cardiovascular effect of reflux, esophageal acid perfusion
produced an increase in cardiac workload in patients with angiographically
proven coronary artery disease. Some patients had ischemic changes on elec-
trocardiography during acid perfusion. This suggests that esophageal and car-
diac disease not only may coexist, but also may interconnect. The standard
clinical approach aimed at distinguishing between esophageal and cardiac pain
may represent a serious oversimplification.
7. Hemorrhage may be the first clinical manifestation of esophagitis. It may be
brisk, bright red, or slow and may result in iron-deficiency anemia.
8. Pulmonary symptoms may be the only manifestation of GER and include
chronic cough, hoarseness of voice, wheezing, hemoptysis, asthma, and recur-
rent aspiration pneumonia. Although it is often assumed by clinicians that pul-
monary symptoms associated with reflux result from aspiration, reflux may
increase airway resistance without aspiration, apparently through vagus-mediated
neural reflexes.
9. Other symptoms such as sleep apnea, poor sleep/insomnia, and daytime
sleepiness may result from nighttime GER.
79466_CH20.QXD 1/2/08 12:15 PM Page 129
Patients follow a normal diet with the exception of foods with pH below 5.0.
Patients are asked to write down their symptoms and their body position
(upright or supine) during the test period. Reflux is defined as the point at
which the pH drops to less than 4.0. Each patient’s reflux status is assessed by a
composite score that incorporates six components:
a. Percentage of time of total acid exposure of the esophagus.
b. Percentage of acid exposure in upright and recumbent positions.
c. Presence of reflux episodes.
d. Total number of reflux episodes.
e. Number of reflux episodes longer than 5 minutes.
f. The longest reflux (time). This test is excellent in identifying acid GER but
does not detect “alkaline” reflux. Prolonged pH monitoring has also been
helpful in documenting the suspected association between GER and pul-
monary disease.
7. Wireless pH monitoring (Brano) device which is placed endoscopically in
the distal esophagus allows for prolonged monitoring (2–4 days) and better
understanding of the day-to-day variability of GER as well as assessing the
effectiveness of acid suppressive therapy without performing a second test.
8. Multichannel intraluminal impedance (MII) is used to assess GER, esopha-
geal bolus transit (peristaltic function), and the proximal extent of the reflux
event. MII may be used in combination with esophageal manometry and
pH testing. It is helpful in documenting GER regardless of the pH of the
refluxate.
9. Esophagogastroduodenoscopy and mucosal biopsy. Flexible fiberoptic
endoscopy has become the most widely used method to examine the mucosal
surface of the esophagus for evidence of esophagitis. Endoscopic forceps biop-
sies are adequate for evaluating histologic changes of GERD. Even when
endoscopic appearance of the esophagus is normal, histologic examination of
the biopsies may confirm the presence of GERD.
Findings of esophagitis by endoscopy are as follows:
a. Mild. Erythema; edema of the mucosa with obliteration of small, linear
blood vessel; mild friability; and increased irregularity of the Z line.
b. Moderate-severe. Round and longitudinal superficial ulcers or erosions,
diffusely hemorrhagic mucosa with exudates, and deep, punched-out
esophageal ulcers and strictures.
10. Histology. In patients with GERD, there is a hyperplasia of the basal cell layer
of the squamous epithelium. This layer constitutes more than 15% of the
epithelial thickness. The dermal papilla extends more than 65% of the dis-
tance to the epithelial surface. Polymorphonuclear leukocytes and eosinophils
may be seen in the lamina propria and may invade the epithelium. Ingrowth
of capillaries is also seen in the lamina propria.
In about 10% to 20% of the patients with chronic GERD, a specialized
columnar metaplastic epithelium (Barrett’s epithelium) is present. Endoscopic
examination of the stomach and the duodenum can rule out other possible
lesions in these areas.
11. Summary. For the diagnosis of GERD, most patients with the classic symp-
toms of GERD of heartburn or regurgitation are given an empiric trial of
medical therapy without further investigation. Endoscopy and mucosal biop-
sies are recommended in patients with refractory symptoms, odynophagia,
dysphagia, and atypical symptoms and in patients when Barrett’s esophagus is
suspected (e.g., those patients with GER symptoms for more than 5 years).
Prolonged pH monitoring and manometry are reserved for patients with atyp-
ical symptoms and pulmonary complaints.
to stricture formation include prolonged GER, reflux while supine, nasogastric intu-
bation, duodenal ulcer disease, gastric hypersecretory states, postgastrectomy states,
scleroderma, and treated achalasia. Ringlike stricture in the distal esophagus at the
E-G junction is called Schatzki’s ring.
1. Location. Strictures are usually located in the distal third of the esophagus. In
the barium esophagogram, they usually have a smooth, tapered appearance
and are of variable lengths. In some instances of Barrett’s esophagus, the stric-
ture is located in the middle third or, less commonly, in the proximal third of
the esophagus.
2. Symptoms. Peptic strictures usually produce no symptoms until the esophageal
intraluminal diameter is decreased to less than 12 mm. Initially dysphagia is
mostly for solids, but with progressive narrowing, swallowing of liquids also
becomes a problem. It is not uncommon for the patient to notice improvement in
the usual reflux symptoms as dysphagia develops with narrowing of the stric-
tured area. Some patients do not recall even having GER symptoms.
3. Treatment. After appropriate diagnostic tests (barium esophagram, endoscopy,
and biopsies) have been performed to ensure that the stricture is not due to a
malignant process, intensive medical therapy is begun for reflux esophagitis.
With resolution of edema and inflammation, some patients may have relief of
their symptoms. However, most patients require additional therapy in the form
of dilatations, surgery, or both.
a. Dilatation. Progressive dilatations with graded mercury-filled rubber bou-
gies (Maloney or Hurst dilators) have been used in the past for symptomatic
relief. Savary dilators passed over a guidewire or inflatable balloon dilators
with endoscopic guidance offer safer and more effective means of dilatation.
Savory dilators come in graded sizes. The guide wire is passed
through the biopsy channel of the endoscope and advanced into the stric-
ture, then into the stomach. The endoscope is then removed. The savory
dilator is then “threaded over” the guide wire and gently advanced into the
lumen of the stricture to dilate it. Afterward, it is withdrawn, then a larger-
sized savory dilator is advanced. Similarly, the process is repeated with
larger dilators until the stricture is adequately dilated and/or “blood” is
seen on the dilator. The dilators should never be forced into the stricture to
avoid perforation. Many gastroenterologists use savory dilators with the
aid of fluoroscopy.
Endoscopic balloon dilators allow endoscopic visualization during
the entire procedure of dilatation. Each balloon catheter can be inflated to
three enlarging sizes that allow progressive dilatation with one insertion.
During the EGD the balloon dilatator catheter is introduced via the biopsy
channel into the esophageal lumen, then into the stricture. The balloon at
the end of catheter is then dilated progressively until the stricture is dilated
to the desired size.
The major complications of dilatations are perforation and bleeding.
Perforations are rare but should be suspected if the patient complains of
persistent pain after dilatation. Perforations are identified by a radiologic
contrast study. Patients are treated with intravenous nutrition and antibi-
otics to cover for organisms from the mouth flora. Surgical drainage and
repair should be considered early, since mortality associated with large
esophageal perforations is high.
Dilatation of the stricture and medical therapy for reflux yield good
results in 65% to 85% of patients. The patency of the esophageal lumen is
maintained by additional dilatations at intervals of weeks to months.
b. Surgery. For the 15% to 40% of patients in whom dilatation and medical
therapy fail, surgery is indicated. The preferred surgical approach to stric-
tures is pre- or intraoperative dilatation combined with an antireflux opera-
tion, such as the Nissen fundoplication. If the stricture cannot be dilated or
is too extensive, resection and end-to-end anastomosis or interposition of a
segment of colon or small bowel may be used. These may be combined with
79466_CH20.QXD 1/2/08 12:15 PM Page 132
b. Drugs that decrease gastric acid output. Those most commonly used
drugs are histamine-2 (H2) blockers. For patients with intermittent, infre-
quent or mild symptoms of GER, H2 blockers may be prescribed. These
drugs are usually effective in controlling symptoms of mild to moderate
GER, but have not been shown to effectively heal erosive esophagitis.
H2-blockers do not suppress gastric secretions completely. They decrease
gastric acid secretion by binding to the histamine receptor on the parietal
cell in a competitive fashion. When their concentration decreases around
the parietal cell, histamine binds to the parietal cell receptor and acid secre-
tion is resumed. Thus, regular and frequent dosing is essential. H2 blockers
also have very limited acid suppressive ability at times of eating and the
gastric pH rarely rises above 2 to 3. Because GER occurs most commonly
during and after eating, acid and pepsin activity is not eliminated with
H2 blockers and their effectiveness is thus limited.
i. Cimetidine (Tagamet), 300 mg q.i.d. or 400 to 800 mg q12h a.c. and h.s.
ii. Ranitidine (Zantac), 150 to 300 mg q12h
iii. Famotidine (Pepcid), 20 to 40 mg at h.s.
iv. Nizatidine, 150 to 300 mg q12h
Note: Tagamet, Zantac, and Pepcid are also available at lower doses as
over-the-counter medications.
c. Proton-pump inhibitors. The final step of gastric acid secretion by the
parietal cell involves the extrusion of a proton or a hydrogen ion (H) into
the gastric lumen in exchange of a potassium ion (K), which enters the
parietal cell via the H K-ATPase or the “proton pump.” PPIs are a group
of drugs designed to inhibit acid secretion by forming a covalent bond
within the proton pump and, thus, inhibiting the exchange of (H) and
(K) ions permanently by that proton pump. These drugs are extremely
effective in inhibition of gastric acid secretion for 19 to 24 hours and allow
the gastric pH to rise above 4 or 5, thus, also eliminating the pepsin activ-
ity. PPIs have been shown to effectively heal erosive esophagitis, diminish
the formation of esophageal strictures, and control most symptoms and
signs of GERD. PPIs, when used continuously, prevent recurrence of ero-
sive esophagitis. As a group of drugs, they are safe and have minimal side
effects. In most instances, they are used as first line therapy for GERD.
Currently, there are six commercially available PPIs approved by the
U.S. Food and Drug Administration (FDA) for healing erosive GERD. These
are omeprazole (Prilosec), 20 to 40 mg p.o. (q.d. or b.i.d.); lansoprazole
(Prevacid), 15 to 30 mg p.o. (q.d. or b.i.d.); rabeprazole sodium (AcipHex),
20 mg orally (once daily); pantoprazole (Protonix), 40 mg p.o. (q.d.);
esomeprazole magnesium (Nexium), 20 to 40 mg p.o. (q.d.); and omepra-
zole sodium bicarbonate (Zegerid), 40 mg p.o. (q.d.). All PPIs are most
effective if taken 15 to 30 minutes before breakfast or dinner. Omeprazole,
lansoprazole, pantoprazole, and esomeprazole magnesium are excreted in the
urine. Rabeprazole sodium is mostly excreted in bile. No dose adjustments
are necessary in renal or hepatic insufficiency. Omeprazole, esomeprazole
magnesium, lansoprazole, and rabeprazole sodium are metabolized by the
p450 system in the liver and may have minor drug–drug interactions with
some drugs; however, these have not been shown to be clinically significant.
Pantoprazole has no known drug–drug interactions. Pantoprazole, esomepra-
zole, and lansoprazole are also available in intravenous (IV) formulation and
have been FDA-approved for in patients with GERD who are unable to
receive medications by the oral route.
Side effects of PPIs are rare and include minor headaches, diarrhea,
and nausea. PPIs have revolutionized the treatment of GERD and most of
its complications. These drugs have been noted to be safe and free of long-
term complications. Initial concerns about gastric bacterial overgrowth,
Vitamin B12 and iron malabsorption, and causation of gastric carcinoid
tumors have not been clinically observed. A retrospective observational
79466_CH20.QXD 1/2/08 12:15 PM Page 135
study suggests that patients taking PPIs for many years may have decreased
calcium absorption. Thus it may be prudent to encourage patients on pro-
longed PPI therapy to take calcium supplements.
d. Drugs that increase LES pressure and esophageal clearance
i. Metoclopramide hydrochloride, a dopamine antagonist, has been
shown to increase LES pressure and to improve esophageal and gastric
emptying. It counteracts the receptive relaxation of the gastric fundus
and increases duodenal and small-bowel motility. It is also a centrally
active antiemetic. This drug is especially helpful in patients with GERD
and abnormalities of gastric emptying. Since metoclopramide crosses
the blood–brain barrier, 10% of patients seem to experience psy-
chotropic side effects (e.g., somnolence, lassitude, restlessness, anxiety,
insomnia, and, rarely, extrapyramidal reactions). These side effects are
reversible with cessation of the drug. Elevated prolactin levels may
occur and cause galactorrhea. The usual dosage is 10 to 20 mg q.i.d.,
15 to 30 minutes a.c. and h.s.
ii. Other prokinetic drugs such as domperidone and cisapride do not
cross the blood–brain barrier and have only the peripheral effects of
metoclopramide. These drugs have excellent promotility qualities and
have been used successfully in the treatment of GERD.
However, cisapride, due to its drug–drug interactions with those
drugs that prolong the Q-T interval (on electrocardiogram) and the pos-
sibility of precipitating cardiac arrhythmias, has been removed from the
market in the United States by the manufacturer. Cisapride and dom-
peridone are available in Canada and other countries.
e. Drugs that enhance mucosal resistance. As the importance of mucosal
resistance is appreciated, drugs that potentiate cytoprotection and mucosal
resistance are being added to the medical armamentarium of acid-peptic
disease.
i. Sucralfate (Carafate), an aluminum sucrose polysulfate shown to be
effective in healing duodenal ulcers due to its cytoprotective action, has
not been shown to be highly effective in patients with esophagitis.
However, sucralfate suspension seems to give symptomatic and possibly
therapeutic benefit to patients with erosive esophagitis. Dose is 1 g 1–4
times daily.
ii. Prostaglandin analogs (e.g., misoprostol) have also been shown to be
cytoprotective and effective in the treatment of peptic ulcer disease, but
have not been found highly effective in the treatment of GERD.
f. The results of maintenance therapy with H2-receptor antagonists are dis-
appointing. Neither twice-daily nor single-dose-bedtime regimens of cimeti-
dine or ranitidine are significantly more effective than placebo in preventing
symptomatic or endoscopic evidence of relapse. However, maintenance ther-
apy with 20 mg of omeprazole daily sustains endoscopic healing in most
patients with severe, recalcitrant esophagitis. In some patients, the dosage
needs to be increased to 40 mg. Significant and persistent elevations in fast-
ing serum gastrin concentrations may occur in a minority of patients. It is
important to note that up to 90% of patients who healed with omeprazole
treatment had recurrence of their esophagitis within 6 months of stopping
treatment, indicating that some form of chronic treatment is needed. Similar
excellent results have also been achieved with the other PPIs (lansoprazole,
rabeprazole sodium, pantoprazole, and esomeprazole magnesium) at the
same doses used to heal erosive esophagitis.
B. Phase III, endoscopic interventional and/or surgical therapy, is reserved for
patients in whom intensive medical therapy has failed and those with complica-
tions such as a nonhealing or bleeding esophageal ulcer or a refractory stricture.
Surgery for Barrett’s esophagus is still controversial.
Endoscopic treatment of GERD has been attractive as an alternative to
surgery by providing a less invasive solution to GER refractory to medical therapy.
79466_CH20.QXD 1/2/08 12:15 PM Page 136
Selected Readings
Abid S, et al. Pill-induced esophageal injury: Endoscopic features and clinical outcomes.
Endoscopy. 2005;37:470–474.
Abou-Rebych H, et al. Long-term failure of endoscopic suturing in treatment of gastro-
esophageal reflux: A prospective follow-up study. Endoscopy. 2005;37(3):213–216.
Boolchard V, et al. Risk for Carcenin Barrett’s esophaus: medical versus surgical therapy. Cur
Gastroenterol Hepatol. 2007;9:189–194.
Byrne P, et al. Comparison of transoesophageal endoscopic plication (TEP) with laparoscopic
Nissen fundoplication (LNF) in the treatment of uncomplicated reflux disease. Am J
Gastroenterol. 2006;101:431–436.
Charbel S, et al. The role of esophageal pH monitoring in symptomatic patients on PPI
therapy. Am J Gastroenterol. 2005;100:283–289.
Chejtec G, et al. Gastroesophageal reflux disease—A review. US Gastroenterol Rev. 2007;
1:84–87.
DeVault KR, et al. Updated guidelines for the diagnosis and treatment of gastroesophageal
reflux disease. Am J Gastroenterol. 2005;100:190–200.
Domagh D, et al. Endoluminal gastroplasty (EndoCinch) versus endoscopic polymer
implantation (Enteryx) for treatments of gastroesophageal reflux disease: 6 month
results of a prospective, randomized trial. Am J Gastroenterol. 2006;101:422–430.
Fass R, et al. Treatment of patients with persistent heartburn symptoms: A double-blind
randomized trial. Clin Gastroenterol. 2006;4:50–56.
Fernando HC, et al. Outcomes of laparoscopic Toupet compared to laparoscopic Nissen
fundoplication. Surg Endosc. 2002;16:905–908.
Hirano I. Review article: Modern technology in the diagnosis of gastroesophageal reflux
disease—Bilitec, intraluminal impedance and Bravo capsule pH monitoring. Aliment
Pharmacol Ther. 2006;23(suppl 1):12–24.
Kikendall JW. Pill induced esophagitis. Gastroenterol Hepatol. 2007;3(4):275–276.
Lagergren J, et al. Symptomatic gastroesophageal reflux as a risk factor for esophageal
adenocarcinoma. N Engl J Med. 1999;340:825.
Martinez SD, et al. Non-erosive reflux disease (NERD), acid reflux and symptom patterns.
Aliment Pharmacol Ther. 2003;17:537–545.
Oleynikov D, et al. Total fundoplication is the operation of choice for patients with
gastroesophageal reflux and defective peristalsis. Surg Endosc. 2002;16:909–913.
Overholt BF, et al. Photodynamic therapy for Barrett’s esophagus follow-up in 100
patients. Gastrointest Endosc. 1999;49:1.
79466_CH20.QXD 1/2/08 12:15 PM Page 137
21 DYSPHAGIA
I. ESOPHAGUS
A. Anatomy
1. The esophagus is a muscular tube measuring about 25 cm (40 cm from the
incisor teeth) extending from the pharynx at the cricoid cartilage to the cardia
of the stomach. It pierces the left crus of the diaphragm and has an intraab-
dominal portion of about 1.5 to 2.5 cm in length.
2. The esophageal mucosa consists of a nonkeratinizing squamous epithelium,
lamina propria extending into the basal layer as rete pegs, and muscularis
mucosa, which is sparse and thin in the upper portion but thicker near the gas-
troesophageal junction.
3. The submucosa contains mucous glands and an extensive lymphatic plexus in
a connective tissue network.
4. Between the submucosa and muscularis propria are the cell bodies of secondary
neurons forming the Auerbach’s plexus.
5. The muscularis propria, the main muscle layers of the esophagus, is composed
of inner circular and outer longitudinal coats. In the upper part, these are stri-
ated. There is a gradual change to smooth muscle in the middle. In the lower
third of the esophagus, both of these coats are entirely composed of smooth
muscle.
6. Between the muscle layers, the myenteric plexus contains the cell bodies of
other secondary neurons.
7. The esophagus does not have a serosal layer.
8. Lower esophageal sphincter. The distal 3 to 4 cm of the esophagus constitutes
a zone of increased resting pressure in an asymmetric fashion. This area, called
the lower esophageal sphincter (LES), behaves both physiologically and pharma-
cologically as a distinct entity from the esophageal smooth muscle immediately
adjacent to it. Basal LES pressure is normally 10 to 25 mmHg higher than intra-
gastric pressure and drops promptly (within 1–2 seconds) with swallowing. The
LES control remains poorly understood but is thought to involve the complex
interaction of neural, hormonal, and myogenic activities.
B. Physiology of esophageal function. The function of the esophagus is to transport
food and secretions from the mouth to the stomach. This coordinated process oper-
ates regardless of the force of gravity.
1. A swallow begins when a liquid or solid bolus is propelled to the back of the
mouth into the pharynx by the tongue. The upper esophageal sphincter (UES),
the cricopharyngeus, which is just below the pharynx, relaxes, allowing the
bolus to pass into the upper esophagus. In response to swallowing, an orderly,
progressive contraction of the esophageal body occurs (primary peristalsis),
propelling the bolus down the esophagus. When the esophagus is distended by
a bolus (i.e., with regurgitation), secondary peristaltic contractions are initi-
ated. The LES relaxes as the bolus reaches the lower esophagus, allowing pas-
sage of the food into the stomach.
2. The relaxation of the UES and peristalsis in the upper esophagus are ini-
tiated by the voluntary act of swallowing, controlled by the swallowing center
in the brainstem and the fifth, seventh, ninth, tenth, eleventh, and twelfth cranial
nerves. These nerves coordinate the movement of the bolus to the hypopharynx,
138
79466_CH21.QXD 1/2/08 12:15 PM Page 139
closure of the epiglottis, relaxation of the UES, and contraction of the striated
muscle of the upper esophagus. The sequential nature of this function is due to
progressive activation of nerve fibers carried in the vagus nerve controlled
through a central mechanism.
3. The peristalsis in the smooth-muscle portion of the esophagus is regu-
lated by activation of neurons located in the myenteric plexus with cholinergic
neural transmission. The vagi innervate the upper esophagus in its striated mus-
cle portion only. If the vagi are cut below the level of mid esophagus, peristalsis
in the lower half of the esophagus and the function of the LES remain intact.
II. DYSPHAGIA
A. Definition
1. Dysphagia is difficulty in swallowing. Clinically, it includes the inability to ini-
tiate swallowing and/or the sensation that the swallowed solids or liquids stick
in the esophagus.
2. Odynophagia refers to pain with swallowing. In some disorders, odynophagia
may accompany dysphagia.
3. Globus hystericus describes the sensation of the presence of “a lump in the
throat” that is relieved momentarily by swallowing.
B. Preesophageal or oroesophageal dysphagia. Patients with this disorder have
problems with the initial steps of swallowing. They may have difficulty in pro-
pelling food to the hypopharynx. If the food passes normally to the hypopharynx,
the presence of pain, intra- or extraluminal mass lesion, or a neuromuscular disor-
der may interfere with the orderly sequence of pharyngeal contraction, closure of
the epiglottis, UES relaxation, and initiation of peristalsis by contraction of the
striated muscle in the upper esophagus.
1. Signs and symptoms. These patients usually cough and expel the ingested
food through their mouth and nose or aspirate when they attempt to swallow.
Their symptoms are worse with liquids than with solids. They may have a
“wet” voice quality, reduced cough, upper airway congestion, and aspiration
pneumonitis.
2. Causes
a. Central nervous system conditions. Cerebral vascular accidents (bulbar
or pseudobulbar palsy), multiple sclerosis, amyotrophic lateral sclerosis,
Wilson’s disease, Parkinson’s disease, Friedreich’s ataxia, tabes dorsalis, brain-
stem tumors, paraneoplastic disorders, reaction to drugs or toxins, other
congenital and degenerative disorders of the central nervous system.
b. Peripheral nervous system conditions. Poliomyelitis (bulbar), diphtheria,
rabies, botulism, diabetes mellitus, demyelinating diseases, Guillain-Barré
syndrome.
c. Disorders of the myoneural junction. Myasthenia gravis, Eaton-Lambert
syndrome.
d. Muscular disorders. Dermatomyositis, muscular dystrophies, myotonic
disorders, congenital myopathies, metabolic myopathies (thyrotoxicosis,
hypothyroidism, hyperthyroidism, steroid myopathy), collagen vascular dis-
eases, amyloidosis.
e. Toxins. Tetanus, botulism, tic paralysis, arsenic, lead, mercury poisoning.
f. Local structural lesions. Conditions involving the mouth, pharynx, and
hypopharynx.
i. Infection or inflammation. Abscess; tuberculosis; syphilis; viral, bacte-
rial, and fungal infections; Lyme disease; diphtheria; rabies.
ii. Space-occupying lesions. Neoplasms, congenital webs, Plummer-
Vinson syndrome.
iii. Extrinsic compression. Cervical spine spurs, lymphadenopathy, thy-
romegaly, Zenker’s diverticulum.
iv. Trauma. Surgical repair, foreign body ingestion, caustic injury.
g. Motility disorders of the upper esophageal sphincter. Hypertensive
UES, hypotensive UES with esophagopharyngeal regurgitation, abnormal
79466_CH21.QXD 1/2/08 12:15 PM Page 140
IV. TREATMENT
A. Oropharyngeal dysphagia. The treatment of oropharyngeal dysphagia depends
on the specific cause.
1. Systemic disease (see section IV.B.1).
2. Neurologically impaired patients require special attention during feedings with
respect to dietary texture; body, head, and neck position; size and frequency of
food bolus administration; and aspiration precautions. Patients should sit fully
upright in bed or in a chair while eating. The bolus size should be small in sips
or bites. Foods with thicker textures (e.g., thick liquids and pudding textures)
are often better tolerated than clear liquids. Spicy, acidic foods and coffee, tea,
and alcohol should be avoided. After meals, patients should remain in the
upright position for an additional 1 to 3 hours to minimize the risk of aspira-
tion. The head of the bed should be elevated during resting and sleeping hours.
B. Esophageal dysphagia
1. Systemic disease. If the disorder is secondary to a systemic disease, the treat-
ment needs to be directed to the primary disease. Infections and inflammatory
lesions of the esophagus are discussed in Chapter 19.
79466_CH21.QXD 1/2/08 12:15 PM Page 146
Selected Readings
Adler DG, et al. Primary esophageal motility disorders. Mayo Clin Proc. 2001;76:195.
Fibbe C, et al. Esophageal motility in reflux disease before and after fundoplication:
A prospective, randomized, clinical and manometric study. Gastroenterology. 2001;121:5.
Fox M, et al. High-resolution manometry predicts the success of oesophageal bolus
transport and identifies clinically important abnormalities not detected by conventional
manometry. Neurogastroenterol Motil. 2004;16:533–542.
Jayaderan R, et al. Dysphagia in the elderly. Pract Gastroenterol. 2001;25:75.
Lee JI, et al. The effect of silderafil on esophageal motor function in health, subjects and
patients with nutcracker esophagus. Neurogastroenterol Motil. 2003;15:617–623.
Oelschlager BK. Surgical options for treatment of esophageal disorders. Gastroenterol
Hepatol. 2007;3(9):687–689.
Oh TL, et al. Dysphagia in inflammatory myopathy: clinical characteristics, treatment
strategies, and outcome in 62 patients. Mayo Clin Proc. 2007;82(4):441–447.
Rubenstein JH, et al. Dysphagia drives doctors to diagnose a disease: Pitfalls in interpreting
observational studies. Gastrointest Endosc. 2005;61:809–811.
Shay S, et al. Twenty-four-hour ambulatory simultaneous impedance and pH monitoring:
A multi-center report of normal values in 60 health volunteers. Am J Gastroenterol.
2004;99:1037–1043.
Straumann A, et al. Natural history of primary eosinophilic esophagitis: A follow-up of
30 adult patients for up to 11.5 years. Gastroenterology. 2003;125:1660–1669.
Tutuian R, et al. Clarification of the esophageal function defect in patients with manometric
ineffective esophageal motility: Studies using combined impedance-manometry. Clin
Gastroenterol Hepatol. 2004;2:230–236.
Varadarjullu S, et al. The yield and the predictors of esophageal pathology when upper
endoscopy is used for the initial evaluation of dysphagia. Gastrointest Endosc. 2005;
61:804–808.
79466_CH22.QXD 1/2/08 12:16 PM Page 147
T he term noncardiac chest pain generally means pain in the chest that mimics or
may be confused with cardiac chest pain. Often the pain is caused by a disorder of the
esophagus, stomach, or gallbladder. Diagnosis is particularly confusing in patients who
have both cardiac and noncardiac chest pain. Much of the diagnostic confusion arises from
the generous overlap in pain sensations entering the spinal cord from the heart, medi-
astinum, stomach, and other upper abdominal organs.
I. The differential diagnosis of noncardiac chest pain is outlined in Table 22-1. Although
this chapter deals mostly with esophageal causes of chest pain, one cannot neglect the
numerous other causes in evaluating patients with chest pain. Also, it seems an inherent
contradiction to say that cardiac chest pain should be a consideration in the differential
diagnosis of noncardiac chest pain; yet some patients who are referred because they
are thought to have noncardiac chest pain, but have not had an adequate cardiac eval-
uation, eventually are found to have cardiac disease or a combination of cardiac and
noncardiac chest pain.
II. DIAGNOSIS. Many patients with noncardiac chest pain already have had an evalua-
tion for cardiac disease. This evaluation may have consisted of an electrocardiogram
(ECG) only or may have been extensive, including stress testing and coronary arteri-
ography. In any event, the patient’s major concern usually is whether he or she has
heart disease. If the pain can be attributed to a noncardiac cause, the patient often
feels better, even though in some instances little can be done to relieve the pain.
Cardiac causes of chest pain must first be excluded. This cannot always be done
with absolute certainty, and in some patients a diagnosis of both cardiac and noncar-
diac chest pain is made. Furthermore, it appears that, in some of these patients,
noncardiac pain can stimulate cardiac chest pain.
A. Clinical presentation
1. The character of the pain may help differentiate cardiac from noncardiac
pain. Cardiac pain typically is aggravated by stress and exercise and radiates to
the neck, shoulder, and left arm. Chest wall and esophageal pain, however, also
sometimes appear to be aggravated by stress and exercise. The pain of gastroe-
sophageal reflux can radiate to the neck and jaw, but it rarely radiates down
the arm.
2. Accompanying symptoms and relation of the pain to other events can
help differentiate the cause. Dysphagia in association with the pain points to an
esophageal origin (see Chapter 21). If the dysphagia is for both liquids and
solids, an esophageal motility disorder is likely. Pain after eating can be cardiac,
but it is more likely to be of esophageal or perhaps gallbladder origin. Pressure
over the site of pain that aggravates the pain suggests a chest wall source such
as costochondritis or trauma, although chest wall tenderness has been described
in cardiac pain.
B. Diagnostic studies
1. Exclusion of cardiac disease
a. X-ray and ECG. All patients should be evaluated with a chest x-ray and
ECG. Some patients require exercise stress testing using standard ECG
monitoring or thallium scanning.
147
79466_CH22.QXD 1/2/08 12:16 PM Page 148
b. Coronary arteriography, CT, or MRI. Not every patient with chest pain
requires coronary arteriography to exclude cardiac disease for practical clinical
purposes. Some patients clearly have a noncardiac problem, such as gastroe-
sophageal reflux or an esophageal motility disorder. In many patients, however,
coronary disease must be excluded by coronary arteriography, CT, or MRI.
2. Esophageal studies
a. Barium swallow radiography (see Chapter 9). X-ray films of the esopha-
gus usually are not of much help in evaluating chest pain that mimics cardiac
pain. However, obstructing lesions or severe esophagitis (reflux, monilial, or
herpetic) may be evident from the static films. Furthermore, fluoroscopy or
videotape of the swallowing function may suggest that a motility disorder is
present.
b. Endoscopy and esophageal mucosal biopsy. Endoscopy is indicated if
a structural abnormality is seen on the barium study. Biopsy of the
mucosa also may provide information that is not evident by radiography
or endoscopy. Endoscopic punch biopsies or suction biopsies of the
esophagus may show inflammation, thinning of the epithelium, hypertro-
phy of the basal regenerative layer, prolongation of the papillae of lamina
propria that project into the epithelium, or presence of eosinophils,
all of which have been correlated with gastroesophageal reflux (see
Chapter 20).
c. Esophageal motility studies (see Chapter 8).
i. Manometric examination of the esophagus may be crucial in the
diagnosis of an esophageal motility disorder. However, correlation
of manometric findings with clinical symptoms may be difficult. Because
symptoms and manometric abnormalities are usually intermittent, a nor-
mal motility tracing does not exclude a motility disorder. Manometric
abnormalities in the absence of symptoms are also difficult to interpret.
79466_CH22.QXD 1/2/08 12:16 PM Page 149
Nutcracker
NEMD Esophagus
(25%–45%) (25%–45%)
DES
Achalasia (10%–15%)
(5%–10%)
Figure 22-1. Manometric diagnoses of esophageal motility disorders in patients with chest pain
of esophageal origin. NEMD, nonspecific esophageal motility disorder; DES, diffuse esophageal
spasm. (From data compiled from several reports. See Benjamin DS, Castell DO. Esophageal causes
of chest pain. In: Castell DO, Johnson LF, eds. Esophageal Function in Health and Disease. New York:
Elsevier; 1983.)
III. TREATMENT
A. Esophageal motility disorders
1. With the exception of achalasia, all the esophageal motility disorders are
treated in roughly the same manner if they are associated with chest pain.
Specific foods or hot or cold beverages associated with symptoms should be
avoided. Agents to relax the esophageal smooth muscle have been used.
Sublingual nitroglycerin may be sufficient for intermittent symptoms, and long-
acting nitrates have been effective in some patients. The calcium channel block-
ing agents (nifedipine, diltiazem, verapamil) have received much attention, but
double-blind, placebo-controlled studies have not shown clear benefit.
2. Differentiation from angina pectoris. The perceptive reader already has dis-
cerned that the pharmacologic treatment of chest pain caused by an esophageal
motor disorder is similar to the treatment of angina pectoris. Thus, one may
ask whether it makes much difference to distinguish between the two. Although
79466_CH22.QXD 1/2/08 12:16 PM Page 151
the treatment may be similar, the prognosis and the patient’s peace of mind cer-
tainly are different when the diagnosis is coronary artery disease.
3. Esophageal dilatation. One medical treatment of esophageal motor disorders
that does not apply to cardiac disease is esophageal dilatation. Simple passive
dilatation of the esophagus with a 30 French (F) to 40 F Maloney dilator has
been shown to provide temporary, sometimes long-standing, relief in some
patients.
4. Myotomy. An occasional patient with a well-documented motility disorder and
severe pain unresponsive to medical treatment may benefit from a long surgical
myotomy of the esophageal muscle coat.
5. Achalasia. The treatment of achalasia traditionally has been either forceful
dilatation or rupture of the hypercontracting LES with a balloon dilator or sur-
gical myotomy. Recently, long-acting nitrates and calcium channel blocking
agents have been successful in some patients.
B. Gastroesophageal reflux. The treatment of gastroesophageal reflux (GER) is
described in Chapter 20. A small minority of patients with GER also has an
esophageal motility disorder that is stimulated by GER. Treatment of the GER
treats the motility disorder in these patients.
C. Nonmotility esophageal disorders. The treatments of caustic ingestions, esophageal
infections, and esophageal cancer are reviewed in Chapters 16, 19, and 23, respectively.
D. Nonesophageal disorders. Identification of a nonesophageal disorder, such as
gallstones or peptic ulcer, in a patient complaining of chest pain does not neces-
sarily implicate that disorder as a cause of the chest pain. However, it is reasonable
to treat the disorder appropriately. If pain persists after successful treatment, addi-
tional esophageal or cardiac diagnostic studies may be indicated.
Selected Readings
Charbel S, et al. The role of esophageal pH monitoring in symptomatic patients on PPI
therapy. Am J Gastroenterol. 2005;1100:283–289.
Hobson AR, et al. Neurophysiologic assessment of esophageal sensory processing in
noncardiac chest pain. Gastroenterology. 2006;130:80–88.
Jones H, et al. Treatment of noncardiac chest pain; a controlled trial of hypastherapy. Gut.
2006;55:1403–1408.
Liuzzo JP, et al. Chest pain from gastroesophageal reflux disease in patients with coronary
artery disease. Cardiol Rev. 2005;13:167–173.
Rodriguez-Stanley S, et al. Effect of tegaserod on esophageal pain threshold, regurgitation,
and symptom relief in patients with functional heartburn and mechanical sensitivity.
Clin. Gastroenteral Hepatol. 2006;4:442–450.
Schey R, et al. Noncardiac chest pain: current treatment. Gastroenterol Hepatol.
2007;3:255–262.
79466_CH23.QXD 1/2/08 12:17 PM Page 152
23 ESOPHAGEAL CANCER
II. PREDISPOSING FACTORS. Chronic use of alcohol and smoking are associated with
esophageal carcinoma. This may be due to chronic irritation of the esophageal mucosa
with these agents. Other conditions with increased prevalence of esophageal carcinoma
are lye strictures, achalasia, previous exposure to ionizing radiation, head and neck
cancer, Plummer-Vinson syndrome, tylosis, celiac sprue, and Barrett’s epithelium.
Squamous cell carcinoma accounts for less than one half of esophageal carcino-
mas. Adenocarcinomas of the esophagus, which used to account for less than 10% of
esophageal cancers, now account for greater than two thirds of all esophageal malig-
nancies in the United States. Adenocarcinoma usually arises from metaplastic columnar
epithelium (Barrett’s epithelium) and rarely from esophageal glands. Adenocarcinoma
of the stomach may spread to the esophagus by extension.
Gastroesophageal reflux disease (GERD) is thought to be the major risk factor
for esophageal adenocarcinoma. Recurrent symptoms of GERD seem to increase the
risk of esophageal adenocarcinoma by eightfold. The annual incidence of cancer in
Barrett’s epithelium is approximately 0.8%. Anticholinergic calcium channel block-
ers, nitrates, and theophyllines, by decreasing the lower esophageal sphincter tone, are
thought to increase the risk of adenocarcinomas of the esophagus. In addition, obe-
sity, which increases intraabdominal pressure and GERD, is an added risk factor.
152
79466_CH23.QXD 1/2/08 12:17 PM Page 153
III. PROGNOSIS. The 5-year survival rate for all patients is approximately 16%. The
5-year survival even with the earliest stages of cancer is only 50% to 80% and with
lymph node involvement it drops to below 25%. With locally advanced esophageal
cancer the survival drops to 5% to 10% (with radiation or surgery alone) and with
chemoradiation and surgery, it may be 25% to 27%.
Esophageal cancers grow extensively locally and invade adjacent structures.
The tumor has the propensity to spread longitudinally via lymphatic channels within the
esophageal wall to mediastinal cervical and celiac lymph nodes. There may also
be hematogenous spread to lungs, liver, and other organs. Esophagobronchial or
esophagopleural fistulas may form and manifest as recurrent pneumonia or abscess.
Erosion into the aorta may result in exsanguination.
IV. CLINICAL FEATURES. Progressive dysphagia for less than a year, first with solids
then with semisolids and liquids, is the most common symptom. Substernal
pain, usually steady, radiating to the back may also be present and may suggest
periesophageal spread of the tumor. Most patients complain of anorexia and pro-
found weight loss. Patients may have iron-deficiency anemia from blood loss from
the lesion, but brisk bleeding is rare. Hoarseness may result from involvement of the
recurrent laryngeal nerve. If the lesion is obstructive, patients may aspirate
esophageal contents and may present with aspiration pneumonia and pleural
effusion. Horner’s syndrome, cervical adenopathy, hpeatoroegalys, boney pain, and
paraneoplastic syndromes including hypercalcemia, inappropriate ACTH, and
gonadotropins may be present.
V. DIAGNOSIS
A. A barium swallow is usually the first noninvasive test ordered to establish the
diagnosis of esophageal carcinoma. A double-contrast study may be helpful in
identifying small, plaquelike lesions. The usual finding is an irregular luminal nar-
rowing. There may be a ridge or shelf at the superior portion of the tumor.
However, the differentiation of the tumor from a benign peptic stricture can be
extremely difficult by radiography.
B. Endoscopy. Fiberoptic endoscopy allows direct localization and inspection of the
lesion. Retroflexion of the endoscope in the fundus of the stomach allows visual-
ization of lesions at the esophagogastric junction and cardia, which could not be
done by rigid scopes. Direct biopsies and brushings provide tissue for histologic
and cytologic examination.
C. Computed tomography (CT) is very helpful in determining the extent and spread
of extramucosal tumor.
D. Endoscopic ultrasonography (EUS). EUS, with its unique ability to define the
anatomy of the gut wall in detail, offers the most accurate method for evaluating
the depth of esophageal cancer invasion and detecting abnormal regional lymph
nodes. Because esophageal carcinoma originates in the mucosa and progressively
invades deeper layers of the esophageal wall, the TNM classification recommended
by the International Union Against Cancer and the American Joint Committee on
Cancer lends itself to accurate staging with EUS. T indicates depth of primary
tumor invasion, N indicates spread of cancer to regional lymph nodes, and M indi-
cates distant metastases.
E. Magnetic resonance imaging (MRI) is a technique in longitudinal and cross-
sectional body imaging. Currently, it offers no advantage over CT in detecting
infiltrating growth.
F. PET scan may improve the detection of stage IV disease.
G. Preoperative thoracoscopy and laparoscopy may help assess extent of local dis-
ease as well as the involvement of regional lymph nodes and celiac and perigastric
nodes.
H. Staging and prognosis. Pathologic stage is the most important factor in progno-
sis. Recurrence and survival are strongly related to depth of tumor invasion, metas-
tases to adjacent lymph nodes and distant organs. TNM system has been revised
and used in prognosis. T1-2, NOMO are potentially curable with surgery alone.
79466_CH23.QXD 1/2/08 12:17 PM Page 154
Tumor invasion into serosa (T3) or regional or distant lymph node metastases (T4)
are associated with significant reduction in survival.
VI. TREATMENT. The therapy for carcinoma of the esophagus is determined by the stage
of the disease. The mainstay of therapy has been surgery, with or without radiother-
apy and chemotherapy. Inoperable tumors have been treated by radiotherapy because
squamous carcinoma of the esophagus is relatively radiosensitive. Chemotherapy
alone has not been very successful. Treatment protocols combining chemotherapy and
radiation therapy before and after surgery offer somewhat better results than single-
modality therapy. Most successful chemotherapeutic agents are the combination of
cisplatin, 5-fluorouracil, paclitaxel, irinotecan hydrochloride, vinorelbine tartrate,
and gemcitabine hydrochloride preoperatively and as neoadjuvant chemotherapy.
A. Surgery. Total esophagectomy is the surgical procedure of choice. Patients with
lesions of the lower third that are less than 5 cm seem to do better. Because the
cure rate has been so dismally low with “curative” surgery, palliative resection has
been used to alleviate symptoms and allow patients to swallow.
The morbidity and mortality with a large thoracotomy are still very high in
these patients. The esophageal resection and esophagogastric anastomosis may be
done with a combined abdominal incision and a right thoracotomy. If the lesion is
low enough, an abdominal approach may suffice. For most lesions, it is preferred
to resect the involved portion of the esophagus with wide margins, bring the stom-
ach into the chest, and create an anastomosis with the remaining esophagus.
Colonic or jejunal segment interposition carries a high complication rate. For pur-
poses of palliation, the stomach may be anastomosed to the esophagus in a side-
to-side fashion to bypass the obstructed area.
B. Radiation therapy
1. Radiation therapy for squamous cell carcinoma of the esophagus has been
used for attempted cure with unsatisfactory results. It is used in protocols before
or after surgery and for palliation.
2. Presurgical radiation therapy alone has not been found to increase the cure
rate.
3. Adenocarcinomas are resistant to radiation therapy.
C. Other palliative measures
1. Mechanical dilatation. When surgery and radiation therapy are contraindi-
cated, or when these treatments have failed, mechanical dilatation of the
esophageal lumen may be attempted with Savary or balloon dilators under endo-
scopic guidance. Because the risk of esophageal perforation is high in these
patients, the dilatation should be done slowly and with great care.
2. Tube placement. If it becomes difficult to maintain a lumen at the area of the
tumor, a stent (plastic or metal) may be placed in the lumen endoscopically. These
tubes may also be used to close off, at least temporarily, a tracheoesophageal fis-
tula. These tubes may erode into the esophageal wall, causing ulceration, bleed-
ing, and perforation.
3. YAG-laser therapy has been found to be quite effective in palliation of patients
with advanced obstructing esophageal tumors. These masses may be “pared
down” by the laser to open the lumen. Lasers may also be useful in the treat-
ment of very early lesions. Additional controlled studies need to be done using
this technique.
4. Injection necrosis of fungating esophageal cancer can be accomplished with
intratumoral injection of absolute alcohol or ethylene glycol with endoscopic
visualization. In instances of luminal narrowing, malignant strictures may be
dilated first, and then injected concentrically to additionally “open up” the
lumen.
5. These palliative measures are usually repeated as needed with progressive,
repeated growth of the tumor.
D. Selection of therapy. Presently, the optimal therapy for esophageal cancer is
unclear. A strong case can be made for establishing comparable diagnostic staging
criteria and treating newly diagnosed patients according to well-designed, established
79466_CH23.QXD 1/2/08 12:17 PM Page 155
research protocols whenever possible. If the use of research protocols is not feasi-
ble, a reasonable approach is to use resectional surgery for “resectable” tumors in
the distal third of the esophagus (T1–3 N1) with pre- and neoadjuvant chemoradi-
ation therapy. For patients who are poor candidates for surgery or who have
obstructing tumors, one or more of the listed palliative measures may be used.
E. Prevention and surveillance. Patients with recurrent symptoms of GERD are rec-
ommended to have endoscopy and biopsies for the diagnosis of Barrett’s esophagus.
Endoscopic surveillance will allow the diagnosis of earlier stage tumors and allow
for improved survival. Patients should be advised to stop smoking and use alcohol
in moderation. Current recommendations for patients with Barrett’s esophagus are
still debated, but most centers offer biannual endoscopic surveillance and annually
if there is low-grade dysplasia. Patients with high-grade dysplasia are recommended
to substantiate the diagnosis by two different pathologists, and then consider under-
going esophagectomy or mucosal ablation with photodynamic therapy.
Selected Readings
Bowrey DJ, et al. Use of alarm symptoms to select dyspeptics for endoscopy causes patients
with curable esophagogastric cancer to be overlooked. Surg. Endosc. 2006;20:I725–8.
Dar M, et al. Can extent of high-grade dysplasia predict the presence of adenocarcinoma
at esophagectomy? Gut. 2003;52:486–489.
Fountoulakis A, et al. Effect of surveillance of Barrett’s esophagus on the clinical outcome
of oesophageal cancer. Br J Surg. 2004;91:997–1003.
Lagergren J, et al. Association between medications that relax the lower esophageal
sphincter and risk for esophageal adenocarcinoma. Ann Int Med. 2000;133:165.
Lagergren J, et al. Symptomatic gastroesophageal reflux as a risk factor for esophageal
adenocarcinoma. N Engl J Med. 1999;340:825–831.
Largi A, et al. EUS followed by EMR for staging of high grade dysplasia and early cancer
in Barrett’s esophagus. Gastrointest Endosc. 2005;62:16.
May A, et al. Accuracy of staging in early esophageal cancer using high resolution
endoscopy and high resolution endosonography: A comparative, prospective trial. Gut.
2004;53:634–640.
Papachritou GI, et al. Use of stents in benign and malignant esophageal disease. Rev.
Gastroenterol. Disord. 2007;7(2):75–88.
Pech O, et al. Long-term results of photodynamic therapy with 5-aminoevulinic acid for
superficial Barrett’s cancer and high-grade intraepithelial neoplasia. Gastrointest
Endosc. 2005;62:24–30.
Playford RJ. New British Society of Gastroenterology (BSG) guidelines for the diagnosis
and management of Barrett’s esophagus. Gut. 2006;55:442–443.
Portale G, et al. Comparison of the clinical and histological characteristics and survival of
the distal esophageal–gastroesophageal junction adenocarcinoma in patients with and
without Barrett’s mucosa. Arch Surg. 2005;140:570–575.
Ross WA, et al. Evolving role of self-expanding metal stents in the treatment of malignant
dysphagia and fistulas. Gastrointest Endosc. 2007;65:70–76.
Sharma P. Low-grade dysplasia in Barrett’s esophagus. Gastroenterology. 2004;127:
1233–1238.
Wang KK, et al. American Gastroenterological Association medical position statement: Role
of the gastroenterologist in the management of esophageal carcinoma. Gastroenterology.
2005;128:1468–1470.
Wong A, et al. Epidemiologic risk factors for Barrett’s esophagus and associated
adenocarcinoma. Clin Gastroenterol Hepatol. 2005;3:1–10.
79466_CH24.QXD 1/2/08 12:18 PM Page 156
T he term peptic ulcer disease (PUD) refers to disorders of the upper gastrointestinal
tract caused by the action of acid and pepsin. These agents not only cause injury them-
selves, but also typically augment the injury initiated by other agents. The spectrum of
peptic ulcer disease is broad, including undetectable mucosal injury, erythema, erosions,
and frank ulceration. The correlation of severity of symptoms to objective evidence of dis-
ease is poor. Some patients with pain suggesting peptic ulcer disease have no diagnostic
evidence of mucosal injury, whereas some patients with large ulcers are asymptomatic.
156
79466_CH24.QXD 1/2/08 12:18 PM Page 157
Figure 24-1. Diagram of factors that promote mucosal injury (Offense) versus those that protect
the mucosa (Defense). On the offensive side, hydrochloric acid (H+) is essential for the action of
pepsin and many ulcerogenic factors. For example, aspirin, bile acids, and the nonsteroidal antiin-
flammatory drugs cause much more mucosal injury in an acid milieu. The roles of corticosteroids,
smoking, and stress are less clear, but these factors probably contribute to mucosal injury. Alcohol
can cause mucosal injury without the assistance of acid. The defense of the mucosa is a complex
phenomenon that involves the interaction of a number of protective mechanisms indicated in the
figure. (See Table 24-1 and text.)
Concepts Mediators
resists the passage of large molecules, such as pepsin; however, H+, other
small particles, and ethanol seem to have little difficulty in penetrating mucus
to reach the mucosal surface.
b. Bicarbonate is produced in small amounts by surface epithelial cells and
diffuses up from the mucosa to accumulate beneath the mucous layer, creat-
ing a thin (several micrometers) layer of alkalinity between the mucus and
the epithelial surface.
c. The hydrophobic layer of phospholipid that coats the luminal membrane
of surface epithelial cells is believed to help prevent the back diffusion of
hydrophilic agents such as hydrochloric acid.
d. Mucosal blood flow is important not only in maintaining oxygenation and
a supply of nutrients, but also as a means of disposing of absorbed acid and
noxious agents.
e. The alkaline tide refers to the mild alkalinization of the blood and mucosa
that result from the secretion of a molecule of bicarbonate (HCO3) by the
parietal cell into the adjacent mucosa for every H+ ion that is secreted into
the gastric lumen. This slight alkalinity may contribute to the neutralization
of acid that diffuses back into the mucosa and may augment the effects of
mucosal blood flow.
f. Epithelial renewal, which involves the proliferation of new cells and sub-
sequent differentiation and migration to replace old cells, but which requires
several days, is necessary in the healing of deeper lesions, such as erosions
and ulcers.
g. Restitution refers to the phenomenon of rapid migration, within minutes, of
cells deep within the mucosa to cover a denuded surface epithelium. This process
probably accounts for the rapid healing of small areas of superficial injury.
h. Prostaglandins, particularly of the E and I series, are synthesized abun-
dantly in the mucosa of the stomach and duodenum. They are known to
stimulate secretion of both mucus and bicarbonate and to maintain mucosal
blood flow. Prostaglandins also may have beneficial protective effects
directly on epithelial cells.
i. Epidermal growth factor (EGF) is secreted in saliva and by the duodenal
mucosa and may exert topical protective effects on the gastroduodenal mucosa.
C. Relation of Helicobacter pylori to peptic disease. Since 1983, evidence has
accumulated to implicate a bacterium, H. pylori, in the pathogenesis of some forms
of peptic disease. The organism is found adherent to the gastric mucosal surface
and, when looked for, has been identified in more than 90% of patients with duo-
denal ulcer. Its presence has also been correlated with gastritis, gastric ulcer, and
gastric erosions and with chronic infection, gastric mucosal atrophy, intestinal
metaplasia, and in a minority of patients, gastric adenocarcinoma and MALT
lymphoma. Some patients with H. pylori are asymptomatic and may even have
79466_CH24.QXD 1/2/08 12:18 PM Page 159
B. Perforation. Duodenal or gastric ulcers may perforate into the peritoneal cavity. In
10% of patients, perforation is accompanied by hemorrhage. Initially, patients feel an
abrupt onset of intense abdominal pain which is then followed by hypotension and
shock as the peritoneal cavity is flooded by gastric juice and contents and peritonitis
develops. Mortality is imminent if appropriate therapy is not initiated immediately.
The risk of perforation is increased in patients with PUD who smoke, use
NSAIDs, and who are elderly. Use of crack cocaine has been associated with per-
forated ulcers in the younger patients, most likely as a result of cocaine-induced
vasoconstriction and ischemia of the gastric or duodenal wall.
Diagnosis is clinical; however, free air is noted within the abdominal cavity
on plain and upright x-rays of the abdomen in approximately 75% of patients. CT
scan offers confirmatory information. Endoscopy should be avoided. Patients
should receive intravenous broad spectrum antibiotics and immediate surgery is
recommended to close the perforation and irrigate the peritoneal cavity.
C. Penetration occurs when PUD burrows through the gut wall into an adjacent organ
(i.e., pancreas, liver, or colon). Occasionally fistulas may develop. Pancreatitis, hem-
orrhage, or peritonitis may develop. Diagnosis is usually by CT scan. Treatment should
be tailored to pathology and may include surgery.
D. Obstruction. PUD of the antrum, pyloric channel, and duodenum may result in
gastric outlet obstruction. Patients may present with early satiety, nausea, vomiting,
epigastric pain, and bloating. Dehydration and electrolyte abnormalities may occur.
Diagnosis may be made by plain x-rays, CT scan, and more definitely by UGI
endoscopy. The narrowed segment may be dilated endoscopically. Some cases
require surgery.
III. DIAGNOSIS
A. Clinical presentation
1. History. The classic symptoms of peptic ulcer disease—epigastric burning pain
on an empty stomach that is relieved by food or antacids—are familiar to most
physicians and laypeople. Sometimes the pain radiates to the back, suggesting
an ulcer of the posterior aspect of the duodenal bulb that may have penetrated
into the pancreas. However, many patients with peptic disease experience non-
specific abdominal discomfort, which broadens the differential diagnostic con-
siderations to include gastroesophageal reflux disease (GERD); gallbladder
disease; pancreatic disorders; cancer of the stomach, pancreas, or biliary system;
mesenteric vascular insufficiency; and irritable bowel syndrome. A minority of
patients has gastrointestinal bleeding, weight loss, or vomiting; the vomiting
may be caused by partial or complete gastric outlet obstruction.
Over the past three decades, the yearly incidence of discrete peptic ulcer
disease appears to have decreased, but peptic disease of other types, such as
gastritis and duodenitis, sometimes related to ingestion of aspirin, nonsteroidal
antiinflammatory drugs, corticosteroids, or ethanol have increased in incidence.
Thus, important historical information includes a record of drug and alcohol
use and a history of smoking, and previous diagnosis of peptic disease.
2. Physical examination. The physical examination of a patient with peptic dis-
ease may be normal. Some patients have upper abdominal tenderness and
guarding. Rigidity of the abdomen and absent bowel sounds suggests perfora-
tion. Stool should be tested for occult blood.
B. Diagnostic studies
1. Most patients with dyspepsia or uncomplicated peptic disease may initially
require no diagnostic study. It may be sufficient to begin empiric treatment H2
blockers or PPIs to control acid secretion. Patients also should be advised
regarding diet, smoking, and lifestyle, as described in section IV. If they do not
respond to treatment within a reasonable time, usually 2 to 4 weeks, endoscopy
should be considered, during which biopsies can be obtained to test for
H. pylori and, in the case of a gastric ulcer, to evaluate for a cancerous lesion.
2. Patients with potential complications. A minority of patients present initially
with signs or symptoms that should act as “red flags” to alert the physician
to the increased possibility of complications of peptic disease or cancer.
79466_CH24.QXD 1/2/08 12:18 PM Page 161
C. Drug therapy
1. Agents that control acid
a. Antacids neutralize gastric acid within the lumen of the esophagus, stom-
ach, and duodenum. They have been used for many years and are effective in
relieving peptic symptoms.
i. Composition. The magnesium-containing antacids are more potent than
those that do not contain magnesium, but they tend to cause diarrhea.
Magnesium-free antacids tend to cause constipation. Sometimes the two
types are alternated to provide adequate acid control and regulate tolerable
bowel habits. Calcium-containing antacids offer the initial acid neutraliza-
tion and stimulate gastric acid secretion. Another consideration is the
sodium content of an antacid when prescribing for a patient who is on a
sodium-restricted diet.
ii. Dosage. Most patients take antacids for amelioration of symptoms only.
To treat peptic conditions with a full antacid regimen, about 30 mL
should be taken 1 and 3 hours after a meal and at bedtime. This regimen
requires a daily antacid intake of more than 200 mL, resulting in a
monthly cost in excess of $50. Because few patients are likely to adhere
to such a program and because of the expense, other medications,
described in the following sections, are preferred.
b. H2 antagonists. Histamine is a potent stimulant of gastric acid secretion. For
this reason, agents that block the histamine receptor on the parietal cell, so-
called H2 antagonists, are effective inhibitors of acid secretion. Interestingly,
the H2 antagonists inhibit not only histamine-stimulated acid secretion, but
also acid that is stimulated by the vagus nerve (acetylcholine) and by gastrin.
This phenomenon suggests an interrelation among these three receptors. The
commercially available H2 antagonists are cimetidine (Tagamet), ranitidine
(Zantac), famotidine (Pepcid), and nizatidine (Axid).
i. Dosage. Typical full therapeutic doses of these agents are cimetidine,
300 mg q.i.d. or 400 mg b.i.d.; ranitidine, 150 mg b.i.d.; famotidine,
20 mg b.i.d. or 40 mg once before bed; and nizatidine, 300 mg once
before bed. Alternate regimens also have been used, such as cimetidine,
800 mg at bedtime; and ranitidine, 300 mg at bedtime.
ii. Side effects. The H2 antagonists are remarkably safe, although side effects
have been reported. These reports include rare instances of leukopenia and
the occasional occurrence of elevated liver enzymes, skin rash, constipation,
and diarrhea. Long-term use of cimetidine has been associated with gyneco-
mastia. Ranitidine appears to offer some advantage over cimetidine in that
it appears to cause less feminization in men and fewer central nervous sys-
tem effects, and it has lesser affinity for the cytochrome-P450 system in the
liver than does cimetidine, resulting in fewer drug interactions. Both cimeti-
dine and ranitidine interfere with gastric mucosal alcohol dehydrogenase
and, if taken before ingestion of alcohol, may increase serum alcohol levels.
This effect on alcohol dehydrogenase does not occur with famotidine.
Famotidine appears to offer many of the advantages of ranitidine and is
slightly longer acting. Nizatidine has a similar profile to ranitidine.
c. Prostaglandins. Prostaglandins are attractive on theoretical grounds
because they affect both sides of the mucosal injury-protection balance. They
inhibit acid secretion and exert a cytoprotective effect on the mucosa.
However, the dose required to elicit the cytoprotective effect is much lower
than the acid-inhibitory dose. When prostaglandins are administered in the
lower, cytoprotective dose, there is little evidence that they have any clinical
benefit.
Prostaglandin analogs (e.g., misoprostol [Cytotec]) have been used to
treat peptic conditions in clinical trials and appear to be as effective as H2
antagonists in healing ulcers when administered in doses that inhibit acid
secretion. However, there is little evidence that synthetic prostaglandins are
clinically effective when given in the lower, purely cytoprotective doses that
79466_CH24.QXD 1/2/08 12:18 PM Page 164
except that the antrum has been removed. In a Billroth II procedure, the food
bypasses a large portion of the proximal intestine, resulting in suboptimal
stimulation of bile and pancreatic secretion and poor mixing of food with these
secretions (see Chapter 28). Whether a Billroth I or Billroth II is performed
depends largely on technical considerations at the time of surgery.
E. Treatment of hypersecretory states
1. Gastrinoma. Current management of gastrinoma, or ZES, includes treatment with
H2 antagonists or PPIs, such as omeprazole, often in high doses. For example, rani-
tidine 1,200 mg per day in divided doses is a typical regimen, but some patients may
require up to 6 g per day. Omeprazole 60 to 80 mg a day, or any other PPI, effec-
tively inhibits acid secretion in most patients who have ZES. Some evidence indicates
that highly selective vagotomy lessens the requirement for H2 antagonists in many of
these patients, but the use of PPI has decreased the need for this operation.
2. Antral G-cell hyperplasia. Patients with antral G-cell hyperplasia have
increased numbers of antral G cells, which results in excessive gastrin secretion.
Whether antral G-cell hyperplasia is a distinct entity or is merely the extreme
end of a range of quantitatively different G-cell populations remains conjec-
tural. The secretin test (see section III.B.5) does not result in a rise in serum gas-
trin, as it does in ZES. Most patients respond well to H2 antagonists. Because
antrectomy removes the source of gastrin in these patients, the operation should
be considered in patients with refractory illness.
3. Retained antrum syndrome. Retained antrum syndrome develops only in
patients who have had an antrectomy and gastrojejunostomy (Billroth II). At
the time of surgery, deformity of the antroduodenal area may make identifica-
tion of the demarcation between antral and duodenal mucosa difficult.
Inadvertently, a small portion of antral mucosa may remain behind and
become incorporated into the duodenal stump after the antrum has been
resected and the proximal duodenal stump has been oversewn. The G cells of
the retained antrum are bathed constantly in an alkaline milieu, which pro-
motes constant, uninhibited secretion of gastrin. Serum gastrin levels may rise
to the range of the ZES. However, the secretin tests (see section III.B.5) should
not produce a rise in serum gastrin as it does in ZES. Patients may respond to
H2 antagonists or omeprazole, but the definitive treatment is reoperation to
remove the small portion of retained antrum.
Selected Readings
Barrison AF, et al. Patterns of proton pump inhibitors in clinical practice. Am J Med. 2001;
111:469–473.
Chan FKL, et al. Celecoxib versus diclofenac and omeprazole in reducing the recurrent ulcer bleeding
in patients with arthritis. N Engl J Med. 2002;347:2104–2110.
Johnson JH. Endoscopic risk factors for bleeding peptic ulcers. Gastrointest Endosc. 2002;56:1–6.
Kahi CJ, et al. Endoscopic therapy versus medical therapy for bleeding peptic ulcers with adherent
clot—a meta analysis. Gastroenterology. 2005;129:855–863.
Law JYW, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment
of bleeding peptic ulcers. New Engl J Med. 2000;343:310–316.
Leontiadis G, et al. Proton pump inhibitor therapy for peptic ulcer bleeding: Cochrane collaboration
meta-analysis of randomized controlled trials. Mayo Clin Proc. 2007; 82(3):286–296.
Leontiadis GI, et al. Systematic review and meta-analysis of proton pump inhibitor therapy in peptic
ulcer bleeding. BMJ. 2005;330:568–570.
Rostom A, et al. Gastrointestinal safety of cyclooxygenase-2 inhibitors: A Cochrane Collaboration
systematic review. Clin. Gastroenterol Hepatol. 2007;7(5):818.
Sridhar S, et al. Hydrogen peroxide (H2O2), H. pylori, NSAIDS, and clot stability: Do they matter?
American College of Gastroenterology Annual Mtg P-37, 2005.
Talley NJ, et al. Practice Parameters Committee of the American College of Gastroenterology.
Guidelines for the management of dyspepsia. Am J Gastroenteral. 2005;I00:2324–37.
Zargar SA, et al. Pantoprazole infusion as adjuvant therapy to endoscopic treatment in patients
with peptic ulcer bleeding: prospective randomized controlled trial. J Gasteroenterol Hepatol. 2006;
21:716–721.
79466_CH25.qxd 1/2/08 12:19 PM Page 167
STRESS ULCER 25
S tress ulcer refers to an ulcer or, more often, multiple ulcers that develop during the
severe physiologic stress of serious illness.
III. Treatment of stress ulceration usually begins with prevention. Careful attention to
respiratory status, acid–base balance, and treatment of other illnesses helps prevent
the conditions under which stress ulcers occur.
Patients who develop stress ulcers typically do not secrete large quantities of gas-
tric acid; however, acid does appear to be involved in the pathogenesis of the lesions.
Thus, it is reasonable either to neutralize acid or to inhibit its secretion in patients at
high risk. In patients admitted to surgical ICUs, hourly antacid titration to keep the
intragastric pH above 4 has been shown to reduce markedly the frequency of acute
bleeding. Because gastric-acid suppression with PPIs is much more profound, allowing
the gastric pH to rise to 5 or 6, PPI infusion IV is more effective in the prevention of
stress ulcers and stress ulcer bleeding.
Not all patients in an ICU require acid suppressive prophylaxis. Patients who
should be considered for a prophylactic regimen include those who are on respirators
and have multisystem disorders and those with a history of peptic ulcer or upper gas-
trointestinal bleeding. Once bleeding has occurred, however, intravenous acid
suppressive therapy with a PPI to keep intragastric pH above 5 remains the most
appropriate therapy (see Chapter 14).
167
79466_CH25.qxd 1/2/08 12:19 PM Page 168
Selected Readings
Chak A, et al. Effectiveness of endoscopy in patients admitted to the intensive care unit
with upper G.I. hemorrhage. Gastrointest Endosc. 2001;53:6.
Dore MP, et al. Ulcers and gastritis. Endoscopy. 2004;36:42–47.
James YW, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic
treatment of bleeding peptic ulcers. N Engl J Med. 2000;343:310–316.
Klebl FH, et al. Therapy insight: prophylaxis of stress-induced gastrointestinal bleeding in
critically ill patients. Nat Clin Pract Gastroenterol Hepatol. 2007;4(10):562–570.
Martindale RG, et al. Contemporary strategies for the prevention of stress-related mucosal
bleeding. Am J Hosp Pract. 2005;62(suppl):S511–S517.
Wolfe MM, et al. Acid suppression: Optimizing therapy for gastroduodenal ulcer healing,
gastroesophageal reflux disease and stress-related erosive syndrome. Gastroenterology.
2000;118(suppl 2):S9.
79466_CH26.qxd 1/2/08 12:20 PM Page 169
HELICOBACTER PYLORI 26
III. INFECTION. Acute infection has been demonstrated dramatically by Warren and
Marshall, who voluntarily ingested a culture of HP and developed acute HP disease.
It is not known how often infection with HP spontaneously clears. In most cases,
infection is lifelong. Most infected persons develop chronic active gastritis. Gastritis
may be confined mostly to the gastric antrum and/or involve the entire stomach. In
169
79466_CH26.qxd 1/2/08 12:20 PM Page 170
persons with predominately antral involvement, gastric acid secretion may be normal
or increased. In such persons, duodenal ulcer and esophagitis secondary to GER may
occur. In persons with diffuse HP infection of the entire stomach, gastric atrophy may
develop and acid secretion may diminish; in fact, these persons may develop hypo-
and achlordria and even vitamin B12 deficiency. In advanced cases, intestinal metapla-
sia and gastric adenocarcinoma may develop. Once patients develop severe atrophy,
intestinal HP metaplasia or gastric adenocarcinoma may not be present in these tis-
sues where there is no acid secretion.
Interestingly, in persons treated with proton pump inhibitors (PPIs) chronically,
there is usually a shift of HP infection from the antrum to the corpus and accelerated
atrophic changes. Thus some experts advise eradication of HP before long-term PPI
therapy is instituted in HP-infected individuals.
In some patients, HP infection arouses a robust induction of lymphocytic infil-
tration of gastric mucosa with lymphocytic follicle formation and in some cases MALT
lymphoma. These have been seen to regress after appropriate HP eradication therapy.
However, B-cell lymphomas that develop in association with HP infection may not
regress after HP eradication and may need oncologic intervention.
IV. DIAGNOSIS
A. Endoscopy and biopsy is the gold standard for the diagnosis of HP. Biopsies
should be obtained from all parts of the stomach since the colonization of HP is
spotty and is affected by the mode of infection, patients’ medications, and whether
patients are on acid-suppressive therapy. Histology not only confirms the presence
of HP, but also gives information on the presence or absence of gastritis, gastric
atrophy, intestinal metaplasia, MALT lymphoma, and cancer. Sensitivity ranges
from 90% to 95% and specificity 95% to 100%. False negative results occur in the
sections of recent GI bleeding, use of Bismuth-containing medications, antibiotics,
sucralfate, and acid-suppressive medications.
CLO test, PyloriTek test uses a biopsy in a medium containing urea which
undergoes color change if HP is present with urease activity.
HP may be cultured from gastric biopsies. This should be considered in
patients who have failed two courses of appropriate antibiotic treatment regimens
and are thought to have resistant HP.
B. Breath tests. Breath testing in patients infected with HP utilizes their ingesting
C14 (radioactive) or C13 containing urea, then blowing into a container to measure
the tagged CO2 released by HP degradation of urea (CONH3) by its urease which
releases CO2 and NH3. Sensitivity of breath testing ranges from 85% to 95% and
is operator-dependent. False negative results can occur in patients with history of
recent antibiotic, bismuth, sucralfate and antisecretionary therapy and in those
with advanced histologic changes such as atrophy and intestinal metaplasia.
C. Stool antigen test. A monoclonal and polyclonal antibody test is available to test
for HP antigen in stool samples. This has a similar sensitivity and specificity as
compared to breath tests. False negative tests are seen as in breath tests as well as
in patients with a history of recent GI bleeding.
Both the breath test and stool antigen test should not be done for at least
4 weeks after HP eradication therapy.
D. Serology. HP antibody may be detected in persons who have been infected with
HP. However, in many persons the IgG HP antibody persists in the plasma even
after proper eradication (at least for 18 months) and thus it is not a good measure
of current infection.
Salivary and urinary assay have similar sensitivity and specificity to serology.
E. When and whom to test. HP testing is recommended for patients who are to be
treated for the eradication of HP infection. In patients with history of PUD and
dyspepsia, if endoscopy is indicated, biopsies should be obtained from multiple
sites. In patients with gastric ulcer, biopsies should also be obtained from the ulcer
to check for cancer.
After eradication therapy, testing for presence of HP is not routinely
practiced. In patients with PUD-related GI hemorrhage or those with recurrent
79466_CH26.qxd 1/2/08 12:20 PM Page 171
dyspeptic, symptoms may be tested for HP after at least 4 weeks have elapsed from
eradication, antibiotic, and acid antisecretory therapy. Stool antigen test, breath
test and, in indicated cases, endoscopy and biopsy may be used.
Selected Readings
Delchier JL. Gastric MALT lymphoma, a malignancy potentially curable by eradication of
Helicobacter pylori. Gastroenterol Clin Biol. 2003;27:453–458.
Duck WM, et al. Antimicrobial resistance incidence and risk factors among Helicobacter
pylori–infected persons, United States. Emerg Infect Dis. 2004;10:1088–1094.
Gisbert JP, et al. Helicobacter pylori “rescue” therapy after failure of two eradication
treatments. Helicobacter. 2005;10:363–372.
Gisbert JP, et al. Review article: C-urea breath test in the diagnosis of Helicobacter pylori
infection—a critical review. Aliment Pharmacol Ther. 2004;15:1001–1017.
Gisbert JP, et al. Stool antigen test for the diagnosis of Helicobacter pylori infection:
A systematic review. Helicobacter. 2004;9:347–368.
Janssen MJ, et al. Meta-analysis: The influence of pre-treatment with a proton pump inhibitor
on Helicobacter pylori eradication. Aliment Pharmacol Ther. 2005;21:341–345.
Jarbol D, et al. Proton pump inhibitor or testing for Helicobacter pylori as the first step for
patients presenting with dyspepsia? A cluster-randomized trial. Am J Gastroenterol.
2006;I0I:1200–8.
Laey C, et al. Helicobacter infection and gastric cancer. Gastroenterology. 2001;120:324.
Laine L, et al. Bismuth-based quadruple therapy using a single capsule of bismuth biskalcitrate,
metronidazole and tetracycline given with omeprazole versus omeprazole, amoxicillin, and
clarithromycin for eradication of Helicobacter pylori in duodenal ulcer patients:
A prospective, randomized, multicenter North American trial. Am J Gastroenterol. 2003;
98:562–567.
Ohata H, et al. Progression of chronic atrophic gastritis associated with Helicobacter
pylori infection increases risk of gastric cancer. Int J Cancer. 2004;109:138–143.
Qasim, A, et al. Rifabutin- and furazolidone-based Helicobacter pylori eradication
therapies after failure of standard first- and second-line eradication attempts in
dyspepsia patients. Aliment Pharmacol Ther. 2005;21:91–96.
Saad RJ, et al. Levofloxacin-based triple therapy versus bismuth-based quadruple therapy
for persistent Helicobacter pylori infection: A meta-analysis. Am J Gastroenterol.
2006;101:488–496.
79466_CH26.qxd 1/2/08 12:20 PM Page 172
GASTRIC NEOPLASMS 27
M ost neoplasms of the stomach are malignant, and most of those are adenocarci-
noma. A minority are lymphoma, leiomyosarcoma, and liposarcoma. Benign neoplasms
include adenomatous, hyperplastic, and hamartomatous polyps; leiomyomas; and lipomas.
Rarely, gastrinomas, carcinoids, vascular tumors, fibromas, and squamous cell carcinomas
occur in the stomach.
Adenocarcinoma of the stomach is the second most frequent cause of cancer death
worldwide. In East Asia, Latin America, and the former Soviet Union, the incidence is up
to eight times higher than in the United States and it remains high among immigrants from
those areas. Although screening programs in high-risk areas such as Japan may result in
the detection of early lesions, in lower risk areas such as the United States and Western
Europe, most cancers are relatively advanced by the time of diagnosis.
I. GASTRIC ADENOCARCINOMA
A. Pathogenesis
1. Pathology
a. Early gastric cancer (EGC) is gastric cancer that has not penetrated the
major muscle layer of the stomach wall. EGC can be divided into three types
based on macroscopic appearance (Fig. 27-1).
Figure 27-1. Classification of early gastric carcinoma according to the Japanese Endoscopy
Society. (From Green PHR, et al. Early gastric cancer. Gastroenterology. 81;247:1981. Reprinted with
permission.)
173
79466_CH27.qxd 1/2/08 12:20 PM Page 174
b. Advanced gastric cancer is gastric cancer that has penetrated the muscle
layer of the stomach. This condition also has been divided into three types:
i. Polypoid or intestinal
ii. Diffuse infiltrating or signet ring type
iii. Ulcerating
2. Risk factors (Table 27-1)
a. Some population groups appear to be at higher risk than others for devel-
opment of gastric adenocarcinoma. For example, the prevalence of gastric
cancer in Japan is about 10 times the prevalence in the United States.
Furthermore, in Japan approximately 30% of gastric cancers at the time of
diagnosis are EGC, whereas in the United States only 5% can be classified
as EGC. Japanese who move away from Japan to a low-risk area have a
similar risk to those who remain in Japan. However, second- and third-
generation Japanese children of these immigrants have a progressively lower
risk that approaches the local population.
b. Several dietary factors have been implicated. Increased consumption of salt
appears to be a consistent finding. Dietary nitrates also may be important.
Cigarette smoking increases risk. However, a diet rich in fresh fruits and
vegetables, daily aspirin use, and COX-II antagonist reduce the risk.
c. Conditions that predispose to achlorhydria, such as pernicious anemia and
atrophic gastritis, carry a higher than average risk of gastric cancer. Whether
this is because the reduced acid allows bacteria that have the capacity to
nitrosate dietary amines to carcinogenic nitrosamines to grow within the
stomach or because of other effects is not clear.
d. Partial gastrectomy 15 or more years in the past was thought to be asso-
ciated with a higher risk of development of adenocarcinoma within the gas-
tric remnant. Recent evidence indicates that this increased risk is lower than
originally anticipated. Virtually all these patients eventually have chronic
gastritis and become hypo- or achlorhydric, which, as indicated previously,
may increase the risk of development of cancer.
e. Helicobacter pylori infection has been associated with gastric adenocarci-
noma. The cancer is thought to arise from gastric intestinal metaplasia that
arises in patients who develop chronic atrophic gastritis with chronic infec-
tion with H. pylori. Especially those strains that are CagA appear to be
more carcinogenic than CagA strains.
B. Diagnosis
1. Clinical presentation
a. EGC typically is asymptomatic and usually is discovered during endoscopy
performed for other complaints. When symptoms have been attributed to
EGC, they usually are vague, such as epigastric discomfort and nausea.
C. Polypoid lesions and benign tumors. The term polyp refers to any protrusion
into the lumen of a viscus and thus does not necessarily connote benign or malig-
nant histopathology. In common medical usage, however, the term polyp usually
refers to a lesion of epithelial origin.
1. Histologic types. Gastric polyps are adenomatous, hyperplastic, or hamar-
tomatous. Of these, only adenomatous polyps and carcinoids appear to have
malignant potential. Other benign lesions that may resemble gastric polyps
grossly are leiomyomas and lipomas. Small polyps normally seen in the gastric
corpus are usually fundic gland polyps that contain dilated gastric glands.
2. Diagnosis. Often, benign polypoid lesions are discovered incidentally during
upper GI x-ray series or endoscopy. The studies may have been performed for
complaints of nausea, abdominal pain, or weight loss, but whether the polyps
are responsible for those complaints is conjectural.
Regardless of how the polypoid lesion has been identified, endoscopy
should be performed to clarify the appearance of the lesion, to obtain biopsies,
and to survey for other lesions. If possible, adenomatous polyps should be
removed endoscopically. Polyps that cannot be removed should be biopsied and
brushed for cytologic study.
3. Treatment. The diagnosis of frank carcinoma, lymphoma, or other malignancy
leads to appropriate treatment of that condition. Removal of an adenomatous
polyp removes the risk of malignant degeneration. The diagnosis of a benign,
nonadenomatous polypoid lesion is reassuring in that the lesion is not cancer-
ous and will not become cancerous.
Selected Readings
Chan AO, et al. Synchronous gastric adenocarcinoma and mucosa-associated lymphoid
tissue lymphoma in association with Helicobacter infection: Comparing reported cases
between the East and the West. Am J Gastroenterol. 2001;96:1922–1924.
Chung DC, et al. A woman with a family history of gastric and breast cancer. N Eng J
Med. 2007;357:283–291.
Delchier JC. Gastric MALT lymphoma: A malignancy potentially curable by eradication of
Helicobacter pylori. Gastroenterol Clin Biol. 2003;27:453–458.
El-Serag HB, et al. Epidemiologic differences between adenocarcinoma of the oesophagus
and adenocarcinoma of the gastric cardia in the USA. Gut. 2002;50:368–372.
El-Zahabi N, et al. The value of EUS in predicting the response of gastric mucosa-
associated lymphoid tissue lymphoma to Helicobacter pylori eradication. Gastrointest
Endosc. 2007;65:89–96.
El-Zimaity HM, et al. Patterns of gastric atrophy in intestinal type gastric carcinoma.
Cancer. 2002;94:1428–1436.
Hwang JH, et al. A prospective study comparing endoscopy and EUS in the evaluation of
GI subepithelial masses. Gastrointest Endos. 2005;62:202–208.
Kurtz RC, et al. Gastric cardia cancer and dietary fiber. Gastroenterology. 2001;120:568.
Laey C, et al. Helicobacter pylori infection and gastric cancer. Gastroenterology. 2001;
20:324–330.
Lynch HT, et al. Gastric cancer: new genetic developments. J Surg Oncol. 2005;90:114–133.
MacDonald JS. Gastric cancer: New therapeutic options. N Engl J Med. 2006;355:76–77.
Meining A, et al. Atrophy-metaplasia-dysplasia-carcinoma sequence in the stomach:
A reality or merely a hypothesis? Best Pract Res Clin Gastroenterol. 2001;15:983–998.
Morita D, et al. Analysis of Sentinel Node involvement in gastric cancer. Clin Gastroenterol
Hepatol. 2007;5:1046–1052.
Noffsinger A, et al. Preinvasive neoplasia in the stomach: Diagnosis and treatment. Clin
Gastroenterol Heptatol. 2007;5:1018–1023.
Sakuramoto S, et al. Adjuvant chemotherapy for gastric cancer with S-1. N Eng J Med.
2007;357:1810–1820.
Usui S. Laparoscopy-assisted total gastroectomy for early gastric cancer: comparison with con-
ventional open toal gastrectomy. Surg Laparosc Endosc Percutan Tech. 2005;15:309–314.
Wagner AD, et al. Chemotherapy for advanced gastric cancer. Cochrane Database Syst
Rev. 2005;2:CD004064.
TABLE 27.1
79466_CH28.qxd 1/2/08 12:21 PM Page 178
28 POSTGASTRECTOMY DISORDERS
G astrectomy usually means removal of part of the stomach and anastomosis of the
gastric remnant with either the duodenum (Billroth I) or a loop of proximal jejunum
(Billroth II) (see Chapter 24). These operations typically are performed as surgical treatment
of peptic ulcer disease or cancer of the stomach. Rarely, the entire stomach is removed.
Removal of part or all of the stomach can be associated with a variety of conse-
quences and complications (Table 28-1). These may range in severity from a simple inabil-
ity to eat large meals, due to loss of the reservoir function of the stomach, to more serious
complications, such as severe dumping and profound nutritional sequelae.
I. DUMPING SYNDROME
A. Pathogenesis. The dumping syndrome develops as a result of the loss of pyloric
regulation of gastric emptying. Thus, strictly speaking, a portion of the stomach
does not necessarily have to be removed; a pyloroplasty alone can lead to the
dumping syndrome. After a pyloroplasty or an antrectomy, hyperosmolar food is
“dumped” rapidly into the proximal small intestine.
1. During the early phase of the dumping syndrome, the hyperosmolar small-
bowel contents draw water into the lumen, stimulate bowel motility, and release
vasoactive agents, such as serotonin, bradykinin, neurotensin, substance P, and
vasoactive intestinal peptide from the bowel wall. Patients experience abdomi-
nal cramps, diarrhea, sweating, tachycardia, palpitations, hypotension, and
light-headedness. These effects typically occur within 1 hour after eating.
2. In the late phase, because of the absorption of a large amount of glucose after
the meal, plasma insulin rises excessively and blood sugar may plummet.
Consequently, the patient may experience tachycardia, light-headedness, and
sweatiness 1 to 3 hours after a meal.
B. Diagnosis. The typical symptoms and signs in the setting of gastric surgery usu-
ally are sufficient to make the diagnosis of dumping syndrome. A blood sugar
I. Dumping syndrome
II. Recurrent ulcer
III. Nutritional and metabolic sequelae
A. Loss of weight
1. Poor food intake
2. Dumping syndrome
3. Malabsorption
B. Malabsorption, with or without weight loss
1. Malabsorption of iron, calcium, folate, vitamin B12
2. Malabsorption of fat, protein, carbohydrate
IV. Afferent loop syndrome
V. (?) Carcinoma in the gastric remnant
178
79466_CH28.qxd 1/2/08 12:21 PM Page 179
determination several hours after a meal when symptoms are at their worst may be
helpful in confirming the late phase.
C. Treatment. Patients are advised to eat small meals six to eight times a day.
Carbohydrates are restricted to minimize glucose absorption. Medications to
reduce bowel motility, such as diphenoxylate or loperamide, may be helpful. In
rare instances, surgical revision with anastomosis of the gastric remnant to an
antiperistaltic segment of jejunum may be necessary.
passage of time. The definitive treatment is surgical and consists of either relieving
adhesive bands or shortening the afferent loop and reforming the anastomosis.
Selected Readings
Dulucq JL, et. al. A completely laparoscopic total and partial gastrectomy for benign and
malignant diseases: a single institute’s prospective analysis. 2005; 200:191–197.
Madura JA. Postgastrectomy problems: remedial operations and therapy. In: Camperon
JL, ed. Current surgical therapy. St. Louis:Mosby;2001:89–94.
Tersmette AC, et al. Long-term prognosis after partial gastrectomy for benign conditions:
survival and smoking-related death of 2633 Amsterdam postgastrectomy patients
followed up since surgery between 1931 and 1960. Gastroenterology. 1991;101:148.
79466_CH29.qxd 1/2/08 12:22 PM Page 182
29 DIARRHEA
D iarrhea can be defined as an increase in the fluidity, frequency, and volume of daily
stool output. The daily stool weight is usually increased over the normal average of 200 g
due to an increase in the stool water above the normal content of 60% to 75%. There may
also be a change in the stool solids.
I. NORMAL INTESTINAL FLUID BALANCE. During fasting, the intestine contains very lit-
tle fluid. On a normal diet of three meals a day, about 9 L of fluid is delivered to the small
intestine. Oral intake accounts for 2,000 mL, and the remainder is secreted from various
parts of the gastrointestinal tract into the lumen (Table 29-1). Of this, 90% is absorbed in
the small intestine. One to two liters is presented to the colon, and approximately 90% is
absorbed. The colon has the capacity to absorb all of the fluid presented to it, but the pres-
ence of unabsorbed osmotically active solutes from the diet (e.g., some carbohydrates) and
from bacterial action prevents complete fluid absorption and desiccation of the fecal mate-
rial. This results in 100 to 200 mL of fluid in the stool. Thus, approximately 98% of the
fluid presented to the intestine each day is absorbed by the small and large intestines.
Feces on the average contain 100 mL of water, 40 mEq/L of sodium (Na+), 90 mEq/L
of potassium (K+), 16 mEq/L of chloride (Cl), and 30 mEq/L of bicarbonate (HCO3).
The remaining anions are organic and result from bacterial fermentation of unabsorbed
carbohydrates. The gastrointestinal tract does not have diluting mechanisms; thus, the
osmolality of the fecal fluid is never less than the osmolality of plasma. In fact, osmolality
of fecal fluid is usually greater than that of plasma due to continuing bacterial fermenta-
tion of unabsorbed carbohydrates into osmotically active particles after defecation.
Water transport across the intestinal epithelium is passive. It is secondary to
osmotic gradients generated by active transport of electrolytes such as Na+ and Cl or
other solutes, such as sugars and amino acids. Intestinal ion absorption occurs mainly
across epithelial cells that reside at the tips of the villi. Crypt cells are involved in ion
secretion: Na+ is the major ion actively absorbed, and Cl is the major ion secreted.
Na+ is actively absorbed throughout the intestinal tract empowered by the Na+
pump in the form of Na+-K–ATPase located in the basolateral plasma membrane of
intestinal epithelial cells. Thus water is absorbed along with Na+.
Oral 2,000
Secretions
Salivary 1,500
Gastric 2,500
Pancreatic 1,500
Biliary 500
Small intestinal 1,000
Total 9,000
182
79466_CH29.qxd 1/2/08 12:22 PM Page 183
I. Carbohydrate malabsorption
A. Disaccharidase deficiency in the small-intestinal mucosa
1. Primary (e.g., idiopathic lactase, sucrase–maltase deficiency)
2. Secondary (e.g., after infectious gastroenteritis and with mucosal inflammation)
B. Ingestion of mannitol, sorbitol (diet candy, soda, chewing gum)
C. Lactulose therapy
II. General malabsorption
A. Sprue (idiopathic gluten enteropathy, tropical)
B. Postradiation, postischemic enteritis
III. Ingestion of sodium sulfate (Glauber’s salt), sodium phosphate, magnesium sulfate
(Epsom salts), milk of magnesia, magnesium-containing antacids
79466_CH29.qxd 1/2/08 12:22 PM Page 184
Enterotoxins
Vibrio cholerae
Escherichia coli
Staphylococcus aureus
Bacillus cereus
Hormonal secretagogues
Vasoactive intestinal polypeptide (pancreatic cholera syndrome, secreting villous adenoma)
Calcitonin (medullary carcinoma of thyroid)
Serotonin (carcinoid)
Prostaglandins, prostanoids (intestinal lymphoma, inflammatory bowel disease)
Gastric hypersecretion
Zollinger–Ellison syndrome
Short-bowel syndrome
Systemic mastocytosis
Basophilic leukemia
Laxatives
Ricinoleic acid (castor oil)
Bisacodyl
Dioctyl sodium sulfosuccinate
Aloe
Senna
Danthron
Phenolphthalein
Bile salts
Terminal ileal disease/resection
Bile duct obstruction
Fatty acids
Pancreatic insufficiency
Small-intestine mucosal disease
79466_CH29.qxd 1/2/08 12:22 PM Page 185
VI. MOTILITY DISTURBANCES. Both reduced and increased motility of the intestine
may result in diarrhea.
A. Increased motility of the small intestine results in decreased contact time of
chyme with absorptive surfaces. Large amounts of fluid delivered to the colon may
overwhelm its absorptive capacity and result in diarrhea. The reduced contact time
in the small intestine may interfere with absorption of fatty acids and bile salts,
allowing them to reach the colon where they induce a secretory diarrhea. Diarrhea
associated with hyperthyroidism, carcinoid, and postgastrectomy dumping syn-
drome are examples.
B. Decreased motility of the small intestine may allow colonization of the small
intestine with colonic-type bacteria. The digestion and absorption of fats, carbo-
hydrates, and bile salts may be affected, resulting in osmotic or secretory diarrhea.
This mechanism is involved in the diarrhea seen with diabetes, hypothyroidism,
scleroderma, amyloidosis, and postvagotomy states.
C. Increased colonic motility with premature emptying of colonic contents is the
major cause of diarrhea in the irritable bowel syndrome.
D. Anal sphincter dysfunction caused by neuromuscular disease, inflammation,
scarring, and postsurgical states may result in fecal incontinence, which may be
interpreted by the patient as diarrhea.
VII. CLINICAL APPROACH. It is helpful to classify diarrhea into clinical categories, tak-
ing into consideration the duration, setting, and sexual preference of the patient.
Diarrhea of abrupt onset of less than 2 to 3 weeks’ duration is called acute diarrhea.
If the diarrheal illness lasts longer than 3 weeks, it is called chronic diarrhea. The
causes of acute and chronic diarrhea are listed in Tables 29-4 and 29-5. If diarrhea
occurs in the setting of antibiotic therapy or after a course of antibiotics, antibiotic-
associated diarrhea and pseudomembranous colitis due to C. difficile cytotoxin should
be considered.
A. Acute diarrhea. The most common cause of acute diarrhea is infection.
1. Food poisoning is produced by a preformed bacterial toxin that contaminates
the food. Bacterial replication in the host is not necessary for the development
of disease. The resultant illness usually has an acute onset and short duration
and occurs in small, well-defined epidemics, without evidence of secondary
spread.
2. Diarrhea resulting from multiplication of organisms in the intestine may be
divided into inflammatory–invasive versus noninflammatory–noninvasive cat-
egories, as seen in Table 29-4. Most of these types of diarrhea result from ingestion
of contaminated food or water after 1 to 2 days of incubation. Animal reservoirs
may exist for some common pathogens, including Salmonella, Campylobacter,
Yersinia, Giardia, Cryptosporidium, and Vibrio parahaemolyticus. Waterborne
disease in which the pathogens are spread from animals or water to humans is
79466_CH29.qxd 1/2/08 12:22 PM Page 186
Viral Infections
Small intestine Mucosal invasion absent Rotavirus (children, adults)
Noninflammatory
Watery diarrhea Norwalk virus
Fecal leukocyte absent Enteric adenovirus
Bacterial infections
Small intestine Mucosal invasion absent Vibrio cholerae
Noninflammatory Toxigenic Escherichia coli
Watery diarrhea
Fecal leukocytes absent
Colon Mucosal invasion present Salmonella
Inflammatory Shigella
Bloody diarrhea Campylobacter
Fecal leukocytes present Yersinia enterocolitica
Invasive E. coli
E. coli O157:H7
Staphylococcus aureu
(toxin)
Vibrio parahaemolyticus
(toxin)
Clostridium difficile (toxin)
Parasitic infections
Small intestine Mucosal invasion absent Giardia lamblia
Noninflammatory Cryptosporidium
Watery diarrhea
Fecal leukocytes absent
Colon Mucosal invasion present Entamoeba histolytica
Inflammatory
Bloody diarrhea
Fecal leukocytes present
Food poisoning
Small intestine Toxin induced Staphylococcus aureus
Mucosal invasion absent Bacillus cereus
Noninflammatory Clostridium perfringens
Watery diarrhea Clostridium botulinum
Drugs
Laxatives
Sorbitol
Antacids (Mg2+, Ca2+ salts)
Lactulose
Colchicine
Quinidine
Diuretics
Digitalis
Propranolol
Theophylline
Aspirin
Nonsteroidal antiinflammatory drugs
Chemotherapeutic agents
Antibiotics
(continued)
79466_CH29.qxd 1/2/08 12:22 PM Page 187
Type Agent
d. Carcinoid
e. Medullary thyroid cancer
f. Hormone-secreting pancreatic tumors
g. Tumors secreting vasoactive intestinal polypeptide
h. Gastrinoma
5. Neoplasms. Villous adenoma, colon cancer with obstruction, and fecal
impaction may present with diarrhea.
6. Drugs and laxatives. Surreptitious use of laxatives and drugs, including those
listed in Table 29-4, should always be considered in the evaluation of chronic
diarrhea.
7. Irritable bowel syndrome is very common and may present with only chronic
intermittent diarrhea, constipation, or a combination of both. Most patients
also complain of abdominal cramps, gas, belching, and mucous stools.
8. Incontinence of stool. Anal sphincter dysfunction due to the presence of fis-
sures, fistulas, perianal inflammation, tears from childbirth, anal intercourse
or other trauma, diabetic neuropathy, or neuromuscular disease may result in
frequent stools, which may be interpreted by the patient as diarrhea.
VIII. DIAGNOSIS
A. History. The physician should find out from the patient an accurate description
of the nature of the diarrhea: the duration, frequency, consistency, volume, color,
and relation to meals. Also, it is important to determine the presence of any under-
lying illnesses or systemic symptoms and to establish the patient’s recent travel
history, use of medications or drugs, and sexual preferences.
1. The history can help determine whether the pathology is in the small or large
bowel. If the stools are large, watery, soupy, or greasy, possibly containing
undigested food particles, the disorder is most likely in the small intestine.
There may be accompanying periumbilical or right lower quadrant pain or
intermittent, crampy abdominal pain.
2. If the disease is in the descending colon or rectum, the patient usually passes
small quantities of stool or mucus frequently. The stools are usually mushy and
brown, and sometimes mixed with mucus and blood. There may be a sense of
urgency and tenesmus. If there is pain, it is usually achy and located in the
lower abdomen, pelvis, or sacral region. The passage of stool or gas may pro-
vide temporary relief of the pain.
3. Blood in the stool suggests inflammatory, vascular, infectious, or neoplastic dis-
ease. The presence of fecal leukocytes indicates mucosal inflammation.
4. Diarrhea that stops with fasting suggests osmotic diarrhea, except that
secretory diarrhea due to fatty acids and bile salt malabsorption may also stop
with fasting. Large-volume diarrhea that continues during fasting is most likely
secretory. Persistence of diarrhea at night suggests the presence of an organic
cause rather than irritable bowel syndrome. Fecal incontinence may be due to
anal disease or sphincter dysfunction.
5. Diet. The correlation of patients’ symptoms with ingestion of milk, other dairy
products, or sorbitol-containing artificially sweetened diet drinks, candy, and
chewing gum should also be noted.
B. Physical examination of the patient should focus on the general condition of the
patient, degree of hydration, presence of fever, and other systemic origins of tox-
icity. A variety of physical findings can be sought in a patient with chronic diar-
rhea and may give clues to the etiology of the diarrheal process. These include
goiter, skin rash, arthritis, peripheral neuropathy, postural hypotension, abdomi-
nal bruit, perianal abscess, fistula, and rectal mass or impaction.
C. Diagnostic tests. The initial laboratory evaluation of the patient should include
a complete blood count with differential, serum electrolytes, blood urea nitrogen,
and creatinine. A chemistry profile and urinalysis may also help to assess the sys-
temic involvement with the diarrheal state.
1. Examination of the stool is the most important diagnostic test in the evalua-
tion of a patient with acute or chronic diarrhea. A fresh sample of stool should
79466_CH29.qxd 1/2/08 12:22 PM Page 190
be examined for the presence of pus (white cells), blood, and bacterial and par-
asitic organisms. Best yield is obtained if the examination is repeated on three
fresh stool samples obtained on three separate days.
a. Presence of white blood cells. Wright or methylene blue stain of the
stool smeared on a glass slide will demonstrate the white cells if they are
present. The presence of fecal leukocytes suggests intestinal inflammation as
a result of a mucosal invasion with bacteria, parasite, or toxin (Table 29-4).
Inflammatory bowel disease and ischemic colitis may also result in white
blood cells in the stool.
b. The absence of fecal leukocytes suggests a noninflammatory noninvasive
process (e.g., viral infection, giardiasis, drug-related); however, it is never
diagnostic because a false-negative result may occur in inflammatory states.
c. Occult or gross blood in the stool suggests the presence of a colonic neo-
plasm, an acute ischemic process, radiation enteritis, amebiasis, or severe
mucosal inflammation.
d. Bacterial and parasitic organisms. Fresh stool samples must be examined
for the presence of ova and parasites. Organisms that colonize the upper
small intestine (Table 29-4) may not be found in stool samples and duodenal
or jejunal aspirates or biopsies or the string test may be required. Stool cul-
tures will help determine the bacterial pathogen in most cases. However,
special techniques may sometimes be necessary, such as for Yersinia,
Campylobacter, Neisseria gonorrhoeae, C. difficile, and E. coli: H7.
e. Fat and phenolphthalein. In the evaluation of chronic or recurrent diarrhea,
stool should also be examined for the presence of fat (qualitative and quanti-
tative) and phenolphthalein. Phenolphthalein is found in many laxative
preparations and gives a red color when alkali is added to the stool filtrate.
2. Sigmoidoscopy or colonoscopy or both should be performed without cleans-
ing enemas. Stool samples may be obtained with a suction catheter for micro-
scopic examination and cultures. Most patients with acute or traveler’s
diarrhea do not need proctosigmoidoscopic examination. Sigmoidoscopy is
especially helpful in the evaluation of:
a. Bloody diarrhea
b. Diarrhea of uncertain etiology
c. Inflammatory bowel disease, pseudomembranous colitis, pancreatic dis-
ease, or laxative abuse (melanosis coli)
3. Radiologic studies. Most causes of diarrhea become apparent after the pre-
ceding tests. However, in chronic or recurrent diarrhea, barium studies of the
large and small intestine may demonstrate the location and extent of the dis-
ease. It should be remembered that once barium is introduced into the bowel,
examination of the stool for ova and parasites and cultures may be fruitless for
several weeks because barium alters the gut ecology.
4. Other tests. In cases of chronic diarrhea, other specialized tests may be nec-
essary to assess for malabsorption, bacterial overgrowth, or abnormal hor-
monal states. These are discussed in appropriate chapters.
IX. TREATMENT. Acute diarrhea with fluid and electrolyte depletion is a major cause of
mortality, especially in children in developing countries of the world. Fluid repletion
by intravenous or oral routes can prevent death. Oral rehydration therapy can be
accomplished with a simply prepared oral rehydration solution. Physiologically, water
absorption follows the absorption of glucose-coupled sodium transport in the small
intestine, which remains intact even in the severest of diarrheal illnesses.
Oral rehydration solution can be prepared by adding 3.5 g of sodium chloride
(or three fourths of a teaspoon or 3.5 g of table salt), 2.5 g of sodium bicarbon-
ate (or 2.9 g of sodium citrate or 1 teaspoon of baking soda), 1.5 g of potassium chlo-
ride (or 1 cup of orange juice or 2 bananas), and 20 g of glucose (or 40 g of sucrose or
4 table spoons of sugar) to a liter (1.05 qt) of clean water. This makes a solution of approx-
imately 90 mmol of sodium, 20 mmol of potassium, 80 mmol of chloride, 30 mmol of
bicarbonate, and 111 mmol of glucose per liter. Not only is this solution lifesaving in
79466_CH29.qxd 1/2/08 12:22 PM Page 191
severe diarrhea, but also it is less painful, safer, less costly, and superior to intravenous
fluids because the patient’s thirst protects against overhydration. It should therefore be
the preferred route of rehydration in conscious adult and pediatric patients in tertiary
and intensive care units. Furthermore, the output of stool can be reduced with food-
based oral rehydration therapy. With the additional sodium-coupled absorption of
neutral amino acids and glutamine (a key mucosal nutrient in the small bowel, analo-
gous to short-chain fatty acids in the colon), oral rehydration therapy can also be used
to speed recovery from small-bowel injury.
The composition of cereal-based oral rehydration solution is like that of
standard oral rehydration solution (3.5 g of sodium chloride, 2.5 g of sodium bicar-
bonate, and 1.5 g of potassium chloride), except that the 20 g of glucose is replaced
by 50 to 60 g of cereal flour (rice, maize, millet, wheat, or sorghum) or 200 g of
mashed, boiled potato; stirred into 1.1 L of water; and brought to a boil. Not only
can oral rehydration therapy (especially with cereal and continued feeding) reverse
the loss of fluid, but also it can prevent the fatal hypoglycemia seen with failure of
gluconeogenesis, a major cause of death in children with diarrhea in developing
areas. Furthermore, simple oral rehydration therapy can be started early in the home
and can prevent most complications of dehydration and malnutrition.
Attention should also be directed at reduction of the patient’s symptoms and dis-
comfort to reduce absenteeism from work or school as well as to improve the sense of
well-being of the patient. Available drugs can be divided into groups based on their
mechanisms of action: absorbents, antisecretory agents, opiate derivatives, anticholin-
ergic agents, and antimicrobial agents.
A. Absorbents (e.g., Kaopectate, aluminum hydroxide) do not influence the course
of the disease but help produce solid stools. This effect may allow the patient to
alter the timing of stooling and permit a more voluntary control of defecation.
B. Antisecretory agents. In most cases of diarrhea, regardless of etiology, invasive
or noninvasive, intestinal secretion contributes greatly to the stool volume.
Antisecretory drugs, including prostaglandin synthesis inhibitors, have been used
to diminish bowel secretion.
Bismuth subsalicylate (Pepto-Bismol) has been shown to block the secre-
tory effects of V. cholerae, enterotoxigenic E. coli, and Shigella as well as to prevent
intestinal infection by these agents if given prophylactically. The usual therapeutic
dosage of Pepto-Bismol is 30 mL every 30 minutes for eight doses. The prophylac-
tic dosage is 60 mL or two tablets q.i.d. for the duration of the prophylaxis (e.g.,
for travelers). Pepto-Bismol tablets are as effective as the liquid preparation.
C. Opiate derivatives are widely used in both acute and chronic diarrhea. By dimin-
ishing peristalsis, they delay gut transit of fluid and allow more time for fluid
absorption. They may be used in patients with moderate symptoms (3–5 stools per
day) but should not be used in patients with fever, systemic toxicity, or bloody
diarrhea. Their use should be discontinued in patients who have not shown
improvement or whose condition has worsened on therapy.
Opiate derivatives include paregoric (tincture of opium), diphenoxylate with
atropine (Lomotil), and loperamide (Imodium). Imodium has two advantages over
Lomotil in that it does not contain atropine and it has fewer central opiate effects.
D. Anticholinergic agents do not appear to be useful in the treatment of most diar-
rheal disorders. Some patients with irritable bowel syndrome may benefit from use
of dicyclomine hydrochloride (Bentyl).
E. Antimicrobial agents. When the diarrhea is severe and the patient has systemic
signs of toxicity, stool cultures should be performed to identify the pathogenic
organism. The most effective antimicrobial agent for the particular pathogen
should be used. In selected cases of severe diarrhea, if a laboratory is not available,
empiric antibiotics with activity against both Shigella and Campylobacter strains
may be administered (e.g., ciprofloxacin or trimethoprim/sulfamethoxazole or
erythromycin). Recently, rifaximin (Xifaxan), a nonabsorbable (gut specific)
antibiotic has become available for the treatment of traveler’s diarrhea. Table 29-6
lists common pathogens and recommended therapeutic agents. Table 29-7 summa-
rizes an approach to therapy of acute diarrhea.
79466_CH29.qxd 1/2/08 12:22 PM Page 192
p.o., by mouth; b.i.d., twice a day; q.i.d., four times a day; t.i.d., three times a day.
79466_CH29.qxd 1/2/08 12:22 PM Page 193
No. of diarrheal
Severity of illness stools/day Therapy
Selected Readings
Al Ghamdi MY, et al. Causation: Recurrent collagenous colitis following repeated use of
NSAIDs. Can J Gastroenterol. 2002;16:861–862.
Alaedini A, et al. Narrative review—celiac disease: Understanding a complex autoimmune
disorder. Ann Intern Med. 2005;142:289–298.
AUS, et al. Giardia intestinalis. Curr Opin Infect Dis. 2003;16:453–460.
Aziz B, et al. 25-year-old man with abdominal pain, nausea and fatigue. Mayo Clin Proc.
2007;82(3):359–362.
Bardhan PK, et al. Screening of patients with acute infectious diarrhoea: Evaluation of clinical
features, faecal microscopy, and faecal occult blood testing. Scand J Gastroenterol. 2000;
35:54–60.
Bengmark S, et al. Prebiotics and synbiotics in clinical medicine. Nutr Clin Pract. 2005;
20:244–261.
Butler T, et al. New developments in the understanding of cholera. Curr Gastroenterol
Rep. 2001;3:315.
Chermesh I, et al. Probiotics and the gastrointestinal tract: Where are we in 2005? World
Gastroenterol. 2006;12:853–857.
Erim TD, et al. Collagenous colitis associated with Clostridium difficile: A cause effect?
Dig Dis Sci. 2003;48:1374–1375.
Guarner F. Enteric flora in health and disease. Digestion. 2006;73(suppl 1):5–12.
Headstrom PD, et al. Chronic diarrhea. Clin Gastroenterol Hepatol. 2005;3:734–737.
Jaskiewicz K, et al. Microscopic colitis in routine colonoscopies. Dig Dis Sci. 2006;
51:241–244.
Lo W, et al. Changing presentation of adult celiac disease. Dig Dis Sci. 2003;4:395–398.
Nyhlin N, Bohr J, Eriksson S, Tysk C. Systematic review: microscopic colitis. Ailment
Pharmacol Ther. 2006;23:1525–1534.
Quigley EMM, et al. Small intestinal bacterial overgrowth: Roles of antibiotics, prebiotics
and probiotics. Gastroenterology. 2006;130:S78–S90.
Sanders JW, et al. Diarrhea in the returned traveler. Curr Gastroenterol Rep. 2001;3:304.
Saulsbury FT. Clinical update: Henoch-Schonlein purpura. Lancet. 2007;669:976–978.
Schiller JR. Chronic diarrhea. Gastroenterology. 2004;127:287–293.
Snelling AM. Effects of probiotics on the gastrointestinal tract. Curr Opin Infect Dis.
2005;18:420–426.
Thomas, PD, et al. Guidelines for the investigation of chronic diarrhea, 2nd. Edition. Gut.
2003;53(supp 1):v1–v15.
Wong CS, et al. The risk of hemolytic uremic syndrome after antibiotic treatment of
Escherichia coli O157:H7 infections. N Engl J Med. 2000;342:1930.
79466_CH30.qxd 1/2/08 12:23 PM Page 195
195
79466_CH30.qxd 1/2/08 12:23 PM Page 196
solutions (e.g., Rehydralylate, Pedialyte, and Rosol) are highly effective, even in
the presence of vomiting. Breastfeeding and transplacentally transmitted mater-
nal antibodies have been found to prevent rotavirus infections. Progress is being
made toward the development of a vaccine.
B. Enteric adenovirus
1. Epidemiology. Adenovirus serotypes 40 and 41 are enteric adenoviruses that
cause gastroenteritis without nasopharyngitis and keratoconjunctivitis. Infection
is transmitted from person to person. Enteric adenovirus is second to rotavirus
as the cause of pediatric viral gastroenteritis, especially in children under the age
of 2. Nosocomial outbreaks occur, but spread to adults is uncommon.
2. Clinical disease. Endemic enteric adenovirus gastroenteritis may occur year-
round without seasonal preference. The incubation period is 8 to 10 days. The
onset is with low-grade fever and watery diarrhea, followed by 1 to 2 days of
vomiting. Illness typically lasts 1 to 2 weeks but occasionally lasts longer.
3. Diagnosis. There are no fecal leukocytes in stool. Diagnosis may require
demonstration of the virus by electron-microscopy in stool samples, use of
nucleic acid hybridization, and radioimmunoassays using monoclonal antibod-
ies specific for adenovirus 40 and 41.
4. Treatment is supportive and symptomatic.
C. Norwalk virus and Norwalk-like viruses
1. Epidemiology. Norwalk virus is the prototype of the Norwalk-like group of
viruses that, unlike rotavirus and enteric adenoviruses, are small and round and
resemble the other small gastroenteritis viruses (e.g., calcii viruses, astroviruses,
and small featureless viruses). These viruses are refractory to in vitro cultivation
and purification. Norwalk virus possesses a single-stranded RNA, which has
been detected in diarrheal stools by immune electromicroscopy. The Norwalk
virus family causes approximately 40% of epidemics of gastroenteritis that
occurs in recreational camps; on cruise ships; in schools, colleges, nursing
homes, hospital wards, cafeterias, and community centers; and among sports
teams and families by the ingestion of contaminated foods such as salad and
cake frosting. Epidemics occur year-round, affecting older children and adults
but not infants and young children. Infection spreads rapidly by the fecal–oral
route, with an incubation period of 12 to 48 hours. In food-borne outbreaks,
infectious virus may be excreted in the feces for at least 48 hours after the
patients have recovered. Airborne transmission by means of droplets of vomit
or through the movement of contaminated laundry also occurs.
2. Histopathology. The infection affects the small intestine, sparing the stomach
and the colonic mucosa. There is blunting of villi and microvilli and cellular
infiltration in the lamina propria, especially in the jejunum. Malabsorption of
d-xylose, lactose, and fat occurs, but absorption returns to normal within 1 to
2 weeks after recovery. It is thought that gastric emptying is delayed, which
would explain the nausea and vomiting that accompanies the watery diarrhea.
3. Clinical disease. The onset is rapid with abdominal pain, low-grade fever,
vomiting, and diarrhea that usually last 48 to 72 hours.
4. Diagnosis. There are no fecal leukocytes. Specific diagnostic techniques are
restricted to a few research laboratories that use both a radioimmunoassay
(RIA) and an enzyme-linked immunosorbent assay (ELISA); the tests may be
used to measure Norwalk viral antigens in stool and antibody in serum.
5. Treatment is supportive.
D. Calcii virus
1. Epidemiology. Human calcii viruses are poorly understood agents that are
related to the Norwalk virus group. They affect mostly infants and young children
but may also infect adults in epidemics.
2. Clinical disease. The incubation period is 1 to 3 days. The clinical presenta-
tion resembles that of rotavirus or Norwalk viral gastroenteritis. The diarrhea
is accompanied by vomiting, abdominal pain, and low-grade fever.
3. Diagnosis. Calcii virus can be detected in stool by electron-microscopy and by
RIA from serum. Serum antibody may be protective against reinfection.
4. Treatment. The treatment is supportive.
79466_CH30.qxd 1/2/08 12:23 PM Page 197
Chapter 30: Viral, Bacterial, and Parasitic Disorders of the Bowel 197
E. Astrovirus
1. Epidemiology. The virus can be cultivated in cell cultures. Astrovirus is a single-
stranded RNA virus with five human serotypes. It causes outbreaks of diarrhea in
children 1 to 7 years of age; in the elderly, especially in nursing homes; and rarely
in young adults, suggesting that adults may be protected by previously acquired
antibodies.
2. Clinical disease. The incubation period is 1 to 2 days. Watery diarrhea may be
accompanied by vomiting.
3. The diagnosis is by stool electron-microscopy and ELISA.
4. Treatment is supportive.
F. Other viruses associated with gastroenteritis. Several other groups of viruses
known to cause diarrhea in animals are associated with gastroenteritis in humans,
but their causative relation to disease is unclear. Coronaviruses are detected by
electron-microscopy in stools of persons living under poor sanitary conditions. It
has been associated with outbreaks in nurseries and with necrotizing enterocolitis
in newborns. Echoviruses and picornaviruses also have been implicated in diar-
rheal disease of the young. Enteroviruses have been shown in controlled epi-
demiologic studies not to be important causes of acute gastroenteritis.
II. BACTERIAL INFECTIONS OF THE BOWEL. Bacterial gastroenteritis may result from
the ingestion of a preformed bacterial toxin present in the food at the time of ingestion,
by the production of a toxin or toxins in vivo, or by invasion and infection of the
bowel mucosa by the bacterial pathogen. In this chapter, the most common bacterial
pathogens affecting immunocompetent hosts are discussed. Enteric bacterial infections
in HIV-infected and other immunocompromised patients are discussed in Chapter 44.
A. Toxigenic bacteria. In general, toxigenic bacteria produce watery diarrhea
without systemic illness. There may be low-grade fever in some patients. Some
microorganisms can produce other toxins in addition to an enterotoxin, for exam-
ple, neurotoxins that can cause extraintestinal manifestations. The stools contain
no blood or fecal leukocytes, which helps to distinguish these diseases from diar-
rheas caused by tissue-invasive organisms. Table 30-1 lists some of the common
causes of toxigenic diarrhea.
1. Staphylococcus aureus
a. Epidemiology. Staphylococcal food poisoning is the most frequent cause of
toxin-mediated vomiting and diarrhea encountered in clinical practice. All
coagulase-positive staphylococci can produce enterotoxins. Staphylococci
are introduced into food by the hands of food-handlers. The organisms mul-
tiply and produce the toxin if the food is kept at room temperature. The
foods most commonly implicated are coleslaw, potato salad, salad dressings,
milk products, and cream pastries. Food contaminated with staphylococci is
normal in odor, taste, and appearance.
b. Clinical disease. Staphylococcal food poisoning is manifested by an abrupt
onset of vomiting within 2 to 6 hours after ingestion of the contaminated
food. The diarrhea is usually explosive and may be accompanied by abdom-
inal pain. Fever is usually absent.
c. The diagnosis is usually suspected from the history. In most instances, the
organism can be cultured from the contaminated food.
d. Treatment. Gastroenteritis resolves with supportive care within 12 to
24 hours. Antimicrobial therapy is not indicated.
2. Bacillus cereus
a. Epidemiology. Bacillus cereus is a common gram-positive, spore-forming
organism found in soil. Contamination of food occurs before cooking.
Vegetative growth continues at temperatures of 30˚ to 50˚C, and spores can
survive extreme temperatures. The spores of the organism germinate and
produce toxins during the vegetative stage.
B. cereus is a frequent cause of food poisoning from many sources, but is
usually associated with contaminated rice or meat from Chinese restaurants.
b. Clinical disease. B. cereus intoxication manifests as two distinct clinical
syndromes. The “emesis syndrome” is caused by the thermostable toxin and
mimics staphylococcal food poisoning. Within 2 to 6 hours after ingestion of
the contaminated food, the patient has severe vomiting and abdominal pain
with or without diarrhea. There is no accompanying fever or systemic mani-
festations. Illness is self-limited and lasts 8 to 10 hours. The “diarrhea syn-
drome” is caused by the thermolabile enterotoxin and occurs after 8 to
16 hours of ingestion of the contaminated food. It is characterized by a foul-
smelling, profuse watery diarrhea, usually accompanied by nausea, abdomi-
nal pain, and tenesmus. Most of the symptoms resolve in 12 to 24 hours.
c. Diagnosis is made by history and stool cultures demonstrating the organism.
d. Treatment is supportive.
3. Vibrio cholerae
a. Epidemiology. V. cholerae is a mobile, gram-negative bacterium with a sin-
gle flagellum and is easily recognizable by a fecal Gram’s stain. It produces a
thermostable enterotoxin, which stimulates the adenylate cyclase in small-
intestinal crypt cells, especially in the jejunum, resulting in profuse secretory
diarrhea. V. cholerae is seen occasionally in the United States, especially along
the Gulf coast. Any fecally contaminated water or food has the potential to
cause cholera, but contaminated saltwater crabs and freshwater shrimp are
responsible for most instances seen in the United States.
b. Clinical disease. The incubation period is 1 to 3 days after ingestion.
Cholera is characterized by an abrupt onset of profuse, large-volume, watery
diarrhea. The stools are isotonic and do not contain blood or mucus. There
is usually no associated fever, abdominal pain, vomiting, or tenesmus.
Hypotension, shock, and death may result if volume depletion is not ade-
quately treated.
c. Diagnosis. Organisms may be demonstrated by dark-field microscopy of
the stool and by stool cultures.
d. Treatment. The mainstay of therapy is volume repletion intravenously or orally
with fluids that contain glucose and electrolytes. If dehydration is adequately
reversed, patients recover in 7 to 10 days without antimicrobial therapy. The
duration of the disease may be shortened to 2 to 3 days with oral tetracycline
500 mg q.i.d. or doxycycline 300 mg in a single dose. In resistant instances, alter-
native antimicrobials are furazolidone 100 mg q.i.d., erythromycin 250 mg q.i.d.,
or trimethoprim/sulfamethoxazole 160/800 mg b.i.d. for 3 days.
4. Enterotoxigenic Escherichia coli
a. Epidemiology. Enterotoxigenic E. coli (ETEC) can cause diarrhea by tissue
invasion or via its enterotoxin. The enterotoxin is thermolabile and produces
diarrhea by the same mechanism as the cholera enterotoxin. The organism is
transmitted from contaminated water and food by the fecal oral route. Even
though it may cause outbreaks in the United States, ETEC is the most
common traveler’s pathogen. A large inoculum is required for disease. The
incubation period is 1 to 3 days.
b. The clinical disease is similar to cholera. The watery diarrhea is profuse
and lasts 3 to 5 days. There may be accompanying mild abdominal pain.
79466_CH30.qxd 1/2/08 12:23 PM Page 199
Chapter 30: Viral, Bacterial, and Parasitic Disorders of the Bowel 199
Viruses
Rotavirus
Norwalk and related viruses
Adenovirus
Bacteria
Vibrio cholerae
Escherichia coli (enterotoxigenic)
All bacteria listed in Table 30-3
Protozoa
Giardia lamblia
Cryptosporidium
c. The diagnosis is made by history and culture or toxin assay from the con-
taminated food or the patient’s blood or stool. Electromyography may be
used to differentiate the disease from Guillain–Barré syndrome.
d. Therapy. When intoxication is suspected, therapy should be started imme-
diately with administration of the polyvalent antitoxin and penicillin.
Gastrointestinal lavage with orally administered solutions (e.g., GoLYTELY
or Colyte) may help eliminate the toxin from the gastrointestinal tract.
Guanidine hydrochloride may be used to reverse the motor weakness. Some
patients may require ventilatory support.
B. Bacteria causing “enteric” infection. The resulting diarrhea may be watery or
bloody. Although watery diarrhea is often associated with infections with viruses,
protozoa, and toxin-producing bacteria such as Vibrio cholerae and enterotoxi-
genic E. coli, invasive bacteria can also cause watery diarrhea (Table 30-2). The
diarrhea is usually greater than 1 L/day. Systemic symptoms such as fever,
headache, myalgia, and arthralgias are usually absent.
Bloody diarrhea, or dysentery, is usually accompanied by abdominal pain,
tenesmus, nausea, vomiting, and systemic symptoms such as fever and malaise.
Bacteria that result in bloody diarrhea are listed in Table 30-3. These enteric infec-
tions cannot be distinguished easily from one another clinically. Diagnosis must be
based on the identification of the infectious agent by appropriate cultures.
1. Campylobacter jejuni and Campylobacter fetus
a. Epidemiology. C. jejuni is the most common bacterial pathogen that causes
bloody diarrhea in the United States. It is implicated in infections in under-
developed countries also. The organism is a microaerophilic, gram-negative
curved rod transmitted to humans from contaminated pork, lamb, beef,
Campylobacter
Escherichia coli (enteropathogenic and invasive)
Shigella
Salmonella
Escherichia coli O157:H7
Yersinia
Vibrio parahaemolyticus
Clostridium difficile
Entamoeba histolytica
79466_CH30.qxd 1/2/08 12:23 PM Page 201
Chapter 30: Viral, Bacterial, and Parasitic Disorders of the Bowel 201
milk and milk products, and water, and from exposure to infected house-
hold pets. The organism is destroyed by appropriate cooking, pasteuriza-
tion, and water purification. The incubation period is 1 to 7 days.
b. Histopathology. The bacterial endotoxin causes mucosal inflammation in the
small and large intestine that resembles the lesions seen in ulcerative colitis
and Crohn’s disease and those seen with Salmonella and Shigella infections.
The infection is usually more severe in the colon than in the small bowel.
c. Clinical disease
i. Enterocolitis. Bowel infections with C. jejuni and rarely C. fetus cause a
diarrheal illness resembling enteritis from Salmonella and Shigella.
Occasionally there is a prodrome of headache, myalgia, and malaise for
12 to 24 hours, followed by severe abdominal pain, high fever, and pro-
fuse watery and then bloody diarrhea. The diarrhea is usually self-limited
and in most instances resolves in 7 to 10 days; however, in one fifth of
the instances, the diarrhea has a protracted or a relapsing course.
ii. Systemic infection. C. fetus and rarely C. jejuni may cause a systemic
infection, especially in elderly, debilitated patients, and in those with
alcoholism, diabetes mellitus, and malignancies. Bacteremia may be tran-
sient or may lead to localized infection such as endocarditis, meningitis,
cholecystitis, and thrombophlebitis. There may or may not be clinically
evident enterocolitis.
d. Complications. Campylobacter infection may be complicated by Reiter’s
syndrome, mesenteric adenitis, terminal ileitis (resembling Crohn’s ileitis),
and rarely an enteric fever like illness.
e. Diagnosis is made by stool and blood cultures. Stool Gram’s stain may
show the organism with its characteristic “gull wings.” In dark-field/phase-
contrast microscopy, the organism shows “darting motility.” Fecal leuko-
cytes are present in 75% of instances.
f. Treatment. In mild cases, supportive therapy is given. In cases with bloody
diarrhea, erythromycin 250 mg p.o. q.i.d. for 5 to 7 days or ciprofloxacin
500 mg p.o. b.i.d. for 3 to 7 days is effective.
2. Salmonella
a. Epidemiology. The three primary species of Salmonella (Salmonella typhi,
Salmonella choleraesuis, and Salmonella enteritidis) may cause disease in
humans. S. enteritidis is a common cause of infectious diarrhea. There are
1,700 serotypes and variants of Salmonella, which are classified into 40
groups. Ninety percent of Salmonella organisms that are pathogenic for human
beings are in groups B, C, and D. The organism is transmitted from fecally
contaminated foods and water with fecal–oral contact. Poultry and poultry
products constitute the major reservoir for the bacteria. A large inoculum
(105 organisms) is required to produce infection. Thus, the incidence is rela-
tively low despite the widespread contamination of commonly ingested foods.
b. Pathology. Salmonella elaborates an enterotoxin, which is responsible for
the watery diarrhea. The organism also adheres to the mucosal surface and
invades the epithelium, resulting in colitis and bloody diarrhea.
c. Clinical disease. Salmonella invades the mucosa of the small and large
intestine and produces an enterotoxin that causes a secretory diarrhea.
Watery diarrhea is more common, but bloody diarrhea may occur. Patients
complain of headache, malaise, nausea, vomiting, and abdominal pain
within 6 to 48 hours after ingesting the contaminated food. The disease is
usually self-limited and resolves in 7 days. Fever and bacteremia occur in
less than 10% of patients. Immunosuppression, malignancy, hemolytic
states, liver disease, achlorhydria, and chronic granulomatous disease of
children predispose patients to progressive salmonellosis with bacteremia,
with localized infection in joints, bones, meninges, and other sites. In 5% of
patients, the bacteria may localize in the reticuloendothelial system and may
cause an enteric fever (especially S. typhi). A carrier state also exists in some
patients, with bacteria carried in the gallbladder or in the urinary tract.
79466_CH30.qxd 1/2/08 12:23 PM Page 202
d. The diagnosis is made by history and stool and blood cultures. A fourfold
rise in serum O and H agglutinin titers 3 to 4 weeks after infection confirms
the diagnosis.
e. Treatment is supportive in most instances. Antimicrobial therapy is contraindi-
cated for most patients because it can increase the carrier state. However,
antimicrobial agents such as ampicillin, chloramphenicol, trimethoprim/sul-
famethoxazole, ciprofloxacin hydrochloride, or third-generation cephalosporins
can be used in young children or in patients who are susceptible to bacteremia
and prolonged salmonellosis (Table 30-4). Patients with bacteremia, enteric
fever, and metastatic infection should be treated with antimicrobial therapy.
Also, patients with underlying acquired immunodeficiency syndrome (AIDS),
hemolytic states, lymphoma, and leukemia, and neonates, the elderly and
chronic carriers should receive antimicrobials. Anticholinergic agents and opi-
ates should not be used because they can prolong the excretion of the bacteria.
3. Shigella
a. Epidemiology. There are four major groups: Shigella dysenteriae, Shigella
flexneri, Shigella boydii, and Shigella sonnei. S. dysenteriae causes the sever-
est form of dysentery. In the United States, 60% to 80% of instances of bacil-
lary dysentery are caused by S. sonnei with a seasonal preference for winter.
In tropical countries, S. flexneri dysentery is more common especially in the
late summer months. It is transmitted by the fecal–oral route. Human beings
Chapter 30: Viral, Bacterial, and Parasitic Disorders of the Bowel 203
are the only natural host for this organism. Enteric infections with Shigella
are most commonly seen in children 6 months to 5 years old, although per-
sons of all ages can become infected. Clinical shigellosis is highly contagious
and can be caused by a very small inoculum: fewer than 200 organisms.
Food, water, and milk can be contaminated, which can result in epidemics.
Incidence of the disease increases in crowded, unsanitary conditions.
b. Pathophysiology. Shigella elaborates an enterotoxin that is responsible for
the watery diarrhea. The organism also adheres to the mucosal surface and
invades the epithelium, resulting in colitis and bloody diarrhea.
c. Clinical disease. The incubation period is 1 to 3 days. In most individuals,
the disease starts as lower abdominal pain and diarrhea. Fever is present in
less than half of the patients. In many patients, there is a biphasic illness that
begins as fever, abdominal pain, and watery diarrhea. In 3 to 5 days, rectal
burning, tenesmus, and small-volume bloody diarrhea characteristic of
severe colitis develop. Toxic megacolon and colonic perforation may recur.
Extraintestinal complications include conjunctivitis, seizures, meningismus,
Reiter’s syndrome, thrombocytopenia, and hemolytic uremic syndrome.
The course of shigellosis is variable. In children, it may resolve in 1 to
3 days and in most adults in 1 to 7 days. In severe instances, it may last
longer than 3 to 4 weeks with associated relapses. It may be confused with
idiopathic ulcerative colitis. A minority of patients become chronic carriers.
d. The diagnosis of shigellosis is made by identification of the gram-negative
bacillus in the stool. Sigmoidoscopic findings are identical to those of idio-
pathic inflammatory bowel disease.
e. Treatment. Patients should receive supportive therapy with antipyretics and
fluids. Antiperistaltic agents such as diphenoxylate hydrochloride (Lomotil)
or loperamide hydrochloride (Imodium) should be avoided. Antimicrobial
therapy decreases the duration of fever, diarrhea, and excretion of the organ-
isms in the stool. Trimethoprim/sulfamethoxazole, tetracycline, and ampi-
cillin (but not amoxicillin) are all effective; however, resistance has been
demonstrated. Ciprofloxacin and norfloxacin are also effective.
4. Escherichia coli. In addition to ETEC, other serotypes of E. coli also cause
diarrhea. These include enteroinvasive E. coli (EIEC), enteropathogenic E. coli
(EPEC), enterohemorrhagic E. coli (EHEC), diffuse adherence E. coli (DAEC),
and enteroaggregating E. coli (E AGGEC). All of these bacteria possess plasmid-
encoded virulence factors. They make specific interactions with the intestinal
mucosa by way of bacterially derived adhesions. Some produce enterotoxins
and cytotoxins. Transmission is fecal–oral.
a. Enteroinvasive E. coli
i. Epidemiology. EIEC is a traveler’s pathogen. Epidemics have been
described resulting from imported cheese. The organism also causes epi-
demics in young children, 1 to 4 years of age.
ii. Clinical disease. Similar to Shigella, EIEC invades and destroys the
colonic mucosal cells and causes, first, watery diarrhea followed by a
dysentery like syndrome. The incubation period is 1 to 3 days. The fever
and diarrhea last 1 to 2 days.
iii. Diagnosis. Fecal leukocytes are present. Serotyping and ELISA are avail-
able only in research settings.
iv. Treatment is supportive. Bismuth subsalicylate, by decreasing colonic
secretions, seems to decrease the diarrhea and other symptoms in all
infections with E. coli species. The antimicrobials used in shigellosis are
effective, as well as rifaximin 200 mg p.o. t.i.d.
b. Enteropathogenic E. coli
i. Epidemiology. EPEC is a major cause of diarrhea in both economically
developed and underdeveloped countries. It commonly causes outbreaks
in nurseries affecting children up to 12 months of age. It may also cause
sporadic diarrhea in adults. The bacteria adhere closely to the enterocyte
membrane via an adherence factor with destruction of microvilli.
79466_CH30.qxd 1/2/08 12:23 PM Page 204
ii. Clinical disease. Disease onset is with fever, vomiting, and watery diar-
rhea. Symptoms may continue for longer than 2 weeks and patients may
relapse.
iii. Diagnosis is by serotype analysis
iv. Treatment. Nonabsorbable antibiotics such as neomycin, colistin, and
polymyxin have been recommended. Ciprofloxacin hydrochloride, nor-
floxacin, and aztreonam are also effective and preferred.
c. Enterohemorrhagic E. coli
i. Epidemiology. EHEC or E. coli O157:07, and rarely E. coli O26:H11
have been detected in contaminated hamburger meat; outbreaks have
occurred in nursing homes, daycare centers, and schools.
ii. Pathogenesis and clinical disease. EHEC elaborates a Shigella-like
toxin (verotoxin 1) that is identical to the neuroenterocytotoxin of
S. dysenteriae and to verotoxin 2 and an adherence factor encoded by a
plasmid. Transmission is fecal–oral. Although the disease is more com-
mon in children, several outbreaks have occurred in adults from conta-
minated beef. The bloody diarrhea may be copious but may show no
fecal leukocytes. It usually lasts 7 to 10 days but may be complicated by
hemolytic uremia syndrome.
iii. Diagnosis. Stool cultures and serotyping sorbitol-negative E. coli iso-
lates may yield the organism.
iv. Treatment. Symptomatic treatment and ciprofloxacin or norfloxacin
may be used in severe illness. Supportive measures are recommended.
d. Diffuse adherence E. coli (DAEC) affects young children, especially in eco-
nomically underdeveloped countries. Diarrhea is usually watery, lasts less
than 2 weeks, and may become persistent.
e. Enteroaggregating E. coli (EAGGEC) has been recognized recently as a
pathogen especially affecting the ileum and the terminal ileum. The aggre-
gating bacteria gather around the villi and cause epithelial destruction. The
pathogenesis is transferred by a plasmoid via fimbriae. It causes persistent
diarrhea in children and is more common in economically underdeveloped
countries. The management of the diarrheal illness is similar to that of other
E. coli species strains.
5. Yersinia enterocolitica
a. Epidemiology. Y. enterocolitica can be found in stream and lake water and
has been isolated from many animals, including dogs, cats, chickens, cows,
and horses. It is transmitted to humans via contaminated food or water, or
from human or animal carriers. It most commonly affects children and
rarely causes disease in adults. It is found worldwide, especially in
Scandinavia and Europe, and may result in epidemics.
b. Pathogenesis and clinical disease. Yersinia causes a spectrum of diseases
ranging from gastroenteritis to invasive ileitis and colitis. The organism is
invasive and elaborates a heat-stable toxin. These properties allow its invasion
into and through the distal small-bowel mucosa and subsequent infection of
the mesenteric lymph nodes. The incubation period is 4 to 10 days. The dis-
ease normally lasts several weeks but can be prolonged for many months.
The manifestation of Yersinia infection is variable. In infants and
young children less than 5 years, it may be febrile gastroenteritis lasting 1
to 3 weeks. In older children, it may mimic acute terminal ileitis, mesenteric
adenitis, or ileocolitis. It may be confused with acute appendicitis. The
enterocolitis presents with bloody diarrhea, fever, and abdominal pain
accompanied by anorexia, nausea, and fatigue. The diarrhea usually lasts 1
to 3 weeks but may be protracted (3 months). Polyarthritis, erythema
multiforme, and erythema nodosum occasionally develop 1 to 3 weeks after
the onset of diarrhea. Bacteremia is rare but may be seen in immunosup-
pressed patients and may result in hepatosplenic abscess, meningitis, and
infections of other organs. Metastatic foci may occur in joints, lungs, and
bones.
79466_CH30.qxd 1/2/08 12:23 PM Page 205
Chapter 30: Viral, Bacterial, and Parasitic Disorders of the Bowel 205
c. Diagnosis can be made by stool and blood cultures using special media and
culture conditions. The laboratory should be notified. Serologic tests have
been useful in Europe and Canada. The serotypes found in the United States
do not give reliable serologic results.
d. Treatment in most instances is supportive. Antimicrobials such as tetracy-
cline, chloramphenicol, and trimethoprim/sulfamethoxazole may be used in
severe illness.
6. Aeromonas hydrophila
a. Epidemiology. A. hydrophila is a member of the Vibrionaceae family.
It is transmitted from contaminated food and water, especially in the
summer months.
b. Pathogenesis and clinical disease. Aeromonas produces several toxins.
The heat-labile enterotoxin and the cytotoxin are implicated in the intestinal
infection. The disease commonly follows ingestion of untreated water just
before the onset of symptoms and consists of fever, abdominal pain, watery
diarrhea, and vomiting lasting about 1 to 3 weeks in children and 6 weeks or
longer in adults. In 10% of instances, diarrhea is bloody and mucoid. Chronic
diarrhea and choleralike presentation have also been described. In immuno-
compromised patients and patients with hepatobiliary disease, bacteremia
may occur.
c. Diagnosis. Stool cultures are diagnostic. Fecal leukocytes may be present in
one third of instances.
d. Treatment is supportive in mild instances. In severe illness and with chronic
diarrhea, antibiotics may shorten the duration of the disease. Aeromonas is
resistant to beta-lactam antibiotics. Trimethoprim/sulfamethoxazole, tetra-
cycline, and chloramphenicol have been effective.
7. Plesiomonas shigelloides
a. Epidemiology. Plesiomonas is another member of the Vibrionaceae family
that causes sporadic diarrheal disease affecting travelers to Mexico, Central
America, and the Far East after ingestion of raw shellfish. It produces a
choleralike toxin but also has invasive potential.
b. Clinical disease. Diarrhea is usually watery, but in one third of the patients
it is bloody. Abdominal pain is usually severe. Vomiting and fever may be
present. Although the disease is usually over in 1 week, it may last longer
than 4 weeks.
c. Diagnosis is by stool culture. Fecal leukocytes may be present.
d. Treatment is supportive. The organism has the same microbial sensitivity as
Aeromonas.
8. Clostridium difficile
a. Epidemiology. C. difficile is a spore-forming obligate anaerobe. It is found
as “normal flora” in 3% of adults, 15% of hospitalized patients, and 70% of
infants in pediatric wards. It may cause disease in people of all ages, but it
most frequently affects elderly and debilitated patients. The transmission is
usually fecal–oral; however, it may be transmitted environmentally by spores
carried on fomites or on contaminated hands of health care workers. The dis-
ease usually follows antibiotic use with disruption of the normal colonic flora.
All antimicrobial agents with the exception of vancomycin and parenterally
administered aminoglycosides have been linked with C. difficile enterocolitis.
In most instances, the ingestion of the antimicrobial agent is within 6 weeks
of the onset of the diarrhea. Case reports of C. difficile disease in patients
who have not received antibiotics include patients with neutropenia or ure-
mia, those undergoing cancer chemotherapy, and homosexual males.
b. Pathogenesis and clinical disease. C. difficile produces two major tox-
ins. Toxin A is an enterotoxin, and toxin B is a cytotoxin used in commer-
cial latex agglutination testing for detection of the infection. Toxin A binds
to receptors on the colonic mucosal surface and causes severe inflammatory
changes. The toxigenic effect is catalyzed by previously present trauma or
injury to the mucosal cells. The severity of the disease varies from watery
79466_CH30.qxd 1/2/08 12:23 PM Page 206
III. PARASITIC INFECTIONS OF THE BOWEL. Numerous parasites infect the human
bowel and cause disease. This chapter discusses only the most common parasitic
infections, namely, protozoan pathogens seen in the United States in normal, immuno-
competent people.
In recent years, there has been an increase in protozoan infections of the bowel
caused by increases in international travel to tropical and subtropical areas of the world,
in male homosexuality, and in AIDS. The protozoan infections seen in AIDS are dis-
cussed in Chapter 43. The antimicrobial therapy is summarized in Table 30-4 (page 204).
A. Giardia lamblia
1. Epidemiology. Giardia is a flagellated protozoan seen worldwide. It is trans-
mitted by the fecal–oral route from fecally contaminated water and food. It is
also a traveler’s pathogen and affects children and adults, especially people with
IgA deficiency, hypochlorhydria, and malnutrition.
2. Pathogenesis and clinical disease. After ingestion of the cysts, excystation
releases the organism in the upper small intestine. Giardia adheres to the brush
border membrane of the enterocytes. Histologically, it may cause a celiac
sprue-like lesion, resulting in lactose deficiency and malabsorption. Incubation
after ingestion of the organisms is 1 week. Most infections produce mild, self-
limiting enteritis with watery diarrhea, abdominal bloating, cramps, and flatu-
lence lasting 1 to 3 weeks. The stools may be bulky and foul smelling. In a
minority of the patients, the infection persists and results in a chronic or recur-
rent disease with weight loss and malabsorption.
3. Diagnosis is made by multiple stool examinations because the shedding of the
protozoan is episodic. In difficult illnesses, duodenal aspirates and touch prepa-
rations made by duodenal biopsy specimens can be used.
4. Treatment is with quinacrine hydrochloride (Atabrine), metronidazole (Flagyl),
or furazolidone (Furoxone). In recurrent illness, combination drug therapy may
be more efficacious.
B. Cryptosporidium
1. Epidemiology. Cryptosporidium is an important coccidian protozoan in veteri-
nary medicine. It is also a ubiquitous human pathogen affecting both immuno-
competent and immunosuppressed patients. There is a high rate of infection
among homosexual men, children in daycare facilities, and immigrants arriving
79466_CH30.qxd 1/2/08 12:23 PM Page 207
Chapter 30: Viral, Bacterial, and Parasitic Disorders of the Bowel 207
Selected Readings
Ali S, et al. Giardia intestinalis. Curr Opin Infect Dis. 2004;16:453–460.
Bardham PK, et al. Screening of patients with acute infectious diarrhea; evaluation of clinical
features, faecal microscopy and faecal occult blood testing. Scand J Gastroenterol.
2000;35:55–60.
Basnyat B, et al. Enteric (typhoid) fever in travelers. Clin Infect Dis. 2005;41:1467–1472.
Gahatsatos P, et al. Meta-analysis of outcome of cytomegalovirus colitis in immuno-
competent hosts. Dig Dis Sci. 2005;50:609–616.
79466_CH30.qxd 1/2/08 12:23 PM Page 208
MALABSORPTION 31
T he term malabsorption connotes the failure to absorb or digest normally one or more
dietary constituents. Patients with malabsorption often complain of diarrhea, and sometimes
the distinction between malabsorption and diarrhea of other causes (see Chapter 29) initially
is difficult. For example, patients with primary lactase deficiency fail to absorb a specific
dietary constituent, lactose, and a watery, osmotic diarrhea develops. However, most patients
with malabsorption present with a syndrome characterized by large, loose, foul-smelling
stools and loss of weight. On additional study, it is found that they cannot absorb fat and
often carbohydrate, protein, and other nutrients also. Table 31-1 indicates that a wide vari-
ety of disorders of the organs of digestion can cause malabsorption or maldigestion.
I. DIAGNOSTIC STUDIES. Before discussing the disorders that may cause malabsorp-
tion, it is useful to review several of the diagnostic studies that are available to aid in
evaluating patients with this condition. The number and order of diagnostic studies
used depends on the clinical signs and symptoms of the patient.
A. Blood tests. The hemoglobin and hematocrit levels may identify an anemia that
accompanies malabsorption. A low mean cell volume (MCV) may be found in iron
deficiency, whereas a high MCV may result from malabsorption of folate or
209
79466_CH31.qxd 1/2/08 12:24 PM Page 210
Figure 31-1. Plain x-ray film of the abdomen in a patient with extensive calcification of the pan-
creas (arrows) and pancreatic insufficiency.
vitamin B12. Serum levels of liver enzymes, protein, amylase, calcium, folate, and
vitamin B12 may be abnormal and should be ordered.
B. Radiographic studies
1. Plain films or computed tomography scan of the abdomen may show
calcification within the pancreas, which indicates chronic pancreatic insufficiency
(Fig. 31-1).
2. A barium examination of the upper gastrointestinal tract, including the small
bowel, usually is one of the first diagnostic studies in the evaluation of malab-
sorption syndrome. Often the findings are nonspecific. The bowel may be dilated
and the barium diluted because of increased intraluminal fluid. A more specific
finding is thickening of the intestinal folds caused by an infiltrative process, such
as lymphoma, Whipple’s disease, or amyloidosis. The narrowed, irregular termi-
nal ileum in Crohn’s disease is virtually diagnostic (Fig. 31-2), although lym-
phoma and other infiltrative disorders also must be considered. Diverticula,
fistulas, and surgical alterations in bowel anatomy also may be evident.
C. Fecal fat determination. Malabsorption of fat (steatorrhea) is common to most
malabsorptive conditions (Table 31-2). Patients should ingest at least 80 g of fat
per day to obtain reliable interpretation of qualitative or quantitative fat determi-
nation. Mineral oil and oil-containing cathartics should be avoided.
1. Qualitative screening test. The Sudan stain for fecal fat is easy to perform
and reasonably sensitive and specific when interpreted by an experienced per-
son. A small amount of fresh stool is mixed thoroughly with normal saline or
water on a glass slide. A drop of glacial acetic acid is added, and the slide is
heated to hydrolyze the fatty acids from the triglycerides in the stool. The Sudan
stain is then added. Increased stool fat is indicated by abnormally large or
increased numbers (100/40 field) of fat droplets.
2. The quantitative determination of stool fat is more accurate than qualitative
screening, but the collection of stool often is disagreeable to patients, family, and
nursing personnel. The stool is collected over 72 hours in a large sealed container,
which can be enclosed in a plastic bag and refrigerated to contain unpleasant
odors. Most normal people excrete up to 6 g of fat per 24 hours on a diet that
contains 80 to 100 g of fat. Stool fat in excess of 6 g per 24 hours can result from a
79466_CH31.qxd 1/2/08 12:24 PM Page 211
Figure 31-2. Small-bowel x-ray series in a patient with Crohn’s disease. Note the narrowed, irreg-
ular contour of the terminal ileum and cecum (arrow). (From Eastwood GL. Core Textbook of
Gastroenterology. Philadelphia: Lippincott Williams & Wilkins; 1984:102. Reprinted with permission.)
Diagnostic Test
dilated or strictured ducts, calcification, and pancreatic masses can imply pan-
creatic disease.
E. Bile acid breath test. Conjugated bile acids that are secreted into the duode-
num are resorbed with about 95% efficiency in the terminal ileum. If radio-
labeled [14C]-glycocholate is given orally to a healthy person, about 5% of it
enters the colon and undergoes bacterial deconjugation. The carbon dioxide
(14CO2) derived from the glycine is absorbed and excreted by the lungs and can
be measured in expired air. Bacterial overgrowth in the small intestine promotes
earlier bacterial deconjugation of the [14C]-glycocholate, and consequently, a
larger amount of 14CO2 is measured in the breath. Similarly, disease or resection
of the terminal ileum allows more bile acids to pass into the colon and undergo
bacterial deconjugation, resulting in an increase in expired carbon dioxide.
F. Xylose tolerance test. D-Xylose is a five-carbon sugar that remains intact when it
is absorbed across intestinal mucosa. Consequently, measurement of xylose absorp-
tion can be used as a screening test for diffuse disease of the small-intestinal mucosa.
The patient drinks 25 g of xylose dissolved in 500 mL of water, and the urine is col-
lected for the next 5 hours. A healthy person absorbs enough xylose to excrete more
than 5 g of xylose. Because low xylose excretion can result from inadequate hydra-
tion, the patient is encouraged to drink an additional 1,000 mL of water during the
5 hours of urine collection. In addition to mucosal disease, a low urinary xylose
excretion can result from small-bowel bacterial overgrowth, decreased circulatory
volume, massive ascites, and renal disease. To avoid the problem of urine collection
in patients with renal disease or who are unable to collect the urine accurately, a
blood xylose level at 2 hours after ingestion of the xylose can be determined. The
normal 2-hour blood xylose level is above 40 mg/dL.
G. Lactose absorption tests. The lactose tolerance test is an indirect measurement
of the activity of intestinal lactase, a brush border enzyme that hydrolyzes lactose
to glucose and galactose. To perform the lactose tolerance test, a fasting blood glu-
cose level is drawn, and the patient swallows 50 g of lactose mixed in 500 mL of
water. The blood glucose level is determined 15, 30, 60, and 90 minutes after
ingestion of lactose. If the patient is lactase deficient, the blood glucose level fails
to rise more than 20 mg/dL above the fasting level (Fig. 31-3).
(mg/dL)
Figure 31-3. Serial blood glucose levels during a lactose tolerance test in a patient with primary
lactase deficiency. The blood glucose failed to rise more than 20 mg/dL above the fasting value.
(From Eastwood GL. Core Textbook of Gastroenterology. Philadelphia: Lippincott Williams & Wilkins;
1984:113. Reprinted with permission.)
79466_CH31.qxd 1/2/08 12:24 PM Page 214
Figure 31-4. Photomicrograph of a normal human small-intestinal mucosal biopsy taken from the
distal duodenum. The villi are tall and straight and the villus-to-crypt ratio is about 5:1. (From
Eastwood GL. Core Textbook of Gastroenterology. Philadelphia: Lippincott Williams & Wilkins;
1984:105. Reprinted with permission.)
79466_CH31.qxd 1/2/08 12:24 PM Page 215
Figure 31-5. Photomicrograph of a mucosal biopsy from the distal duodenum in a patient with
celiac sprue. The villi are severely blunted, the crypts are elongated, the surface epithelial cells are
flattened, and the lamina propria contains a dense inflammatory cell infiltration.
Disorder Examples
Anemia
Iron deficiency Primary biliary cirrhosis
Folate deficiency Primary sclerosing cholangitis
Vitamin B12 deficiency
Prolonged prothrombin time
Osteopenic bone disease Autoimmune cholangitis
Short stature
Idiopathic hepatits
and cirrhosis
Infertility
Neuropsychiatric disorders
Peripheral neuropathy Lymphocytic gastritis
Ataxia Lymphocytic colitis
Seizures
Cognitive deficits
Hyperactivity-attention deficit disorder
IgA deficiency Non-Hodgkin’s lymphoma
IgA nephrolpathy Intestinal T-cell lymphoma
Intestinal adenocarcinoma
Type I diabetes mellitus
Oropharyngeal and esophageal squamous
cell carcinoma
Autoimmune thyroid disease
Autoimmune adrenal disease
Sjögren’s syndrome
Systemic lupus erythematosis
Rheumatoid arthritis
b. Diagnosis. Stool fat is increased, but the small-bowel x-ray series and the
xylose tolerance test are normal. Serum cholesterol and triglycerides are low,
and beta-lipoproteins are absent. A small-bowel biopsy is diagnostic, show-
ing villous epithelial cells distended with fat, but the appearance is otherwise
normal.
c. Treatment. There is no specific treatment of the underlying disease. The fat
malabsorption improves with restriction of dietary long-chain triglycerides
and substitution of medium-chain triglycerides, which do not require chy-
lomicron formation but rather are absorbed directly from the villous epithe-
lial cells into the blood. Fat-soluble vitamins also are indicated.
V. LYMPHATIC DISORDERS
A. Pathogenesis. Obstruction of lymphatic drainage from the gut causes dilatation of
the lymphatics (lymphangiectasia) and loss of fat and protein in the stool. In some
patients, the disorder appears to be congenital or idiopathic. In others, there is an
association with Whipple’s disease, congestive heart failure, right-heart valvular dis-
ease, or frankly obstructive lesions, such as abdominal lymphoma, retroperitoneal
fibrosis, retractile mesenteritis, mesenteric tuberculosis, and metastatic cancer.
B. Diagnosis. Patients commonly seek treatment for weight loss, diarrhea, and
edema caused by the decrease in plasma proteins. Some patients have chylous
ascites. Barium examination of the small intestine may be normal, may have an
appearance of nonspecific malabsorption, or may indicate a nodular mucosal pat-
tern caused by distended or infiltrated villi. Steatorrhea usually is mild. The xylose
tolerance test should be normal unless the underlying disease (e.g., lymphoma) has
infiltrated the mucosa. Small-bowel biopsy should confirm the diagnosis by iden-
tifying dilated lymphatics within the cores of the villi.
C. Treatment. In addition to undergoing treatment of any associated disorder that
may be responsible for the lymphatic obstruction, patients with lymphatic disease
should receive medium-chain triglycerides, limit their intake of long-chain triglyc-
erides, and take supplemental fat-soluble vitamins.
Selected Readings
Buchman AL, et al. AGA technical review on short bowel syndrome and intestinal
transplantation. Gastroenterology. 2003;124:1111–1134.
79466_CH31.qxd 1/2/08 12:24 PM Page 221
Catassi C, et al. Association of celiac disease and intestinal lymphomas and other cancers.
Gastroenterology. 2005;128:S79–S86.
Chitkara DK, et al. Gastrointestinal complications of cystic fibrosis. Clin Perspect
Gastroenterol. July/August 2000:201.
Cureton P, The gluten-free diet: can your patient afford it? Pract Gastroenterol. 2007;
xxxi(4):75–84.
Desai AA, et al. Bacterial overgrowth syndrome. Curr Treatment Options Infect Dis.
2003;5:189–196.
Dewar DH, et al. Clinical features and diagnosis of celiac disease. Gastroenterology.
2005;128:S19–S24.
Ginsburg PM, et al. Malabsorption testing: A review. Curr Gastroenterol Rep. 2000;2:370.
Graham DY, et al. Visible small intestinal mucosal injury in chronic NSAID users. Clin
Gastroenterol Hepatol. 2005;3:55–59.
Kagnoff MF. Overview and pathogenesis of celiac disease. Gastroenterology. 2005;128:
S10–S18.
Kupper C. Dietary guidelines and implementation for celiac disease. Gastroenterology.
2005;128:S121–S127.
Leung WK, et al. Small bowel enteropathy associated with chronic low-dose aspirin
therapy. Lancet. 2007;369:614.
Rostom A, et al. The diagnostic accuracy of serologic tests for celiac disease: A systemic
review. Gastroenterology. 2005;128:S38–S46.
Snow CF. Laboratory diagnosis of vitamin B12 and folate deficiency: A guide for the
primary care physician. Arch Intern Med. 1999;159:1289.
Southerland JC, et al. Osteopenia and osteoporosis in gastrointestinal diseases: Diagnosis
and treatment. Curr Gastroenterol Rep. 2001;3:399.
Sundaram A, et al. Nutritional management of the short bowel syndrome in adults. J Clin
Gastroenterol. 2003;34:207–210.
Swartz MN. Whipple’s disease—past, present, and future. N Engl J Med. 2000;342:647.
TABLE 31.1
79466_CH32.qxd 1/2/08 12:30 PM Page 222
32 SMALL-INTESTINAL NEOPLASMS
AND CARCINOID TUMORS
N eoplasms of the small intestine, either benign or malignant (Table 32-1), are unusual
but not rare, comprising less than 5% of all gastrointestinal tumors. Because they are
uncommon and relatively inaccessible to standard diagnostic studies, the diagnosis of
small-bowel tumors is sometimes delayed.
I. DIAGNOSIS
A. Clinical presentation. Small-intestinal (SI) tumors usually occur in people over
age 50. The presenting signs and symptoms are similar whether the tumors are
benign or malignant. Small-bowel obstruction, either partial or complete, mani-
fested by abdominal pain or vomiting or both, is a frequent presentation. Chronic
partial obstruction may predispose to stasis and bacterial overgrowth, leading to
bile acid deconjugation and malabsorption (see Chapter 31). Bleeding from the
tumor or ulceration in association with the tumor also is common. Perforation of
the bowel is rare. If a duodenal tumor is located in the vicinity of the ampulla of
Vater, obstruction of the common bile duct may develop, resulting in biliary stasis
and jaundice. Weight loss commonly accompanies malignant tumors.
The physical examination usually is nondiagnostic. Signs of small-bowel
obstruction may be evident, such as a distended, tympanic abdomen and high-
pitched bowel sounds. Occasionally malignant small-bowel tumors can be pal-
pated. Stool may be positive for occult blood.
B. Laboratory and other diagnostic studies
1. Blood studies. Anemia may develop because of blood loss or malabsorption.
Hypoalbuminemia can result from malabsorption, extensive metastatic liver
disease, or chronic illness. Evaluation of serum alkaline phosphatase or biliru-
bin levels may indicate biliary obstruction or metastatic liver disease.
2. An upright plain x-ray film of the abdomen may show air-fluid levels in
patients with small-bowel obstruction. If the obstruction is acute and does not
resolve with nasogastric suction and intravenous fluids, surgery may be
required without additional diagnostic testing.
3. Barium-contrast x-ray studies. The patient who has a small-bowel tumor with-
out frank obstruction but perhaps with colicky pain or bleeding typically undergoes
Malignant Benign
Adenocarcinoma Adenoma
Lymphoma Leiomyoma
Leiomyosarcoma Lipoma
Carcinoid Hamartoma
Metastatic tumors Neurogenic tumors
Melanoma Endometrioma
Kaposi’s sarcoma Inflammatory polyp
222
79466_CH32.qxd 1/2/08 12:30 PM Page 223
Figure 32-1. Small-bowel x-ray series showing a polypoid filling defect in the proximal small
intestine (arrow). At surgical resection, the lesion was found to be a histiocytic lymphoma.
barium-contrast x-ray studies. The small-bowel series may show a polypoid lesion
(Fig. 32-1) or the typical “napkin ring” deformity of adenocarcinoma encircling the
bowel. Often the routine upper gastrointestinal x-ray series with small-bowel fol-
low-through is nondiagnostic because the lesion is obscured by the large amount of
barium in the small intestine. In these instances, a small-bowel enema, or entero-
clysis, may delineate the lesion. This procedure is performed by passing a small tube
into the proximal duodenum and instilling a small amount of barium with air. This
provides excellent air–barium contrast of the small intestine and enhances the diag-
nostic capability. For example, a diffuse nodular appearance of the mucosa on
barium-contrast study may suggest lymphoma.
4. Endoscopy and biopsy. If a lesion is within the duodenum, endoscopic exam-
ination and biopsy are indicated.
5. Capsule endoscopy may be used to directly visualize SI tumors. There is risk
of the capsule being retained if the tumor is large.
6. Ultrasound and computed tomography scan. In patients with biliary obstruc-
tion, abdominal ultrasound may help define the lesion and identify dilated bile
ducts. These patients also should undergo endoscopic retrograde cholangiopan-
creatography (ERCP) or percutaneous transhepatic cholangiography (PTC).
Computed tomography (CT) scanning of the abdomen is useful in delineating
mass lesions and surveying for metastatic disease to the liver, lymph nodes, and
mesentery.
7. Selective arteriography of the celiac axis and superior mesenteric artery may
be helpful to the surgeon in planning appropriate surgery.
II. TREATMENT
A. Surgery. The treatment of most small-bowel tumors, benign or malignant, is surgi-
cal. Even malignant tumors that have metastasized may require surgical palliation
79466_CH32.qxd 1/2/08 12:30 PM Page 224
III. CARCINOID TUMORS. In the United States, the incidence of carcinoid tumors is
approximately 8 per 100,000. Since most cases are asymptomatic and follow an indo-
lent course, the true incidence may be higher.
Carcinoid tumors are most commonly found in the appendix, ileum, rectum,
stomach, and lungs.
Carcinoid tumors are commonly classified according to their derivation from the
embryonic gut: foregut (bronchial and gastric), midgut (small intestine [SI] and
appendiceal), and hindgut (rectal). The clinical presentation and management of these
tumors varies according to their site of origin.
Of the foregut carcinoids, gastric carcinoids are usually asymptomatic and are
found incidentally; however, patients with bronchial carcinoids may present with cough,
hemoptysis, postobstructive pneumonia, Cushing’s syndrome, or carcinoid syndrome.
Midgut carcinoids, especially small-intestinal carcinoids, may cause inter-
mittent SI obstruction or mesenteric ischemia. Appendiceal carcinoids are usually
found incidentally. If there are metastases, carcinoid syndrome may occur.
Hindgut or rectal carcinoids are usually found incidentally, but may cause consti-
pation and rectal bleeding. Carcinoid syndrome is rarely seen even if there are metastases.
A. Bronchial carcinoid tumors comprise about 2% of primary lung tumors. They
usually present in the fifth decade of life and are rarely associated with carcinoid
syndrome; however, they have been associated with ectopic ACTH secretion result-
ing in Cushing’s syndrome. One third of patients, especially smokers, may present
with atypical carcinoids which are much more aggressive and usually metastasize
to the mediastinal lymph nodes. Surgery may be indicated.
B. Gastric carcinoid tumors make up less than 1% of gastric neoplasms. They are
classified into three distinct groups:
1. Type I: Associated with chronic atrophic gastritis
2. Type II: Associated with the Zollinger-Ellison syndrome and ectopic gastric
secretion
3. Type III: Sporadic gastric carcinoids
Both type I and type II gastric carcinoids are associated with
achlorhydria and hypergastrinemia, which is thought to result in hyperplasia
of the enterochromaffin cells in the stomach, leading to small, multiple carci-
noid tumors. These tumors generally follow an indolent course and are rarely
invasive. The small tumors may be resected endoscopically. Larger or recur-
rent tumors may require more extensive surgery. In patients with atrophic
gastritis, antrectomy has been used to eliminate the source of the gastrin pro-
duction to achieve tumor regression. In patients with Zollinger-Ellison syn-
drome, the use of somatostatin analogs has resulted in tumor regression.
Between 15% and 25% of gastric carcinoids are sporadic and develop
in the absence of hypergastrinemia. These are usually solitary and greater than
1 cm in size. They are frequently invasive, metastatic, and tend to pursue an
aggressive clinical course. Sporadic gastric carcinoids have been associated with
79466_CH32.qxd 1/2/08 12:30 PM Page 225
2. Treatment
a. Surgical resection of the involved liver segment(s), when possible, may
provide long-term symptomatic relief and prolonged survival times.
b. Liver transplantation may be offered to patients with liver-isolated
metastatic disease; however, the role of OLT in such patients is unclear.
c. Hepatic artery embolization may be used as a palliative procedure in
patients with liver metastases who are not candidates for surgical resection.
However, the duration of response may be brief, ranging from 4 to 24
months. Side effects may include renal failure, hepatic necrosis, and sepsis.
d. Radiofrequency ablation and cryoablation (either alone or in tandem
with surgical resection) may be offered as less invasive procedure; however,
efficacy, especially in patients with extensive hepatic metastases, has not
been well studied.
e. Systemic therapy
i. Somatostatin analogs (i.e., octreotide) at a dose of 150 g t.i.d.
improves the symptoms of carcinoid syndrome in nearly 90% of patients.
Long-acting octreotide at a dose of 20 mg intramuscularly may be
administered once monthly with gradual increase in dose as needed to
control symptoms. Patients may also use additional short-acting
octreotide for breakthrough symptoms. Lanreotide (another somato-
statin analog) has similar clinical efficacy. These drugs are well tolerated
by patients; however, possible side effects include injection-site reactions,
stearrhea, and hyperglycemia.
ii. Interferon- (IFN) may be used alone or in combination with somato-
statin analogs. The addition of IFN- therapy to somatostatin analogs
has been reported to be effective in controlling symptoms in patients
with carcinoid syndrome who may be resistant to somatostatin analogs
alone. The combination of these drugs may also significantly slow the
rate of tumor progression in the majority of patients. Side effects of IFN-
therapy include bone marrow suppression (especially myelosuppression),
autoimmune thyroid disease, fatigue, and depression.
iii. Chemotherapy. Streptozocin combined with fluorouracil (5-FU),
cyclophosphamide, or doxorubicin have been used in patients with
metastatic carcinoid disease with poor results. Even though there may be
slight survival benefit with streptozocin and 5-FU, the renal toxicity,
myelosuppression, nausea, vomiting, and fatigue limits the use of these
drugs as first-line therapy.
iv. Newer agents in the treatment of metastatic carcinoid include radiola-
beled somatostatin analog drugs, inhibiting binding of proangiogenic
growth factor (vascular endothelial growth factor, or VEGF) to its
receptor (vascular endolethial growth factor receptor, or VEGFR)
with blocking antibodies such as bevacizumab (Avastin) and sunitinib
(Sutent).
Selected Readings
Eamonn MM, et al. Small intestinal transplantation. Curr Gastroenterol Rep. 2001;3:408.
Gill SS, et al. Small intestinal neoplasms. J Clin Gastroenterol. 2001;33:267–282.
O’Neil BH. Management of carcinoid tumors and the carcinoid syndrome. Clin Perspect
Gastroenterol. 2001;4:279..1
79466_CH33.qxd 1/2/08 12:30 PM Page 227
T he irritable bowel syndrome (IBS) is the most common of all digestive disorders,
affecting nearly everyone at one time or another and accounting for up to 50% of patients
referred to a gastroenterology practice. Although characterized as a disorder of bowel
motility, in many patients it usually is an exaggeration of normal physiologic responses and
possibly heightened perception of pain.
Numerous terms have been used to describe the syndrome (Table 33-1). Irritable
bowel syndrome seems to be the most appropriate. Terms that include the words colon
or colitis are inaccurate because the condition is not limited to the colon, and inflamma-
tion is not a feature. Furthermore, use of the term colitis leads to confusion with ulcerative
colitis and conveys an inaccurate impression to the patient.
In many patients IBS may be characterized as diarrhea predominant (IBS-D),
constipation predominant (IBS-C), and for some patients, an alternation of diarrhea
with constipation (alternators).
227
79466_CH33.qxd 1/2/08 12:30 PM Page 228
asymptomatic individuals and may even become stronger. Emotional stress also
induces colonic motor activity, both in healthy individuals and in patients with
IBS, but it is possible that symptoms are perceived to a greater degree in patients
with IBS. Balloon distention of the rectosigmoid colon in patients with IBS causes
spastic contractions of greater amplitude than in asymptomatic subjects.
Furthermore, there is evidence that patients with IBS who complain of gaseous
distention and abdominal cramps cannot tolerate quantities of small-bowel intra-
luminal gas that are easily tolerated by healthy individuals (see Chapter 34).
II. DIAGNOSIS
A. Clinical presentation
1. Symptoms. Patients with IBS typically complain of crampy abdominal pain,
diarrhea, or constipation. In some patients, chronic constipation is punctuated
by brief episodes of diarrhea. A minority of patients have only diarrhea.
Symptoms usually have been present for months to years, and it is common for
patients with IBS to have consulted several physicians about their complaints
and to have undergone one or more gastrointestinal evaluations.
2. Timing of symptoms. The patient may be able to correlate symptoms with
emotional stress, but often such a relation is not evident or becomes apparent
only after careful questioning as the physician becomes acquainted with the
patient. If abdominal cramps are a feature, they often are relieved temporarily
by defecation. Bowel movements may be clustered in the morning or may occur
throughout the day, but rarely is the patient awakened at night. Stools may be
accompanied by an excessive amount of mucus, but blood is not present unless
there is incidental bleeding from hemorrhoids.
3. The differential diagnosis is broad, including most disorders that cause diar-
rhea and constipation (see Chapters 28, 29, 30, and 35). However, there are several
features that suggest the diagnosis of IBS (Table 33-2). Several organic disorders
may mimic IBS and, in fact, may be unrecognized for years in patients who
mistakenly have been diagnosed as having IBS. Patients with lactose intoler-
ance typically have postprandial diarrhea associated with crampy pain (see
Chapter 31). They are healthy in all other respects. A careful history and a trial
of a lactose-free diet usually are sufficient to make a diagnosis. Celiac sprue,
Crohn’s disease, and endometriosis also can masquerade as IBS because of the
vagueness of the symptoms in many patients. A clinical history of postprandial
abdominal pain suggests the possibility of gallbladder, pancreatic, or peptic
disease. Because IBS may affect the entire digestive tract, belching and symp-
toms of gastroesophageal reflux and dyspepsia are common in patients with IBS.
Anorexia, weight loss, fever, rectal bleeding, and nocturnal diarrhea
all suggest a cause other than IBS for the patient’s symptoms. The physician
Characteristic Uncharacteristic
patients with IBS experience no relief merely from reassurance. Many have carried
the diagnosis of IBS for years and continue to experience distressful symptoms
despite supportive reassurance and diet and drug therapy. Although these patients
often understand that they have a “nervous bowel,” that understanding does little
to alleviate symptoms, and they continue to seek treatment. Stress reduction pro-
grams may be helpful.
B. Diet and fiber therapy. The commonsense approach to diet therapy is the most
appropriate. There is no need for bland or highly restrictive diets in the treatment
of IBS. Patients should avoid foods that they find cause symptoms. If lactose-
containing foods produce cramps and diarrhea, these should be eliminated from
the diet.
The role of fiber in the treatment of IBS has been controversial. However,
clinical experience suggests that a high-fiber diet and/or fiber supplements provide
symptomatic relief in some patients. Patients with crampy abdominal pain and
constipation seem most likely to benefit, although sometimes patients with watery
diarrhea also experience a firming of their stools after the fiber content of the diet
has been increased. However, fiber supplements may result in increased gas and
bloating due to bacterial fiber.
C. Drug therapy
1. Antispasmodics. Unfortunately, drug therapy of IBS often is empiric. Patients
with diarrhea and abdominal pain may benefit from a so-called antispasmodic.
These drugs are anticholinergic in their mode of action, but whether they actually
relieve spasm is conjectural. A reasonable choice is dicyclomine hydrochloride,
10 to 20 mg three to four times daily because it is less likely than others to
cause unpleasant nongastrointestinal anticholinergic side effects. Regardless of
what preparation is used, patients should be cautioned about the possibility of
the development of dry mouth, blurred vision, and urinary retention.
2. Laxatives should be used judiciously in the treatment of the IBS with constipa-
tion (IBS-C). However, many patients with constipation become dependent on
the long-term use of laxatives and may need to be withdrawn from these agents.
3. Tegaserod (Zelnorm), a partial 5-HT4 agonist, has clinical efficacy in the treat-
ment of IBS-C in female patients up to age 65. It facilitates the secretion of
serotonin within the intestinal wall and thus enhances peristalsis through the
GI tract. It is contraindicated in patients with cardiovascular risk factors.
(Zelnorm has been removed from the U.S. market by the manufacturer for fur-
ther evaluation of its safety in long-term use.)
79466_CH33.qxd 1/2/08 12:30 PM Page 231
Selected Readings
Cash BD, et al. Fresh perspectives in chronic constipation and other functional bowel
disorders. Rev. Gastroenterol Disord. 2007;7(3):116–133.
Hammer J, et al. Disturbed bowel habits in patients with non-ulcer dyspepsia. Ailment
Pharmacol Ther. 2006;24:405–I0.
Henningsen P, et al. Management of functional somatic syndromes. Lancet. 2007;369:
946–955.
Johanson JF, et al. A dose-ranging, double-blind, placebo-controlled study of lubiprostone
in subjects with irritable bowel syndrome and constipation (c-IBS). Gastroenterology.
2006;130(4 suppl 2):A-131.
Longstreth GF, et al. Functional bowel disorders. Gastroenterology. 2006;130:1480–1491.
Marshall JK, et al. Post-infectious irritable bowel syndrome (IBS) after the Walkerton
outbreak of waterborne gastroenteritis (GE). Gastroenterology. 2006;130(4 suppl 2):
A-T1160.
Pimentel M, et al. A 10 day course of rifaximin, a nonabsorbable antibiotic, produces a
durable improvement in all symptoms of irritable bowel syndrome: A double-blind,
randomized controlled study. Gastroenterology. 2006;130(4 suppl 2):A-134.
Quigley EMM. The use of probiotics in functional bowel disease. Gastroenterol Clin
North Am. 2005;34:533–545.
Saier MH Jr, et al. Priobiotics and prebiotics in human health. J Mol Microbiol Biotechnol.
2005;10:22–25.
Sander DS, et al. Association of adult coeliac disease with irritable bowel syndrome:
A case-controlled study in patients fulfilling ROME II criteria referred to secondary
care. Lancet. 2001;358:1504–1508.
Spiller R. Clinical update: irritable bowel syndrome. Lancet. 2007;369:1586–1588.
Tack J, et al. A randomized controlled trial assessing the efficacy and safety or repeated
tegaserod therapy in women with irritable bowel syndrome with constipation. Gut.
2005;54(12):1707–1713.
Tack J, et al. Functional gastroduodenal disorders. Gastroenterology. 2006;I30:I466–79.
79466_CH34.qxd 1/2/08 12:31 PM Page 232
G as occurs normally within the gastrointestinal tract, yet many patients complain of
excessive gas. The complaint of gas has no uniform connotation. Some patients mean that
they belch too much, others experience abdominal discomfort and attribute it to gas,
whereas still others regard the amount of flatus passed as being excessive.
232
79466_CH34.qxd 1/2/08 12:31 PM Page 233
Figure 34-1. Diagram of the gastrointestinal tract, indicating the sources and composition of
intestinal gas. (From Eastwood GL. Core Textbook of Gastroenterology. Philadelphia: Lippincott
Williams & Wilkins; 1984:188. Reprinted with permission.)
II. PATHOGENESIS. Patients with gaseous symptoms may complain of belching, abdom-
inal pain and bloating, or of excessive flatus. The pathogenesis of each condition is
different. Also, there is considerable overlap in the patients who complain of gas and
bloating and those with irritable bowel syndrome (see Chapter 33).
A. Belching is caused by the eructation of swallowed air. If the normal volume of
swallowed air cannot be passed into the proximal small bowel because of a motil-
ity disorder, gastroparesis or gastric outlet obstruction, or due to an incompetent
lower esophageal sphincter (LES), belching may develop. Thus, patients with gas-
troesophageal reflux disease (GERD), gastric carcinoma, peptic ulcer disease, or
uremia may complain of belching.
Patients with gallbladder disease often belch for unknown reasons.
Sometimes belching is a nervous habit, and the swallowed air may not even reach
the stomach before eructation takes place. Rarely, belching of feculent-smelling gas
indicates chronic gastric stasis or a gastrocolic fistula.
79466_CH34.qxd 1/2/08 12:31 PM Page 234
B. Abdominal pain and bloating are often attributed to an excess of gas. In some
instances, failure to adequately digest a component of food (e.g., lactase defi-
ciency) with consequent bacterial degradation and gas production is responsible
for the abdominal pain and bloating. In other patients, the cause of symptoms is
unclear. When patients with abdominal pain and bloating who did not have spe-
cific food intolerance were compared to healthy people, they did not differ with
respect to volume or content of intestinal gas, either under fasting conditions or
after a meal. However, patients with gaseous complaints seem to have more reflux
of gas from the duodenum into the stomach and experience painful symptoms
after infusion of gas into the proximal small bowel in volumes that are well toler-
ated by healthy people. Thus, patients who complain of bloating and abdominal
gas may be suffering from a defect in gut motility and a decreased threshold to
pain rather than from too much gas. As indicated previously, some of these patients
may suffer from irritable bowel syndrome.
C. Excessive flatulence may result from a disorder of intestinal motility but usually
is a consequence of increased amounts of gas produced by the action of bacteria
on dietary substrates. Lactase deficiency is a good example, although the list of
dietary substrates that could be implicated is virtually endless. Bacterial degrada-
tion of undigested carbohydrates and simple sugars also cause gas and bloating.
III. DIAGNOSIS
A. Belching. A careful history may identify nervous air swallowing or ingestion of
excessive quantities of carbonated beverages. An abdominal ultrasound examina-
tion to evaluate the gallbladder and either an upper gastrointestinal (GI) x-ray series
or endoscopy to evaluate the proximal gastrointestinal tract may be indicated.
B. Abdominal pain and bloating. Patients who complain of abdominal pain and
bloating present a different diagnostic problem. Some have intolerance for a
specific dietary component, such as lactose. Other organic diseases must be con-
sidered, such as celiac sprue, regional enteritis, recurrent partial small-bowel
obstruction, or even giardiasis. The evaluation of such patients usually includes
ultrasound of the abdomen, Esophagogastroduodenoscopy (EGD), biopsy,
upper GI and small-bowel x-ray series, and, if stools are loose, stool examina-
tion for ova, parasites, and bacterial pathogens and possibly colonoscopy and
biopsy.
C. Excessive flatus. Complaints of excessive flatus should be evaluated by a search
for a specific food or group of foods that may be causing symptoms.
IV. TREATMENT. Unless a specific diagnosis such as gallstones, peptic ulcer, or lactase
deficiency is made, the treatment of patients who complain of gas is often unreward-
ing. Patients with nervous air swallowing may improve simply by becoming aware of
their air swallowing. Elimination of milk, legumes, cabbage, and similar foods, and
white-sugar- and flour-containing foods may be effective in patients who complain of
abdominal pain, gas, bloating, or flatus. Extensive elimination diets are helpful in
occasional patients who have the perseverance to eliminate one dietary constituent at
a time each week for several weeks.
In patients who do not have a demonstrable, treatable disorder or in whom a spe-
cific food has not been implicated, so-called antispasmodic agents, such as dicyclomine
hydrochloride, or bulking agents sometimes are useful. However, bulking agents,
because they contain nondigestible substrates, have the potential for increasing
flatus. Some patients have responded to stress-reduction techniques or psychological
counseling.
Selected Readings
Camilleri M. Treating irritable bowel syndrome: Overview, perspective and future
therapies. Br J Pharmacol. 2004;141:1237–1248.
Comilleri M. Diabetic gastroparesis. New Engl J Med. 2007;356:820–829.
Floch MH. Use of diet and probiotic therapy in the irritable bowel syndrome: Analysis of
the literature. J Clin Gastroenterol. 2005;39(4 suppl 3):S243–S246.
79466_CH34.qxd 1/2/08 12:31 PM Page 235
Levitt MD, et al. The relation of passage of gas and abdominal bloating to colonic gas
production. Ann Intern Med. 1996;124:422.
Levitt MD, et al. Evaluation of an extremely flatulent patient, case report and proposed
diagnostic and therapeutic approach. Am J Gastroenterol. 1998;93:2276.
Pimentel M, et al. Eradication of small intestinal bacterial overgrowth reduces symptoms
of irritable bowel syndrome. Am J Gastroenterol. 2000;95(12):3503–3506.
Poredenoord AJ, et al. Physiologic and pathologic belching. Clin Gastroenterol Hepatol.
2007;5(7):772–775.
Reddymason S, et al. New methodology in assessing gastric emptyizing and gastrointestinal
transit. US Gastroenterol Rev. 2007;1:19–22.
Reid G. Porbiotics in gastrointestinal management—what’s new. US Gastroenterol Rev.
2007;1:66–75.
Serra J, et al. Intestinal gas dynamics and tolerance in humans. Gastroenterology. 1998;
115:542.
Spiller R. Clinical update: irritable bowel syndrome. Lancet. 2007;369:1586–1588.
Spiller RC. Postinfectious irritable bowel syndrome. Gastroenterology. 2003;124:1662–1671.
Suarez FL, et al. An understanding of excessive intestinal gas. Curr Gastroenterol Rep.
2000;2:413.
Suarez FL, et al. Intestinal gas. Clin Perspect Gastroenterol. July/August 2000:209.
79466_CH35.qxd 1/2/08 12:32 PM Page 236
35 DIVERTICULAR DISEASE
T he term diverticular disease generally refers to diverticulosis of the colon and its
complications. However, diverticula can occur throughout the gastrointestinal tract. In the
esophagus, diverticula may be associated with dysphagia (see Chapter 21). Diverticula of
the stomach usually are asymptomatic. In the small intestine, they can predispose to bac-
terial overgrowth and malabsorption (see Chapter 31). In this chapter, we confine the dis-
cussion to diverticula of the colon.
Colonic diverticulosis is exceedingly common in westernized countries. The preva-
lence of diverticula increases with age: About 30% of the general population at age 60 and
about 80% at age 80 have colonic diverticula. Although 90% of people with diverticula
remain asymptomatic, because of the high prevalence of the condition, symptomatic diver-
ticular disease occurs frequently.
I. DEFINITIONS. Several of the terms used to describe colonic diverticula and their com-
plications have caused confusion or have been misused. Thus, it is important to review
some terminology (Table 35-1).
A single outpocketing from the bowel wall is called a diverticulum. Because
diverticulum is a Latin neuter word, it ends in um, and its plural form ends in a. Thus
several outpocketings are referred to as diverticula, not diverticuli or diverticulae.
The presence of a diverticulum or diverticula is called diverticulosis. Diverticu-
losis does not imply a pathologic condition or set of symptoms. Diverticulitis means
inflammation in one or more diverticula. Sometimes the diagnosis of diverticulitis is
assumed on the basis of clinical symptoms (see section IV.A.1) that may overlap with
the symptoms of irritable bowel syndrome (see Chapter 33) and gaseousness (see
Chapter 34). In the absence of firm evidence for diverticulitis, perhaps a more general
term should be used, such as diverticular disease or symptomatic diverticulosis.
II. PATHOGENESIS. True diverticula, that is, those in which the walls contain all layers
of the bowel, are found occasionally in the colon. However, the most prevalent colonic
diverticula are pseudodiverticula. These are herniations of mucosa and submucosa
through the muscularis propria at the sites of penetration of the nutrient arteries. The
development of diverticula seems to be related to increased pressure within the lumen
of the bowel, which over time causes the herniations. Most diverticula occur in the
sigmoid and descending colons, although they also may be found more proximally.
Epidemiologically, the prevalence of colonic diverticulosis is related to age and
diet. The condition is much more frequent in populations in which dietary fiber has
been replaced by refined carbohydrates.
Diverticulum—singular
Diverticula—plural
Diverticulosis—the presence of one or more diverticula
Diverticulitis—inflammation in one or more diverticula
Diverticular disease—any complication of diverticulosis
236
79466_CH35.qxd 1/2/08 12:32 PM Page 237
Figure 35-1. Barium enema x-ray examination showing extensive diverticulosis throughout the
colon. Note the numerous barium-filled pockets protruding outside the bowel lumen. The distortion
of the normal mucosal architecture by diverticulosis sometimes makes it difficult to exclude carci-
noma or polyp by barium enema examination.
79466_CH35.qxd 1/2/08 12:32 PM Page 238
3. The differential diagnosis includes the irritable bowel syndrome, frank diver-
ticulitis (see section IV), carcinoma of the colon, Crohn’s disease or proctocol-
itis, urologic disorders, and gynecologic disorders.
B. Treatment. In many respects, the treatment of uncomplicated symptomatic diver-
ticulosis resembles the treatment of irritable bowel syndrome (see Chapter 33).
A high-fiber diet has been shown to alleviate the discomfort of diverticular disease.
Fiber also can be added in the form of unprocessed bran and other hydrophilic
bulk laxatives, such as Metamucil. Antispasmodic–anticholinergic drugs, such as
dicyclomine hydrochloride, may be useful in the treatment of crampy abdominal
pain. In constipated patients, lubiprostone (Amitiza) may be preferred to induce
softer stools. Cathartic laxatives should be avoided.
IV. DIVERTICULITIS
A. Diagnosis
1. Clinical presentation. The point at which symptomatic diverticulosis
becomes diverticulitis may be difficult to determine in some patients. Fully
developed diverticulitis is characterized by acute lower abdominal pain,
fever, and tachycardia. The lower abdomen is tender to palpation, and there
may be rebound tenderness. The patient may present with an acute abdomen
(see Chapter 13); more typically, symptoms evolve over several hours or
days. A mass in the lower abdomen may connote an abscess or inflammatory
phlegmon. Bowel sounds may be active if partial or complete obstruction
has occurred, or they may be hypoactive or absent if peritonitis has devel-
oped. A rectal examination may help localize the abscess or inflammatory
mass.
In some patients with chronic diverticulitis, chronic inflammatory changes
may develop that resemble inflammatory bowel disease (IBD) or Crohn’s disease.
2. Diagnostic studies
a. Blood studies. The white blood cell count typically is elevated in acute diver-
ticulitis, with a predominance of immature forms (left shift) in the differential
count. Hemoglobin and hematocrit levels may reflect hemoconcentration.
b. A urinalysis may show white blood cells and red blood cells. An unusual
complication of diverticulitis is a colonic–urinary bladder (colovesical) fis-
tula; in this condition, the urine contains large numbers of white blood cells
and bacteria and possibly feces. Patients with colovesical fistulas frequently
complain of pneumaturia.
c. Plain abdominal x-ray films, both supine and upright, should be obtained.
Air-fluid levels suggest ileus or obstruction. Free air in the abdomen, indi-
cating a perforated diverticulum, may be evident on lateral decubitus
abdominal films or under the diaphragm on the upright chest x-ray film.
d. Computed tomography (CT) scan and/or ultrasound of the abdomen and
pelvis are helpful in identifying the inflammatory mass or an abscess cavity
and by demonstrating other conditions in the differential diagnosis, such as
an ovarian cyst.
e. Sigmoidoscopy, colonoscopy, and barium enema x-ray examinations.
Sigmoidoscopy or colonoscopy may be performed cautiously if perforation
is not suspected. However, it is best to delay these tests until symptoms have
subsided. The radiologic diagnosis of diverticulitis by barium enema
requires evidence of perforation of a colonic diverticulum by demonstrating
either a fistulous tract or an abscess cavity. Although these findings are
unequivocal, they are not necessary to make the clinical diagnosis of diver-
ticulitis. Acute lower abdominal pain in association with fever and an ele-
vated white blood cell count in a person with demonstrated diverticula are
sufficient.
3. The differential diagnosis includes inflammatory bowel disease, infectious
C. difficile or ischemic colitis, carcinoma of the colon, other causes of bowel
obstruction, gynecologic disorders (e.g., ruptured ovarian cyst), and urologic
disorders (e.g., renal colic).
79466_CH35.qxd 1/2/08 12:32 PM Page 239
B. Treatment
1. General. Patients with severe acute diverticulitis are best treated by allowing
nothing by mouth and administering intravenous fluids and electrolytes.
Intravenous broad-spectrum antibiotics, such as the combination of ampicillin,
gentamicin, and ciprofloxacin hydrochloride or second- or third-generation
cephalosporin and metronidazole, are indicated and should be continued for 10
to 14 days. If an abscess is identified by ultrasonography or CT scan, percuta-
neous drainage under ultrasound or CT guidance should be considered.
Patients whose symptoms are less severe may be treated with oral antibiotics
(e.g., ciprofloxacin hydrochloride 500 mg p.o. b.i.d. or Levaquin 500 mg q.d.
and metronidazole 250–500 mg p.o. t.i.d. for 10–14 days).
2. Surgery. Few would argue that generalized peritonitis, with or without evi-
dence of free perforation, should be treated surgically. Unresolved obstruction
and colovesical fistula also are indications for surgical treatment. Because most
patients with uncomplicated diverticulitis recover with medical treatment and
do not have recurrences of acute disease, surgery is not recommended routinely.
However, failure to improve after several days of medical treatment or recur-
rence after successful treatment are indications for surgery in a patient whose
operative risk is reasonable.
A one-stage procedure with resection and primary anastomosis is ideal. If
active infection is evident, a primary resection with a proximal colostomy, fol-
lowed several months later by reanastomosis, may be indicated. Alternatively,
the surgeon may elect to perform a proximal diverting colostomy as the pri-
mary operation, allowing the infection and inflammation to resolve, before
proceeding at a later date with the resection and anastomosis.
Selected Readings
Anaya DA, et al. Risk of emergency colectomy and colostomy in patients with diverticular disease. Arch
Surg. 2005;140:681–685.
Chapman JR, et al. Diverticulitis: a progressive disease? Do multiple recurrences predict less favorable
outcomes? Ann Surg. 2006;243:876–883.
Constantinide VA, et al. Primary resection with anastomosis is Hartman’s procedure in nonelectric surgery
for acute colonic diverticulitis: a systematic review. Dis Colon Rectum. 2006;49:966–981.
Jacobs DO. Diverticulitis. N Eng J Med. 2007;357:2057–2066.
Jensen DM, et al. Urgent colonoscopy for diagnosis and treatment of severe diverticular hemorrhage.
N Engl J Med. 2000;342:78.
Korzenik JR. Case closed? Diverticulitis: epidemiology and fiber. J Clin Gastroenterol. 2006;40:
Suppl 3:S112–S116.
Kumar RR, et al. Factors affecting the successful management of intraabdominal abcesses with antibiotics
and the need for percutaneous drainage. Dis Colon Rectum. 2006;49:183–189.
Mueller MH, et al. Long-term outcome of conservative treatment in patients with divrticulitis of the
sigmoid colon. Eur J Gastroenterol Hepatol. 2005;17:649–654.
Parra-Blanco A. Colonic diverticular disease: pathophysiology and clinical picture. Digestion. 2006;
73(suppl)1:47–57.
Rafferty J, et al. Buie and the Standards Committee of the American Society of Colon and Rectal
Surgeons. Practice parameters for sigmoid diverticulitis. Dis. Colon Rectum. 2006;49:939–944.
Ramirez FC, et al. Successful endoscopic hemostasis of bleeding colonic diverticula with epinephrine
injections. Gastrointest Endosc. 1996;43:167.
Salzman H, et al. Diveticular disease: diagnosis and treatment. Am Fam Physician. 2005;72:1229–1234.
Stollman N, et al. Diverticular disease of the colon. Lancet. 2004;363:631–639.
Zaidi E, et al. CT and clinical features of acute diverticultis in an urban U.S. population: rising frequency
in young, obese adults. AJR Am J Roentgenol. 2006;187:689–694.
79466_CH36.qxd 1/2/08 12:33 PM Page 240
C onstipation, like diarrhea, is difficult to define with precision due to the wide vari-
ation in normal bowel habits. However, because 95% of people have at least three bowel
movements per week, for practical purposes constipation can be defined as a condition in
which fewer than three stools per week are passed. In addition, patients may experience
difficulty in passing stools, may use manual maneuvers (i.e., digital disimpaction, pressing
on the perineum and/or lower abdomen), and may complain of hard and lumpy stools.
The economic costs of constipation are impressive. In the United States, more than
$250 million is spent annually on laxatives. Additional costs of unknown magnitude are
incurred in the evaluation of patients for underlying disorders that may predispose to
constipation.
II. DIAGNOSIS
A. Clinical presentation
1. History. To paraphrase a common saying, “One man’s constipation is another
man’s diarrhea.” Thus it is important that the physician determine what the
patient means by constipation. How frequently are stools passed? What is their
240
79466_CH36.qxd 1/2/08 12:33 PM Page 241
“Functional” constipation
Conditions in which constipation is part of the symptom complex
Irritable bowel syndrome
Symptomatic diverticular disease
Disorders of bowel muscle function (e.g., pseudoobstruction, myotonic dystrophy,
systemic sclerosis)
Disorders that can cause constipation
Endocrine–metabolic conditions (e.g., diabetes, hypokalemia, hypothyroidism,
hypercalcemia, pregnancy)
Bowel obstruction (e.g., neoplasm, benign stricture, ischemic colitis, surgical stenosis)
Disorders causing painful passage of stool (e.g., proctitis, anal fissure)
Neurogenic disorders (e.g., Hirschsprung’s disease [aganglionosis], autonomic neuropathy,
multiple sclerosis, Parkinson’s disease)
Drugs (e.g., calcium channel blockers, nitrates, anticholinergics, antidepressants, opiates)
of action results in soft stools and predictable bowel function. In the studies,
there were no electrolyte or fluid imbalances or drug–drug interactions.
Recommended dose is 24 g p.o. b.i.d. Clinical studies show that it is also
effective in the treatment of IBS-C.
IV. FECAL IMPACTION. Impaction of a firm, immovable mass of stool is found most
often in the rectum but may occur within the sigmoid or descending colon. Fecal
impaction typically develops in elderly, inactive patients, but the differential diagnosis
is the same as for ordinary constipation, ranging from functional constipation to
hypotonic bowel disorders to distal bowel obstruction (Table 36-1).
Regardless of the underlying cause, the treatment consists of several
approaches. First, the impaction may be broken manually during digital rectal exam-
ination. If that attempt is not completely successful, the mass can be softened and
evacuated by warm water or saline lavage through a sigmoidoscope or rectal tube.
Sometimes glycerine suppositories or mineral oil enemas are useful. Oral mineral oil
may be administered if there is no risk of aspiration. Occasionally dilatation of the
anus under general anesthesia is used to gain access to the fecal impaction. Rarely,
surgical removal of the impaction is necessary.
Selected Readings
Brandt LJ, et al. Systematic review on the management of chronic constipation in North
America. Am J Gastroenterol. 2005;100:S5–S22.
Camilleri M, et al. Effect of a selective chloride activator, lubiprostone, on gastrointestinal
transit, gastric sensory and motor functions in healthy volunteers. Am J Physiol
Gastrointest Liver Physiol. 2006;290:G942–G947.
Cash BD, et al. Fresh perspectives in chronic constipation and other functional bowel
disorders. Rev. Gastroenterol Disord. 2007;7(3):116–133.
Chiaroni G, et al. Biofeedback is superior to laxatives for normal transit constipation due
to pelvic floor dyssynergia. Gastroenterology. 2006;130:675–664.
Dukas L, et al. Association between physical activity, fiber intake, and other lifestyle variable
and constipation in a study of women. Am J Gastroenterol. 2003;98:1790–1796.
Ehrenpreis E. Constipation, colonic inertia and colonic marker studies. Pract Gastroenterol.
2001;24:18.
Fernandez-Banares F. Nutritional care of the patient with constipation. Best Pract Res Clin
Gastroenterol. 2006;20:575–587.
Kamm MA, et al. Tegaserod for the treatment of chronic constipation: A randomized,
double-blind placebo-controlled multinational study. Am J Gastroenterol. 2005;100:
362–372.
Kamm, MA. Clinical case: Chronic constipation. Gastroenterology. 2006;131:233–239.
Lembo A, et al. Chronic constipation. N Engl J Med. 2003;349:1360–1368.
Muller-Lissner SA, et al. Myths and misconceptions about chronic constipation. Am J
Gastroenterol. 2005;100:232–242.
Rao SS. Constipation: Evaluation and treatment. Gastroenterol Clin North Am. 2003;32:
659–683.
Talley NJ. Definition, epidemiology, and impact of chronic constipation. Rev Gastroenterol
Disord. 2004;4(suppl 2):S3–S10.
Ueno R, et al. Evaluation of safety and efficacy in a twelve-month study of lubiprostone for
the treatment of chronic idiopathic constipation. Am J Gastroenterol. 2006;101:S491.
Abstract 1269.
Wald A. Chronic constipation: advances in management. Neurogastroenterol Motil. 2007;
19:4–10.
79466_CH37.qxd 1/2/08 12:35 PM Page 244
I nflammatory bowel disease (IBD) refers to the idiopathic inflammatory bowel dis-
orders, ulcerative colitis, Crohn’s disease, and microscopic or lymphocytic and collagenous
colitis. Clearly, a number of other conditions also are associated with inflammation, includ-
ing bacterial and parasitic infections, ischemic bowel disease, and radiation colitis.
Nevertheless, until the causes of ulcerative colitis and Crohn’s disease are identified, the
term inflammatory bowel disease serves a useful purpose in distinguishing these conditions
from other bowel disorders.
IBD has a prevalence of 0.3% to 0.5% in the adult U.S. population with a slight
female preponderance. It is most commonly seen in young patients between the ages of
15 and 25; however, there is second peak in the incidence of IBD at 40 to 60 years.
Approximately 15% of patients with IBD have close relatives who also have IBD.
Pain crampy, lower abdominal, relieved by Pain constant, often in right lower quadrant,
bowel movement not relieved by bowel movement
Bloody stool Stool usually not grossly bloody
No abdominal mass Abdominal mass, often in right lower quadrant
Affects only colon May affect small and large bowel, occasionally
esophagus and stomach
Mucosal disease (granulomas are Transmural disease (granulomas found in a
not a feature) minority of patients)
Continuous from rectum May be discontinuous (skip areas)
244
79466_CH37.qxd 1/2/08 12:35 PM Page 245
varies with the site and degree of involvement. It may be gradual in onset or may pre-
sent with recurrent episodes of abdominal pain, diarrhea, and/or low-grade fever.
Physical examination may reveal a right lower quadrant mass or tenderness, anal fis-
sures, perianal abscess, or fistulas. Some 10% to 15% of patients with either ulcerative
colitis or Crohn’s disease have extraintestinal manifestations of the disease (Table 37-2).
In most patients, the two disorders can be distinguished on clinical, radiologic,
and pathologic grounds. However, in about 20% of patients with IBD affecting the
colon, it is impossible to make a definitive diagnosis (indeterminate colitis).
II. ETIOLOGY. Because the causes of IBD are not known, the pathophysiology of the
disorders is incompletely understood. An immunologic mechanism in the pathogenesis
is assumed, but the inciting causes are incompletely understood. The intestinal flora,
various cytokines including tumor necrosis factor (TNF) and some of the interleukins,
and other factors are thought to play a role in the ongoing inflammatory process.
Hereditary factors appear to play a role; patients with ulcerative colitis or
Crohn’s disease have a 10% to 15% chance of having a first- or second-degree rela-
tive who also has one or the other type of IBD.
abdominal pain, fever, bloody diarrhea with mucus, and many of the endoscopic,
radiologic, and histologic features of ulcerative colitis. The disease typically sub-
sides within several days but may run a protracted course, in which case treat-
ment with erythromycin or ciprofluxin may provide relief of symptoms.
Another pathogen that may complicate the diagnosis of ulcerative colitis is
C. difficile, the bacterial agent that has been implicated in antibiotic-associated
pseudomembranous colitis. C. difficile may be relevant to ulcerative colitis in two
ways. First, antibiotic-associated colitis may be confused with ulcerative colitis.
Second, C. difficile may be responsible for exacerbations of preexisting ulcerative
colitis. When chronic ulcerative colitis is in remission, the demonstration of C. dif-
ficile toxin in the stool is probably related to recent treatment with sulfasalazine
or antibiotics. Colonic infection with enteroinvasive E. coli, especially E. coli
O157:H7, may resemble ulcerative colitis and present with similar findings.
C. Flexible sigmoidoscopy or colonoscopy is indicated in the evaluation of rectal
bleeding of any cause. The normal rectal and colonic mucosa appears pink and
glistening. When the bowel is distended by insufflated air, the submucosal vessels
can be seen. Normally, there is no bleeding when the mucosa is stroked with a cot-
ton swab or touched gently with the tip of the sigmoidoscope.
In ulcerative colitis, the mucosal surface becomes irregular and granular. The
mucosa is friable, meaning that it bleeds easily when touched. With more severe
inflammation, bleeding may be spontaneous. These findings are nonspecific and
may be seen in most of the conditions listed in Table 37-3. In some patients with
chronic ulcerative colitis, pseudopolyps develop. The rectal mucosa is normal in
patients who have Crohn’s disease without rectal involvement. If the disease does
affect the rectum, the appearance may be similar to that of ulcerative colitis or may
include aphthous, deep or linear ulcerations and fissures.
D. Mucosal biopsy. Sigmoidoscopic or colonoscopic mucosal biopsies in patients
with IBD generally are safe; however, they should not be performed if toxic
megacolon is suspected. In ulcerative colitis, the histopathology of the rectal
mucosa may show a range of abnormal findings. These include infiltration of the
mucosa with inflammatory cells, flattening of the surface epithelial cells, a
decrease in goblet cells, thinning of the mucosa, branching of crypts, and crypt
abscesses (Fig. 37-1). All of these findings, including crypt abscesses, are non-
specific and may be seen in other colitides, including Crohn’s disease, bacterial
colitis, and amebiasis. Because endoscopic biopsies include mucosa and a vari-
able proportion of submucosa, the transmural nature of Crohn’s disease cannot
be appreciated. However, substantial submucosal inflammation or fissuring of
the mucosa may suggest Crohn’s disease. The finding of noncaseating granulo-
mas also strongly favors a diagnosis of Crohn’s disease (Fig. 37-2), but granulo-
mas are identified infrequently in mucosal biopsies from patients with established
Crohn’s disease and may accompany other conditions, such as tuberculosis and
lymphogranuloma venereum. The identification of amebic trophozoites by
biopsy confirms that diagnosis. Large numbers of mucosal eosinophils are typi-
cal of eosinophilic colitis.
E. Radiography
1. The plain film of the abdomen usually is normal in patients with mild-to-
moderate IBD. Air in the colon may provide sufficient contrast to indicate loss
of haustral markings and shortening of the bowel in ulcerative colitis or nar-
rowing of the bowel lumen in Crohn’s disease.
In severe colitis of any cause, the transverse colon may become dilated
(Fig. 37-3). When this finding is accompanied by fever, elevated white cell
count, and abdominal tenderness, toxic megacolon is likely (see section VII.A).
The plain film of the abdomen should be repeated once or twice a day in
patients with toxic megacolon to follow the course of colonic dilatation.
2. Computed tomography (CT) of the abdomen and pelvis may be very informa-
tive in patients presenting with chronic or recurrent abdominal pain and suspicion
of IBD. Abdominal masses and abscesses, fistulas, and, most commonly, inflam-
matory thickening of the involved bowel wall may facilitate the diagnosis of IBD.
79466_CH37.qxd 1/2/08 12:35 PM Page 249
Figure 37-1. Photomicrograph of a rectal biopsy from a patient with ulcerative colitis.
Inflammatory cells predominate in the lamina propria. The surface epithelial cells are flattened, and
the number of goblet cells appears to be decreased. A crypt abscess is evident (arrow).
Figure 37-2. Photomicrograph of the mucosa and submucosa within a resected portion of termi-
nal ileum from a patient with Crohn’s disease. A fissure extends through the mucosa to the submu-
cosa. At the base of the fissure is a granuloma. (From Eastwood GL. Core Textbook of Gastroenterology.
Philadelphia: Lippincott Williams & Wilkins; 1984:137. Reprinted with permission.)
79466_CH37.qxd 1/2/08 12:35 PM Page 250
Figure 37-3. Plain abdominal x-ray film of a patient with toxic megacolon showing dilatation of the
transverse colon.
3. CT enterography and colography are much more sensitive than UGI, SBFT,
and barium enema and should be preferred when available.
4. Barium enema should not be performed in patients who are acutely ill with
colitis because of the possibility that the preparation for barium enema or the
procedure itself may precipitate toxic megacolon. Even in patients with mild-to-
moderate colitis, vigorous cathartic preparation for barium enema should be
avoided. Rather, oral PEG electrolyte solutions are preferable to prepare
patients for barium enema or colonoscopy.
Some patients with early ulcerative colitis have normal findings on bar-
ium enema examination. Double-contrast studies, however, usually reveal a
diffuse granular appearance of the mucosa. Loss of haustration, ulcerations,
pseudopolyps, and shortening of the bowel are later developments (Fig. 37-4).
Sometimes an area of narrowing requires differentiating between a benign
stricture and carcinoma. Reflux of barium into the terminal ileum may show
dilatation and mild mucosal irregularity for several centimeters, the so-
called backwash ileitis associated with ulcerative colitis.
The diagnosis of Crohn’s disease can be inferred on the basis of several
radiologic findings (Fig. 37-5; see also Fig. 31-2). Narrowing of the bowel
from fibrosis or edema and formation of fistulas reflect the transmural
nature of the disease. Involvement of the terminal ileum and presence of
skip areas in either the large or the small bowel strongly favor a diagnosis
of Crohn’s disease rather than ulcerative colitis. Finally, mucosal changes of
deep ulcers and linear fissures are characteristic of Crohn’s disease.
5. An upper GI and small-bowel series may be of diagnostic help in the evalu-
ation of Crohn’s disease. It is important to remember that, in the radiologic
evaluation of a patient with chronic or recurrent abdominal pain, a routine
79466_CH37.qxd 1/2/08 12:35 PM Page 251
Figure 37-4. Barium enema x-ray examination showing ulcerative colitis involving the entire colon
from rectum to cecum. The mucosal pattern is irregular, the haustral markings are absent from the
left colon, and several pseudopolyps are evident (arrows). (From Eastwood GL. Core Textbook of
Gastroenterology. Philadelphia: Lippincott Williams & Wilkins; 1984:132. Reprinted with permission.)
Figure 37-5. Barium enema x-ray examination of a patient with Crohn’s disease showing a stric-
ture in the sigmoid colon. Several years before, the patient had undergone resection of the terminal
ileum and right colon with an ileum–transverse colon anastomosis. (From Eastwood GL. Core
Textbook of Gastroenterology. Philadelphia: Lippincott Williams & Wilkins; 1984:143. Reprinted with
permission.)
G. Pill-cam imaging has become a newer useful tool in the diagnosis of small-
intestinal Crohn’s disease. The pill may become lodged in narrow structures and
may require surgical removal. Nonspecific superficial ulcers may signify Crohn’s
ulcers, or may have resulted from use of aspirin or nonsteroidal anti-inflammatory
drugs. Thus, clinical judgement is needed to make the appropriate diagnosis.
VI. TREATMENT
A. Diet and nutrition. Patients with mild symptomatic IBD usually are able to take
food orally. The diet should be nutritious. Traditionally, fiber has been restricted
during periods of active symptoms. Some patients cannot tolerate milk, which may
or may not be related to lactase deficiency. Patients with Crohn’s disease who have
terminal ileal involvement and steatorrhea may require supplemental fat-soluble
vitamins, medium-chain triglycerides, and parenteral vitamin B12. Replacement
iron may be indicated in patients who are iron-deficient. Patients in remission do
not require any restriction of fiber or other dietary constituent except as dictated
by his or her own experience.
In severe IBD, an elemental oral diet or nothing by mouth with total parenteral
nutrition (TPN) has been recommended when patients are hospitalized. The use of
TPN to treat fistulas of Crohn’s disease is controversial. Administration of TPN over
79466_CH37.qxd 1/2/08 12:36 PM Page 253
Azulfidine 500-mg tablets 1–2 tablets global p.o. q.i.d (1–4 g/d)
Dipentum 500-mg tablets 2 capsules p.o. b.i.d. (1 g/d)
Asacol 400-mg tablets 2–4 tablets p.o. t.i.d (2.4–4.8 g/d)
Pentasa 250-mg capsules 4 capsules p.o. t.i.d. (3–4 g/d)
Colazal 750-mg capsules 3 capsules p.o. t.i.d. (1,250 mg/d)
Rowasa enema 4 g-unit dose/60-mL 1–2 daily (preferably at bedtime)
enema
Canasa suppositories 500-mg rectal 1–2 times daily
suppositories
ASA, aminosalicylic acid; p.o., orally; q.i.d., four times daily; t.i.d., three times daily.
several weeks has been associated with the closure of a high proportion of fistulas.
However, the fistulas commonly recur after reinstitution of oral feedings.
B. Drugs. Because the etiology of IBD is incompletely understood, drug treatment is
aimed at alleviating and reducing inflammation.
1. Sulfasalazine and aminosalicylates (mesalamines) (Table 37-4).
Sulfasalazine (Azulfidine) historically has been the most commonly used drug
in the treatment of colitis. It has been show to be efficacious in the treatment of
UC and Crohn’s disease when the colon is involved. The drug consists of sul-
fapyridine linked to 5-aminosalicylic acid (5-ASA or mesalamine) via an azo
bond. Intestinal bacteria break the azo bond and release the two components.
The sulfapyridine moiety is systemically absorbed and excreted in the urine and
the 5-ASA moiety, the active component, stays in the intestinal lumen in con-
tact with the mucosa and eventually is excreted in the feces.
a. Side effects. Abdominal discomfort is common and is attributed to the
effects of the salicylate portion of the drug on the upper GI tract. This prob-
lem is minimized by ingestion of the sulfasalazine after eating. Patients also
may become folate-deficient because of competition between folate and sul-
fasalazine for absorption. Other side effects, such as skin eruptions and bone
marrow suppression, are less common and are attributed to the sulfa portion.
b. Dosage. The initial daily dose is low (1 g) to minimize GI side effects. A
therapeutic dose of 3 to 4 g per day is appropriate. A CBC and liver chem-
istry tests should be obtained before starting therapy, every 1 to 3 months
initially, and every 6 to 12 months during long-term treatment.
c. Maintenance treatment with sulfasalazine has been shown to reduce the
frequency of exacerbations of ulcerative colitis. The usual maintenance
dosage is 2 to 3 g per day in divided doses, although an occasional patient
may benefit from 1 g per day (i.e., 500 mg twice daily).
2. Other 5-ASA (mesalamine) preparations. Because the serious side effects of
sulfasalazine are related to the sulfa portion, there has been much interest in
developing similar drugs that retain the salicylate portion but replace the sul-
fapyridine. Several oral 5-ASA preparations are available (Table 37-4).
Olsalazine (Dipentum), which consists of two 5-ASA molecules joined
by an azo bond such as sulfasalazine, requires bacterial degradation in the
colon. It is as effective against active UC and in maintaining remission. High
incidence of diarrhea as a side effect of Dipentum limits its use in most patients.
Asacol is a controlled-release tablet form of 5-ASA, which is encapsu-
lated by an acrylic resin that dissolves at a pH higher than 6.0. It has been
shown to be effective in both active and remitted UC and Crohn’s disease,
79466_CH37.qxd 1/2/08 12:36 PM Page 254
especially when the ileum is involved. The usual dose is 2.4–4.8 mg daily in two
or three divided doses.
Pentasa is a controlled-release formulation of 5-ASA that is encapsulated
in ethylcellulose microgranules. It is effective in active and remitted UC and in
active Crohn’s disease regardless of disease location. Pentasa appears to help in
maintaining remission in both small-bowel and colonic Crohn’s disease. The
recommended dose is 1 g q.i.d.
Balsalazide disodium (Colazal) is a 5-ASA preparation containing an
azo bond. The active 5-ASA moiety is released in the colon by cleavage of the
azo bond by colonic bacteria. It is primarily effective in treating IBD involving
the entire colon, as well as left-sided colitis and proctitis. The recommended
dose is 750 mg three tablets t.i.d. one a day.
Once daily dosing with mesalamine (Liallda) has recently been made avail-
able. The tablets contain a novel matrix which protects the mesalamine from
degredation until it reaches the colon, then allows its gradual release through the
entire length of the colon. The recommended dose is 1–2 g 2–4 tablets daily. The
topical 5-ASA preparations include Rowasa enema and Canasa rectal supposito-
ries. These are effective in active and remitted distal UC and ulcerative proctitis.
The 5-ASA preparations are also recommended for patient with Crohn’s dis-
ease to prevent postoperative recurrence of Crohn’s disease, especially in patients
with colitis and to some extent in patients with small-bowel involvement. The
5-ASA preparations and sulfasalazine are safe to use in pregnant women.
3. Corticosteroids. Historically, corticosteroids have been used in patients with
severe UC or Crohn’s disease to induce a remission. Intravenous (IV) steroids (i.e.,
hydrocortisone 100 mg IV q6h or methylprednisolone 10–30 mg IV q6–8h) are
usually used in such patients. When patients can take oral medications, pred-
nisone at doses of 40 to 60 mg per day is usually given. If symptoms improve, the
drug is tapered and withdrawn over a period of several weeks to months. Steroids
are not recommended for maintenance therapy of UC or Crohn’s disease, because
steroids do not prevent relapse of Crohn’s disease or UC and they have major side
effects. However, many patients become steroid dependent and experience recur-
rence of symptoms when the dose of prednisone is reduced to less than 15 mg per
day. One strategy is to use 6-mercapopurine (6-MP), or azathioprine (Imuran)
for steroid-dependent patients to help taper them off steroids. Biologic agents
such as Remicade and Humira may be used instead of corticosteroids.
Several corticosteroid enema preparations are available for the treat-
ment of proctitis and distal colitis. These are usually administered once or
twice a day.
Budesonide (Entecort) is an orally administered corticosteroid that is
released in the ileum and right colon at a pH of about 5.5. It has low systemic
effects due to less than 20% absorption into the systemic circulation and a very
efficient first-pass metabolism in the liver. It is effective in treating Crohn’s dis-
ease involving the ileum and right colon, as well as microscopic (lymphocytic)
colitis. The dose is 9 mg daily in active disease. It is usually tapered from 6 to
3 mg daily after 2 to 6 months. It is helpful in inducing remission. The drug is
then stopped. Due to its low systemic bioavailability, it has minimal side effects
of glucocorticoid steroids. However, bone mineral density may diminish during
therapy. Patients receiving glucocorticoid therapy should be offered supplemen-
tal calcium and vitamin D.
4. Antibiotics. Most patients with mild-to-moderate IBD are successfully main-
tained on therapy with either sulfasalazine or 5-ASA and occasionally may
require systemic or topical steroid therapy to treat disease relapses. However,
about 20% to 25% of patients with UC and one third of patients with Crohn’s
disease require additional therapy for refractory disease, steroid dependency,
and fistulas. Bacteria is known to play an important role in the pathogenesis of
Crohn’s disease and it may play a role in UC. In Crohn’s diseases, the indica-
tions for the use of antibiotics include perianal disease, localized peritonitis due
to microperforation, bacterial overgrowth secondary to a chronic stricture, and
as an adjunct to drainage therapy for abscesses and fistulas. Antibiotics should
79466_CH37.qxd 1/2/08 12:36 PM Page 255
the short-term response rates are impressive, over half of these patients must
undergo colectomy during 6 months of follow-up. In patients with Crohn’s
disease, cyclosporin has not been consistently effective.
Oral cyclosporin at low doses (5 mg/kg per day) is not efficacious in
IBD. Higher doses (8–10 mg/kg per day) of cyclosporin may be more effec-
tive; however, the side effect profile (hypertension, and nephro- and bone
marrow toxicities) limits its usefulness. Cyclosporin should not be used for
the maintenance of remission for UC or Crohn’s disease.
Clinical response in patients who respond to cyclosporin therapy is rapid
(within 2 weeks). The total duration of therapy should not exceed 4 to 6
months. Monitoring of nephrotoxicity is required. The dose should be adjusted
downward whenever the baseline serum creatinine level increases by 30%.
d. Infliximab (Remicade) is a chimeric monoclonal antibody to human tumor
necrosis factor-2 (TNF-2), a proinflammatory cytokine that plays an important
role in the pathogenesis of Crohn’s disease. Controlled trials have demon-
strated high efficacy for infliximab in moderate to severely active Crohn’s dis-
ease, in fistulizing Crohn’s disease, and in severe UC with sufficient evidence
for its safety. Clinical response is rapid (within 1–2 weeks) and the duration of
response ranges from 8 to 12 weeks per infusion. Infliximab has probably
supplanted the role of cyclosporin and corticosteroids in Crohn’s disease.
Infliximab is administered IV (5 mg/kg) over 2 hours. Infusions for active or
fistulizing Crohn’s disease are given as three doses at 0, 2, and 6 weeks for a
starting dose. Infusion reactions may be minimized by pretreatment with an
antihistamine and a steroid preparation. To minimize symptomatic disease
recurrence, the IV infusions are repeated at 8-week intervals.
Toxicities observed for infliximab therapy include formation of human
antichimeric antibodies (HACA), autoantibodies and a serum sickness–like
delayed hypersensitivity reaction in some patients 2 to 4 years after initial
treatment. There may be activation of dormant tuberculosis (TB). Patients
should be tested for TB prior to initiation of therapy with infliximab. The risk
of development of lymphoma in treated patients seems to be slightly increased.
e. Adalimumab (Humira) (Human antitumor necrosis factor [TNF] mono-
clonal antibody) has been approved by the FDA for the treatment of rheuma-
toid arthritis and Crohn’s disease. It has also been shown to be effective in
the treatment of Crohn’s disease unresponsive to conventional therapies and
also to a percentage of patients who have not responded to Remicade or
who have lost their response to Remicade. The advantage of adalimumab
over Remicade is that it has less immunogenicity and that it is administered
subcutaneously (sc) 40 mg every 2 weeks after a starting dose of 80 mg sc.
The dose may be increased to weekly in refractory patients.
f. Certolizumab pegol (Cimzia) is a humanized TNF alpha Fab monoclonal
antibody fragment linked to polyethylene glycol (PEG). It is waiting for FDA
approval for the treatment of moderate to severe Crohn’s disease. This bio-
logic drug is administered subcutaneously at weeks 0, 4, and 8 weeks, then
once monthly at the same dose (400 mg). It has been shown to be effective
in inducing response and remission in adult patients with active Crohn’s dis-
ease unresponsive to conventional therapy as well as other biologic agents.
g. Thalidomide has shown anti-TNF activity in a subset of patients with
Crohn’s disease; however, due to its highly teratogenic potential, it is not
suitable for routine use.
h. Nicotine given in enema form has shown some benefit in patients with dis-
tal UC but due to its side effect profile, it is not widely used.
6. Prebiotics and probiotics. It is believed that luminal bacterial flora influ-
ences the development and/or progression of IBD.
Prebiotics are nutrients utilized by specific microorganisms and support the
growth of these organisms that interact with the host to improve the host barrier
function and impact the innate immune response. Prebiotics include carbohy-
drates resistant to digestion (i.e., inulin, lactulose, and other oligosaccharides).
79466_CH37.qxd 1/2/08 12:36 PM Page 257
Probiotics are live microorganisms that are important for the develop-
ment of a healthy innate immune system. Commonly used probiotics include
Lactobacillus, Bifidobacterium, and other organisms including yeast such as
Saccharomyces and combination VSL#3.
The data whether pre- and probiotics are helpful in IBD are controver-
sial. However, probiotics have been shown to be effective in the treatment of
pouchitis.
7. Antidiarrheal medications. If diarrhea does not improve with the previously
described medical therapy in patients with mild-to-moderate IBD, treatment
with an antidiarrheal drug may be helpful. Codeine is most effective, although
loperamide (Imodium) or diphenoxylate (Lomotil) may be preferred because
of their lower addictive potential. These and other opiate derivatives should
not be used in patients with severe IBD because of the possibility of inducing
toxic megacolon.
Nonsteroidal antiinflammatory drugs (NSAIDs) should not be used in
patients with IBD. The role of cyclooxygeranase (Cox)-2 NSAIDs in pain con-
trol for IBD is not clear.
8. Bile acid-binding resins and medium-chain triglycerides. Because many
patients with Crohn’s disease have involvement of the terminal ileum with con-
sequent bile acid malabsorption, diarrhea, and steatorrhea, treatment with a
bile acid-binding resin such as cholestyramine or with medium-chain triglyc-
eride supplements may be indicated (see Chapter 30) in addition to other
antidiarrheal drugs.
In summary, patients with mild and moderate IBD are usually started
on a 5-ASA, and clinical response is determined within 2 to 4 weeks. If patients
continue to be symptomatic, an AZA is added to the treatment. Patients who
achieve remission are followed at 1 to 3 months clinically. Those who do not
reach remission are then offered pulse steroid therapy or a biologic agent.
C. Psychotherapy. Formal psychotherapy appears to be of little benefit in the pri-
mary treatment of IBD. However, emotional support of the patient by the physi-
cian, family, clergy, and other concerned people is important. Psychotherapy may
be indicated in some patients to help them cope with living with a chronic disease.
D. Surgery
1. Ulcerative colitis. The standard noncontroversial indications for surgery in
ulcerative colitis include toxic megacolons perforation, abscess, uncontrollable
hemorrhage, unrelieved obstruction, fulminating disease, carcinoma, and high-
grade dysplasia on colonic biopsies. Usually a total proctocolectomy is per-
formed. Historically, this operation has been accompanied by a permanent
ileostomy. Currently rectal sphincter-saving operations are the norm. In this
operation, the rectal mucosa is removed, but the muscular wall of the rectum is
left intact. The terminal ileum is anastomosed to the anus, usually with the for-
mation of a reservoir pouch. If the operation is performed in an acutely ill
patient with severe colitis, it is usually carried out in two steps: the colectomy
first and the reanastomosis several months later.
Despite increased surgical experience and patient appeal, problems with the
ileal pouch–anal anastomosis remain. Technical failure occurs in approximately
5% to 8% of patients within the first 5 to 10 years. Pouchitis occurs acutely in
40% to 50% of patients. Most of these patients respond to a course of antibiotic
therapy with metronidazole or ciprofloxacin hydrochloride, or both. Xifaxan has
also been shown to be effective in some patients. Approximately 5% to 8% will
develop chronic pouchitis and require maintenance suppressive therapy or possi-
ble pouch takedown and formation of a permanent ileostomy. There are reports
of the development of neoplasia in the pouch after a number of years.
2. Crohn’s disease. Surgery in Crohn’s disease is indicated for perforation,
abscess, obstruction, unresponsiveness to treatment, intractable disease, and
some fistulas. Because removing a segment of diseased bowel does not “cure”
the patient of Crohn’s disease as proctocolectomy cures a patient with ulcera-
tive colitis, the dictum is to remove as little bowel as is necessary to correct the
79466_CH37.qxd 1/2/08 12:36 PM Page 258
problem. Patients with a single, short segment of Crohn’s disease respond the
best to surgery. The rectal sphincter-saving operation is not an option in
Crohn’s patients because of the possibility that anorectal Crohn’s disease may
develop later. Stricturoplasty may be a surgical option in some patients with
stricturing Crohn’s disease.
a. The recurrence rate after surgery of Crohn’s disease is high: 30% after
5 years, 50% after 10 years, and 70% after 15 years. Whether this is because
of some deleterious effect of surgery on the bowel or merely represents the
natural history of patients with severe Crohn’s disease is difficult to deter-
mine. In any event, the aim of surgery is to remove grossly diseased bowel
and preserve as much normal-appearing bowel as possible. If the terminal
ileum has been removed, the patient should undergo a Schilling test 3 to 6
months after surgery to survey for vitamin B12 malabsorption. Most patients
will require monthly vitamin B12 injections.
b. Fistulas often respond to medical treatment, and even those that persist do
not always require surgery. However, enterocutaneous fistulas that are
poorly tolerated by the patient and all enterovesical fistulas may require sur-
gical correction. Also, fistulas between loops of bowel should be corrected
surgically if clinically significant malabsorption occurs.
with preserved crypts, the colitis is called lymphocytic colitis. However, if there
is a collagen band thicker than 10 mm below the epithelium, the condition is
called collagenous colitis. The etiology of microscopic colitis is not known.
Treatment is usually initiated with a 4- to 8-week trial of bismuth subsal-
icylate two tablets three to four times a day. If symptoms do not abate, a 5-ASA
compound and/or budesonide may be used to achieve remission. Antibiotics
such as ciprofloxacin 500 mg b.i.d. or metronidazole 250–500 mg t.i.d. or
Xifaxan 200 mg one or two tablets t.i.d. may be used for 1 to 3 months. In
some cases, corticosteroids and immunomodulators may be necessary to induce
remission.
Selected Readings
Baumgart DC, et al. Inflammatory bowel disease: cause and immunobiology. Lancet.
2007;369:1627–1640.
Baumgart DC, et al. Inflammatory bowel disease: clinical aspects and established and
evolving therapies. Lancet. 2007;369:1641–1658.
Beaugerie L, et al. Drug-induced microscopic colitis: Proposal for a scoring system and
review of the literature. Aliment Pharmacol Ther. 2005;22:277–284.
Biancone L, et al. Treatment with biologic therapies and the risk of cancer inpatients with
IBD. Nat Clin Pract Gastroenterol Hepatol. 2007;4:78–79.
Bonderup OK, et al. Budesonide treatment of collagenous colitis: A randomized, double-
blind, placebo-controlled trial with morphometric analysis. Gastroenterology. 2003;52:
248–251.
Bosch X, et al. Antineutrophilcytoplasmic antibodies. Lancet. 2006;368:404–418.
Caprilli R, et al. Current management of severe ulcerative colitis. Nat Clin Pract
Gastroenterol Hepatol. 2007;4:92–101.
Colombel JF, et al. Adalimumab for maintenance of clinical response and remission in
patients with Chrohn’s disease: The CHARM trial. Gastroenterol. 2007;132:51–65.
Feagon BG, et al. Mesalamine maintenance therapy for Crohn’s disease. Gastroenterology.
2001;120:585.
Fernandez-Banares F, et al. Collagenous and lymphocytic colitis: Evaluation of clinical and
histological features, response to treatment and long term follow-up. Am J Gastroenterol.
2003;98:340–347.
Fichera A, et al. Surgical treatment for Crohn’s disease. J Gastroent. Surg. 2007;11(6):
791–803.
Hanauer SB, et al. Human anti-tumor necrosis factor monoclonal antibody (adalimumab)
in Crohn’s disease: The CLASSIC-1 trial. Gastroenterology. 2006;130(2):323–333.
Itzkowitz SH, et al. Colorectal cancer screening and surveillance in inflammatory bowel
disease. Inflam. Bowel Dis. 2005;11(3):314–327.
Jarnerot G, et al. Infliximab as rescue therapy in severe to moderately severe ulcerative colitis:
A randomized, placebo-controlled study. Gastroenterology. 2005;128:1805–1811.
Lemann M, et al. A randomized, double-blind controlled withdrawal trial in Crohn’s
disease patients in long-term remission on azathioprine. Gastroenterology. 2005;
128(7):1812–1818.
Lemann M, et al. Infliximab plus azathioprine for steroid-dependent Crohn’s disease patients:
A randomized placebo-controlled trial. Gastroenterology. 2006;130(4):1054–1061.
Leung WK, et al. Small bowel enteropathy associated with chronic low-dose aspirin
therapy. Lancet. 2007;369:614.
Lichtenstein GR, et al. American Gastroenterological Association Institute medical position
statement on corticosteroids, immunomodulators, and infliximab in inflammatory bowel
disease. Gastroenterology. 2006;130(3):935–939.
Lichtenstein GR, et al. American Gastroenterological Association Institute technical review
on corticosteroids, immunomodulators, and infliximab in inflammatory bowel disease.
Gastroenterology. 2006;130(3):940–987.
Lichtenstein GR, et al. Infliximab maintenance treatment reduces hospitalization, surgeries
and procedures in Chron’s disease. Gastroenterol. 2005;128:862–869.
Loftus EV, Jr. Clinical epidemiology of inflammatory bowel disease: incidence, prevalence,
and environmental influences. Gastroenterology. 2004;126:1504–1517.
79466_CH37.qxd 1/2/08 12:36 PM Page 262
Marshall JK, et al. Putting rectal 5-aminosalicylic acids in its place: The role in distal
ulcerative colitis. Am J Gastroenterol. 2000;95:1628.
Robert M. Pathology of microscopic, lymphocytic, and collagenous colitis. J Clin
Gastroenterol. 2004;38:517–526.
Rodemann JF, et al. Incidence of clostridium difficile infection in inflammatory bowel
disease. Clin Gastroenterol Hepatol. 2007 Mar;5:339–344.
Rutgeerts P, et al. Infliximab for induction and maintenance therapy for ulcerative colitis.
Nat Clin Pract Gastroenterol Hepatol. 2007;4:92–101.
Sabery N. Use of serologic markers as a screening tool in inflammatory bowel disease
compared with elevated erythrocyte sedimentation rate and anemia. Pediatrics.
2007;119:e193–199.
Sandborn WJ, et al. Optimizing anti-TNF treatment in inflammatory bowel disease.
Gastroenterology. 2004;126:1593–1610.
Sartor RB. Therapeutic manipulation of the enteric microflora in inflammatory bowel
diseases: antibiotics, probiotics, and prebiotics. Gastroenterology. 2004;126:1620–1633.
Schiller L. The clinical spectrum of microscopic colitis. J Clin Gastroenterol. 2004;38:
527–530.
Schreiber S, et al. A randomized placebo-controlled trial of certolizumab pegol (CDP870)
for treatment of Crohn’s disease. Gastroenterology. 2005;129(3):807–818.
Scheibers S, et al. Maintenance therapy with certolizumab pegol for Crohn’s disease.
N Eng J Med. 2007;357:239–350.
Seidman EG. Clinical use and practical application of TPMT enzyme and 6-mercaptopurine
metabolic monitoring in IBD. Rev. Gastroenterol Disorder. 2003;3(suppl 1):S1–S9.
Xavier RJ, et al. Unraveling the pathogens of inflammatory bowel disease. Nature. 2007;
448(7142):427–434.
Yang YX. Methotrexate for the maintenance of remission in Crohn’s disease. Gastroenterology.
2001;120:1553.E 37.1
79466_CH38.qxd 1/2/08 12:37 PM Page 263
263
79466_CH38.qxd 1/2/08 12:37 PM Page 264
Figure 38-1. Plain x-ray film of the abdomen in an elderly patient with nonocclusive ischemic
colitis. Air within the bowel provides a natural contrast medium. The characteristic “thumbprint”
indentations are evident in the upper descending colon (arrows).
Figure 38-2. Barium enema x-ray film of the region of the splenic flexure in a patient with nonoc-
clusive ischemic colitis. The “thumbprint” pattern is present in the descending colon (arrows).
79466_CH38.qxd 1/2/08 12:37 PM Page 267
II. ANGIODYSPLASIA
A. Pathophysiology. Angiodysplasias are ecstatic, vascular lesions within the submu-
cosa consisting of arterial, venous, and capillary elements. They may be found
throughout the GI tract but occur most frequently in the cecum and ascending
colon. Angiodysplasias are clinically important because they are a cause of acute
and chronic GI bleeding, particularly in elderly patients.
Most angiodysplasias are thought to develop as a consequence of normal
aging. Aortic stenosis occurs in up to 15% of patients with bleeding angiodys-
plasias, but pathogenic association is uncertain. In some patients, angiodysplastic
lesions may occur throughout the GI tract as part of the hereditary disorder known
as Osler-Weber-Rendu disease. In this disease, telangiectatic lesions also are found
on the skin, in the nail beds, and in the mucosa of the mouth and nasopharynx.
B. Clinical presentation. Angiodysplasias cause no symptoms until they bleed. In
patients who ultimately are diagnosed as having bled from an angiodysplasia,
therefore, the only common history is that of GI bleeding. The usual presentation
is one of acute lower GI bleeding. Often a cause of bleeding is not found or the
bleeding is attributed to another common, coexisting disorder, such as peptic dis-
ease or diverticulosis. Patients may experience several bleeding episodes, some-
times over a period of months, before the correct diagnosis is made.
C. Diagnostic studies. The two methods of diagnosing angiodysplasias are colonoscopy
and angiography.
1. Colonoscopy. An angiodysplastic lesion appears as a submucosal red blush
resembling a spider angioma of the skin. To optimize the chance of visualizing
the lesion by colonoscopy, however, the bowel must be free of blood and debris,
which may not be possible during acute bleeding.
2. Angiography. The lesions may be identified angiographically by the following
findings:
a. An early-filling vein
b. A vascular tuft of dilated vessels
c. A slowly emptying vein
D. Treatment. The traditional treatment of a bleeding angiodysplasia has been to
resect the segment of bowel that contains the lesion, usually the ascending colon.
Up to 30% of patients so treated bleed again, either from a new angiodysplastic
lesion or from a lesion that was not resected. An alternative to surgical resection is
colonoscopic electrocoagulation if the lesions can be identified by colonoscopy.
Selected Readings
Bismar MM, Sinicrope FA. Radiation enteritis. Curr Gastroenterol Rep. 2002;4:361–365.
Brandt IJ, et al. AGA technical review on intestinal ischemia. Gastroenterology. 2000;
118:195.
Chang L, et al. Incidence of ischemic colitis and serious complications of constipation
among patients using alosetron: Systematic review of clinical trials and post-marketing
surveillance data. Am J Gastroenterol. 2006;101:1069–1079.
Deana DG, et al. Reversible ischemic colitis in young women. Association with oral
contraceptive use. Am J Surg Pathol. 1995;19:454.
Dowd J, et al. Ischemic colitis associated with pseudoephedrine: Four cases. Am J
Gastroenterol. 1999;943:2430.
Flobert C, et al. Right colonic involvement is associated with severe forms of ischemic
colitis and occurs frequently in patients with chronic renal failure requiring
hemodialysis. Am J Gastroenterol. 2000;95:195.
Nehme OS, et al. New developments in colonic ischemia. Curr Gastroenterol Rep. 2001;3:416.
Noyer CM, et al. Colon ischemia: Unusual aspects. Clin Perspect Gastroenterol.
November/December 2000:315.
Schuller JG, et al. Cecal necrosis: Infrequent variant of ischemic colitis: Report of five
cases. Dis Colon Rectum. 2000;43:708.
Shrestha S, et al. Henoch Schonlein purpura with nephritis in adults: adverse prognostic
indicators in a UK population. QJM. 2006;99:253–265.
Uzoigwe CE, et al. A surgical solution for vasculitis? Lancet. 2007;369:1054.
79466_CH39.qxd 1/2/08 12:38 PM Page 269
C olorectal cancer (CRC) is the second leading cause of cancer mortality in the United
States. In men, CRC is second in prevalence only to lung cancer, and in women, it is third
behind breast and lung cancer. More than 95% of the cancers are thought to have their ori-
gin in adenomatous polyps. In the United States, the prevalence of CRC is 30 to 40 per
100,000 and it increases with age.
I. COLONIC POLYPS
A. Pathogenesis. The term polyp refers to any protrusion into the lumen of the gas-
trointestinal (GI) tract. A sessile polyp is a raised protuberance with a broad
base. A pedunculated polyp is attached to the bowel wall by a stalk that is nar-
rower than the body of the polyp. Submucosal polyps are lipomas, leiyomyomas,
hemangiomas, fibromas, lymphoid tissue, endometriomas, melanomas, or metasta-
tic lesions. Most submucosal polyps are benign; however, many patients with
carcinoid metastatic lesions, melanomas, lymphomas, and Kaposi’s sarcomas have
malignant polyp formation in the colon.
Polyps may be benign or malignant. In the colon, polyps can be described
as adenomatous, hamartomatous, hyperplastic, or inflammatory, according to
their histopathologic appearance. Hyperplastic and inflammatory polyps are
usually benign, but are also at risk for carcinoma. Adenomatous polyps can be
classified as tubular, villous, or tubulovillous, depending on whether their his-
tologic appearance is primarily glandular, villous, or mixed, respectively.
Although most adenomatous polyps are benign, some may contain carcinoma
and others may degenerate later to carcinoma. The risk of a polyp containing
carcinoma increases directly with the size of the polyp (Table 39-1). Cancer is
more likely to occur in villous adenomas than the other types, and benign ade-
nomatous polyps may coexist with adenocarcinoma elsewhere in the bowel. In
short, a colon that has a tendency to produce polyps also is at higher risk of the
development of cancer.
B. Clinical presentation. Because colonic polyps are so common—some estimates of
prevalence are as high as 50% in people over age 50—they are an important risk
factor in the development of CRC in the general population.
Most patients with polyps are asymptomatic. In those instances, the polyps
remain undiscovered or are diagnosed during surveillance examination. Sometimes
polyps cause occult or gross bleeding. Occasionally patients complain of abdominal
1 cm 0–2
1–2 cm 10–20
2 cm 30–50
269
79466_CH39.qxd 1/2/08 12:38 PM Page 270
Figure 39-1. Air-contrast barium enema showing a sigmoid polyp on a long stalk (arrows).
Extent of involvement
of polyp by cancer Implication Recommendations
Cancer involves only the The cancer has been Surgery not indicated;
mucosa without penetration cured by polypectomy. repeat colonoscopy in 1 y.
of muscularis mucosae
(carcinoma in situ).
Cancer (moderately or well The cancer probably Surgery not indicated
differentiated) has penetrated has been cured. because the surgical risk
the muscularis mucosae but outweighs the risk of
does not involve blood vessels residual cancer; repeat
or lymphatics in the stalk. colonoscopy in 1 y.
Cancer has penetrated the Residual cancer may Segmental resection of
muscularis mucosae and be present in the colon is indicated if
involves blood vessels or the colon. the patient is a good
lymphatics, or the cancer is operative risk. If residual
poorly differentiated, whether cancer is found in the
or not it involves blood patient at surgery, repeat
vessels or lymphatics. colonoscopy in 3–6 mo; if
no residual cancer, repeat
colonoscopy in 6–12 mo.
Cancer involves the resectional Cancer remains Segmental resection of
margin of the polyp stalk. in the bowel. the colon is indicated if
the patient is an operative
candidate. Repeat
colonoscopy in 3–6 mo.
a. The cancerous change involves only the mucosa and does not penetrate
the muscularis mucosa into the stalk of the polyp. This condition is
sometimes called carcinoma in situ or high-grade dysplasia. Colonoscopic
resection of the polyp in this instance is regarded as curative. No surgical
treatment is indicated. The patient should be scheduled to return in 1 year for
follow-up colonoscopy.
79466_CH39.qxd 1/2/08 12:38 PM Page 272
b. The cancer penetrates the muscularis mucosae of the polyp into the
stalk but does not involve blood vessels or lymphatics within the resected
portion of the stalk, and the cancer is moderately to well differentiated.
Although a small number of patients with this finding has cancerous
involvement of the bowel or local lymph nodes, the mortality (2%) and
morbidity of surgery to resect the portion of the colon that contained the
polyp exceeds the risk of residual cancer. Thus, additional surgery for these
patients is not recommended. Repeated colonoscopy should be scheduled
for 1 year later.
c. The cancer not only penetrates the muscularis mucosae of the polyp but
also has invaded blood vessels or lymphatics within the stalk. In these
patients, cancer also is likely to be present in the bowel or local lymph nodes.
Also in this category are patients with poorly differentiated cancers that do
not involve blood vessels or lymphatics. If operative risk is not prohibitive,
these patients should undergo resection of the segment of colon that con-
tained the polyp. Repeated colonoscopy to examine the anastomosis for
recurrent tumor should be scheduled for 3 to 6 months if residual cancer is
found at surgery. If no residual cancer is found, colonoscopy should be
scheduled for 6 to 12 months.
d. The cancer involves the resectional margin of the polyp, indicating that
residual cancer remains in the patient. If the patient is an operative candi-
date, segmental resection of the colon is indicated. These patients should
undergo follow-up colonoscopy in 3 to 6 months to check for residual tumor
at the anastomosis and for other polyps.
273
79466_CH39.qxd 1/2/08 12:38 PM Page 274
Advancing age
Family history of colorectal or polyps
High-fat, low-fiber diet
Bowel disorders
Inflammatory bowel disease (ulcerative colitis, Crohn’s disease)
Adenomatous polyps
Some polyposis syndromes (see Table 39-3)
Familial colon cancer syndrome
Genital tract cancer in women
First, merely growing older increases the risk. Over age 40, the annual inci-
dence of colon cancer begins to accelerate, doubling every decade. Second,
dietary factors, such as a high-fat, low-fiber diet, may increase the risk of devel-
oping CRC. Third, bowel conditions such as ulcerative colitis, Crohn’s disease,
adenomatous polyps, and some of the polyposis syndromes (see sections I and II)
predispose to colon cancer. Finally, there seems to be a hereditary predisposi-
tion to the development of CRC. The probability of CRC developing in a person
who has a first-degree relative with CRC is more than 15%, compared to a 6%
risk in the general population. For hereditary nonpolyposis colon cancer syn-
drome (HNPCC) (see section IV), a high risk of development of colon cancer is
inherited in an autosomal dominant manner. These patients may have coexisting
adenomatous polyps and may also have melanoma or cancer of the uterus or
ovaries. Patients with breast cancer may also have increased risk of developing
colon cancer.
CRC may develop in any part of the colon. Approximately 45% of CRC
is seen in the rectosigmoid and 25% in the ascending colon and cecum. Genetic
examination of the tumors shows mutations in K-ras, APC, VCC, and P53
genes.
B. Clinical presentation
1. History. The most common presenting sign is lower GI bleeding, and the most
common symptom is change in bowel habit. Unfortunately, these are both late
manifestations of the disease. Bleeding may be gross or occult. Occult bleeding
typically is detected on routine rectal examination or stool screening tests. A
change in bowel habit may be manifested as a decrease in the caliber of the
stool if the lesion is distal and constricts the bowel lumen. Patients may com-
plain of constipation. Sometimes diarrhea develops around a partially obstruct-
ing lesion. Mass lesions of the ascending colon, because of the larger diameter
of the ascending colon, may grow to a considerable size before symptoms
develop. Other possible consequences include anemia, weight loss, anorexia,
malaise, abdominal mass, and enterovesical or enterocutaneous fistula. Rarely,
the patient may seek treatment initially for symptoms of metastatic disease,
such as jaundice or bone pain.
2. Physical examination may reveal the effects of weight loss, muscle wasting, or
anemia. A mass may be evident in the abdomen or by rectal examination. Stool
may be positive for gross or occult blood.
C. Diagnostic studies
1. A complete blood count should be obtained to evaluate for anemia. Serum
iron studies confirm iron deficiency. An elevated serum alkaline phosphatase
level, if confirmed to originate in the liver either by fractionation or by the find-
ing of an elevated 5-nucleotidase, suggests liver metastases. Sometimes an
isolated elevation of serum lactic dehydrogenase or G-glutamyltranspeptidase
is the only indicator of liver involvement. Hyperbilirubinemia in conjunction
79466_CH39.qxd 1/2/08 12:38 PM Page 275
annual fecal occult blood testing decreased the 13-year cumulative mortality
from CRC by 33%.
3. Colonoscopy, sigmoidoscopy, barium enema x-ray, CT colographic exam-
ination. See Chapters 5 and 9.
C. In the general population. See Figure 39-3.
D. In patients with colonic polyps. See Figure 39-4.
E. For people with a family history of CRC. People who have a parent, a sibling,
or a second-degree relative with colorectal carcinoma are at higher risk of devel-
opment of CRC, albeit much less than those who have familial colon cancer
syndrome. Such people also are at higher risk of the development of CRC and
polyps. People who have one first-degree relative with CRC should undergo CRC
screening (preferably with colonoscopy) starting at age 40, or 10 years before the
age of the onset in the affected relative. For people with two affected first-degree
relatives, or one first-degree relative and one or more second-degree relatives, ini-
tial colonoscopy at age 40 may be justified, followed by routine yearly stool occult
blood testing and periodic colonoscopy every 3 to 5 years.
F. For postoperative colonoscopic follow-up of patients with resected CRC.
Patients who have had a colorectal carcinoma are at greater risk of the develop-
ment of another colorectal carcinoma. Patients who have had resection of the
colon for cancer should undergo colonoscopy 12 months after surgery to look for
recurrence at the anastomosis and elsewhere in the colon. Subsequent colono-
scopic examinations should be every 2 to 3 years.
G. Chemoprevention of prevention of CRC. Compliance to screening programs for
CRC remains poor and alternative approaches are being sought. Chemoprevention
of CRC involves the long-term use of a variety of oral agents to prevent the devel-
opment of adenomatous polyps and their subsequent progression to CRC.
Molecular analyses of adenomas and CRCs have led to a genetic model of colon
carcinogenesis in which the development of cancer results from the accumulation of
a number of genetic alterations. By interfering with these molecular events, chemo-
prevention could inhibit or reverse the development of adenomas or the progression
from adenoma to cancer. Recent observations suggest that patients taking inhibitors
such as celecoxib, aspirin, NSAIDs, cyclooxygenerase-2 (COX-2), supplemental
folate, calcium, and estrogen, as seen in postmenopausal women taking hormone
replacement therapy, experience a chemopreventive benefit. Other potential
chemopreventive agents as ursodiol, eflornithine, and oltipraz are undergoing eval-
uation in clinical studies.
Chemoprevention should not replace periodic fecal occult blood tests and
endoscopic screening as well as modification in known risk factors of CRC such as
reduction in the intake of red meat, appropriate exercise, smoking cessation, and
weight control.
79466_CH39.qxd 1/2/08 12:38 PM Page 278
Figure 39-4. Recommendations for cancer surveillance in patients with colonic polyps.
79466_CH39.qxd 1/2/08 12:38 PM Page 279
Selected Readings
Anderson JC, et al. Prevalence of colorectal neoplasia in smokers. Am J Gastroenterol.
2003;98:2777–2783.
Barclay RL, et al. Colonoscopic withdrawal times and adenoma detection during screening
colonoscopy. N Engl J Med. 2006;355:2533–2541.
Chan AT, et al. Aspirin and the risk of colorectal cancer in relation to the expression of
COX-2. N Engl J Med. 2007 May 24;356:2131–2134.
Cho E, et al. Alcohol intake and colorectal cancer: A pooled analysis of 8 cohort studies.
Ann Intern Med. 2004;140:603–612.
Cole BF, et al. Folic acid for the prevention of colorectal adenomas: A randomized clinical
trial. JAMA. 2007 June 6;297:2351–2359.
Coluca G, et al. Phase III randomized trial of FOLFIRI versus FOLFAX 4 in the treatment
of advanced colorectal cancer. J Clin Oncol. 2005;23(22):4866–4875.
DeJong A, et al. Decrease in mortality in Lynch syndrome families because of surveillance.
Gastroenterology. 2006;130:665–671.
Dove-Edwin I, et al. Prevention of colorectal cancer by colonoscopic surveillance in
individuals with a family history of colorectal cancer: A 16 year prospective follow-up.
BMJ. 2005;331:1047–1052.
Eng C, et al. Impact of quality of life of adding cetuximab to irontecan in patients who have
failed prior oxaliplatin-based therapy:the EPIC trial. J Clin Oncol. 2007;25:suppl:18s.
Giardiello FM, et al. AGA technical review on hereditary colorectal cancer and genetic
testing. Gastroenterology. 2001;121:198.
Goldberg RM, et al. A randomized controlled trial of fluoro acid plus leukovour, Irinotecan
and oxiplatin combinations in patients with previously untreated metastatic colorectal
cancer. J Clin Oncol. 2004;22(1):23–30.
Hur C, et al. The management of small polyps found by virtual colonoscopy: results of a
decision analysis. Clin Gastroenterol Hepatol. 2007;5:237–244.
Kapiteijn E, et al. Preoperative radiotherapy combined with total mesorectal excision for
resectable rectal cancer. N Engl J Med. 2001;345:638.
King JE, et al. Care of patients and their families with familial adenomatous polyposis.
Mayo Clin Proc. 2000;75:57.
Latreille MW, et al. Colonoscopy screening for detection of advanced neoplasia. N Engl
J Med. 2007;356:632–634.
Lierbeman DA, et al. Risk factors for advanced colonic neoplasia and hyperplastic polyps
in asymptomatic individuals. JAMA. 2003;290:2959–2967.
Offit K, et al. Reducing the risk of gynecologic cancer in the Lynch syndrome. N Engl
J Med. 2006;354:292–294.
Pickhardt P, et al. Computed tomographic virtual colonoscopy to screen for colorectal
neoplasia in asymptomatic adults. N Engl J Med. 2003;349:2191–2200.
Poston GJ, et al. Onco Surge: A strategy for improving resectability with curative intent in
metastatic colorectal cancer. J Clin Oncol. 2005;23(9):2038–2048.
Ratto C, et al. Combined modality therapy in locally advanced primary rectal cancer. Dis
Colon Rectum. 2003;46:59–67.
Rex DK, et al. ACG colorectal cancer prevention action plan: update on CT-colonography.
Am J Gastroenterol. 2006;I0I:1410–1403.
Rex DK, et al. Guidelines for colonoscopy surveillance after cancer resection: A consensus
update by the American Cancer Society and the U.S. Muti-Society Task Force on
Colorectal Cancer. Gastroenterology. 2006;130:1865–1871.
Rex DK, Maximizing detection of adenomas and cancers during colonoscopy. Am J
Gastroenterol. 2006;101:2866–2877.
Tonkef DJ, et al. Cetuximab for the treatment of colorectal cancer. N Engl J Med.
2007;357:2040–2048.
Ulrich DM and Potter JD. Folate and cancer—Timing is everything. JAMA. 2007 Jun 6;
297:2408–2409.
Winawer S, et al. Colorectal cancer screening and surveillance: Clinical guidelines and
rationale-update based on new evidence. Gastroenterology. 2003;124:544–560.
Winawer SJ, et al. Guidelines for colonoscopy surveillance after polypectomy: A consensus
update by the U.S. Multi-Society Task Force on Colorectal Cancer and the American
Cancer Society. Gastroenterology. 2006;130:1872–1885.L
79466_CH40.qxd 1/2/08 12:43 PM Page 280
40 OSTOMY CARE
T he word ostium means “opening.” Thus the words ileostomy and colostomy refer
to openings into the ileum and colon, respectively, which are the subjects of this chapter.
Other ostomies include gastrostomies and jejunostomies, which usually are formed for the
purpose of alimentation, and ileal loop urostomies, which are formed to replace the urinary
bladder.
An ileostomy is usually the end result of a total proctocolectomy. A colostomy may
be formed after a partial colon resection, typically for cancer, diverticulitis, or ischemic
disease.
280
79466_CH40.qxd 1/2/08 12:43 PM Page 281
explore the patient’s concerns about it, review possible consequences and compli-
cations, and reassure the patient that after full recovery from surgery most patients
are able to conduct normal lives, including participation in normal physical, sex-
ual, and social activities. The physician, surgeon, and ostomy nurse all play impor-
tant roles in the preoperative preparation of the patient. Patients almost always
benefit from talking with another person who has an ostomy. Such people may be
well known to the physician or the ostomy nurse or they may be contacted through
the local Ostomy Association, which is listed in the telephone book of most
medium-to-large communities. The United Ostomy Association (36 Executive
Park, Suite 120, Irvine, CA 92714; telephone 1-800-826-0826) is a source for
informational material.
B. Skin care. The skin surrounding an ostomy is at risk for injury. Patients with an
ileostomy are likely to have more skin irritation than patients with a colostomy
because ileostomy stool is liquid and contains digestive enzymes. Tape and ill-
fitting appliances can also contribute to skin excoriation.
Nondetergent soap and water are the most appropriate peristomal skin-
cleansing agents. A variety of skin-conditioning agents and seals for appliances are
available. However, an improperly fitting appliance will negate the best skin care.
C. Fluids, salt, and diet
1. Fluids and salt. Because patients with ileostomies may lose up to a liter of
water and 30 mEq of sodium in the stool per day, they are encouraged to drink
2 to 3 L of water per day and not to restrict salt. Mild-to-moderate diarrhea
may be treated by increasing fluid and sodium intake and by adding bicarbon-
ate of soda and orange juice to provide potassium. More severe diarrhea may
require intravenous fluids and electrolytes. Patients with colostomies generally
do not have problems with excessive loss of fluid and electrolytes because suf-
ficient colon remains to maintain water- and electrolyte-conserving function.
2. Diet. Patients should be encouraged to eat a normal diet, with some modifica-
tion. Gas-producing foods may cause discomfort and embarrassment. Fresh
fruits may promote loose stools. Most patients discover through trial and error
the diet to which they are best suited.
3. Vitamin and mineral supplements. Some patients require vitamin and min-
eral supplements, particularly patients who have steatorrhea or vitamin B12
malabsorption.
D. Odors. Disagreeable odors from the ostomy bag can be distressful. The odors are
due to gases that are derived from the action of bacteria on intestinal substrates
(see Chapter 34). The problem is treated by emptying the bag frequently, avoiding
gas-forming foods, and adding a deodorant to the bag, such as chlorine tablets or
sodium benzoate.
E. Ostomy dysfunction. Partial obstruction of the ostomy, usually an ileostomy,
may result from recurrent disease, impaction of nondigestible material, or kinking
of a loop of bowel just proximal to the stoma. Patients experience abdominal
cramping pain and increased ostomy effluent.
1. Diagnosis. Gentle examination of the stoma with the small finger is indicated.
This may be followed by endoscopic examination, perhaps with a pediatric
sigmoidoscope or fiberoptic gastroscope. Retrograde barium-contrast radio-
logic studies through the stoma also may be helpful.
2. Treatment is dictated by the diagnosis. Recurrent inflammatory bowel disease
may respond to a course of treatment. Some patients require surgical revision
of the stoma.
Selected Readings
Bradley M, et al. Essential elements of ostomy care. Am J Nurs. 1997;97:38.
Catanzaro J, et al. High-tech wound and ostomy care in the home setting. Crit Care Nurs
Clin North Am. 1998;10:327.
Erwin-Toth P. The effect of ostomy surgery between the ages of 6 and 12 years on psychosocial
development during childhood, adolescence and young adulthood. J Wound Ostomy
Continence Nurs. 1999;26:77.
79466_CH40.qxd 1/2/08 12:43 PM Page 282
Evans JP, et al. Revising the troublesome stoma: Combined abdominal wall recountering
and revision of stomas. Dis Colon Rectum. 2003;46:122–126.
Mitchel JV. A clinical pathway for ostomy care in the home: Process and development.
J Wound Ostomy Continence Nurs. 1998;25:200.
Turnbull GB, et al. Ostomy care: Foundation for teaching and practice. Ostomy Wound
Manage. 1999;45(suppl 1A):23.
79466_CH41.qxd 1/2/08 12:45 PM Page 283
N early everyone has experienced anorectal discomfort. Our low-fiber diet, which
results in small, hard stools; our lifestyle, which restricts the opportunities for defecation;
and our erect posture, which promotes engorgement of the hemorrhoidal plexus all com-
bine to make anorectal problems virtually ubiquitous. The anorectum also is the site of
local manifestations of more generalized disorders, such as inflammatory bowel disease.
Finally, because the rectum is a sexual organ for some people, sexually transmitted diseases
may occur at that site.
I. HEMORRHOIDS
A. Pathogenesis. Hemorrhoids are dilated veins within the anal canal and distal rec-
tum. External hemorrhoids are derived from the external hemorrhoidal plexus
below the dentate line and are covered by stratified squamous epithelium. Internal
hemorrhoids are derived from the internal hemorrhoidal plexus above the dentate
line and are covered by rectal mucosa.
Hemorrhoids are thought to develop in most instances as a consequence of
erect posture, straining at stool, heavy lifting, or childbirth. In some patients, por-
tal hypertension predisposes to hemorrhoids; rarely, hemorrhoids develop as a
result of an intraabdominal mass.
B. Clinical presentation
1. History. Hemorrhoids typically cause bleeding, which is detected as streaks of
red blood on the stool and toilet paper. Patients also may complain of anal itch-
ing or pain. However, severe pain is an unusual symptom unless the hemorrhoid
is thrombosed.
2. Physical examination. Inspection of the anus may reveal bluish, soft, bulging
veins indicative of external hemorrhoids or prolapsed internal hemorrhoids.
Nonprolapsed internal hemorrhoids cannot be seen externally and are difficult
to distinguish from mucosal folds by digital rectal examination unless they are
thrombosed. Thrombosed hemorrhoids usually are exquisitely tender.
C. Diagnostic studies. The anal canal and rectum should be examined by anoscopy
and sigmoidoscopy. Symptomatic hemorrhoids usually are accompanied by vary-
ing degrees of inflammation within the anal canal. At sigmoidoscopy, the anus and
rectum can be evaluated for other conditions in the differential diagnosis of rectal
bleeding and discomfort, such as anal fissure and fistula, proctitis and colitis, rec-
tal polyp, and cancer. Barium enema x-ray examination or colonoscopy should be
performed in patients over age 50 and in patients of any age whose stool remains
positive for occult blood after appropriate treatment for hemorrhoids.
D. Treatment. A high-fiber diet, stool softeners, and avoidance of straining at stool
and heavy lifting may be sufficient to treat mild hemorrhoidal symptoms. Warm
baths twice a day and anal lubrication with glycerine suppositories provide further
comfort. Addition of medicated suppositories, such as Anusol-HC (containing
hydrocortisone), may help reduce associated inflammation. However, steroid-
containing medications should be limited to 2 weeks of continuous use to avoid
atrophy of the anal tissues.
Additional treatment usually requires the expertise of a gastroenterologist or
surgeon. Rubber-band ligation is usually the first definitive treatment. The proce-
dure requires no anesthesia and produces excellent results in most patients.
283
79466_CH41.qxd 1/2/08 12:45 PM Page 284
IV. PROCTITIS
A. Pathogenesis. Proctitis is an inflammation of the rectal mucosa. It may be idio-
pathic or related to a specific cause, such as radiation or gonococcal infection.
B. Clinical presentation. Patients with proctitis typically complain of pain and
bleeding on defecation. Stools may be loose, but often they are well formed. In
fact, some patients move their bowels infrequently to avoid pain and thus become
constipated, which aggravates symptoms even more. Digital rectal examination
may reveal a small amount of bloody stool or mucus.
Idiopathic proctitis is a chronic, relapsing, localized condition that progresses
to ulcerative colitis in some patients. The appearance of an erythematous, friable,
sometimes hemorrhagic mucosa, however, is indistinguishable from that of ulcer-
ative colitis. Small ulcerations and pus suggest the possibility of gonococcal
proctitis. Ulcers and vesicles of the distal rectum, which may extend to the peri-
anal skin, suggest herpes proctitis.
C. Diagnostic studies. Sigmoidoscopy shows the inflammation to be limited to the
rectum, sometimes within several centimeters of the anus. Mucosal biopsy and
culture of rectal secretions should be obtained. More extensive bowel involvement
may be documented by barium enema or colonoscopy.
D. Treatment
1. Idiopathic proctitis. Treatment of idiopathic proctitis usually consists of
mesalamine enemas (i.e., Rowasa) or suppositories (i.e., Canasa) or steroid ene-
mas once or twice daily for 4 to 6 weeks. Patients should be instructed to lie in
the left decubitus position and gently insert a suppository or an enema before
bedtime and another in the morning after a bowel movement. Some patients
require one or two mesalamine or steroid enemas per day for prolonged periods
to control symptoms. Patients who have frequent recurrences of proctitis may
respond to prophylactic treatment with sulfasalazine (1 to 2 g b.i.d.) or a
mesalamine preparation (e.g., Asacol 800 mg t.i.d. or Colazal 2.25 g t.i.d.).
Steroid or mesalamine enemas also are somewhat effective in treating radiation
proctitis.
2. Infectious proctitis. Treatment of infectious proctitis is directed at the
causative agent. For example, gonococcal proctitis typically responds to aque-
ous procaine penicillin G, 4.8 million units intramuscular (IM), plus 1 g of oral
probenecid. For penicillin-allergic patients, tetracycline (orally, 1.5 g followed
by 0.5 g q.i.d. for 4 days) is effective.
V. RECTAL PROLAPSE
A. Pathogenesis. Rectal prolapse ranges in severity from prolapse of a small portion
of rectal mucosa to protrusion of the entire rectal wall through the anus (proci-
dentia). Straining at stool is thought to be causative, but pelvic surgery,
childbearing, and weak pelvic musculature are contributive factors.
B. Clinical presentation and diagnosis. Patients with rectal prolapse complain of
bleeding, passage of mucus, and irritation of the exposed mucosa. In some
patients, prolapse of the rectal mucosa or of a substantial portion of the rectum
may be observed. Other patients must strain to produce the prolapse. Poor anal
sphincter tone usually is evident on digital examination. Sigmoidoscopy and, in
most patients, barium enema are indicated to define the extent of irritated mucosa
and to rule out associated conditions.
C. Treatment. If the prolapse is mild and confined to the mucosa, the treatment
regimen is the same as for hemorrhoids. However, frank procidentia requires
operative treatment if the surgical risk is good.
B. The diagnosis is made by the typical history of rectal pain and absence of physi-
cal findings. Rectal and sigmoidoscopic examinations should be performed to rule
out other treatable disorders.
C. Treatment is symptomatic, consisting of warm baths and, if symptoms are
frequent, analgesics and muscle relaxants. Because of the transient nature of the
disorder, however, it is often difficult to determine whether treatment has been
effective. The condition usually resolves spontaneously.
VIII. TUMORS
A. Pathogenesis. Tumors of the anus sometimes are confused with other inflamma-
tory and infectious lesions. The most common anal tumor is epidermoid carci-
noma, but adenocarcinoma and malignant melanoma also can occur.
B. Clinical presentation and diagnosis. Patients complain of anal pain, bleeding,
itching, or presence of a mass. They may think that they are having more trouble
with their hemorrhoids. The diagnosis of neoplasm is suspected by visual inspec-
tion, digital examination, and anoscopy; it is confirmed by biopsy. A search for
local extension and metastatic disease is warranted. This typically includes routine
liver tests, chest x-ray, and computed tomography of the abdomen and pelvis.
C. Treatment. Small anal tumors can be treated by local excision. More extensive lesions
require the addition of radiation therapy or chemotherapy, or an abdominoperitoneal
resection of the rectum and anus.
Selected Readings
Beets-Tan RGH. Preoperative MR imaging of anal fistulas: Does it really help the surgeon?
Radiology. 2001;218:75.
Cheung O, et al. The management of pelvic floor disorders. Ailment Pharmacol Ther.
2004;19:481–495.
Diamant NE, et al. AGA technical review on anorectal testing techniques. Gastroenterology.
1999;116:735.
Dorsky R. Coping with the pain and annoyance of hemorrhoids. Dis Health Nutr. January/
February 2000:21.
Gosens MJEM, et al. Improvement of staging by combining tumor and treatment
parameters: the value for prognostication in rectal cancer. Clin Gastroenterol Hepatol.
2007;5(8):997–1003.
79466_CH41.qxd 1/2/08 12:45 PM Page 287
Jones MP, Post JP, Crowell MD. High-resolution manometry in the evaluation of anorectal
disorders: a simultaneous comparison with water-perfused manometry. Am J
Gastroenterol. 2007;102:1–6.
Kang YS, et al. Pathology of the rectal wall in solitary ulcer syndrome and complete rectal
prolapse. Gut. 1996;38:587.
Koslin DB. Anal fistulae. Rev Gastroenterol Disord. 2001;1:56.
Madoff RD, et al. AGA technical review on the diagnosis and care of patients with anal
fissure. Gastroenterology. 2003;124:235–245.
Masderstein EL, et al. Surgical management of rectal prolapse. Nat Clin Pract Gastroenterol
Hepatol. 2007;4(10):552–561.
Rao SC. How useful are manometric tests of anorectal function in the management of
defecation disorders? Am J Gastroenterol. 1999;116:735.
Rao SC. Dyssynergic: disorders of the anorectum. Gastroenterol Clin North Am. 2001;
31:97–114.
Rao SSC. Diagnosis and management of fecal incontinence. Am J Gastroenterol. 2004;
99:1585–16.
Sentovic SM. Fibrin glue for anal fistulas: Long term results. Dis Colon Rectum. 2003;46:
498–502.
Sobhani I, et al. Prevalence of high-grade dysplasia and cancer in anal canal in human
papilloma virus-infected individuals. Gastroenterology. 2001;120:857.
Sotlar K, et al. Human papilloma virus type 16-associated primary squamous cell carcinoma
of the rectum. Gastroenterology. 2001;120:988.
Wald A, Fecal incontinence in adults. N Engl J Med. 2007;356:1648–1655.
79466_CH42.qxd 1/2/08 12:46 PM Page 288
42 SEXUALLY TRANSMITTED
ENTERIC DISORDERS
II. DIAGNOSIS. Knowledge of a patient’s sexual practices can be helpful but is not
necessary to make an etiologic diagnosis of one of the conditions listed in Table 42-1.
288
79466_CH42.qxd 1/2/08 12:46 PM Page 289
A. Oropharyngeal disorders
1. Gonococcal pharyngitis can present with sore throat, exudate of the pharynx
and tonsils, and ulcerations of the tongue and buccal mucosa. However, some
patients with gonococcal infection of the oropharynx may be asymptomatic but
transmit the disease to others. If gonococcal infection is suspected, the lesions
should be cultured immediately.
2. Syphilis. The oral lesions of syphilis are elevated, round, sometimes ulcerated,
and usually painless. They occur most frequently on the lips but also may be
found on the tongue or tonsils, and elsewhere within the mouth and pharynx.
These lesions heal within several weeks but are superseded by the systemic signs
and symptoms of secondary syphilis, namely, fever, sore throat, lymphadenopa-
thy, pruritus, and skin lesions. A nonspecific pharyngitis may also be present.
The darkfield examination of the primary oral lesions may be confused by the
presence of normal oral spirochetes. Serologic tests may not be positive until the
secondary form of the disease appears. Aspiration of enlarged lymph nodes for
darkfield examination may give an early diagnosis.
3. Herpes pharyngitis. Patients with herpes pharyngitis have erythema and ulcer-
ations of the mouth, tongue, gingiva, and pharynx. Exudate and lym-
phadenopathy also may be present. The presence of a herpes infection in the
patient’s sexual partner makes the diagnosis more likely. Facilities to culture the
virus may not be available. The diagnosis can be inferred from serologic tests
directed against the herpes simplex virus.
B. Gastric disorders. Although the infectious agents that cause sexually related
enteric diseases may pass through the stomach, gastritis per se is not a feature of
those diseases. However, secondary syphilis, a rare complication, can involve the
stomach. In syphilitic gastritis, the mucosa becomes ulcerated and infiltrated with
chronic inflammatory cells. Patients complain of abdominal pain and vomiting.
Loss of weight is common. Spirochetes can be identified by darkfield examination
of mucosal biopsies. The diagnosis can be confirmed by serologic tests for syphilis.
The stomach also can be involved with lymphoma or with Kaposi’s sarcoma when
these conditions affect AIDS patients.
C. Enteric disorders. Patients with sexually transmitted enteric infections appear sim-
ilar clinically to those who contract the infection in some other manner. The diagno-
sis and management of infectious diarrheal conditions are discussed in Chapter 29.
D. Anorectal disorders
1. Gonococcal proctitis may present with anorectal pain, tenesmus, and a
mucopurulent discharge. At sigmoidoscopy, the rectal mucosa appears red and
contains pus and small ulcers. Mucosal biopsy shows nonspecific inflammation.
The diagnosis is made by Gram’s stain and culture of rectal aspirates or mucosal
biopsies. Many patients with anorectal gonococcal infection are asymptomatic
but are a source of infection to others.
2. Anorectal syphilis is characterized by a painless chancre, which often is mis-
taken for an anal fissure. The diagnosis of syphilis is based on a high index of
suspicion in a susceptible patient, darkfield examination of the lesion, and sero-
logic follow-up.
3. Chlamydia trachomatis can infect the rectum and intestine with either the
lymphogranuloma venereum (LGV) serotype or the non-LGV serotype. The
non-LGV infections are similar to infections caused by gonococci. They are
usually mildly symptomatic with anorectal discharge, tenesmus, anorectal pain,
and mild mucosal inflammation. In contrast, LGV infections typically cause
severe proctocolitis. Anorectal pain is severe and is accompanied by bloody
purulent discharge, tenesmus, and diarrhea. The rectal mucosa is friable and
ulcerated. The histologic appearance may be similar to that in Crohn’s disease,
with diffuse inflammation and granulomas. Stricturing and fistula formation
add to the confusion with Crohn’s disease. C. trachomatis can be isolated from
the rectum. Serologic tests of LGV infection confirm the diagnosis. Response to
treatment with tetracycline also is confirmatory of the diagnosis.
79466_CH42.qxd 1/2/08 12:46 PM Page 290
Gonococcus
Oropharyngeal APPG, 4.8 million units TS (80 mg/400 mg),
Rectal IM probenecid, 1 g p.o. 9 tablets daily 5 d
APPG, 4.8 million units Spectinomycin, 2 g IM
IM probenecid, 1 g p.o.
Syphilis Benzathine penicillin G, Tetracycline, 500 mg p.o.
2.4 million units IM q.i.d. 15 d
Chlamydia
LGV Tetracycline, 500 mg p.o. Erythromycin, 500 mg p.o.
q.i.d. 21 d q.i.d. 21 d
Non-LGV Tetracycline, 500 mg p.o. Erythromycin, 500 mg p.o.
q.i.d. 7 d q.i.d. 7 d
Herpes simplex Acyclovir, 1 tablet Supportive therapy
5 times/d 10 d
Papilloma virus (condyloma) Podophyllin (10%–25%) Cryotherapy or surgery
on lesion every other day
Campylobacter Erythromycin, 500 mg Tetracycline, 500 mg
p.o. q.i.d. 7 d p.o. q.i.d. 7 d
Salmonella Antibiotic treatment not Ampicillin, 1 g intravenous
necessary except for q4h 10 d; TS, 2
severe cases tablets p.o. q12h 10 d
Shigella TS, 160 mg/800 mg Ampicillin, 500 mg
p.o. b.i.d. 7 d p.o. q.i.d. 7 d
Giardia Metronidazole, 250 mg Quinacrine hydrochloride,
p.o. t.i.d. 7 d 100 mg t.i.d. 7 d
Entamoeba Metronidazole, 750 mg Diiodohydroxyquin, 650 mg
p.o. t.i.d. 7 d p.o. t.i.d. 20 d
Candida (esophagitis) Fluconazole, Ketoconazole, 200 mg
50–100 mg 10 d p.o. b.i.d.
Cryptosporidium Supportive therapy only
in immunocompetent
patients
Spiramycin, 1 g p.o. q.i.d.
in patients with AIDS
Isospora Supportive therapy only
in immunocompetent
patients
Spiramycin, 1 gm
p.o. q.i.d. in patients
with AIDS
Mycobacterium Multidrug antituberculous
avium-intracellulare therapy
4. Herpes can involve not only the anus and rectum but also the perianal skin, with
the typical herpetic vesicles and ulcerations. Patients complain of pain, rectal
discharge, and bloody stool. The diagnosis often can be made on clinical presenta-
tion alone. Histologic examination of rectal mucosal biopsies may reveal intranu-
clear inclusion bodies in addition to focal ulcers and perivascular mononuclear cell
infiltrates. Culture of the virus is diagnostic but may not be available. The serologic
diagnosis requires seroconversion or a fourfold or greater rise in antibody titer.
5. Condylomata acuminata, or anal warts, are caused by the human papilloma
virus and are common in people who practice anal intercourse. They appear as
small brownish papules around the anus.
6. Traumatic injury may have many causes and may take many forms. Dilatation
and stretching of the anus from anal intercourse can cause fissures and tearing of
the mucosa and underlying structures. The practice of inserting fingers, hands,
arms, or foreign objects into the rectum increases the likelihood of anal and rectal
trauma. The diagnosis of anorectal trauma is based on historical information unless
a foreign object is apparent. Traumatic and infectious disorders can coexist.
E. Liver disorders
1. Gonococcal perihepatitis is a consequence of spread of infection from the
fallopian tubes in women and through lymphatics or blood in both men and
women. The signs and symptoms are similar to those of acute cholecystitis,
namely, acute right upper abdominal pain that may radiate to the shoulder,
nausea and vomiting, and fever. Although the patient may or may not have
symptomatic pelvic gonorrhea, cultures of the uterine cervix for gonococcus
typically are positive. The condition must be differentiated from acute chole-
cystitis and other causes of acute abdominal pain (see Chapter 13).
2. Syphilitic hepatitis is an unusual manifestation of secondary syphilis. It is
thought to be due to an infiltration of the liver by spirochetes, which arrive
through the portal system from primary lesions in the rectum. The clinical picture
is one of hepatomegaly and obstructive jaundice. Liver biopsy is diagnostic when
it shows granulomas and spirochetes within the liver.
3. The hepatitis viruses and cytomegalovirus can be transmitted by sexual con-
tact. The diagnosis and management of these infections are discussed in Chapter 50.
III. TREATMENT. Effective treatment of the infectious causes of sexually related enteric
disease depends on the identification of the offending agent or agents and use of the
appropriate antibiotic regimen (Table 42-2). Supportive treatment may involve intra-
venous fluids and nourishment in patients who are unable to swallow or who have
severe diarrhea. There is no known effective therapy for Kaposi’s sarcoma. Lymphoma
may respond to chemotherapy or radiation therapy (see Chapter 32).
Selected Readings
Jones DJ, Goorney BP. ABC of colorectal diseases: Sexually transmitted diseases and anal
papillomas. BMJ. 1992;305:820.
Laughon BE, et al. Prevalence of enteric pathogens in homosexual men with and without
acquired immunodeficiency syndrome. Gastroenterology. 1988;94:984.
Owen WF Jr. The clinical approach to the male homosexual patient. Med Clin North Am.
1986;70:499.
Quinn TC. Clinical approach to intestinal infections in homosexual men. Med Clin North
Am. 1986;70:611.
Smith PD, et al. Gastrointestinal infections in AIDS. Ann Intern Med. 1992;116:63.
Surawicz CM, et al. Anal dysplasia in homosexual men: Role of anoscopy and biopsy.
Gastroenterology. 1993;105:658.
Weller ID. The gay bowel. Gut. 1985;26:869.
79466_CH43.qxd 1/2/08 12:53 PM Page 292
292
79466_CH43.qxd 1/2/08 12:53 PM Page 293
Site Organism
HIV, human immunodeficiency virus; HSV, herpes simplex virus; CMV, cytomegalovirus;
MAC, Mycobacterium avium complex.
with esophageal ulcers. Kaposi’s sarcoma (KS) lesions may be seen in the
oropharynx and are associated with GI KS.
3. Laboratory tests. The stage of immunodeficiency determines the differential
diagnosis of esophageal disease. The levels of HIV viremia and CD4 lympho-
cyte count are the two most important laboratory tests. OIs of the esophagus
are uncommon until the CD4 count falls below 200/mm3. IEU and CMV
esophagitis are rarely seen until the CD4 count drops below 100/mm3.
4. Empiric therapy. Because candidiasis is the most common cause of esophageal
disease in HIV-infected patients, an empiric trial of antifungal therapy is reason-
able. Further diagnostic testing is then based on the clinical response. Empirical
therapy with fluconazole 200 mg per day followed by 100 mg per day for a 7- to
10-day treatment course is recommended. The clinical response of Candida
esophagitis to the treatment with fluconazole is rapid (within 3 days). If no sub-
stantial improvement occurs in 3 to 5 days, endoscopy is recommended. Most
patients who fail antifungal therapy do not have candidiasis, but rather esophageal
ulcers. Additional empirical trials such as antiviral therapy with acyclovir sodium
or ganciclovir sodium are discouraged. In patients with CD4 counts higher than
200/mm3 and with symptoms typical of GERD, a trial of gastric acid suppressive
therapy with high-dose, proton pump inhibiters is recommended.
5. Barium swallow/upper GI (UGI) series may document esophageal candidia-
sis or ulceration in symptomatic patients. However, because there are many
causes of esophageal ulcers in AIDS, patients will require endoscopic examina-
tion for biopsies.
6. Endoscopy is a means of directly visualizing the UGI mucosa as well as for
obtaining biopsies for histopathologic examination of the lesions visualized.
The endoscopic appearance of a lesion often suggests the diagnosis. Candidiasis
appears as cottage cheese–like plaques and coats most of the esophageal
mucosa. However, because Candida coexists with other lesions at least in 25%
of cases, multiple biopsies are required.
Viral esophagitis may present as diffuse esophagitis or small superficial
ulcers with HSV and as one or more large, well-circumscribed ulcers with
CMV. Biopsies of the ulcer margins and base are required for histopathologic
diagnosis. Immunohistochemical stains of the biopsies will increase the diag-
nostic accuracy. IEU may resemble CMV ulcers. Multiple biopsies of the ulcer
base and margins are required to exclude the presence of CMV. KS appears as
elevated blue-violatious nodules.
7. Treatment is to be directed to the etiology of the esophageal lesion. Esophageal
candidiasis responds well to fluconazole or itraconazole. Both agents are also
available in liquid formulation for patients who cannot swallow tablets or
capsules.
HSV esophagitis is treated with acyclovir sodium or valacyclovir
hydrochloride. For CMV disease, ganciclovir sodium is effective. The newer
drug, cidofovir, may be administered by a once-weekly injection. IEU may be
treated with either prednisone or thalidomide: The response rate is higher
than 90%. Patients should be treated or continued on HAART to help expedite
healing and to prevent relapse.
c. Endoscopic examination of the UGI tract and the colon are invaluable in
the evaluation of diarrhea in patients with AIDS. In addition to direct visu-
alization, biopsies are obtained for histologic examination.
4. Treatment. Drug-induced diarrhea must always be included in the differential
diagnosis of diarrhea in patients undergoing HAART. If diarrhea stops when
the retroviral agents are stopped and returns when they are restarted, one or
more of these drugs is implicated.
Even though CMV colitis responds to ganciclovir sodium in over 50% of
patients, effective therapy for many of the OIs in AIDS is lacking. So far, there
is no effective treatment for microsporidia and Cryptosporidia infection.
However, improvement of the immune function by HAART is essential and
may lead to remission of Cryptosporidia- and microsporidia-induced diarrhea.
Bacteria
Campylobacter Erythromycin 250–500 mg p.o. q.i.d. for 7 d or ciprofloxacin
500 mg p.o. b.i.d. for 7 d
Salmonella Amoxicillin 1 g p.o. t.i.d for 3–14 d or trimethoprim/sulfamethoxazole
DS p.o. b.i.d. for 14 d or ciprofloxacin 500 mg p.o. b.i.d. for 7 d
Shigella Trimethoprim/sulfamethoxazole DS p.o. b.i.d. for 5–15 d or ampicillin
500 mg p.o. q.i.d. for 5 d or ciprofloxacin hydrochloride 500 mg
p.o. b.i.d. for 7 d
Clostridium difficile Metronidazole 500 mg p.o. t.i.d. for 7–10 d or vancomycin
hydrochloride 125–500 mg p.o. q.i.d. for 7–10 d
Fungi
Candida Ketoconazole 100–200 mg b.i.d. for 10–14 d or fluconazole
500–100 mg b.i.d. for 10–14 days for amphotericin B
0.3–0.6 mg/kg/body weight daily
Viruses
Herpes simplex Acyclovir 200–800 mg 5 times a day for 7 d or 5–12 mg/kg q8h for 7 d
Cytomegalovirus Ganciclovir 5 mg/kg IV b.i.d. for 14–21 d or foscarnet sodium
60 mg/kg IV q8h for 14 d, then 90–120 mg/kg daily; maintenance
6 mg/kg IV daily 5 d per week
Protozoa
Giardia Metronidazole 250 mg p.o. t.i.d. for 5 d or quinacrine hydrocholoride
100 mg t.i.d. for 5 d
Entamoeba Metronidazole 750 mg p.o. t.i.d. for 10 d, then iodoquinol 650 mg
t.i.d. for 20 d
Isospora Trimethoprim/sulfamethoxazole DS p.o. t.i.d for 10 d then twice
weekly for 3 weeks
Cryptosporidium Unknown (trials with azithromycin hydrochloride, letrazuril, and
paromomycin)
Microsporidia Unknown (trials with metronidazole and albendazole)
Mycobacteria
Mycobacterium Unknown (trials with clarithromycin 500–1000 mg p.o. b.i.d. for 2–4
avium-complex weeks and ethambutin 15–20 mg/kg/d, rifabutin 300–600 mg daily,
clofazimine 100–200 mg daily, ciprofloxacin hydrochloride 750 mg
daily, and amikacin 7.5 mg/kg daily or b.i.d.)
HIV, human Immunodeficiency virus; p.o., per os; b.i.d., twice daily; q.i.d., four times daily; t.i.d., three
times daily; IV, intravenous.
79466_CH43.qxd 1/2/08 12:53 PM Page 297
IV. HEPATOBILIARY DISEASE IN AIDS. Abnormal results of liver chemistry tests occur
in about 60% of patients with AIDS. Up to 80% of these patients have hepatomegaly,
and nearly 85% have histologic changes in hepatic parenchyma.
The spectrum of hepatobiliary disease in patients with AIDS is summarized in
Table 43-3. This spectrum includes viral hepatitis; granulomatous liver disease sec-
ondary to drugs; fungal, protozoan, bacterial, and mycobacterial infections; steatosis;
nonspecific portal inflammation; sinusoidal abnormalities including peliosis hepatis;
neoplasms such as KS and non-Hodgkin’s lymphoma; and biliary diseases including
acalculous cholecystitis, ampullary stenosis, and sclerosing cholangitis. These disor-
ders may be superimposed on previous hepatobiliary disease resulting from alco-
holism, intravenous (IV) drug abuse, and viral hepatitis.
A. Disorders
1. Viral infections (see also Chapter 50)
a. Hepatitis A virus (HAV) is transmitted through the fecal-oral route. It is
prevalent in both IV drug users and homosexual men. Many anti-HIV–
positive patients have anti-hepatitis A (IgG), which is the serologic evidence
of past exposure to hepatitis A virus, with full recovery. There is no chronic
form of HAV and no evidence that the clinical course of an acute HAV infec-
tion is altered in patients with AIDS. Treatment is supportive.
b. Hepatitis B virus (HBV) is transmitted parenterally by contaminated needles
and sexually from infected people. IV drug users and homosexual men are at
high risk of the development of HBV infection. In fact, approximately 90% of
patients with AIDS have serologic evidence of HBV infection, and 10% to 20%
are chronic carriers. Patients with AIDS and previous HBV infection have nor-
mal or slightly elevated serum transaminase levels. This is because HBV is not
directly cytopathic to the infected hepatocytes and the degree of inflammatory
response and liver damage is largely dependent on the host’s immunologic status.
Patients with HIV-induced immune suppression are likely to have less
inflammatory response and an improvement in the biochemical and histologic
features of chronic HBV infection. However, it has been noted that there
is increased replication of HBV in HIV-infected individuals, determined by an
increase in the HBV deoxyribonucleic acid (DNA) polymerase activity and an
increase in HBe antigen levels and HB core antigen–positive hepatocyte nuclei.
Patients with AIDS who have acute hepatitis B have increased viremia
and an increased risk of the development of chronic hepatitis B. Patients
with hepatitis B who are also infected with HIV respond poorly to interferon
therapy, even in the absence of AIDS. The presence of HIV antibodies is also
associated with a suboptimal response to HBV vaccination in terms of both
the level of anti-HBs and the percentage of patients responding to the vac-
cine. Higher doses of the vaccine may need to be used in this population.
Measurement of HB surface antibody titers is recommended in HIV-positive
people to determine whether the desirable titer of greater than 10 milli-
International Units per liter (mIU/L) is present.
c. Delta hepatitis. Hepatitis delta virus (HDV) is a hepatotropic RNA virus
dependent on HB surface antigen for its replication and expression. It is
known to cause coinfection with HBV or superinfection in patients with
chronic hepatitis B. Patients with HIV infection have slightly higher serum
aminotransferase levels with HDV infection. Reactivation of HDV after HIV
infection has been reported.
d. Hepatitis C virus (HCV) infection and positive serologic studies for HCV anti-
body are often seen in patients infected with HIV. Most patients have HCV,
chronic active hepatitis, or cirrhosis. The response to antiviral therapy with
ribovirin and pegylated interferon has been poorer in these patients, as
compared to patients who are not co-infected with HIV. However, HCV
co-infection should be treated in the apprapriate patient.
e. Herpes simplex virus. More than 95% of homosexual men with AIDS have
serologic evidence of herpes simplex virus (HSV) infection. Patients with AIDS
may have HSV encephalitis, esophagitis, or orolabial or genital HSV infections
with pain, ulceration, and progressive tissue destruction. Hepatitis may occur
with widely disseminated HSV infection. In most of these conditions, patients
manifest orocutaneous or genital vesicles or both, ulcers, fever, hepatomegaly,
and leukopenia. Fulminant hepatitis may develop with overwhelming infection,
and patients may have coagulopathy, hepatic encephalopathy, and shock.
Diagnosis may be established by histopathologic examination of a liver biopsy
specimen. The virus may be cultured from the blood, urine, cutaneous lesions, or
liver. Mortality is very high despite therapy with acyclovir sodium or vidarabine.
f. Epstein-Barr virus infection. The course of hepatitis in patients with AIDS
who also have the Epstein-Barr virus has not been well characterized.
g. CMV infection usually produces subclinical disease in immunocompetent
adults. Occasionally patients have fever, hepatomegaly, and slightly elevated
aminotransferase levels. CMV infection may remain latent after primary
infection and recur with immunocompromise.
Approximately 95% of homosexual men have serologic evidence of
previous CMV infection. In HIV-infected patients, CMV may produce coli-
tis, esophagitis, pneumonitis, and retinitis. It is usually disseminated when
the liver is involved. In patients with CMV hepatitis, the serum alkaline
phosphatase and aminotransferase levels are moderately increased. Hepatic
involvement ranges from the asymptomatic carrier state to fulminant
hepatic necrosis. The diagnosis is made by liver biopsy in affected patients.
CMV commonly produces a parenchymal and portal mononuclear cell infil-
trate and focal hepatic necrosis. Occasionally granulomas are present.
Cytoplasmic inclusion bodies seen in the hepatocytes and in situ hybridiza-
tion and immune fluorescence techniques can rapidly detect CMV. CMV can
be cultured from urine, blood, and tissue from infected sites.
Treatment is with IV ganciclovir sodium, which may stabilize the clinical
course of patients with CMV infection but may cause neutropenia. Foscarnet
may be used as an alternative therapy without associated neutropenia.
2. Bacterial and mycobacterial infections
a. Mycobacterium avium-intracellulare (MAI) is the most common oppor-
tunistic pathogen causing hepatic infection in AIDS. It is usually found in
patients with AIDS who have had previous opportunistic infections. The pre-
sentation is often with fever, malaise, anorexia, weight loss, diarrhea,
79466_CH43.qxd 1/2/08 12:53 PM Page 299
the liver chemistry tests and clinical improvement. A liver biopsy or other
appropriate workup may be necessary to determine the precise nature of the
hepatobiliary abnormality in patients in whom the suspected drug is needed or
in those in whom withdrawal of the drug does not result in improvement.
6. Hepatic neoplasms
a. KS is the most common neoplasm found in patients with AIDS. It occurs pri-
marily in homosexual men, in whom it behaves as an aggressive malignancy
with visceral and cutaneous lesions. About one half of the patients have GI
lesions, which appear as violaceous macules on endoscopy. The lesions are
generally submucosal and may present with bleeding or obstruction.
About one third of the patients have hepatic involvement. The liver
chemistry tests may be normal, or there may be an elevation of the serum
alkaline phosphatase levels. On CT scan, hepatic lesions have a nonspecific
appearance. Unguided percutaneous liver biopsy is insensitive in the diagno-
sis of KS. Occasionally KS lesions can be seen on laparoscopy if the lesions
are superficial and anteriorly located. In general, hepatic involvement is
rarely documented antemortem.
Macroscopically, the lesions are multifocal and may occur at subcapsular
hilar and intrahepatic locations. Histologically, KS is seen as multifocal areas of
vascular endothelial cell proliferation with pleomorphic spindle-shaped cells
and extravasated red blood cells. Sinusoidal dilatation with vascular lakes may
be present. The clinical spectrum may include peliosis hepatis and angiosarco-
matous lesions.
Treatment includes radiotherapy or chemotherapy with vinblastine,
vincristine, or etoposide. Interferon therapy also may produce considerable
tumor reduction.
b. Lymphoma. In HIV-infected patients, the development of lymphoma is
considered a criterion for AIDS. Patients with AIDS, like other immunosup-
pressed patients, have an increased risk of the development of non-
Hodgkin’s lymphoma, often of B-cell origin. Patients with AIDS in whom
lymphoma develops are usually homosexuals.
Lymphoma in patients with AIDS is often extranodal with involvement
of unusual sites such as the central nervous system or rectum. Most patients
have multiorgan involvement. Primary hepatic lymphoma also may occur.
Patients initially may have lymphadenopathy, hepatomegaly, jaundice,
right upper quadrant pain, and systemic symptoms such as fever, malaise,
and night sweats.
Hyperbilirubinemia and considerable elevations of serum alkaline
phosphatase levels usually occur in advanced illness. CT and ultrasound are
helpful imaging techniques in the diagnosis of hepatic lymphoma. The
lesions are usually multifocal and may obstruct the biliary tract and result in
ductal dilatation.
Histologic examination of liver biopsy specimens obtained with CT or
laparoscopic guidance confirms the diagnosis. The lymphomas are usually
high grade and respond less well to chemotherapy than they do in immuno-
competent patients.
c. Other malignancies. Hepatic metastasis from cancers developing in other
sites has been seen in patients with AIDS. These include malignant
melanoma, adenocarcinoma, and small-cell cancers. The immunodeficiency
may permit dissemination to distant sites, including the liver.
7. Biliary tract disease. In addition to diseases of the biliary tract seen in
immunocompetent patients, opportunistic infections involving the gallbladder
and the biliary tract occur in patients with AIDS and may present with atypical
findings, sepsis, or acute abdomen.
a. Acalculous cholecystitis is rarely seen in immunocompetent people. It
is sometimes associated with total parenteral nutrition and biliary sludge.
In patients with AIDS, it may have a subacute clinical course or may present
with fever and right upper quadrant abdominal pain. Patients may also have
79466_CH43.qxd 1/2/08 12:53 PM Page 302
Selected Readings
Abou Lafia DM. AIDs-related non-hodgkins lymphoma. Infect Med. 2007;24(11):470–481.
Bivi EJ, et al. Diagnostic yield and cost effectiveness of endoscopy in chronic human immuno-
deficiency virus-related diarrhea. Gastrointest Endosc. 1998;48:354.
Blanshard C, et al. Investigation of chronic diarrhea in acquired immunodeficiency syndrome.
A prospective study of 155 patients. Gut. 1996;38:824.
Carr A, et al. Treatment of HIV associated microsporidiosis and cryptosporidiosis with combination
antiretroviral therapy. Lancet. 1996;38:824.
Dietrich DT, et al. Diagnosis and treatment of esophageal diseases associated with HIV infection.
Am J Gastroenterol. 1996;91:2265.
Koziel MJ, et al. Viral hepatitis in HIV infection. N Engl Med. 2007;356:1445–1457.
Poles MA, et al. HIV-related diarrhea is multifactorial and fat malabsorption is commonly present,
independent of HAART. Am J Gastroenterol. 2001;96:1831.
Re VI, et al. Prevalance, risk factors, and outcomes for occult hepatitis B infection among
HIV-infected patients. J Acquir Immune Defic Syndr. 2007;44:315–320.
Smith JO, et al. Hepatitis C and HIV. Curr Gastroenterol Rep. 2007;9:83–90.
Soriano V, et al. Core of patients coinfected with HIV and hepatitis C virus. Aids. 2007;21(9):1073–1089.
Wilcox CM, et al. Fluconazole compared with endoscopy for human immunodeficiency virus
infected patients with esophageal symptoms. Gastroenterology. 1996;110:1803.
Wilcox CM, et al. Review article: The therapy of gastrointestinal infections associated with acquired
immunodeficiency syndrome. Aliment Pharmacol Ther. 1997;11:435.
Wilcox CM. Current concepts of gastrointestinal disease associated with human immune deficiency
virus infection. Clin Perspect Gastroenterol. January/February 2001:9.
79466_CH44.qxd 1/2/08 12:54 PM Page 303
OCCULT OBSCURE
GASTROINTESTINAL BLEEDING 44
303
79466_CH44.qxd 1/2/08 12:54 PM Page 304
result for occult blood because ascorbic acid inhibits the pseudoperoxidase
activity of heme. Aspirin and other nonsteroidal antiinflammatory drugs
may injure the gastric mucosa and cause bleeding, thus making the occult
blood test positive.
3. After 3 days of this regimen, the patient should smear stool on the two test win-
dows of a diagnostic card each day for the next 3 days (or the next three stools,
if stools are less frequent than daily) and return the cards to the physician.
Rehydrating the cards before testing increases the diagnostic yield.
II. DIAGNOSIS
A. Clinical presentation. Patients with occult blood in the stool typically feel well.
They may have symptoms related to the cause of the bleeding, for example, peptic
complaints, or abdominal discomfort due to a GI tumor, but they usually do not
have symptoms attributable to blood loss. They may be anemic (hemoglobin level
of less than 12 g/dL for women and less than 13 g/dL for men and serum ferritin
level of less than 45 mg/L) due to either the chronic loss of blood or, more likely,
the underlying disorder, or they may have normal levels of hemoglobin and hema-
tocrit. The presence of occult blood may be detected by routine screening exami-
nation or in response to the patient’s symptoms.
B. Diagnostic studies
1. Examination of the colon by colonoscopy is the preferred diagnostic study
in the evaluation of patients for occult GI bleeding. The use of flexible
sigmoidoscopy and air-contrast barium enema has been used as an alterna-
tive to colonoscopy; however, the sensitivity and specificity is less. Virtual
colonoscopy is in the process of development as an alternative to colonoscopy.
2. Examination of the upper GI tract. If the examination of the colon is nondi-
agnostic, examination of the upper GI tract is indicated. The clinician may elect
to study the UGI tract first if the patient’s history suggests UGI disease or lesion.
Endoscopy is the preferred method of studying the UGI tract for mucosal lesions.
3. Examination of the small intestine may be accomplished by radiologic or
endoscopic techniques. The most widely available technique is barium upper GI
and small-bowel series. Unfortunately, the diagnostic yield of this technique is
inferior to small-bowel enema or enteroclysis, which is more arduous and
not widely available. Enteroclysis is performed by passing a tube by mouth into
the proximal duodenum and instilling barium followed by air. This technique
provides an excellent air–barium contrast and is effective in detecting mass
lesions of the SI, but is ineffective in detecting mucosal lesions.
4. Enteroscopy is indicated for the evaluation of the SI for the detection of
mucosal lesions. Push enteroscopy is the usual technique used and entails a
perioral insertion of a push enteroscope or a pediatric colonoscope via the UGI
tract into the distal duodenum, jejunum and possibly into the ileum. Biopsy and
endoscopic therapy can be performed when a lesion is found. Sonde
enteroscopy involves the passage of a long enteroscope into the SI via the
upper GI tract and allowing peristalsis to carry it into the distal SI. The mucosa
is visualized as the sonde endoscope is slowly removed. It is a very uncomfort-
able procedure and does not allow biopsy or endoscopic therapy of the lesion.
Double balloon enteroscopy is available at specialized medical centers to
visualize the SI and obtain biopsies or apply therapeutic interventions.
Interoperative enteroscopy permits visualization of the SI by the use of
enteroscope or pediatric colonoscope that is advanced through the SI during
laparotomy. Push enteroscopy and interoperative enteroscopy have been
reported to detect abnormalities in up to 70% of patients.
5. More recently, the mucosa of the SI has been visualized by Pill-cam or capsule
endoscopy. The patient swallows the capsule that takes numerous pictures
along its course down the SI. These pictures are then retrieved by downloading
them via a computer.
6. Special studies. A small number of patients continue to bleed, usually
intermittently, from a source that defies identification, even after repeated
79466_CH44.qxd 1/2/08 12:54 PM Page 305
III. The treatment of occult GI bleeding depends on the cause. See the appropriate chapter.
Vascular ectasias of the SI are the most common source of bleeding in patients
with obscure GI bleeding. Endoscopic and surgical therapy is most successful with
large and discrete vascular ectasias. The treatment of diffuse and multiple vascular
ectasias are difficult and less successful. Hormonal therapy with estrogen and proges-
terone has been tried and has been noted to be successful in some reported cases, but
not in controlled studies.
Selected Readings
Amaro R, et al. Diagnostic and therapeutic options in obscure gastrointestinal blood loss.
Curr Gastroenterol Rep. 2000;2:395.
de Leusse A, et al. Capsule endoscopy or push enteroscopy for first-line exploration of
obscure gastrointestinal bleeding? Gastroenterology. 2007 Mar;132:855–862.
Goldstein JL, et al. Small bowel mucosal injury is reduced in healthy subjects treated with
celecoxib compared with ibuprofen plus omeprazole, as assessed by video capsule
endoscopy. Aliment Parmacol Ther. 2007 May 15;25:1211–1222.
Graham DY, et al. Visible small-intestinal mucosal injury in chronic NSAID users. Clin
Gastroenterol Hepatol. 2005;3(1):55–60.
Manner H, et al. Push-and-pull enteroscopy using the double balloon technique (double-
balloon enteroscopy) for the diagnosis of Meckel’s diverticulum in adult patients with
GI bleeding of obscure origin. Am J Gastroenterol. 2006;101:1152–1154.
Pennazio M, et al. Outcome of patients with obscure gastrointestinal bleeding after capsule
endoscopy: Report of 100 consecutive cases. Gastroenterology. 2004;126:643–653.
Qureshi WA. Current and future applications of the capsule camera. Nat Rev Drug Discov.
2004;3:447–450.
Rockey DC. Occult gastrointestinal bleeding. N Engl J Med. 1999;341:39.
Tiester SL, et al. A meta-analysis of the yield of capsule endoscopy compared to other diagnostic
modalities in patients with obscure gastrointestinal bleeding. Am J Gastroenterol. 2005;
100(11):2407–2418.
Yamamoto H, et al. Clinical outcomes of double-balloon endoscopy for the diagnosis and
treatment of small-intestinal diseases. Gastroenterol Hepatol. 2004;2:1010–1016.
79466_CH45.qxd 1/2/08 12:55 PM Page 306
45 ACUTE PANCREATITIS
II. COMPLICATIONS
A. Spread of the inflammatory process. The retroperitoneal location of the pan-
creas and the absence of a well-developed pancreatic capsule allow the inflamma-
tory process to spread freely. The activated pancreatic enzymes dissect through
the tissue planes and affect any of the following organs: the common bile duct,
duodenum, splenic artery and vein, spleen, pararenal spaces, mesocolon, colon,
mesentery of the small bowel, celiac and superior mesenteric ganglia, lesser omen-
tal sac, posterior mediastinum, and diaphragm. The peritoneal surfaces may be
involved with the inflammatory process, leading to exudation and fluid accumu-
lation in the peritoneal cavity (pancreatic ascites) and the lesser omental sac.
Involvement of the diaphragmatic lymphatics may lead to sterile pleural effusion
and pneumonitis.
Local effects of pancreatic enzymes and vasoactive materials include an
intense chemical burn of tissue leading to leakage of protein-rich fluid from the
systemic circulation into peritoneal and retroperitoneal spaces. This phenomenon
may lead to hypovolemia. Systemic effects of these circulating materials include
cardiovascular instability, respiratory failure, and renal failure.
B. Hemorrhage within or around the gland may dissect along tissue planes and lead
to a bluish discoloration in the periumbilical area (Cullen’s sign) or in the cos-
tovertebral angle (Turner’s sign).
C. Pseudocysts. Necrotic tissue, blood, pancreatic juice, and fat from disrupted
cells may accumulate within or adjacent to the pancreas, forming pseudocysts.
Pseudocysts may resolve spontaneously as the inflammation subsides. If the
pseudocyst is large or if there is a communication between the pseudocyst and a
ruptured pancreatic duct with continued secretion of pancreatic juices into the
enclosed space, the pseudocyst may not resolve.
D. Pancreatic abscesses. The secondary infection of a pancreatic phlegmon or
pseudocyst by enteric flora results in pancreatic abscess.
E. Fat necrosis may occur in peritoneal, retroperitoneal, and distant locations such
as in subcutaneous or intramedullary areas. Calcium salts (soaps) of free fatty
acids liberated from fatty tissue may precipitate in these areas.
F. Polyserositis and the adult respiratory distress syndrome. The entry of the
activated pancreatic enzymes (e.g., trypsin, elastase, phospholipase A) into the cir-
culation allows these potent digestive enzymes to attack distant sites.
306
79466_CH45.qxd 1/2/08 12:55 PM Page 307
III. ETIOLOGY. There are many conditions implicated as causative factors in the patho-
genesis of acute pancreatitis (Table 45-1).
A. Alcoholism and biliary tract disease. The two most common etiologic factors
associated with pancreatitis are alcoholism and biliary tract disease (gallstones).
These two factors account for 75% to 85% of all cases. In countries in which the
incidence of alcoholism and excessive alcohol use is high, such as the United States,
Australia, and South Africa, alcohol is the etiologic factor in more than 50% of
patients. In contrast, in countries in which alcoholism is less prevalent, such as
Britain and Israel, biliary tract disease is the most common cause of acute pancre-
atitis. The mortality of gallstone-associated pancreatitis is approximately 8% dur-
ing the first attack and 1% during subsequent attacks. Chronic pancreatitis with
pancreatic insufficiency rarely, if ever, occurs, even after multiple episodes of pan-
creatitis associated with gallstones.
IV. PATHOPHYSIOLOGY. The exact mechanism of the events that trigger the intrapancre-
atic activation of zymogens to active enzymes leading to autodigestion and inflammation
remains unknown. However, there is evidence that more than one mechanism is
involved. It is thought that ischemia, anoxia, trauma, infections, and endo- and
exotoxins set the stage for activation of trypsinogen to trypsin with further activation
of other zymogens to active enzymes, including phospholipase A, elastase, and lipase.
The enzymes, along with the detergent effect of bile acids, then digest cell membranes
and elastic fibers of blood vessels, leading to vascular damage with interstitial edema,
hemorrhage, and parenchymal cell and fat necrosis of the pancreatic and peripancreatic
tissues. Vasoactive substances such as bradykinin and histamine that are released lead to
vasodilatation and increased permeability resulting in more edema and inflammation.
79466_CH45.qxd 1/2/08 12:55 PM Page 310
One popular theory to explain the initiation of the autodigestion cascade sug-
gests obstruction to the outflow of pancreatic juice (i.e., with gallstone impaction at
the ampulla resulting in intraductal hypertension) and inhibition of acinar cell zymo-
gen secretion by exocytosis. In this setting, the zymogen-containing vacuoles in the
acinar cells fuse with lysosomes, which contain proteases. This results in enzyme acti-
vation within the combined vacuoles. These vacuoles rupture into the acinar cell,
resulting in acinar cell injury and death with consequent intra- and extraparenchymal
injury. Reflux of duodenal contents into the pancreatic duct, presence of a common
channel (communicating pancreatic and common bile duct) with or without impacted
gallstone, and occlusion of pancreatic blood vessels are other popular theories. Other
forms of ischemia may also lead to intracellular zymogen activation and result in
parenchymal damage.
On Admission
Age 55 y
WBC 16,000/L
Blood glucose 200 mg/dL (no diabetic history)
Serum LDH 350 IU/L (normal up to 225)
SGOT 250 Sigma Frankel units/L (normal up to 40)
Within 48 h
Age 55 y
WBC 15,000/L
Blood glucose 180 mg/dL (no diabetic history)
Serum urea 16 mmol/L (no response to IV fluids)
BUN rise 5 mg/dL
PaO2 60 mmHg
Serum calcium 8.0 mg/dL
Hematocrit fall 10%
Base deficit 4 mEq/L
Fluid sequestration 6 L
Serum albumin 3.2 gm/dL
Serum LDH 600 units/L (normal up to 255 units/L)
AST or ALT 200 units/L (normal up to 40 units/L)
WBC, white blood count; LDH, lactic dehydrogenase; SGOT, serum glutamic-oxaloacetic transami-
nase; BUN, blood urea nitrogen; AST, aspartate aminotransferase; ALT, alanine aminotransferase;
IV, intravenous; Pao2, partial pressure of arterial oxygen.
From Imrie CW, et al. A single centre double-blind trial of trasylol therapy in primary acute pancreatitis.
Br J Surg. 1978;65:337; and Ranson JH, et al. Prognostic signs and the role of operative management
in acute pancreatitis. Surg Gynecol Obstet. 1974;139:69. Reprinted with permission.
79466_CH45.qxd 1/2/08 12:55 PM Page 311
BP, blood pressure; ECG, electrocardiogram; PO2, partial pressure of oxygen; BUN, blood urea
nitrogen; CNS, central nervous system.
VI. DIAGNOSIS
A. Clinical presentation
1. History. Abdominal pain is the most common complaint in acute pancreatitis.
It is usually located in the epigastrium, left upper quadrant, or periumbilical
area, and often radiates to the back, chest, flanks, and lower abdomen.
The pain is steady, dull, and boring in character. It is usually more intense
when the patient is supine and may lessen in the sitting position with the trunk
flexed forward and the knees drawn up. Patients also complain of nausea, vom-
iting, and abdominal distention secondary to ileus.
2. Physical examination. Patients with acute pancreatitis initially may present
with fever, tachycardia, and hypotension. Shock is common in severe instances
due to hypovolemia caused by third-space fluid sequestration (in retroperi-
toneal and other spaces) with increased vascular permeability and vasodilata-
tion and other systemic effects of proteolytic and lipolytic enzymes released
into the circulation. Jaundice may occur, due to obstructive cholelithiasis or,
more commonly, the compression of the intrapancreatic portion of the common
bile duct with edema of the head of the pancreas. Abdominal tenderness and
rigidity may be present. Bowel sounds are diminished or absent. The presence
of a bluish discoloration around the umbilicus (Cullen’s sign) and at the flanks
(Turner’s sign) suggests hemoperitoneum and results from hemorrhagic necro-
tizing pancreatitis.
Other findings such as pleural effusion (especially on the left side), pneu-
monitis and other pulmonary findings, such as ARDS, and subcutaneous fat
necrosis resembling erythema nodosum may be present. Tetany due to hypocal-
cemia is a rare finding.
B. Diagnostic studies
1. Laboratory studies
a. Serum amylase. Even though there is no definite correlation between the
severity of pancreatitis and the degree of serum amylase elevation, serum amy-
lase elevation is commonly equated to the presence of pancreatitis. However,
hyperamylasemia may be present in many other conditions, as summarized in
Table 45-5. Amylase is found in many organs, including salivary glands, liver,
small intestine, kidney, and fallopian tubes, and in various tumors such as car-
cinoma of the esophagus, lung, and ovary. In 75% of the patients with acute
pancreatitis, the serum amylase is elevated. Hyperamylasemia is noted within
the first 24 hours and persists for 3 to 5 days. Amylase levels normalize unless
there is extensive pancreatic necrosis, ductal obstruction, or pseudocyst
79466_CH45.qxd 1/2/08 12:55 PM Page 312
When ileus is present, the US waves are scattered by the air in the
intestinal lumen, and a clear picture of the pancreas cannot be obtained.
However, US is the test of choice for documenting cholelithiasis and biliary
ductal dilatation. CT scan of the abdomen gives a clearer picture of the pan-
creas and the surrounding areas than US does and should be obtained when
there is a likelihood of severe pancreatitis (e.g., multiple positive Ranson’s
signs during the first 48 hours). If the CT scan is normal or shows mild pan-
creatic edema, the likelihood of a severe complication is remote even if many
of Ranson’s signs are positive. When a markedly swollen pancreas is seen
with or without fluid collections, the presence of extensive necrotic areas
within the pancreas (necrotizing pancreatitis) predisposes the patient to
secondary infection and higher risk of mortality. The distinction between
acute interstitial and necrotizing pancreatitis can be made if a CT scan is
obtained following the administration of IV contrast medium (dynamic CT
scan). In acute interstitial pancreatitis, the pancreas is well perfused and
uniformly enhanced by the intravascular contrast agent. If necrosis is pre-
sent, the areas of devitalized tissue are not perfused and are not enhanced.
The presence of air bubbles in the pancreatic and peripancreatic region
is strong evidence of pancreatic infection. When fever, elevated white blood
count, and clinical toxicity are associated with CT scan evidence of fluid col-
lections or necrotizing pancreatitis, CT-guided percutaneous aspiration,
staining, and culture of the obtained fluid may help to distinguish sterile
pancreatitis from pancreatic infection. When infection is suspected or
proved, surgical debridement may be lifesaving.
ERCP is generally contraindicated in patients with acute pancreatitis
except when an impacted common bile duct stone may be the cause of the
pancreatitis. Endoscopic sphincterotomy in these very ill patients may be
immediately therapeutic. ERCP is also useful in establishing the diagnosis
of pancreas divisum; annular pancreas; pancreatic cancer; periampullary,
ampullary, and pancreatic ductal abnormalities; and the possible communi-
cation of the pancreatic duct with pseudocysts. These studies are usually
performed after the patient has been clinically stabilized.
C. Differential diagnosis. Table 45-6 summarizes the differential diagnosis of
acute pancreatitis. It may be difficult to distinguish between acute cholecystitis,
ascending cholangitis, and pancreatitis, because they may present with similarly
elevated serum amylase and abnormal liver tests. US and dimethylphenylcar-
bamylmethyliminodiacetic acid (HIDA) scans may be helpful in differentiating
between these diseases. Gut ischemia, infarction, viscus perforation, aortic dis-
section, mechanical intestinal obstruction, myocardial infarction, and acute
appendicitis need to be promptly diagnosed and surgically treated.
D. Complications. See Table 45-7.
VII. TREATMENT. In most patients (85%–90%) with acute pancreatitis, the disease
is self-limited and resolves spontaneously. These patients are medically treated with
Pancreatic Cardiovascular
Phlegmon Cardiogenic shock
Abscess Increased cardiac index
Pseudocyst Decreased systemic vascular resistance
Nonpancreatic Pulmonary
Inflammatory involvement of contiguous Hypoxemia
organs Pleural effusions
Obstructive jaundice Pulmonary infiltrates, atelectasis
Pancreatic ascites Adult respiratory distress syndrome
Intraperitoneal hemorrhage Renal
Thrombosis of splenic vein Decreased glomerular filtration rate and
Bowel infarction renal plasma flow
Gastrointestinal bleeding Acute tubular necrosis
Acute renal failure
Systemic
Hematologic
Metabolic
Disseminated intravascular coagulation
Hypocalcemia
Thromboses
Hyperlipidemia
Increased factor VII or fibrinogen
Hyperglycemia
Skin and musculoskeletal
Diabetes ketoacidosis
Erythema nodosum–like lesions
Nonketotic diabetic coma
Angiopathic retinopathy
Pancreatic encephalopathy
Polyarthritis
VIII. Pancreatic pseudocysts are collections of necrotic tissue, fluid, and blood that develop
in or near the pancreas over a period of 1 to 4 weeks after the onset of acute pancreati-
tis. They do not have a true capsule with an epithelial lining. A connection may be pre-
sent to a disrupted pancreatic duct. Most pseudocysts (90%) are solitary lesions located
in the body or tail of the pancreas. The main symptom is abdominal pain.
A. Pathogenesis. Most pseudocysts form during a severe episode of acute pancreati-
tis. Persistence of an elevated serum amylase level for more than a week after the
onset of pancreatitis may signal the formation of a pseudocyst. Other causes of
pancreatic pseudocyst include abdominal blunt trauma with disruption of a pan-
creatic duct, inadvertent surgical ductal trauma, and chronic pancreatitis.
Neoplastic cysts, such as cystadenoma or cystadenoma-carcinoma, account for
10% of cystic pancreatic masses.
B. Diagnostic studies. Abdominal US and CT scans are the best imaging techniques
used in the diagnosis of pancreatic pseudocysts. Serial scans help in following the
size and course of the cysts. ERCP may be required to assess the possible connec-
tion of the pseudocyst with the pancreatic duct.
C. Complications and treatment. Most pseudocysts tend to resolve spontaneously.
Symptomatic pseudocysts require decompression. This can be achieved by surgery
or by percutaneous catheter drainage or by endoscopic techniques. Endoscopic
and radiologic methods should be reserved for those institutions with extensive
experience in these techniques. The serious complications of pancreatic pseudo-
cysts are infection, perforation, and hemorrhage.
79466_CH45.qxd 1/2/08 12:55 PM Page 316
1. Infection. Patients with a pancreatic pseudocyst who have pain, fever, and
leukocytosis need to be evaluated for infection. Percutaneous aspiration under
US or CT guidance with Gram’s stain and culture of the aspirate helps in
confirming the diagnosis. Infected pseudocyst or abscess should be drained
externally.
2. Perforation. Rupture or leak of a pseudocyst into the peritoneal cavity or
retroperitoneum may result in shock and requires emergency surgery. The mor-
tality from this complication is very high.
3. Hemorrhage. A pseudocyst may erode into a viscus (e.g., stomach, small or
large bowel) or a blood vessel with subsequent hemorrhage. Angiography is
often necessary prior to surgery for proper diagnosis of this complication.
Selected Readings
Amar S, et al. Sorofenibinducod pancreatitis. Mayo Clin Proc. 2007;82(4):516.
Badalov N, et al. Drug induced acute pancreatitis: An evidence-based review. Clinical
Gastroenterol Heptatol. 2007;5(6):648–661.
Baron TH, et al. Acute necrotizing pancreatitis. N Engl J Med. 1999;340:1412.
Beger HG, et al. Surgical management of necrotizing pancreatitis. Surg Clin North Am.
1999;79:783.
Brown A, et al. Hemoconcentration is an early marker for organ failure and necrotizing
pancreatitis. Pancreas. 2000;20:367.
DeWaele JJ, et al. Emergence of antibiotic resistance in infected pancreatic necrosis. Arch
Surg. 2004;139(12):1371–1375.
Frakes JT. Biliary pancreatitis: A review emphasizing appropriate endoscopic intervention.
J Clin Gastroenterol. 1999;28:97.
Freeman ML. Pancreatic stents for prevention of post ERCP pancreatitis. Clin Gastroenterol
Hepatol. 2007;5(11):1354–1365.
Freeman ML, et al. Prevention of post-ERCP pancreatitis: A comprehensive review.
Gastrointest Endosc. 2004;59:845–864.
German Antibiotics in Severe Acute Pancreatitis Study Group. Prophylactic antibiotic
treatment inpatients with predicted severe acte pancreatitis. A placebo-controlled,
double-blind trial. Gastroenterology. 2004;126(4):997–1000.
Jacobsen BC, et al. A prospective, randomized trial of clear liquids versus low fat solid diet
as the initial meal in mild pancreatitis. Clin Gastroenterol Hepatol. 2007;5(8):946–951.
McClave SA, et al. Nutrition support in acute pancreatitis: A systematic review of the
literature. JPEN J Parenter Enteral Nutr. 2006;30:143–156.
Meier R, et al. ESPEN guidelines on enteral nutrition: pancreas. Clin Nutr. 2006;25:275–284.
Shankar S, et al. Imaging and percutaneous management of acute complicated pancreatitis.
Cardiovasc Intervent Radiol. 2004;27:567–580.
79466_CH46.qxd 1/2/08 6:14 PM Page 317
CHRONIC PANCREATITIS 46
C hronic pancreatitis results from progressive destruction and fibrosis of the pancreas
with ongoing inflammatory lesions. The exocrine pancreatic tissue and function are lost
in the earlier stages, followed by the loss of endocrine parenchyma and function. The dis-
ease frequently is complicated in the early stages of its evolution by attacks of acute
pancreatitis, which are responsible for recurrent pain. After several years of ongoing
inflammation and fibrosis, pancreatic insufficiency develops, with resulting malabsorption,
steatorrhea, and diabetes mellitus. Acute attacks decrease and pain usually disappears.
I. CLASSIFICATION. Chronic pancreatitis may be classified into two forms that present
with specific lesions and have different causes.
A. Obstructive chronic pancreatitis is caused by the occlusion of pancreatic ducts,
which precedes the onset of pancreatitis. The occlusion may be the result of
tumors, scars of parenchymal inflammation, necrotic pseudocysts, or congenital
anomalies (e.g., annular pancreas, pancreas divisum). The lesions are found in
the part of the pancreas encompassing the occluded ducts. The ductal epithelium is
relatively preserved, and intraductal protein plugs and stones are not present.
Infiltrative and autoimmune diseases such as hemochromatosis and Sjögren’s
syndrome may also involve the pancreas, resulting in pancreatic insufficiency.
B. Chronic calcifying pancreatitis (CCP) is the most frequent cause (95% of all
instances) of chronic pancreatitis. It is significantly associated with chronic alcohol
consumption and is exacerbated by cigarette smoking and by diets high in protein
and high or low in fat.
Less frequently, CCP occurs with hyperparathyroidism and hypercalcemia, and
in some tropical countries (South India, Zaire, Nigeria, Brazil) it occurs in nonalco-
holic young people (average age, 12–20 years) of both sexes living in areas where
protein- and fat-poor diets are consumed. There is also a hereditary autosomal-
dominant form of CCP with variable penetrance. Pancreatic insufficiency from
cystic fibrosis may resemble CCP in morphology and presentation.
1. Morphology. CCP is characterized by the lobular, patchy distribution of lesions
of different intensity in neighboring lobules. Protein plugs are always found in
the ductal and acinar lumina, and in the later stages these form calcifications,
or calculi (pancreatic calcification). Atrophy of the epithelium and stenosis
of the ducts are common. Recurrent attacks of acute pancreatitis, retention
cysts, pseudocysts, and perineural inflammation are frequently associated.
The first visible lesions are protein precipitates or plugs in the lumina of
ducts and acini, which later calcify forming pancreatic stones. The ductal
epithelium in contact with the protein plugs or stones loses its basement mem-
brane, and the duct cells atrophy and disappear with the growth of periductal
connective tissue and fibrosis leading to fibrotic strictures. Distal to the strictures,
the exocrine tissue atrophies and disappears due to plugs, stones, and fibrosis.
When a partially obstructed duct is distended by pancreatic juice under pressure
of secretion, it may form an intrapancreatic cyst. These retention cysts may
grow and extend into peripancreatic tissue, forming retention pseudocysts.
Thus all lesions of chronic calcifying pancreatitis are thought to be secondary to
the formation of protein plugs and stones in the pancreatic ducts and ductules,
resulting in ductal obstruction, parenchymal inflammation, atrophy, and fibrosis.
317
79466_CH46.qxd 1/2/08 6:14 PM Page 318
II. DIAGNOSIS. The insidious nature of chronic pancreatitis delays the early diagnosis of
this disorder in many patients. Patients usually come to medical attention after con-
siderable damage has occurred to the gland. In most cases, it is difficult to differenti-
ate acute relapsing pancreatitis, in which the permanent pancreatic damage is mild to
moderate, from chronic relapsing pancreatitis.
A. Clinical presentations. Chronic pancreatitis is an insidious process of parenchy-
mal damage with necrosis and fibrosis of the gland. Approximately one half of the
patients present with episodes of acute pancreatitis superimposed onto the
damaged organ. One third of the patients may present with only abdominal pain.
Other patients may have jaundice, weight loss, malabsorption, steatorrhea,
79466_CH46.qxd 1/2/08 6:14 PM Page 319
B. Diagnostic studies
1. Serum chemistry profile. There are no specific findings in the serum chemistry
profile in patients with chronic pancreatitis. Even during acute attacks, the
serum amylase or lipase level may not be elevated. The serum chemistry profile
may reflect concomitant liver disease. With obstruction of the common bile
duct, a cholestatic liver chemistry profile may emerge and should be confirmed
by imaging techniques.
2. Stool fat. Steatorrhea is best confirmed by 72-hour stool collection for quanti-
tative determination of fat. Most patients consuming 100 g of fat per day
excrete more than 10 g of fat per day. In advanced pancreatic insufficiency, stool
fat may reach 30 to 40 g per day.
3. Radiologic studies
a. Pancreatic calcifications are present on plain abdominal radiographs in
one third of patients with chronic pancreatitis.
b. Ultrasound and, especially, computed tomography (CT) scan provide for
more sensitive detection of pancreatic calcifications. The presence of pseudo-
cysts, ductal dilatation, and tumors can also be delineated.
c. Angiography. When a tumor is suspected, angiography may be helpful in
identifying neovascularity of the tumor, deviation of normal vascularity due
to tumor, or the lack of vascularity of the cystic lesions. When splenic vein
thrombosis has occurred, angiography demonstrates the site of occlusion and
the presence of gastric or esophageal varices.
d. Endoscopic retrograde cholangiopancreatography. Pancreatic ductal
anatomy is best delineated with endoscopic pancreatography: Dilatation,
cystic changes, strictures, and calculi are demonstrated. Most patients with
advanced chronic pancreatitis have a dilated common pancreatic duct with
intermittent sites of narrowing, creating the “chain of lakes” appearance on
the pancreatograms. The damage is also noted in the secondary ducts, with
dilatation and blunting leading to loss of the fine “acinarization.”
Endoscopic retrograde cholangiopancreatography (ERCP) may
also be helpful in the differentiation of pancreatic cancer from chronic pan-
creatitis. In pancreatic cancer, only that part of the duct involved with the
tumor is abnormal, in contrast to generalized abnormal changes seen with
chronic pancreatitis.
e. Endoscopic ultrasonography (EUS) may be used in instances of CCP in
which a tumor is suspected. The differentiation of sclerotic pancreatic cancer
from atrophic and fibrotic pancreatic parenchyma may be difficult with this
technique.
4. Tests of pancreatic exocrine function. Chronic pancreatitis or pancreatic
insufficiency can be determined with certainty only by demonstrating dimin-
ished exocrine function during adequate stimulation of the pancreas. These tests
may be necessary in situations in which no calcification is seen on radiographic
studies and the ducts appear normal on ERCP.
a. Secretin stimulation test. In this test, the duodenum is intubated under
fluoroscopy and the duodenal contents are aspirated, collected, and analyzed
after the patient receives intravenous secretin (1 U/kg of body weight) to
stimulate pancreatic water and bicarbonate secretion. The total output of
bicarbonate correlates well with the extent of pancreatic damage.
The addition of cholecystokinin to secretin, to stimulate exocrine
enzyme secretion (e.g., amylase or lipase), does not seem to increase the diag-
nostic accuracy. The secretin stimulation test is difficult to perform and
requires technical expertise. It is reliably performed in specialized medical
centers. The overall sensitivity is 74% to 90%.
b. The Lundh test meal provides for endogenous stimulation of pancreatic
exocrine secretion. After the patient is given a standard test meal, the
proximal jejunal fluid is aspirated for 2 hours and analyzed for trypsin.
This test is not as sensitive as the secretin stimulation test (sensitivity,
60%–90%).
79466_CH46.qxd 1/2/08 6:14 PM Page 321
IV. The treatment of chronic pancreatitis is mainly supportive and directed at the
complications.
A. Management of pain
1. In alcohol-related disease, alcohol and tobacco consumption should be
stopped. Patients should be advised to follow a diet with moderate fat and pro-
tein and high carbohydrate content.
2. In some instances pain relief may be obtained by small feedings and anal-
gesics. Many patients require narcotics.
3. Feedback control. Oral treatment with large doses of pancreatic enzymes has
been shown to decrease the abdominal pain experienced by patients with
chronic pancreatitis by inhibiting pancreatic exocrine secretion and allowing
the pancreas to rest. The presence of the proteases trypsin and chymotrypsin
within the lumen of the proximal duodenum exerts a feedback control on pan-
creatic exocrine secretion. The patients in whom pain responds to pancreatic
enzymes most readily are those with mild-to-moderate exocrine impairment.
The use of pancreatic enzymes has also been shown to heal pancreatic fistulas
and to decrease the frequency of attacks of acute, recurrent pancreatitis. The
doses used are similar to those used for the treatment of malabsorption (see
section IV.B).
4. Percutaneous injection of alcohol to destroy the celiac plexus has been
reported to relieve pain in some patients for as long as 6 months. Long-term
effects of this treatment have not been determined.
5. Surgery for relief of pancreatic pain is reserved for patients with intractable
and disabling pain unresponsive to any other mode of therapy. The surgical
procedures used depend on pancreatic and ductal anatomic abnormalities deter-
mined by preoperative CT scan and ERCP findings. In most patients, pain relief
is achieved for the short term. The long-term effectiveness of these procedures
in relieving pain is debated.
Drainage procedures are used when there is generalized or localized
pancreatic ductal dilatation, and partial resection of the gland is used when
there is no ductal dilatation or the abnormality is confined to a segment of the
pancreas.
a. Longitudinal pancreaticojejunostomy (the Puestow procedure). In this
procedure, the dilated pancreatic duct is filleted open longitudinally over the
79466_CH46.qxd 1/2/08 6:14 PM Page 322
Pancrelipase
Cotazym C 8,000 30,000 30,000
Cotazym-S ECMS 5,000 20,000 20,000
Festal II ECT 6,000 20,000 30,000
Ku-Zyme HP C 8,000 30,000 30,000
Pancrease ECMS 4,000 25,000 25,000
Pancrease MT4 ECMT 4,000 12,000 12,000
Pancrease MT10 ECMT 10,000 30,000 30,000
Pancrease MT16 ECMT 16,000 48,000 48,000
Pancrease MT25 ECMT 25,000 75,000 75,000
Viokase T 8,000 30,000 30,000
Zymase ECS 12,000 24,000 24,000
Pancreatin
Creon ECMS 8,000 13,000 30,000
Creon 25 ECMS 25,000 62,500 74,700
Pancreatin 8 T 22,500 180,000 180,000
Entozyme T 600 7,500 7,500
C, capsule; T, tablet; ECT, enteric-coated tablet; ECS, enteric-coated sphere; ECMT, enteric-coated
microtablet; ECMS, enteric-coated microsphere.
Selected Readings
Adler DG, et al. The role of endoscopy in patients with chronic pancreatitis. Gastrointest
Endosc. 2006;63:933–937.
Cahen DL, et al. Endoscopic versus surgical drainage of the pancreatic duct in chronic
pancreatitis. N Engl J Med. 2007;356:676–684.
Dragonuv P, et al. Idiopathic pancreatitis. Gastroenterology. 2005;128(3):756–763.
Dragonuv P, et al. A 54-year-old man with abdominal pain attributed to chronic
pancreatitis. Clin Gastroenterol Hepatol. 2007;5(3):302–306.
Eleftheriadis N, et al. Long term outcome after pancreatic stenting in severe chronic
pancreatitis. Endoscopy. 2005;37:223–230.
Elta GH. Is there a role for the endoscopic treatment of pain from chronic pancreatitis?
N Engl J Med. 2007;356:727–729.
Etemad B, et al. Chronic pancreatitis: Diagnosis classification and new genetic developments.
Gastroenterology. 2001;120:682–707.
Frey CF, et al. Comparison of local resection of the head of the pancreas combined with
longitudinal pancreatic jejunostomy (Frey procedure) and duodenum-preserving
resection of the pancreatic head (Berger procedure). World J Surg. 2003;27:1211–1230.
Gabrielli A, et al. Efficacy of main pancreatic duct endoscopic drainage in patients with
chronic pancreatitis, continuos pain, and dilated duct. Gastrointest Endosc. 2005;61:
576–581.
Lankish MR, et al. The effect of small amounts of alcohol on the clinical course of chronic
pancreatitis. Mayo Clin Proc. 2001;76:242–251.
Lowenfel AB, et al. Risk factors for cancer in hereditary pancreatitis study group. Med
Clin North Am. 2000;84:565–593.
Morinville V, et al. Recurrent acute and chronic pancreatitis: Complex disorders with a
genetic basis. Gastroenterol Hepatol. 2005;1(3):195–205.
Norton ID, et al. The role of transabdominal ultrasonography, helical computed
tomography and magnetic resonance cholangiopancreatography in diagnosis and
management of pancreatic disease. Curr Gastroenterol Rep. 2000;2:120.
79466_CH46.qxd 1/2/08 6:14 PM Page 325
Palanivelu C, et al. Laparoscopic lateral pancreatic jejunostomy: A new remedy for an old
ailment. Surg Endosc. 2006;458–461.
Pickartz T, et al. Autoimmune pancreatitis. Nat Clin Pract Gastroenterol Heptatol.
2007;4(6):314–323.
Pitchumoni CS. Pathogenesis of alcohol-induced chronic pancreatitis: Facts, perceptions
and misperceptions. Surg Clin North Am. 2001;81:379–389.
Ralmondo M. What is the role of EUS in screening for chronic pancreatitis? Nat Clin Pract
Gastroenterol Hepatol. 2007;4(10)530–532.
Toskes PP. Alcohol consumption and chronic pancreatitis. Mayo Clin Proc. 2001;76:241.
Whitcom DC, et al. Mechanisms of disease: Advances in understanding the mechanisms
leading to chronic pancreatitis. Nat Clin Pract Gastroenterol Hepatol. 2004;1(1):46–52.
79466_CH47.qxd 1/2/08 1:29 PM Page 326
E xocrine pancreatic cancer accounts for 95% of all the cancers that arise in the
pancreas. Between 75% and 95% of exocrine pancreatic cancers arise from the ductular
epithelium. Other cancers, such as acinar cell carcinoma, giant cell carcinoma, adenosqua-
mous carcinoma, mucinous carcinoma, cystadenocarcinoma, papillary cystic tumor, muci-
nous ductal ectasia, intraductal papillary neoplasm, fibrosarcoma, leiomyosarcoma, and
lymphoma are rare and account for less than 10% of the exocrine tumors.
Islet cell tumors of the pancreas make up approximately 5% of the carcinomas of
the pancreas. These tumors often manifest themselves by the hormones they secrete.
Tumors secreting gastrin, insulin, glucagon, vasoactive intestinal polypeptide (VIP), pan-
creatic polypeptide (PIP), neurotensin, and somatostatin may present as single tumors or as
part of multiple neoplasm syndromes.
326
79466_CH47.qxd 1/2/08 1:29 PM Page 327
nervous plexuses, especially in the celiac and mesenteric areas. The most common
sites of extralymphatic metastasis are the liver, peritoneum, lungs, intestines,
adrenals, kidneys, bones, and the diaphragm. Pancreatic tumors other than ductal
adenocarcinomas (e.g., cystadenocarcinomas, islet cell tumors) often have a more
indolent course. Tumors such as carcinomas of the breast, lung, thyroid, kidney,
ovary, uterus, and prostate, and melanomas may metastasize to the pancreas and
present as mass lesions in the organ.
C. Diagnosis
1. Clinical presentation
a. History. The early symptoms of pancreatic cancer are vague and nonspe-
cific. The most common symptoms are abdominal pain, back pain, weight
loss, anorexia, nausea, jaundice, diarrhea, malabsorption, depression, and
abdominal mass.
i. An insidious weight loss with anorexia and nausea, accompanied by
upper abdominal pain radiating to the back, is the most common pre-
sentation. Greater than 90% of the patients initially have jaundice.
Common bile duct obstruction by a tumor in the head of the pancreas
may result in jaundice while the mass is still small. Tumors located in the
body and tail of the organ may result in jaundice in later stages either
by extension or due to metastasis to the porta hepatis or the liver
parenchyma.
ii. Up to 70% of the patients may present with diabetes mellitus or glucose
intolerance. The decreased or delayed insulin secretion is thought to
arise from loss of B cells due to the desmoplastic reaction of the tumor.
iii. Migratory thrombophlebitis (Trousseau’s sign) may be a mode of
presentation. However, this entity is not specific for pancreatic cancer. It
may occur with other malignancies such as carcinomas of the stomach,
colon, ovary, and lung.
iv. A minority of the patients may also present with a picture of acute
pancreatitis, cholangitis, gastrointestinal bleeding, polyarthritis,
and skin nodules due to fat necrosis.
b. The physical examination in most instances is not helpful. The major find-
ings in a subpopulation of patients are jaundice, palpable gallbladder, epi-
gastric mass, and nodular liver if metastases are present.
c. Warning signs for early diagnosis. The initial symptoms of pancreatic
cancer are usually ignored by the patient (patient delay) and the physician
(physician delay). The mean duration of symptoms before diagnosis in most
series is 3 to 4 months. Most of the tumors are unresectable and, therefore,
the disease is fatal. The following warning signs may facilitate an early diag-
nosis of this malignancy:
i. Recent upper abdominal or back pain consistent with retroperitoneal
lesion.
ii. Recent upper abdominal pain or discomfort with negative gastrointestinal
investigations.
iii. Jaundice with or without pruritus.
iv. Weight loss greater than 5% of normal body weight.
v. Unexplained acute pancreatitis.
vi. Unexplained onset of diabetes mellitus.
d. Differential diagnosis. A variety of malignant and benign disorders of other
organs may present with features similar to pancreatic cancer. Also, it is
important to remember that pancreatic cancer may coexist in a patient with
a common benign disorder such as gallstones or peptic or diverticular dis-
ease, and normal contrast studies of the gastrointestinal tract, serum
chemistries, and hemogram do not rule out the presence of pancreatic cancer,
especially if the tumor is small.
2. Diagnostic studies
a. Laboratory tests. There are no specific laboratory tests for the early detec-
tion of pancreatic cancer. If there is involvement of the liver or the biliary
79466_CH47.qxd 1/2/08 1:29 PM Page 328
tract, this will be reflected in the serum chemistries. The serum amylase and
lipase in most instances are normal. A subgroup of patients has elevated
blood glucose levels.
b. Tumor markers. Various serologic tumor markers including tumor associated
antigens, enzymes, and hormones have been investigated for early detection of
pancreatic cancer. These are carcinoembryonic antigen (CEA), CA 19-9,
alpha-fetoprotein, pancreatic oncofetal antigens, pancreatic ribonuclease, and
galactosyl transferase isoenzyme II. The sensitivity and specificity of these
assays have not been adequate for early diagnosis of this disease.
The most extensively studied serum marker is CA 19-9, which is used
widely. It is not specific for pancreatic cancer, however, because it also can
be elevated in other gastrointestinal tumors such as those in the bile ducts
and colon. Because levels of CA 19-9 are frequently normal in the early
stages of pancreatic cancer, the test is not reliable for use in screening. The
presence of high levels may help differentiate between benign diseases of the
pancreas and pancreatic cancer. When the pancreatic cancer is completely
resected, the CA 19-9 levels fall, suggesting that it is a useful marker for
follow-up surveillance.
The ratio of testosterone to dihydrotestosterone is below 5 (normal is 10)
in more than 70% of men with pancreatic cancer, presumably because of
increased conversion of testosterone by the pancreatic tumor. This ratio may
be more sensitive than CA 19-9 in detecting smaller pancreatic cancers and
more specific than the other markers.
c. Ultrasound and computed tomography. Ultrasonography usually is the
first examination for suspected pancreatic cancer. Computed tomography
(CT) with intravenous (IV) contrast is used when satisfactory imaging is not
obtained with ultrasound (US). These two techniques are by far the most
sensitive and specific for pancreatic disease. They both demonstrate
enlargement of the gland, alteration in contour or consistency of the gland,
the presence of masses, and biliary or pancreatic duct dilatation. CT scans
may also delineate peripancreatic nodal enlargement as well as invasion of
other organs and vessels. Metastasis to the liver and porta hepatis may be
detected.
US and CT are complementary in imaging the pancreas. The lesions in
the head of the pancreas are seen well by US, whereas those in the body and
tail are detected better by CT scan. However, small lesions, especially in the
body or tail, may be missed by both techniques. Helical thin section CT scan
with IV contrast and CT angiography (CTA) increase the diagnostic yield.
d. Magnetic resonance imaging, contrast-enhanced magnetic resonance
imaging (MRI) using IV gadolinium–DPTA is useful for detecting small
pancreatic tumors. Ductal size is evaluated by magnetic resonance
cholangiopancreatography (MRCP). MR arteriography (MRA) has obvi-
ated the need for angiography and improves the examination of the pancreas
for tumor. Fat-saturation MRI is especially valuable in looking for suspected
tumors in a pancreas that is not enlarged.
e. Endoscopic retrograde cholangiopancreatography (ERCP). The diagno-
sis of pancreatic cancer by ERCP depends on radiographic demonstration of
pancreatic duct stenosis or obstruction caused by the tumor. An accompa-
nying cholangiogram may further delineate abnormalities along the course
of the common bile duct. It can also visualize and differentiate ampullary
and duodenal carcinomas. In experienced hands, it has greater than 90%
sensitivity and specificity in providing a definitive diagnosis of pancreatic
cancer.
In addition, biopsies of periampullary tissue and cytologic examina-
tion of aspirated pancreatic juice may increase the diagnostic yield further.
ERCP is usually performed if an abnormality is noted on the US or CT scan
or if an abnormality is suspected but cannot be demonstrated by these
methods.
79466_CH47.qxd 1/2/08 1:29 PM Page 329
3. Radiation therapy alone has not been unequivocally successful in the treatment
of nonresectable localized cancer of the pancreas. However, the combination of
radiation therapy (4,000–6,000 rad) with chemotherapy (using 5-FU) has been
found to result in enhanced radiation therapeutic efficiency. In one study, the
survival was increased from 5.5 months to 11 months. Newer radiation tech-
niques using iodine 125 directly implanted into the tumor tissue and neutron
beam radiation show promise in the treatment of unresectable cancer of the
pancreas.
4. Intraoperative radiation therapy offers the possibility of delivering higher
doses of radiation to the pancreatic cancer with less risk of injury to the adjacent
organs. In most centers, a single dose of 20 Gy is delivered to the surgically
exposed cancer by an electron beam through a field-limiting cone. Median sur-
vivals of 13 to 16 months have been reported with excellent local control. Five
percent of patients have lived 3 to 8 years. Relief of pain has been achieved in
50% to 90% of these patients. Unfortunately, in 30% of these patients unavoid-
able irradiation of the duodenum may result in bleeding, obstruction, and per-
foration. In most patients, therefore, protective gastrojejunostomy is performed
at the same time as the intraoperative radiation.
III. Cystic tumors of the pancreas are a heterogeneous group of pancreatic neoplasms
that include mucinous cystic neoplasms (50%), serous cystoadenomas (30%), intra-
ductal papillary mucinous neoplasms (12%), papillary cystic tumors (3%), and other
tumors (5%).
A. Mucinous cystadenomas are cystic lesions more commonly seen in the body
and tail of the pancreas and occur more commonly in women. They are often
asymptomatic and may be found during an abdominal US or CT examination. If
there is a mass in the wall of the cystic lesion, CT or EUS guided fine-needle aspi-
ration may be necessary for a diagnosis of cystadenocarcinoma. The aspirated
fluid is usually viscous. Elevated levels of carcinoembryonic antigen (CEA) in the
cyst fluid indicate malignancy. Surgical resection is indicated for malignant
lesions.
B. Serous cystadenomas may occur in any part of the pancreas. They are well-
circumscribed, solitary, cystic lesions. On EUS or CT guided fine-needle aspira-
tion the fluid has low viscosity and low levels of amylase and CEA. Elevated CEA
indicates malignancy. In some cases, CT shows a cystic lesion, especially in the
head and neck of the pancreas, that has a honeycomb appearance with an area of
central fibrosis and calcification (stellate, or star shaped). Small serous cystade-
nomas may be monitored; larger ones and those which are malignant should be
resected.
79466_CH47.qxd 1/2/08 1:29 PM Page 332
ulcer disease (PUD), GERD, diarrhea, and abdominal pain. A positive secretin
test (see Chapter 24) may be necessary to confirm the diagnosis. The majority
of gastrinomas are found in the “gastrinoma triangle” which is an area
bounded by the cystic and the common bile ducts, the duodenum, the sur-
rounding lymph nodes, or in the pancreas. EUS and somatostatin scintigraphy
may help in the localization of the tumor(s) preoperatively; however, intraoper-
ative palpation and duodenectomy is usually necessary. Proton pump inhibitors
are highly effective in controlling the symptoms of associated gastric acid
hypersecretion.
4. VIPomas present with the Verner–Morrison syndrome or pancreatic cholera
characterized by profound diarrhea, hypokalemia, and achlorhydria. The
syndrome is due to the ectopic secretion of vasoactive intestinal peptide (VIP).
VIP causes enterocyte intercellular elevation of cyclic AMP, resulting in intesti-
nal smooth muscle relaxation, inhibition of electrolyte absorption, and pro-
found secretory diarrhea.
VIPomas often present in fifth decade of life. The diagnosis is based on the
presence of elevated serum VIP levels. Most tumors are visualized with CT scan
or MRI, EUS, or somatostatin scintigraphy. Surgical resection is used for local-
ized disease to reduce tumor burden in patients with metastases. Somatostatin
analogs are effective in suppressing VIP secretion and controlling the secretory
diarrhea.
5. Somatostatinomas and PPOmas may be associated with diabetes mellitus,
hypochloridemia, and diarrhea. PPOmas are associated with high serum levels
of PP. Although high levels of PP do not cause symptoms, the tumors cause
symptoms from tumor bulk. Most patients have metastatic disease at the time of
diagnosis; however, surgical resection is curative in patients in the early stage of
the disease.
Selected Readings
Alberts SR, et al. Treatment options for hepatobiliary and pancreatic cancer. Symposium
of solid tumors. Mayo Clin Proc. 2007;82(5)628–637.
Arvanitakis M, et al. Predictive factors for pancreatic cancer in patients with chronic
pancreatitis in association with K-RSA gene mutation. Endoscopy. 2004;36(6):
535–542.
Brugge WR, et al. Cystic neoplasms of the pancreas. N Engl J Med. 2004;351:1218–1226.
Brugge WR, et al. Diagnosis of pancreatic cystic neoplasms: A report of the cooperative
pancreatic cyst study. Gastroenterology. 2004;126:1330–1336.
Dewitt J, et al. Comparison of endoscopic ultrasound and computed tomography for the
preoperative evaluation of pancreatic cancer: a systematic review. Clin Gastroenterol
Hepatol. 2006;4:717–725.
Farnell MB, et al. Pancreas cancer working group. A prospective randomized trial
comparing standard pancreatoduoderectomy with extended lymph aderectomy in
resectable pancreatic head adenocarcinoma. Surgery. 2005;138:618–628.
Giovannini M, et al. Endoscopic ultrasound-guided cystogastrostomy. Endoscopy.
2003;35:239–245.
Handrich SJ, et al. The natural history of the incidentally discovered small simple
pancreatic cyst: Long-term follow-up and clinical implications. Am J Roentgenol.
2005;184:20–23.
Kalady MF, et al. Pancreatic duct strictures: Identifying risks of malignancy. Ann Surg
Oncol. 2004;11(6):555–557.
Kim JE, Lee KT, et al. Clinical usefulness of carbohydrate antigen 19-9 as a screening test
for pancreatic cancer in an asymptomatic population. J Gastroenterol Hepatol.
2004;19:182–186.
Li D, et al. Pancreatic cancer. Lancet. 2004;363:1049–1057.
Maire, F, et al. Differential diagnosis between chronic pancreatitis and pancreatic cancer:
Value of the detection of KRAS II mutations in circulating DNA. Br J Cancer. 2002;
87(5):551–554.
79466_CH47.qxd 1/2/08 1:29 PM Page 334
Maire F, et al. Intraductal papillary mucinous tumors of the pancreas: The perioperative
value of cytologic and histopathologic diagnosis. Gastrointest Endosc. 2003;58:
701–706.
Norton ID, et al. The role of transabdominal ultrasonography, helical computed
tomography and magnetic resonance cholangiopancreatography in diagnosis and
management of pancreatic disease. Curr Gastroenterol Rep. 2000;2:120.
Somagyi L, et al. Diagnosis and staging of islet tumors of the pancreas. Curr Gastroenterol
Rep. 2001;2:159.
79466_CH48.qxd 1/2/08 1:30 PM Page 335
L aboratory tests have an important role in the recognition of liver disease and deter-
mination of the nature and extent of the liver dysfunction. There is no one specific test for
assessment of liver disease. However, the combination of a number of tests that assess dif-
ferent parameters of liver physiology obtained serially over time and interpreted within the
clinical context may serve in establishing the diagnosis and prognosis and help in follow-
ing the course of the hepatic dysfunction. The most useful laboratory tests in liver disease
may be grouped into three categories. These are tests that reflect (a) liver cell injury and
necrosis, (b) synthetic function of the liver, and (c) cholestasis from intra- or extra-
hepatic biliary obstruction or infiltrative processes in the liver, or both.
335
79466_CH48.qxd 1/2/08 1:30 PM Page 336
The degree of elevation of the serum transaminase activity has a low prog-
nostic value. Rapid recovery may occur in cases of toxic hepatitis, in shock-related
liver cell injury, or with relief of acute CBD obstruction, even with values greater
than 3,000 U/L. In contrast, in most patients with cirrhosis and in those with ter-
minal liver failure, values may be near normal.
Transaminase elevations are not specific for liver disease; they may also be
elevated in patients with cardiac and skeletal muscle damage as well as after stren-
uous exercise such as jogging, running, and working out. The extent of enzyme
elevation with muscle disease is usually less than 300 U/L except in acute rhab-
domyolysis. However, with severe muscle injury, other enzymes such as aldolase
and creatine phosphokinase (CPK) are also elevated.
Uremia may depress aminotransferase levels. This effect is reversible after
dialysis, suggesting that a dialyzable inhibitor of the aminotransferase reaction is
in the serum of uremic patients.
Lactic dehydrogenase (LDH) is a cytoplasmic enzyme found in most normal
and malignant tissues. Of the five isoenzymes of LDH (1–5), the electrophoretically
slowest one (LDH-5) corresponds to the liver. LDH is much less sensitive than the
aminotransferases in measuring liver cell injury, even when isoenzyme analysis is
used. It is most sensitive in revealing myocardial infarction and hemolysis.
Unconjugated Conjugated
red cell survival to one-half normal does not cause elevation of serum bilirubin. A
sixfold increase in red cell destruction results in serum bilirubin elevation to less
than 5 mg/dL.
In hemolytic hyperbilirubinemia, there may be a concomitant increase in con-
jugated bilirubin levels. However, if conjugated bilirubin is in excess of 15% of the
total bilirubin level, hepatic dysfunction must also be present.
B. Ineffective erythropoiesis. Patients with hematologic disorders characterized by
abnormalities of heme biosynthesis have increased bilirubin turnover without
increased extramedullary red cell destruction. These disorders include iron-deficiency
anemia, pernicious anemia, thalassemia, sideroblastic anemia, lead poisoning, and
erythropoietic porphyria.
Extrahepatic Intrahepatic
Biliary Drugs
Gallstone Hormones
Stricture Pregnancy
Cyst Viral hepatitis
Diverticula Alcohol hepatitis
Carcinoma Hodgkin’s disease, lymphoma
Bile duct Sarcoidosis
Ampulla of Vater Primary biliary cirrhosis
Lymph node involvement Sclerosing cholangitis
Pancreas Sepsis
Carcinoma Parenteral nutrition
Pseudocyst Postoperative
Chronic pancreatitis Dubin-Johnson syndrome
Rotor syndrome
3. Many drugs produce cholestasis and liver cell injury, which may be accompa-
nied by allergic manifestations such as fever, rash, arthralgia, and eosinophilia.
D. Postoperative jaundice. The cause of postoperative jaundice is multifactorial.
Most patients have bilirubin pigment overload from blood transfusion with
decreased red cell survival and resorption of blood from hematomas and large
ecchymoses. Concurrent use of drugs may cause hepatic dysfunction, injury, or
cholestasis. Hypotension, hypoxemia, sepsis, and shock contribute to impaired
hepatic function. Renal insufficiency may decrease urinary excretion of conjugated
bilirubin and enhance the degree of jaundice.
The hyperbilirubinemia may reach 30 to 40 mg/dL, with most of the bilirubin
being conjugated. Serum alkaline phosphatase level may be elevated up to 10-fold,
but the transaminases are usually only moderately elevated. Liver biopsy in most
instances shows intrahepatic cholestasis. The course of the jaundice depends on the
general condition of the patient. As the entire organ systems recover, the jaundice
subsides and liver function returns to normal.
E. Sepsis from any source in the body may result in conjugated hyperbilirubine-
mia and mild-to-moderate elevation of the transaminases and alkaline
phosphatases.
F. Hepatocellular disease. The most common disorders associated with jaundice
are hepatitis and cirrhosis. With injury to the hepatocytes, all the steps in bilirubin
metabolism are affected. Because secretion is the rate-limiting step, most of the
bilirubin is conjugated. The hyperbilirubinemia in hepatocellular disease usually
does not plateau and may exceed 60 mg/dL.
1. The major causes of intrahepatic cholestasis are
a. Alcohol-related liver disease.
b. Drugs (phenothiazines, sulfonylureas, allopurinol, azathioprine, thiazides,
acetaminophen, and aspirin).
c. Viral hepatitis (acute and chronic A, B, non-A, non-B, delta, Epstein-Barr
virus, cytomegalovirus, and others).
d. Toxic hepatitis.
e. Sepsis.
f. Infiltrative disorders (sarcoid, lymphoma, tuberculosis, primary or metastatic
malignancy, sickle cell disease).
2. Laboratory studies
a. In intrahepatic cholestasis, the laboratory tests reflect abnormal liver
function (Table 48-4).
b. In viral hepatitis, the transaminases may be elevated to 10 to 50 times nor-
mal (see Chapter 49).
c. In alcoholic liver disease, the alkaline phosphatase usually rises up to
about five times normal, SGOT (AST) is elevated less than 10 times normal,
and the SGPT (ALT) is lower than the SGOT (AST). The SGOT/ SGPT ratio
is usually 2:1 to 3:1 (see Chapter 51).
d. In drug-induced cholestasis, bilirubin may not be high, but there is
usually a dramatic rise in alkaline phosphatase with a slight rise in transam-
inases (see Chapter 53).
3. The course of the jaundice depends on the general condition of the patient.
As the entire organ systems recover, the jaundice subsides and liver function
returns to normal.
G. Extrahepatic cholestasis
1. Causes
a. Extrahepatic biliary obstruction due to stones, strictures, lymphadenopa-
thy, or tumors can occur anywhere along the route of the bile ducts from the
hilum of the liver to the duodenal papilla. Gallstone disease accounts for
most of the benign extrahepatic obstruction. Most of these patients have an
abrupt onset of jaundice. The disease ranges from biliary colic to acute chole-
cystitis and ascending cholangitis, especially with CBD stones.
b. Cancer. Pancreatic cancer, cholangiocarcinoma, adenocarcinoma of the
duodenum and ampulla of Vater, metastatic or primary liver tumors, and
79466_CH48.qxd 1/2/08 1:30 PM Page 345
Alcoholic 0–20 5 nl 10 nl 2 nl nl or sl ↓
liver disease
Acute viral 0–20 nl–3 nl 10–50 nl 10–50 nl nl
hepatitis
Drug-induced 5–10 2–10 nl nl–5 nl 10–50 nl nl
intrahepatic
cholestasis
Common bile duct 0–10 nl–10 nl nl–10 nl nl–10 nl nl
obstruction
Malignant 5–20 2–10 nl nl nl nl
common bile
duct obstruction
enlarged nodes at the porta hepatis are common causes of extrahepatic bil-
iary obstruction and jaundice.
2. Laboratory studies
a. In general, patients with gallstone disease have less hyperbilirubinemia
than those with intrahepatic cholestasis or extrahepatic malignant obstruc-
tion. The serum bilirubin is usually less than 20 mg/dL. The alkaline phos-
phatase may be elevated up to 10 times normal. The transaminases may
abruptly rise about 10 times normal and decrease rapidly once the obstruc-
tion is relieved.
b. In pancreatic cancer and other obstructive cancers, the serum bilirubin
may rise to 35 to 40 mg/dL, the alkaline phosphatase may rise up to 10
times normal, but the transaminases may remain normal. See Table 48.4 for
additional patterns of laboratory studies.
VII. DIAGNOSTIC STUDIES. The history, physical examination, and laboratory tests are
usually not sufficient to make the diagnosis of the underlying disorder causing the jaun-
dice. Additional diagnostic procedures are needed to arrive at a definitive diagnosis.
A. Noninvasive techniques
1. An abdominal flat plate (kidneys, ureters, bladder [KUB]) obtained in the
radiology department is the first test to be done on a patient with jaundice.
Aside from providing information on other organs in the abdomen, it may
show calcified gallstones, a “porcelain” gallbladder, air in the biliary tract, or
air in the gallbladder wall.
2. Ultrasonography. In most patients, the ultrasound should be the first proce-
dure performed. It identifies with 95% accuracy the presence of extrahepatic
bile duct obstruction, because the ducts proximal to the obstruction are usu-
ally dilated. It is a sensitive test for revealing stones in the gallbladder, but it
fails to show small stones or strictures in the bile ducts. CBD stones are visu-
alized reliably by ultrasound only in about one third of such patients.
Ultrasound may also demonstrate tumors, cysts, or abscesses in the pancreas,
liver, and other structures near the biliary tract. Ultrasound is limited in
patients who are obese, who have had a recent barium study, or who have a
79466_CH48.qxd 1/2/08 1:30 PM Page 346
for staging of malignant lymphoma and confirming the diagnosis and assessing
the severity of suspected alcoholic liver disease. A 2-cm piece of liver tissue is
needed to ensure accurate diagnosis; however, a sampling error can be expected
10% of the time despite an adequate amount of tissue. Liver biopsy may be
very helpful in evaluating the patient with cholestatic jaundice but only after
extrahepatic bile duct obstruction has been ruled out. Contraindications include
an uncooperative patient, hydatid cyst disease, suspected vascular lesion of
the liver, right-sided pleural effusion, infection of the biopsy site, and clinically
significant coagulopathy (i.e., PT 4 seconds prolonged and platelet count
75,000).
2. Laparoscopy (peritoneoscopy) can be performed with local anesthesia and
allows direct visualization of the liver. It may be helpful in the diagnosis of por-
tal hypertension, cirrhosis, and liver tumors in difficult cases and also allows for
visually directed liver biopsies.
3. Percutaneous transhepatic cholangiography (PTC) is performed in the radi-
ology department. Contrast material is injected directly under fluoroscopy into
the intrahepatic biliary tract through a 22- to 23-gauge, 15-cm-long Chiba or
skinny needle passed percutaneously from a right lateral intracostal approach.
It allows the visualization of the intrahepatic bile ducts. An adequate study
should be obtained in up to 75% of the patients with nondilated ducts and in
more than 90% of patients with dilated ducts as diagnosed by a previous ultra-
sound or CT scan. The complication rate ranges from 1% to 10%.
The patients must be cooperative and have good bleeding parameters with
a PT within 3 seconds of control and a platelet count greater than 50,000.
Prophylactic antibiotics should be used in patients with suspected obstruction
and infection. After the procedure, patients must be monitored for possible
bleeding or leakage of bile into the peritoneum.
This procedure may be used therapeutically to decompress the biliary tract
nonsurgically with the placement of a stent through an area of obstruction from
malignancy or benign stricture of the bile duct. The drainage may be external
or internal into the small bowel. The success rate is variable, and the procedure
should be considered palliative for poor-risk patients or those with nonre-
sectable masses.
4. Endoscopic retrograde cholangiopancreatography (ERCP) is the procedure
of choice when obstruction of the pancreatic or distal CBD is suspected by
ultrasound or CT scan. The duodenal lumen and papilla are visualized, and
contrast material is injected into the pancreatic and bile ducts under fluoro-
scopic guidance. The technique is successful greater than 90% of the time
whether the ducts are dilated or not. Visualization of the pancreatic duct may
reveal chronic pancreatitis, pseudocyst, or tumor causing the obstruction;
stones, strictures, and tumors causing obstruction of the bile ducts also may be
delineated.
ERCP allows for biopsies of the periampullary duodenum and the
papilla for tissue diagnosis. It may also be used as a therapeutic procedure in
patients with recurrent or retained CBD stones. A sphincterotomy can be
performed by cautery with incision of the papilla, relieving the obstruction
and allowing the gallstones to pass into the duodenum. Retrieval of gall-
stones may also be accomplished with intraductal balloon or baskets or after
intraductal crushing using special catheters. Nasobiliary stents may be placed
in the CBD to allow drainage, and permanent biliary stents may be placed in
the obstructed ducts as palliation in instances of inoperable carcinomatous
obstruction.
5. Angiography. Visualization of the hepatic arterial and venous circulation is
helpful in evaluating portal hypertension and determining the vascular supply,
vascularity, and surgical resectability of a mass lesion in the liver.
VIII. SUMMARY. The algorithm shown in Figure 48-2 summarizes the diagnostic workup
for the evaluation of jaundice.
79466_CH48.qxd 1/2/08 1:30 PM Page 348
Figure 48-2. Interpretation of laboratory liver tests in patients with jaundice. (R/o, rule out or
exclude.)
Selected Readings
Balisteri WF, et al. Intrahepatic cholestosis: summary of an American Association for the
study of liver diseases single topic conference. Hepatol. 2005;42:222–235.
Heathcote EJ. Diagnosis and management of cholestatic liver disease. Clin Gastroenterol
Hepatol. 2007;5(7):776–782.
Mendes FD, et al. Abnormal hepatic biochemistries in patients with inflammatory bowel
disease. Am J Gastroenterol. 2007;102(2):344–350.
Pratt DS, et al. Evaluation of abnormal liver enzyme results in asymptomatic patients.
N Engl J Med. 2000;342:1266.
Sorbi D, et al. An assessment of the role of liver biopsies in asymptomatic patients with
chronic liver test abnormalities. Hepatology. 1999;30(suppl):487A.
Vuppalanchi R, et al. Etiology of new-onset jaundice: How often is it caused by
idiosyncratic drug-induced liver injury in the United States? Am J Gastroenterol. 2007
Mar;102:558–562.
Wong K, et al. The diversity of liver diseases associated with an elevated serum ferritin.
Can J Gastro. 2007;20:467–470.
79466_CH49.qxd 1/2/08 1:31 PM Page 349
GALLSTONES 49
II. Gallstone disease is a major health problem in the United States. It affects approxi-
mately 20% of adult Americans. Gallstones are formed by the precipitation of insol-
uble bile constituents: cholesterol, polymerized bilirubin, bile pigments, calcium salts,
and proteins. Gallstones are classified into cholesterol, black pigment, and brown pig-
ment stones. Cholesterol stones are most frequent in industrialized societies. Black
pigment stones occur in patients with chronic hemolytic disorders, and brown pig-
ment stones are associated with impaction in the biliary tract. These stones are more
prevalent than cholesterol stones in the Far East. Cholesterol stones may be pure,
large (2.5 cm), solitary or mixed (70% cholesterol), multiple, smooth, and faceted.
Black and brown stones contain less than 25% cholesterol and are multiple and irreg-
ular. They contain polymerized bilirubin and calcium salts (bilirubinate, phosphate,
and fatty acids). All types of gallstones may become calcified. The calcification is usu-
ally central in pigment stones and peripheral in cholesterol stones.
A. Pigment stones (black and brown)
1. Epidemiology. Clinically, black pigment stones are more prevalent in three
major settings: hemolytic states, cirrhosis, and the elderly. In the United States,
approximately 30% of gallstones are pigment (mostly black) stones. The inci-
dence is age-dependent: Pigment stones are more common in the sixth and sev-
enth decades. In Asia, biliary tract parasitism with Clonorchis sinensis and
Ascaris lumbricoides, biliary ductal stasis, and chronic or repeated spasm at the
sphincter of Oddi as a result of widespread use of opium may contribute to the
increased prevalence (approximately 70%) of brown pigment cholelithiasis.
349
79466_CH49.qxd 1/2/08 1:31 PM Page 350
d. Gallstone dissolution with oral bile acid therapy is based on the theory
that cholesterol stones should dissolve in bile rendered unsaturated with
respect to cholesterol by increasing the concentration of bile salts. Gallstone
clearance may not depend completely on stone dissolution. Cholesterol-rich
stones can disintegrate as they dissolve, and the resultant fragments might
pass out of the gallbladder in the bile via the cystic duct and into the CBD
and duodenum.
The first bile acid used orally was chenodeoxycholic acid (CDCA).
CDCA decreases the cholesterol saturation of bile by lowering cholesterol
secretion, thus bringing about a gradual dissolution of cholesterol stones. In
a controlled trial using 12 to 15 mg/kg per day (approximately 1 g daily),
the stones dissolved in 40% to 60% of the selected patients during 2 years
of continuous therapy.
i. Contraindications. Dissolution therapy is contraindicated in the presence
of any of the following:
a) Pigment stones
b) Calcified stones
c) Stones larger than 1.5 to 2.0 cm
d) Multiple stones
e) Nonopacifying gallbladder on oral cholecystogram
f) Obesity
g) Pregnancy or women who may become pregnant
h) Concomitant liver disease
i) Severe symptoms
j) Nonresponders by oral cystography after 9 months of therapy
k) Poor patient compliance
7. Side effects. The most common side effect is secretory diarrhea induced by
the bile acid in the colon. Occasionally, gallstones become small enough during
therapy to pass into and obstruct the cystic or CBD, resulting in inflammation.
Minor liver enzyme elevations may occur in 7% of the patients without any
significant structural changes in liver histology. There may be a modest rise in
plasma and low-density lipoprotein (LDL) cholesterol.
8. Maintenance of therapy. CDCA therapy needs to be maintained indefinitely
in all patients because the bile reverts to its previous supersaturated state in
1 to 3 weeks after cessation of the therapy, and stones recur within 6 to
48 months.
9. Ursodeoxycholic acid (UDCA) is more potent than CDCA in lowering biliary
cholesterol secretion and saturation. It also has fewer side effects. It does not
affect the serum LDL levels or liver chemistry tests. A UDCA dosage of
10 mg/kg per day is optimal and equivalent to a CDCA dosage of 15 mg /kg per
day. In some patients, stone rim calcifications may occur and limit dissolution.
10. The combination of UDCA and CDCA is at least as effective and free of side
effects as monotherapy with UDCA and is less expensive. The dosage of
CDCA is reduced to 7.5 mg / kg per day, and UDCA is used at a dosage of
5 mg /kg per day.
III. CHOLECYSTITIS
A. Acute calculous cholecystitis. Inflammation of the gallbladder is associated with
gallstones in more than 90% of cases. It is a common problem, presenting as an
acute abdomen, especially in middle-aged women. Acute calculous cholecystitis is
caused by obstruction of the cystic duct either by an impacted stone or by the edema
and inflammation caused by the passage of a stone to the CBD and duodenum. The
obstructed gallbladder becomes distended, and the walls become edematous,
ischemic, and inflamed. Secondary infection with enteric organisms complicates the
inflammation and may lead to cholangitis and sepsis.
1. Diagnosis
a. Clinical presentation. The pain of acute cholecystitis usually starts in the
right upper quadrant or epigastrium as a colicky pain followed by local signs
79466_CH49.qxd 1/2/08 1:31 PM Page 354
patency of the biliary system and the presence of possible residual stones. If
these abnormalities are detected, cholecystectomy should be performed
when possible.
3. Complications
a. Perforation is the most serious complication of acute cholecystitis. It may
be localized or may extend into the peritoneal cavity with subsequent peri-
tonitis or into an adjacent hollow organ, such as the stomach, duodenum, or
colon with formation of a cholecystenteric fistula. Surgical intervention is
necessary in all cases.
b. Gallstone ileus is a form of mechanical intestinal obstruction caused by the
impaction of a large gallstone that has entered into the intestine from a
cholecystenteric fistula. The obstruction may be intermittent as the stone
moves along the intestine until permanent obstruction occurs. Most obstruc-
tions occur in the ileum. Colonic obstruction is rare except at sites of
previous narrowing due to another disease process, such as diverticular or
inflammatory bowel disease. Gallstone ileus requires prompt diagnosis and
laparotomy.
c. Mirizzi syndrome. Rarely, a gallstone impacted in the cystic duct or the
neck of the gallbladder may cause a localized obstruction of the common
hepatic duct from direct pressure or inflammatory changes around the duct.
The obstruction can cause right upper quadrant pain, jaundice, recurrent
cholangitis, and possibly a fistula between the two ducts.
Ultrasound examination may show dilated ducts above the point of
obstruction as well as the stone. ERCP or percutaneous transhepatic cholan-
giography (PTC) confirms the site of obstruction. CT scan may be helpful in
defining the stone and differentiating it from tumor or mass. Surgery is
required in most instances.
B. Acute acalculous cholecystitis is a particularly severe form of inflammation of
the gallbladder that occurs in the absence of cholelithiasis. There is a high inci-
dence of necrosis, gangrene, and perforation of the gallbladder in this group of
patients. The mortality may be as high as 50% if diagnosis is delayed and prompt
therapy not instituted.
Most of the patients are elderly or debilitated as a result of coexisting disease
or trauma. The condition is also seen in patients of all ages in the intensive care
unit, after surgery, or on total parenteral nutrition. Absence of oral intake associ-
ated with gallbladder stasis, sludge formation, and increased biliary pressure due
to narcotic drugs that increase the tone at the sphincter of Oddi may contribute to
its pathogenesis.
1. Diagnosis
a. Clinical presentation. Bile is usually infected with enteric bacteria, which
may lead to sepsis. The clinical presentation may be nonspecific, and the
diagnosis requires a high index of suspicion. Most patients complain of
abdominal pain, nausea, and vomiting. Fever and chills may be present.
Serum bilirubin, alkaline phosphatase, ALT, and AST may be elevated. There
is usually a moderate leukocytosis (10,000–20,000 cells/L) with a left shift.
Serum amylase may be elevated.
b. Diagnostic studies. The diagnostic test of choice in acute acalculous
cholecystitis is ultrasound of the gallbladder, which identifies a distended
gallbladder with thickened walls and biliary sludge. Nuclear scans with
HIDA or PIPIDA may give equivocal results in these debilitated patients and
are not reliable.
2. Treatment. Successful management of acute acalculous cholecystitis
requires prompt diagnosis and early surgical intervention. Patients should be
treated with antibiotics to cover enteric organisms and enterococci.
Cholecystectomy is the surgical procedure of choice. Cholecystostomy is not
recommended because, in most cases, the gallbladder is necrotic or gan-
grenous and its total removal is necessary to prevent perforation and other
complications.
79466_CH49.qxd 1/2/08 1:31 PM Page 356
IV. CHOLEDOCHOLITHIASIS
A. Pathogenesis. Stones in the bile ducts may be primary (develop in the duct) or
secondary (originate in the gallbladder). If they are discovered after cholecystec-
tomy, they may have been overlooked (retained) or may have formed after the
surgery (recurrent).
1. Primary bile duct stones are rare in Western countries. They are more com-
mon in the Orient and are often associated with biliary infections and parasitic
infestations. They are usually pigment stones.
2. Secondary bile duct stones. Because 10% to 15% of patients with choles-
terol gallstones also have stones in the CBD, it is thought that most CBD stones
in the Western countries originate from the gallbladder. In fact, 95% of patients
with ductal stones also have stones in the gallbladder.
3. Stones found in the bile ducts after cholecystectomy may be retained or may
have formed de novo. Bile stasis associated with partial obstruction or marked
dilatation of the duct may promote choledocholithiasis. Recurrent stones are
often formed from bile pigments.
B. Diagnosis
1. Clinical presentation
a. Choledocholithiasis may present in the following ways:
i. Biliary colic–abdominal pain
ii. Obstructive jaundice
iii. Cholangitis
iv. Pancreatitis
v. Hemobilia
b. The biliary obstruction caused by cholelithiasis and the ensuing increased
biliary pressure and diminished bile flow result in the morbidity associated
with duct stones. The rate of progression of the obstruction, its degree, and
concomitant bacterial contamination of the biliary tract determine the sever-
ity of the syndrome. Thus acute obstruction usually causes colic with or
without concomitant pancreatitis; gradual obstruction may present as jaun-
dice. Cholangitis and abscess formation may follow if obstruction is not
relieved. Chronic biliary obstruction, if not relieved, may give rise to sec-
ondary biliary cirrhosis resulting in hepatic failure and portal hypertension.
c. Signs and symptoms. The most common complaint is right upper abdom-
inal or epigastric pain, which is usually associated with nausea and vomit-
ing. Jaundice, which may be fluctuating or progressive, is also common. If
obstruction is severe, dark urine and pale stools may develop. Fever and
chills, if present, will signal cholangitis or abscess formation.
2. Diagnostic studies
a. Laboratory studies. Elevation of alkaline phosphatase and bilirubin levels
is the hallmark of ductal obstruction. Serum amylase may be elevated with-
out concurrent pancreatitis. Elevation of transaminases (ALT, AST) may be
seen transiently with passage of a stone. If it persists, along with leukocyto-
sis, cholangitis is suspected.
b. Ultrasonography is the initial diagnostic test of choice in the workup of
gallstone disease. Aside from the presence of stones in the gallbladder, dilata-
tion of the biliary tract secondary to obstruction of the bile ducts is clearly
seen on ultrasound. If the obstruction occurs acutely, dilatation may not be
present.
In patients who have undergone cholecystectomy, slight dilatation of
the CBD (up to 0.8 cm) may be acceptable without the presence of distal
obstruction.
c. Biliary scintigraphy using 99mT c-labeled HIDA or PIPIDA may show
obstruction of the CBD in 85% to 90% of cases. In a positive scan, if the
cystic duct is patent, the passage of the radionuclide into the gallbladder and
the major ducts but not into the small bowel is noted within 1 to 4 hours.
d. CT scan is an excellent method to demonstrate CBD stones, especially those
with calcium.
79466_CH49.qxd 1/2/08 1:31 PM Page 357
e. MRCP has excellent sensitivity and specificity in visualizing the CBD for
stones as well as detecting other structural abnormalities such as ductal dila-
tion, sclerosing cholangitis and cystic abnormalities.
f. ERCP demonstrates the location of the stone or stones in the bile ducts and
is preferred in patients with suspected CBD obstruction without intrahepatic
ductal dilatation. Endoscopic examination of the upper gastrointestinal tract
and the duodenal ampullary orifice helps rule out pathology in these areas.
Endoscopic sphincterotomy has replaced operative sphincteroplasty in
patients with retained or recurrent CBD stones discovered after cholecystec-
tomy. In most cases, CBD stones smaller than 1.5 cm spontaneously pass
into the duodenum after endoscopic sphincterotomy. Stones larger than
1.5 cm can be fragmented and removed with special ERCP catheters, baskets,
and balloons. This technique can also be used therapeutically in debilitated
patients with CBD stones and intact gallbladders when cholecystectomy and
bile duct exploration are medically contraindicated. If an impacted CBD
stone is the cause of the pancreatitis, endoscopic removal by sphincterotomy
is the preferred immediate mode of therapy. In all cases in which an
obstructed CBD is manipulated, broad-spectrum IV antibiotic coverage must
be provided to the patient to prevent sepsis.
g. PTC may be used diagnostically and occasionally therapeutically in some
patients. If obstruction of the CBD and dilatation of the intrahepatic biliary
tract has been demonstrated by ultrasound, the location and the nature of
the obstruction can be delineated by PTC. Furthermore, it is possible to
relieve the obstruction, even if temporarily, by the insertion of a stent, espe-
cially in debilitated patients in whom surgery is contraindicated. Dissolution
of CBD cholesterol stones by infusion of solvents such as monooctanoin
with a catheter percutaneously placed above the stone or attempts at dis-
lodging and mobilizing the stone and facilitating its passage into the duode-
num or withdrawing it percutaneously may be considered for therapy for
such patients. Patients should be treated with IV broad-spectrum antibiotics
before such attempts.
C. Treatment. In symptomatic patients presenting with stones in the gallbladder and
the CBD, the treatment of choice is laparoscopic cholecystectomy and endoscopic
stone extraction via ERCP pre- or postoperatively. Cholecystectomy and CBD
exploration are reserved for patients with contraindications to the laparoscopic
procedure or who require abdominal exploration. If stones are found in the CBD
during CBD exploration, they should be removed, and a drainage procedure such
as a sphincteroplasty or choledochoenterostomy may be performed to allow the
passage of any residual stones into the gut. In these instances, a T tube is placed in
the CBD to decompress the biliary duct and to allow the performance of postop-
erative cholangiograms.
In approximately 2% of the patients, a residual CBD stone is demonstrated
on postoperative cholangiograms. These residual stones may be extracted either
percutaneously through the T tube or endoscopically by means of ERCP. In situ
dissolution of cholesterol stones with monooctanoin or MBTE infusion has also
been successful in selected patients. If these methods fail, endoscopic sphinctero-
tomy and stone removal or reoperation may be necessary.
abscess formation. Recurrent infection and inflammation of the bile ducts may result
in strictures, areas of dilatation, and intraductular calcium bilirubinate stone formation.
Secondary biliary cirrhosis and portal hypertension may be late sequelae.
A. Diagnosis
1. Clinical presentation. Patients usually present with intermittent abdominal
pain, fever, chills, and jaundice. Dark urine and pale stools suggest the presence
of biliary obstruction.
2. Diagnostic studies. Leukocytosis with a shift to the left and elevation of
serum alkaline phosphatase and bilirubin levels are common. Serum ALT, AST,
and amylase levels are usually elevated.
B. Treatment. When cholangitis is suspected, prompt diagnosis and relief of the
obstruction is essential. Patients should be given IV antibiotics for enteric organ-
isms to prevent sepsis.
Ultrasonography, ERCP, PTC, and abdominal CT scan may be necessary
to diagnose the extent, nature, and location of the obstruction. The choice of defin-
itive treatment for relief of the obstruction depends on the findings. Surgery or
other less invasive techniques such as PTC or ERCP may be required.
Selected Readings
Agrawal S, et al. Gallstones, from gallbladder to gut: Management options for diverse
complications. Postgrad Med. 2000;108:143.
Chuang CZ, et al. Physical activity, biliary lipids and gallstones in obese subjects. Am
J Gastroenterol. 2001;96:1860.
deLedinghen V, et al. Diagnosis of choledocholithiasis: EUS or magnetic resonance
cholangiography? A prospective controlled study. Gastrointest Endosc. 1999;49:26.
Donovan JM, et al. Physical and metabolic factors in gallstone pathogenesis. Gastroenterol
Clin N Am. 1999;28:75–97.
Germanos S, et al. Clinical update: surgery for acute cholecytitis. Lancet. 2007;369:
1774–1776.
Ko C, et al. Gallbladder disease. Clin Perspect Gastroenterol. March /April 2000:87.
Kolla SB, et al. Early vs delayed laparoscopic cholecystectomy for acute cholecystitis:
a prospective, randomized trail. Surg Endos. 2004;18:1323–1327.
Mendez SN, et al. Intestinal motility and bacterial overgrowth in patients with gallstones.
Gastroenterology. 2001;120:1310.
Mulholland MW. Progress in understanding acalculus gallbladder disease. Gastroenterology.
2001;120:570.
Peng WK, et al. Role of laparoscopic cholecystectomy in the early management of acute
gallbladder disease. Br J Surg. 2005;92:586–591.
Schirmer BD. Cholelithiasis and cholecystitis. J Long Term Eff Med Implants. 2005;
15:329–338.
Shaffer EA. Gallbladder sludge: What is its clinical significance? Curr Gastroenterol Rep.
2001;3:166.
Throwridge RL, et al. Does this patient have acute cholecystitis? JAMA. 2003;299:80–86.
Tsai CJ, et al. Long-term intake of dietary fiber and decreased risk of cholecystectomy. Am
J Gastroenterol. 2004;99:1364–1370.
Tsai CJ, et al. Dietary protein and risk of cholecystectomy in a cohort of US women: The
Nurses Health Study. Am J Epidemiol. 2004;160:11–18.
Tsai CJ, et al. Glycemic load, glycemic index, and carbohydrate intake in relation to risk
of cholecystitis in women. Gastroenterology. 2005;129:105–112.
Verma D, et al. EUS vs MRCP for detection of choledocholithiasis. Gastrointest Endosc.
2006;64:248–254.
79466_CH50.qxd 1/2/08 1:32 PM Page 359
VIRAL HEPATITIS 50
H epatitis can be defined as the constellation of symptoms and signs resulting from
hepatic inflammation and hepatic cell necrosis. If the insult is acute and occurs in a previ-
ously asymptomatic individual, the term acute hepatitis can be applied. The most com-
mon causes of acute hepatitis are viruses, toxins, and alcohol. Occasionally other disease
entities such as Wilson’s disease, leukemias, and lymphomas with acute infiltration of the
liver may give rise to a clinical picture of acute hepatitis. Viruses, however, are the major
etiologic agents of acute liver injury.
Systemic infection with several viruses results in hepatic inflammation and cell death.
Viruses that cause hepatitis have been classified as hepatitis A (HAV), B (HBV), C (HCV),
delta (HDV), and E (HEV). However, in some individuals, infection with the Epstein-Barr
virus (EBV) or cytomegalovirus (CMV) also results in acute hepatitis.
In most patients, acute viral hepatitis presents as an acute illness characterized by the
abrupt onset of malaise, fever, anorexia, nausea, headache, abdominal discomfort, and pain.
Jaundice, itching, dark-colored urine, and light-colored stools often cause the patient to seek
medical attention. At this stage, the disease caused by different viruses is usually indistin-
guishable; serologic studies and viral DNA or RNA determination by polymerase chain reac-
tion (PCL) may provide the only means of identification. Pertinent factors regarding the five
forms of acute and chronic viral hepatitis are summarized in Tables 50-1 and 50-2.
359
79466_CH50.qxd 1/2/08 1:32 PM Page 360
Hepatitis
A B C D E
HAV, hepatitis A virus; HBV, hepatitis B virus; HCV, hepatitis C virus; HDV, hepatitis D virus; HEV,
hepatitis E virus; ssRNA, single-stranded RNA; dsDNA, double-stranded DNA; kb, kilobase; Ag, antigen.
From Hoofnagle JH, DiBisceglie AM. Serologic diagnosis of acute and chronic viral hepatitis. Semin
Liver Dis. 1991;11:74. Reprinted with permission.
HAV, hepatitis A virus; HBsAg, hepatitis B surface antigen; HBc, hepatitis B core; HCV, hepatitis C
virus; HDV, hepatitis D virus.
*Initially four tests should be obtained: IgM anti-HAV, IgM anti-HBc, HBsAg, and anti-HCV. In some
situations, further testing for anti-HDV and anti-HCV are needed (see text).
From Hoofnagle JH, DiBisceglie AM. Serologic diagnosis of acute and chronic viral hepatitis. Semin
Liver Dis. 1991;11:11. Reprinted with permission.
79466_CH50.qxd 1/2/08 1:32 PM Page 361
3. Liver disease
a. Primary CMV infections. In patients with primary CMV infections, sub-
clinical liver involvement may be common with a mild rise in liver enzyme
levels. In others, a mild, self-limited hepatitis with a mild or moderate rise
in liver enzymes is present. There may be accompanying hemolysis and atyp-
ical mononuclear cells in the peripheral blood smear.
b. CMV from blood transfusions. In these instances, the virus is thought to
be present in blood leukocytes including neutrophils. The onset of disease is
3 to 6 weeks after transfusions. Hepatitis in most patients is accompanied
by a mononucleosislike disease.
c. Transplant patients. In most patients who have undergone transplanta-
tion, CMV is found to be a common cause of acute hepatitis. CMV may
enter the patient through the transplanted organ or transfusions, or it may
be reactivated during immunosuppression. In these patients, chronic infec-
tion of the liver may also exist without producing disease.
d. Liver biopsy specimens may be cultured for CMV. Histologically, liver
parenchyma contains the owl-eye inclusions in the hepatocytes and may
contain noncaseating granulomas or granulomatous changes.
e. It is debatable whether CMV can cause massive hepatic necrosis. There is
no evidence that CMV causes chronic liver disease in normal hosts.
However, in immunocompromised patients (e.g., in patients with AIDS), the
hepatobiliary infection may be extensive and chronic. Concomitant liver
infections with CMV and HBV have been described.
C. Treatment. In proven cases, intravenous (IV) therapy with ganciclovir sodium
may be helpful. Doses used vary from 2.5 mg/kg q8h for 20 days to 5 mg/kg q12h
for 14 days. Treatment may need to be continued indefinitely (see Chapter 43) in
patients with AIDS.
III. HEPATITIS A. Infections with HAV account for approximately 25% of the clinical
hepatitis cases diagnosed in industrialized nations. It occurs both sporadically and in
epidemics. In general, it causes less morbidity and mortality than hepatitis virus types
B, C, and delta.
A. Pathogenesis. HAV is endemic in underdeveloped countries, where infections
usually occur in children and are clinically inapparent. The outcome of the infec-
tion seems to depend on the age of the patient and the infecting dose of the virus.
The disease is of shorter duration and milder in children. Adults may present with
clinically significant disease; fulminant hepatitis occurs with a frequency of 1 to 8
per 1,000 cases. HAV is a 27-nm, nonenveloped RNA enterovirus and belongs to
the group picornavirus. All strains of the virus identified to date belong to one
serotype. The infection is acquired by the fecal-oral route and can be isolated from
the liver, bile, stools, and blood during the late incubation period and acute pre-
icteric phase of the disease. Fecal viral shedding and viremia diminish once jaun-
dice occurs.
B. Diagnosis
1. Clinical presentation. The typical course of a case of acute hepatitis A is
shown in Figure 50-1. Hepatitis A has an incubation period of from 2 to 7
weeks with a mean of 4 weeks. In clinically apparent cases, there is an abrupt
onset of symptoms. Malaise, fever, anorexia with aversion to food and ciga-
rettes, nausea, and abdominal pain are frequent. Within a few days after the
onset of symptoms, there is a marked rise in serum alanine aminotransferase
(ALT), aspartate aminotransferase (AST), and bilirubin levels, resulting in jaun-
dice. In approximately 25% of the cases, hepatosplenomegaly is noted.
2. The duration of the illness is usually less than 1 month, but elevated serum
transaminases have been recorded for as long as 6 months.
3. Complications from hepatitis A are rare. A cholestatic phase may occur in
some patients and may mimic obstructive jaundice. Relapses have been seen
within 4 to 15 weeks of recovery from the disease and may be precipitated by
exertion or alcohol intake. The disease may last 16 to 40 weeks. There may be
79466_CH50.qxd 1/2/08 1:32 PM Page 363
Figure 50-1. Typical course of acute hepatitis A. ALT, alanine aminotransferase; HAV, hepatitis A
virus; anti-HAV, antibody to HAV. (From Hoofnagle JH, DiBisceglie AM. Serologic diagnosis of acute
and chronic viral hepatitis. Semin Liver Dis. 1991;11:74. Reprinted with permission.)
2. Travelers to endemic areas should receive prophylaxis, with the dose of the
immunoglobulin dependent on the intended period of stay. For travel of less
than 2 months’ duration, 0.02 mL/kg should be given, and 0.06 mL/kg for
longer periods.
3. Vaccine. Over the last several decades, the incidence of HAV infection has been
decreasing in industrialized countries. Because infection in early childhood is
becoming less common, a population of susceptible adolescents and adults, in
whom hepatitis A can be more severe, is emerging. Passive immunization with
serum immune globulin (IG) has a limited duration of protection of approxi-
mately 3 months. For a longer duration of protection, active immunization with
a vaccine is required. Clinical trials with a killed whole virus vaccine have been
successful, and the vaccine is now available. The vaccine is administered in three
doses intramuscularly into the deltoid muscle at 0, 1, and 6 months. A recom-
binant complementary DNA vaccine for this RNA virus is also available.
A combination vaccine (Twinrit-Galaxo Smith Kline) containing 20 g of
HBsAg protein (Energix –B) and 720 ELISA units of inactivated Hep A virus
(HAV VIX) provides dual protection with three injections at 0, 1, and 6 months.
IV. HEPATITIS B (HB). There are about 2.5 billion people worldwide infected with
HB virus (HBV), of these about 350 million persons are chronic carriers and 4 mil-
lion new HBV infections occur yearly. The highest prevalences of HBV (about
15%) are in the Far East and Southeast Asia, Middle East, Africa, and among
Alaskan natives and Pacific Islanders. HBV causes acute and chronic liver disease
and liver cancer.
HBV is spread parenterally (horizontal transmission) or by intimate contact
since HBV is found not only in blood, but also in semen, saliva, and other body secre-
tions. It is believed that transmission in Asia is perinatal (vertical transmission)
from mother to infant.
The risk factors for transmission include high-risk sexual activity (multiple sex-
ual partners, homosexual activity as in men having sex with men), injection drug use,
hemodialysis, living or being born in an endemic area, and working in the health care
profession.
There are eight genotypes (A–H) of HBV. Genotype A is found in Europe and
North America, genotype B and C in the Far East and Southeast Asia, and genotype D
mainly in southern Europe, Africa, and India. The clinical significance of genotypes
are enfolding (i.e., genotype A and B are more sensitive to treatment with interferon
than genotypes C and D).
A. HBV is a DNA virus belonging to a class of animal viruses called hepadna viruses.
These viruses are hepatotropic, tend to cause persistent infections, and have been
associated with the development of hepatocellular carcinoma. HBV is unique among
human viruses in its genomic and antigenic structure and its replicative cycle.
The structure of HBV is shown in Figure 50-2. It consists of an outer shell
made up of a protein (HBsAg) and a complex inner core. The complex inner core
or the nucleocapsid core is a 27-nanometer (nm) icosahedral structure that consists
of 180 copies of viral core protein (HBcAg) surrounding the viral DNA (genome)
and virally encoded DNA polymerase. HBcAg protects the viral genome from
Figure 50-3. Structure of the HBV genome. The thick lines are the DNA strands, and the open
reading frames that code for proteins are indicated outside the DNA. Black boxes indicate the start
sites of the open reading frames. (From Foster GR, Carman WF, Thomas HC. Replication of hepati-
tis B and delta viruses: Appearance of viral mutants. Semin Liver Dis. 1991;11:122. Reprinted with
permission.)
degradation by exogenous nucleases. Figure 50-3 depicts the structure of the HBV
genome.
The HBV genome is composed of circular DNA of approximately 3,200 base
pairs with a complete negative strand and an incomplete complementary positive
strand. The negative strand contains overlapping genes that encode structural pro-
teins (surface proteins and their derivatives and core) and two replicative proteins
(polymerase and X). HBV is unique among DNA viruses in that it replicates in a
way similar to that of the RNA retroviruses such as the human immunodeficiency
virus (HIV) via an RNA intermediate.
Once HBV is blood borne in the host, the replication cycle begins with the
attachment of HBV to the hepatocyte cell membrane and entry into the hepatocyte
cytoplasm. The virus is then uncoated and the nucleocapsid and the viral DNA are
transported into nucleus. Once in the nucleus, the viral genome is repaired by fil-
ing in the gap in the positive-strand DNA and forming the covalently closed circu-
lar DNA (cc cDNA.) Thus, complexly double-stranded HCV DNA is formed.
Synthesis of cc cDNA is catalyzed by the viral DNA polymerase.
79466_CH50.qxd 1/2/08 1:32 PM Page 366
Figure 50-4. Typical course of acute hepatitis B. Initially, HBV-DNA can be detected by blot
hybridization, but as the disease resolves, only low levels can be detected using polymerase chain
reaction. ALT, alanine aminotransferase; HBsAg, hepatitis B surface antigen; HBeAg, hepatitis B e
antigen; HBV-DNA, hepatitis B virus DNA; anti-HBc, antibody to hepatitis B core antigen; anti-HBe,
antibody to HBeAg; anti-HBS, antibody to HBsAg; PCR, polymerase chain reaction. (From Hoofnagle
JH, DiBisceglie AM. Serologic diagnosis of acute and chronic viral hepatitis. Semin Liver Dis.
1991;11:75. Reprinted with permission.)
E. Acute hepatitis. The outcomes of HBV infection are outlined in Figure 50-5.
About one fourth of individuals with HBV infection develop clinically apparent
acute hepatitis. The incubation period, or time between exposure and onset of
symptoms, is 1 to 6 months. During this time, there is active viral replication, and
the patient’s serum becomes positive for HBsAg, DNA, and HBcAb IgM.
1. Clinical presentation. Before jaundice and typical clinical findings of hepatitis
become apparent, these patients may present with rash, neuralgia, arthralgia,
arthritis, glomerulonephritis, and polyarteritis nodosa, vasculitis, mixed cryo-
globulinemia, pericarditis, pancreatitis, and aplastic anemia. These disease
states are believed to result from circulating antigen-antibody complexes.
2. Laboratory tests
a. As the patient becomes symptomatic, there is a concomitant rise in serum
ALT and AST (5–20 times normal) and a moderate elevation of the serum
alkalie phosphatase (2–10 times normal). These enzymes represent hepato-
cellular damage. Serum bilirubin may reach very high levels (30 mg/dL).
Prothrombin time (PT) and partial thromboplastin time (PTT) levels
may also become abnormal depending on the severity of the liver disease.
b. Serum HBV-DNA becomes undetectable in these patients within 1 to 8 weeks
after the onset of symptoms. HBcAg also disappears soon after the peak of
serum transaminase. HBsAg usually remains detectable in the serum through-
out the illness and may persist even into convalescence. This is because the
initial HBsAg titers are very high, and because HBsAg has a long half-life of
8 days, it may take months to clear it to undetectable levels.
c. HBcAb is present in high titers in the serum of infected patients with HBV
when jaundice appears. The initial antibody is of IgM type; with recovery,
the IgM type disappears, and the IgG HBcAb titers reach high levels and
persist for life.
79466_CH50.qxd 1/2/08 1:32 PM Page 368
0%
70
-3
%
%
10
-9
0%
Figure 50-5. Outcomes of acute hepatitis B virus infection in adults. (From Hoofnagle JH, Schafer DF.
Serologic markers of hepatitis B virus infection. Semin Liver Dis. 1986;11:1. Reprinted with
permission.)
d. HBeAb appears when HBeAg becomes negative and often disappears within
a few months or years.
e. HBsAb arises during recovery after HBsAg has cleared. There is usually a
“window” period between the disappearance of HBsAg and the appearance
of HBsAb. The only way to make the diagnosis of acute hepatitis in these
individuals is to show the presence of HBcAb (IgM type). A minority of the
patients (5%–15%) who clear HBV and recover normally never develop
HBsAb. However, most of these individuals have positive IgM HBcAb.
f. In some patients (10%), HBV clearance is very rapid, and HBsAg levels may
be absent at the onset of symptoms. These acutely ill patients have positive
IgM HBcAb titers and may have detectable HBeAg levels. These variations
are more common in both mild and fulminant disease.
g. HBV-DNA can be detected and measured quantitatively with specific assays.
The hybridization assay measures replicating HBV-DNA whereas the ampli-
cation assay performed using the polymerase chain reaction (PCR) measures
HBV-DNA nonspecifically.
3. Prognosis. In patients who can mount a vigorous immune response, the virus
is cleared, and recovery is within a few months (1–6 months). A minority of the
patients with acute hepatitis (1%–5%) develop fulminant hepatic failure (FHF)
due to massive hepatic necrosis. The prognosis in these patients is poor and
depends on hepatic regeneration.
79466_CH50.qxd 1/2/08 1:32 PM Page 369
The risk of HBV infection becoming chronic varies with the age of the
patient at the initial infection. In an immunocompetent adult, the risk is 5%,
in an immune compromised adult 50%, in early childhood 50%, and in the
newborn 90%.
F. Chronic HBsAg carrier state and chronic hepatitis
1. Clinical course and serology. About one tenth of adult patients infected with
HBV do not clear the virus and remain HBsAg positive. Some of these patients
develop chronic progressive hepatitis, and others may remain in a clinically qui-
escent “carrier” state. Figure 50-6 summarizes the serologic course of chronic
type B hepatitis in immune competent adults.
a. In patients who develop chronic type B hepatitis, the initial pattern of HBV
markers is similar to that in patients with acute, self-limited hepatitis.
However, in patients with ongoing active disease, HBeAg and HBV-DNA
persist and accompany elevated serum transaminases even after 6 months of
symptomatic disease.
b. In most patients, the initial disease is mild, and some patients may be
anicteric and symptomatic. These patients may present with only nonspecific
symptoms of anorexia and fatigue and mild-to-moderate elevation of liver
tests.
c. IgM HBcAb levels remain elevated in chronic active type B hepatitis. As the
disease wanes, the titers diminish with an increase in IgG HBcAb.
d. Patients who remain HBsAg positive do not produce specific HBsAb.
However, in 20% to 40% of HBsAg carriers, there may be low levels of
HBsAb directed toward HBsAg subdeterminants not present in the serum.
e. The course of chronic HBV infection varies. The activity of the liver disease
and the serologic markers change over time. In approximately one half of
these patients, HBeAg disappears and is replaced by HBeAb. Concomitantly,
there is a flare of the hepatitis with elevated ALT and AST levels and loss of
Figure 50-6. Representative course of chronic hepatitis B in which acute infection is followed by
chronic infection. Ultimately, there is a remission in disease when seroconversion from HBeAg to anti-
HBe occurs. ALT, alanine aminotransferase; HBeAg, hepatitis B e antigen; HBV-DNA, hepatitis B virus
DNA; HBsAg, hepatitis B surface antigen; PCR, polymerase chain; anti-HBc, antibody to hepatitis B
core antigen; anti-HBe, antibody to HBeAg. (Hoofnagle JH, DiBisceglie AM. Serologic diagnosis of
acute and chronic viral hepatitis. Semin Liver Dis. 1991;11:75. Reprinted with permission.)
79466_CH50.qxd 1/2/08 1:32 PM Page 370
Another mutant of HBV has been identified in patients who have active
hepatitis B but lack HBeAg. These hepatitis B virions have a point mutation in
the precore gene. This mutation prevents the synthesis of the precore-core pro-
tein and abolishes the formation of HBeAg, which modulates viral infection.
Lack of HBeAg expression on the infected hepatocytes prevents immunologic
recognition and destruction of these infected cells, resulting in chronic active
hepatitis.
5. Superinfection. HBV carriers may develop hepatic superinfections with other
hepatotropic viruses. Sudden increases in the serum transaminases may repre-
sent superinfection with HAV, HCV, or HDV.
a. HAV infection may be a superinfection or simultaneous coinfection and is
usually associated with more severe and fulminant hepatitis.
b. HCV superinfections may be difficult to document accurately due to the
delay of detectability of HBcAb in acute HCV infections. However, the diag-
nosis can be made in the face of increased serum transaminases with reduc-
tion of HBsAg titers due to viral interference, and negative tests for IgM
HBcAb, IgM HAV, and anti-HDV. HCV DNA titers should be measured in
ongoing hepatitis.
c. HDV superinfection is discussed in section V.
d. Nonviral causes. It should always be kept in mind that sudden increases in
the serum transaminases may result from nonviral causes, such as drug and
alcohol hepatotoxicity, shock, congestive heart failure, right ventricular fail-
ure, and extrahepatic biliary obstruction.
6. The differential diagnosis of HBsAg-positive acute viral hepatitis is sum-
marized in Table 50-3.
7. Primary hepatocellular carcinoma (HCC). In the parts of the world where
HBV infection is endemic (e.g., Far East, sub-Saharan Africa), PHC is the lead-
ing cause of death from cancer. It appears that persistent HBV infection is the
leading cause of PHC. This suggests that HBV is an oncogenic virus. The pres-
ence of cirrhosis is not necessary for transformation into HCC.
a. Predisposing factors. Some of the predisposing factors for development
of HCC are race (e.g., Asians, Inuits), age at infection (especially infancy
and early childhood), chronic-persistent infection, and the presence of envi-
ronmental cocarcinogenic factors, such as ingestion of ethanol, cigarette
smoking, and possibly exposure to aflatoxin.
HBc, hepatitis B core; HDV, hepatitis delta virus; HAV, hepatitis A virus; HCV, hepatitis C virus; HBV,
hepatitis B virus.
Adapted from Hoofnagle JH, Schafer DF. Serologic markers of hepatitis B virus infection. Semin Liver
Dis. 1986;6:1. Reprinted with permission.
79466_CH50.qxd 1/2/08 1:32 PM Page 372
TABLE 50-4
Clinical Characteristics of Three Phases
of Chronic HBV Infection
Immune reactive
Immune
Clinical profile Inactive carrier tolerant HBeAg HBeAg Mutant
HBeAg
HBeAb
HBV-DNA by 104 105 105 104
PCR
(copies/mL)
ALT Normal Normal Elevated Elevated
Liver biopsy Inactive Inactive Active Active
Treatment Not Not Recommended Recommended
recommended recommended
79466_CH50.qxd 1/2/08 1:32 PM Page 373
If the patient appears to be in the immune tolerant phase and has normal
transaminases, HBV-DNA 100,000 copies/mL, but has evidence of necroin-
flammation or fibrosis on liver biopsy, antiviral therapy is recommended.
There are data suggesting that high HBV-DNA levels may be associated
with increased risk of hepatocellular carcinoma (HCC) even in patients with no
active inflammation in the liver. This may be a reason to treat patients with
HBV-DNA greater than 10,000 copies/mL. However, this is still controversial
since there is no evidence currently that lowering HCV-DNA levels decreases
the risk of HCC in patients with normal transaminases. A family history of
HCC should lower the threshold for initiation of therapy in these patients.
In the immune reactive phase patients have variable degrees of elevation
of the transaminases, evidence of necroinflammatory changes on liver biopsy, and
often high HBV-DNA levels (105 copies/mL). Patients in this phase are the ones
most likely to benefit from antiviral therapy since, if left untreated, the ongoing
necroinflammatory activity will more likely progress to cirrhosis and HCC.
In the immune reactive phase, most patients are HBeAg positive and HBeAb
negative. However, there is a subgroup of patients who are HBeAg negative and
HBeAb positive despite high levels of HBV-DNA. This type of infection is referred
to HBeAg mutant infection. The mutation in these viruses occurs in the precore
or core promoter region of the HBV genome giving rise to a strain of HBeAg that
is incapable of producing HBeAg despite ongoing viral replication. Evolution of
this strain usually follows a period of active disease. Approximately 30% to 50%
of these patients intermittently have normal transaminases and low HBV-DNA
levels. Patients infected with the HBeAg mutant virus are less likely to transition
to an inactive state and appear to be at a greater risk of chronic liver disease com-
plications even with lower HBV-DNA levels (104 copies/mL). Antiviral therapy
in these patients often results in lowering HBV-DNA and transaminase levels and
results in histologic improvement. However, disease activity returns if antiviral
therapy is stopped. Thus, most of these patients require lifelong therapy.
It is important to differentiate between patients infected with HBeAg
mutant virus from inactive carriers who are also HBeAg negative and HBeAb
positive. Inactive carriers have persistently low HBV levels (104 copies/mL)
and normal transaminases.
Inactive carrier state is characterized by normal transaminases, positive
HBeAb, negative HBeAg, no evidence of inflammatory activity on liver biopsy,
and very low nondetectable levels (104 copies/mL) of HBV-DNA.
In this phase progression of liver damage does not occur and reactivation
is uncommon. These patients do not require antiviral therapy. However, these
patients should be followed clinically every 6–12 months.
Reactivation of viral replication and active HBV hepatitis may occur if the
patients are treated with chemotherapy, corticosteroids, or other agents that
may cause immunosuppression. Antiviral treatment of HBV should be insti-
tuted promptly with active disease.
The goal of therapy for chronic HBV infection ideally is to “cure” the
patient by completely eradicating the HB virus. This is seldom possible.
However, in a minority of patients, conversion from HBsAg to HBsAb may
occur spontaneously or during other antiviral therapies.
Achievable qualitative goals include symptomatic improvement, prevention
of long-term complications of chronic liver disease such as cirrhosis and HCC,
reduction of morbidity and mortality, and prevention of spread of infection of
HBV to others. Quantitatively, the goals of therapy include the suppression or
loss of HBV-DNA, seroconversion from HBeAg positivity to HBeAb positivity,
normalization of transaminases, and histologic improvement on liver biopsy.
a. Interferons (IFN) are endogenous glycoproteins with antiviral, antiprolifera-
tive, and immunoregulatory properties. Interferon-, derived from leuko-
cytes, and interferon-, derived from fibroblasts, are similar and occupy the
same receptor on the target cells. Interferon-, derived from T cells, is differ-
ent and occupies a different receptor. Interferon has properties that have an
79466_CH50.qxd 1/2/08 1:32 PM Page 374
impact on both viral replication and host HLA class I expression. These
effects enhance the recognition of the infected hepatocytes by cytolytic T
lymphocytes. Thus, both the antiviral and the immunoregulatory effects of
interferons are necessary for effective antiviral therapy of chronic hepatitis B.
b. Recombinant interferon- has been used with some success in the treat-
ment of chronic HBV. In several studies, when 5 million units of IFN- were
given subcutaneously (SC) every day or 10 million units three times weekly
for 4 to 6 months, approximately 28% to 35% of patients experienced a
sustained loss of HBV-DNA, compared with none of the untreated, control
patients. The loss of HBV-DNA is noted to occur after 6 to 12 weeks of
interferon therapy and is accompanied by an acute hepatitislike elevation of
aminotransferase activity, which suggests enhanced cytolytic T-cell (CTL)
activity against virus-infected hepatocytes.
Outside of the above studies, treatment of patients with chronic HBV
with recombinant IFN- has not been very successful. The drug and treat-
ment are poorly tolerated by patients and the above results have not been
duplicated.
c. In more recent studies, treatment with Pegasys or PEG-IFN2b (polyethylene
glycol interferon-2b) at a dose of 180 g/week administered subcuta-
neously for 12 months gave more promising results (32% HBeAg serocon-
version to HBeAb and 3% HBsAg conversion to HBsAb). Patients with
HBV genotypes A and B seem to have greater success with PEG-IFN ther-
apy than with the other HBV genotypes. Also, combination therapy or the
addition of a nucleoside (i.e., Lamivudine, to PEG-IFN therapy) did not
increase the therapeutic yield.
d. Side effects of interferon therapy. The side effects of interferon therapy
are systemic flulike symptoms, bone marrow suppression, autoimmune
phenomena (e.g., autoimmune thyroiditis, asymptomatic appearance of
autoantibodies), and infections related to granulocytopenia, and psychiatric
disturbances, especially depression.
e. Nucleoside and nucleotides. Lamivudine is the first oral nucleoside
analog approved by the FDA to treat HBV infection. It works by inhibiting
HBV-DNA replication when HBeAg-positive patients are treated with
lamivudine at a dose of 100 mg/d over 1 year. HBeAb seroconversion is
observed in approximately 15% to 18% of patients. Longer duration of
therapy has been associated with higher HBeAb conversion rates. However,
the treatment with lamivudine has limited efficacy due to the development
of HBV-YMDD mutants that are resistant to the drug. The rate of resis-
tance is 20% at the end of the first year of treatment and goes up to 70%
in 3 or 4 years. Relapses of active hepatitis are common with the develop-
ment of YMDD mutants and/or after discontinuation of therapy. In addi-
tion, the development of lamivudine resistance also negatively impacts the
ability of the virus to respond to newer, more potent antiviral drugs.
Adefovir dipivoxil is an oral nucleotide that inhibits HBV-DNA repli-
cation by suppressing HBV-DNA polymerase. One-year therapy of HBeAg-
positive patients with adefovir 10 mg daily results in 12% seroconversion to
HBeAb, and the seroconversion increases up to 40% after 4 years of therapy.
Determination of early response to therapy (by 6 months) is recommended
since 50% of patients may not respond as indicated by a decrease of HBV-
DNA viral load to 4/log copies/mL.
Resistance to adefovir is less than that toward lamivudine (1.5%–2%
per year); however, it rises to 20% to 25% after 5 years of treatment.
Fortunately, adefovir-resistant mutants are sensitive to lamivudine and ade-
fovir is active against lamivudine-resistant strains of HBV. Resistant patients
are treated with combination therapy. Adefovir as well as lamivudine may
be used effectively in the treatment of HBeAg-negative patients. Monitoring
of renal function is recommended during therapy.
Entecavir is a newer nucleoside analog approved by the FDA for use
in the treatment of both HBeAg-positive and -negative patients. It has more
79466_CH50.qxd 1/2/08 1:32 PM Page 375
potent antiviral activity against HBV than lamivudine and adefovir at doses
of 0.5 mg or 1 mg daily taken orally. Approximately 20% of HBeAg-positive
patients achieve HBe antibody seroconversion by the end of 1 year of treat-
ment. The response increases up to 54% by the end of the second year of
therapy with no emergence of resistance. Histologic improvement has also
been demonstrated in these patients. In patients infected with lamivudine-
resistant HBV mutants, 1-mg daily dose is effective. Unfortunately, up to 7%
of patients infected with HBV resistant to lamivudine may also be cross-
resistant to entecavir as well. Entecavir has no antiviral activity against HIV.
Tenofovir is a nucleotide analog which has been used in HIV treat-
ment. It has been shown to be effective for the treatment of HBV and
lamivudine-resistant HBV. It has a more potent treatment potential than
adefovir with no side effects. It is ideal for treatment of patients coinfected
with both HBV and HIV. However, resistant HBV develops similarly to
Lamivudine after 2 years of therapy.
Newer agents include telbivudine 600 mg daily, an oral nucleoside
analog that has been shown to have superior therapeutic benefit compared
to lamivudine 100 mg daily in HBeAg-positive and -negative patients. So for
no resistance has been seen after 2 years of therapy.
Clevudine and pradefovir are also close to being approved by FDA
for the treatment of chronic HBV infection.
In summary, there is currently no standard therapy for patients with
chronic HBV infection who are in immune reactive phase. In selected
patients, especially those who were infected after early childhood and with
no contraindication for interferon therapy, a one-year course of therapy
with pegylated interferon alfa-2-a may provide a good chance of HBeAg
seroconversion with a finite duration of treatment. This treatment has been
shown to be more effective in patients infected with HBV genotype A.
In others, the most potent oral antiviral agent available with no or low
resistance profile should be selected. In patients who clear HBV-DNA and
achieve seroconversion of HBeAg to HBeAb, 6 to 12 months of further ther-
apy is recommended. For patients with HBeAg-negative chronic infection,
oral therapy with a nucleotide that has the least resistance profile should be
selected since the treatment is expected to be indefinite.
Patients who have developed lamivudine resistance should be treated
with combination therapy of adefovir or tenofovir or switched to entecavir
at 1 mg daily. It is expected that combination therapy will be used in most
patients in the future.
H. Hepatitis B vaccine and prophylaxis
1. Hepatitis B vaccine. HBV infection as acute or chronic disease is a major
public health problem throughout the world. In Asia and Africa, HBV infection
occurs primarily perinatally or in childhood, resulting in a 10% to 15% rate of
chronic infection in the adult population. In the developed countries where the
infection occurs mostly in adults, the chronic infection rate is lower (1%–10%)
but still results in sizable morbidity and mortality.
Hepatitis B vaccine is indicated for immunization against infection caused
by all known subtypes of HBV. Because hepatitis D (caused by the delta virus)
does not occur in the absence of HBV infection, it can be expected that hepati-
tis D will also be prevented by hepatitis B vaccination. Immunization is recom-
mended for people of all ages, especially those who are or will be at increased
risk of exposure to HBV (Table 50-4).
a. The first-generation hepatitis B vaccines are derived from the plasma of
chronic HBsAg carriers who are HBeAg negative. The vaccines are composed
of purified HBsAg and induce protective HBsAb but no antibodies to the other
HBV antigens. They have been available since 1980 and have been considered
both effective and safe. Due to new concerns about the safety of this human
plasma-derived vaccine, however, it is no longer in use in the United States.
b. The second-generation hepatitis B vaccines are manufactured using
other methods (i.e., using HBsAg produced by other cell lines or organisms
79466_CH50.qxd 1/2/08 1:32 PM Page 376
Recombivax HB
1/2/08
Group Formulation 0 mo 1 mo 6 mo
Birth through 10 years of age 10 g/mL 0.25 mL (2.5 g) 0.25 mL (2.5 g) 0.25 mL (2.5 g)
11–19 years of age 10 g/mL 0.5 mL (5 g) 0.5 mL (5 g) 0.5 mL (5 g)
1:32 PM
20 years of age 10 g/ml 1.0 mL (10 g) 1.0 mL (10 g) 1.0 mL (10 g)
Hemodialysis patients 40 g/mL 1.0 mL (40 g) 1.0 mL (40 g) 1.0 mL (40 g)
Infants born to HBsAg-positive mothers 10 g/mL 0.5 mL (5 g) 0.5 mL (5 g) 0.5 mL (5 g)
0.5 mL HBIG
Page 377
Engerix-B
Birth through 10 years of age 20 g/mL 0.5 mL (10 g) 0.5 mL (10 g) 0.5 mL (10 g)
11 years of age 20 g/mL 1.0 mL (20 g) 1.0 mL (20 g) 1.0 mL (20 g)
Hemodialysis patients 20 g/mL 2.0 mL (40 g) 2.0 mL (40 g) 2.0 mL (40 g)
Infants born to HBsAg-positive mothers 20 g/mL 0.5 mL (10 g) 0.5 mL (10 g) 0.5 mL (10 g) 0.5 mL (10 g)
20 g/mL 0.5 mL HBIG
HBIG prophylaxis-postexposure 0.06 mL/kg
377
79466_CH50.qxd 1/2/08 1:32 PM Page 378
V. HEPATITIS D (THE DELTA AGENT). The hepatitis D virus (HDV) was discovered in
Italy by Mario Rizzetto in 1977 during an investigation of the distribution of the HBV
antigens in liver biopsy specimens of patients chronically infected with HBV. He
described a new antigen in the nuclei of infected hepatocytes that was obligatorily asso-
ciated with HBsAg. This new antigen was named delta. Subsequently, large amounts of
this antigen were extracted from the liver of a patient with chronic HBV infection and
the delta agent. Using this antigen, a specific and sensitive radioimmunoassay was
developed for the detection of both the delta antigen in liver biopsy specimens and the
antibody directed at the delta antigen in the serum of infected individuals.
A. Epidemiology. HDV has a global distribution. It has been found to be endemic in the
Mediterranean basin, southern Italy, eastern Europe, the Middle East, Asia, western
Africa, Australia, New Zealand, islands in the South Pacific, and the Amazon basin.
Epidemics of HDV have been reported in North and South America, Colombia, and
Venezuela. In endemic regions, transmission occurs from close person-to-person con-
tact with transmucosal exchange of body fluids. Perinatal transmission is rare. In
nonendemic areas such as northern Europe and North America, transmission is by
direct inoculation from infected blood products. Any population that has a high fre-
quency of HBV infection is vulnerable to HDV infection, especially IV drug abusers,
hemophiliacs, hemodialysis patients, homosexual men, and prisoners.
B. Pathogenesis. HDV is a defective hepatotropic RNA virus that requires the pres-
ence of HBV as a “helper virus” for its pathogenicity. It replicates only in hosts
who have a concomitant HBV infection. HDV is a 36-nm spherical particle con-
taining the D antigen and a single-stranded RNA molecule in the interior, which is
coated by HBsAg on the exterior. The structure of HDV is shown in Figure 50-7.
Figure 50-7. The delta agent. HBsAg, hepatitis B surface antigen; RNA, ribonucleic acid. (From
Hoofnagle JH. Delta hepatitis and the delta agent in the American Association for the Study of Liver
Diseases: Hepatobiliary disease–current concepts and controversies [postgraduate course] 1983.)
79466_CH50.qxd 1/2/08 1:32 PM Page 379
HDV seems to have a direct cytopathic effect on hepatocytes rather than causing
immune-mediated hepatocyte damage, as results from HBV infection.
Individuals who are immune to HBV are protected from HDV infection.
However, persons who are susceptible to both HBV and HDV infection (negative for
all markers of HBV and HDV) and who are carriers of HBsAg may get hepatitis D.
C. Acute coinfection. In patients simultaneously exposed to HBV and HDV, coin-
fection occurs. The disease is usually self-limited, and the clinical course resembles
that of classic acute hepatitis from HBV. Occasionally, coinfection results in exten-
sive liver damage and FHF. This occurs more commonly in IV drug abusers. The
diagnosis of coinfection with HBV and HDV is best made by the detection of
positive titers for IgM HDV antibody, HBsAg, and IgM HBcAb, HBV-DNA and
HDV-RNA (available only in research labs) in the patient’s serum.
The coinfection may manifest itself as a single bout of elevated transaminases
or more frequently as a biphasic illness with two separate transaminase peaks sep-
arated by weeks or even months. Expression of the disease due to HDV must wait
for hepatocytes to be infected first by HBV. If the inoculations with the two viruses
are close in time, HDV overwhelms the synthesis of HBV gene products in the
HBV- and HDV-infected cells and causes the first bout of hepatitis. The second
bout of hepatitis is caused by HBV expression. If the two peaks are very close in
time, massive hepatic necrosis may occur and may be fatal to the patient. If the two
bouts are separated by months, the first bout is usually from HDV. In either event,
both viruses usually are cleared, and the patient completely recovers and develops
both an HDVAb and HBsAb. Figure 50-8 summarizes the course of a case of acute
HDV coinfection.
D. Superinfection. In chronic carriers of HBsAg, HDV superinfection usually causes
severe liver disease, even FHF. In fact, 60% of FHF cases thought to be from HBV
are actually due to HDV superinfection. Because most of the hepatocytes are
already infected by HBV and contain HBsAg, HDV disseminates extensively and
Figure 50-8. Typical course of acute delta hepatitis coinfection. ALT, alanine aminotransferase;
HBsAg, hepatitis B surface antigen; HDV-RNA, hepatitis delta virus ribonucleic acid; anti-HDV,
antibody to HDV; anti-HBs, antibody to HBsAg. (From Hoofnagle JH, DiBisceglie AM. Serologic diag-
nosis of acute and chronic viral hepatitis. Semin Liver Dis. 1991;11:79. Reprinted with permission.)
79466_CH50.qxd 1/2/08 1:32 PM Page 380
causes massive hepatic necrosis. Clinically, the disease may resemble fulminant type
B hepatitis but can be recognized to be D hepatitis superinfection with the serologic
presence of a strong IgM and IgG HDAb response and negative IgM HBcAb.
Anti-HDV arises late during acute HDV and is usually not detectable in serum
at the onset of symptoms. In some cases of self-limited hepatitis D, anti-HDV is not
detectable. For these reasons the serologic diagnosis of acute hepatitis D is often
unsatisfactory and requires testing of both acute- and convalescent-phase serum
samples. HDV antigen is present in both the serum and the liver of infected patients.
Immune blotting for serum HDV antigen is available only in the research setting.
However, the HDV antigen is detectable in the nuclei of hepatocytes and weakly in
the cytoplasm in almost all patients with chronic hepatitis D. It has been detected
in many pathology laboratories by a variety of immunostaining techniques, which
can be applied to formalin-fixed, paraffin-embedded sections.
The hepatitis in these individuals tends to become chronic in almost all cases.
In fact, very few patients clear both HBV and HDV after the superinfection. In
HDV superinfection of HBsAg carriers, patients with negative HBcAg serology are
probably the only ones in whom chronic HDV infection develops.
E. Chronic hepatitis D is characterized by a history of acute hepatitis. If the patient
was a “silent carrier” of HBsAg, this acute illness may be the first episode of
hepatitis for the patient. If the patient had chronic hepatitis from HBV, the disease
may be thought of as an exacerbation or a superimposed hepatitis. Figure 50-9
summarizes the typical course of acute delta hepatitis superinfection.
Unfortunately, chronic HDV infection accelerates the progression of liver dis-
ease both in patients with ongoing chronic liver disease from HBV and in silent
carriers of HBsAg. Most patients develop chronic active hepatitis, which often
progresses to cirrhosis over a short span of time. The detection of high titers of
anti-HDV is strongly suggestive of chronic HDV.
Figure 50-9. Typical course of acute delta hepatitis superinfection. ALT, alanine aminotransferase;
HDV-RNA, hepatitis delta virus ribonucleic acid; HBsAg, hepatitis B surface antigen; anti-HDV, anti-
body to HDV. (From Hoofnagle JH, DiBisceglie AM. Serologic diagnosis of acute and chronic viral
hepatitis. Semin Liver Dis. 1991;11:79. Reprinted with permission.)
79466_CH50.qxd 1/2/08 1:32 PM Page 381
F. Treatment. Currently there is no specific therapy for delta hepatitis, and the treat-
ment for all forms of the disease is supportive. Steroids and other immunosup-
pressive drugs have not been effective. Interferon- 9 MU three times weekly
results in initial response in 50% of patients, but biochemical and virologic
responses are rarely sustained. Monitoring of interferon- therapy in chronic HDV
is best accomplished by sequential testing for HDV-RNA in serum or in the liver.
Currently, the control of delta infection depends on its prevention. Because
HDV infection requires the helper function of HBV, prevention of HBV infection
and the HBsAg carrier state will also prevent HDV infection. Thus, the hepatitis B
vaccine helps prevent new HBV and HDV infections. However, the large number
of people carrying the HBsAg remains at risk for development of delta hepatitis.
A vaccine against HDV is not yet available.
VI. HEPATITIS C
A. Epidemiology. Hepatitis C virus (HCV), identified in 1988 is a single-stranded
enveloped RNA virus that belongs to the Flaviviridae family. It appears to be
responsible for most instances of parenterally transmitted non-A, non-B hepatitis.
The virus seems to mutate frequently and appear in many subtypes. Table 50-6
summarizes the differential diagnosis of elevated liver tests suggesting non-A, non-B
hepatitis other than hepatitis C.
HCV is associated with transfusions of contaminated blood and blood prod-
ucts such as plasma, factor VIII, factor IX, fibrinogen, cryoprecipitate, and immune
globulin prior to 1990. HCV is also transmitted by IV drug abuse, hemodialysis,
and organ transplantation. It appears to be transmitted rarely by familial, sexual,
or maternal–infant exposure. Heterosexual transmission seems to be much less
frequent than homosexual transmission of the virus.
Health care workers exposed to a patient or the blood of a patient infected
with HCV may acquire hepatitis C either from an accidental needle stick or with-
out such an incident; however, the risk in such cases seems to be less than 10%.
This occurrence has been documented in dialysis and oncology units and in
plasmapheresis centers. Sporadic instances of hepatitis C occur and may account
for 6% to 36% of the sporadic cases of hepatitis seen in urban areas. There may
be unnoted percutaneous exposure among such patients (i.e., with use of intranasal
cocaine, tattooing, body piercing, and exposure to blood through sharing tooth-
brushes, razors, or other personal items).
Anti-VCA
antibody to the viral capsid antigen of Epstein-Barr virus.
From Hoofnagle JH, Schafer DF. Serologic markers of hepatitis B virus infection. Semin Liver Dis.
1986;6:1. Reprinted with permission.
79466_CH50.qxd 1/2/08 1:32 PM Page 382
Figure 50-10. A case of acute hepatitis C that progressed to chronic infection and disease. ALT,
alanine aminotransferase; HCV-RNA, hepatitis C virus ribonucleic acid; anti-HCV, antibody to hepati-
tis C virus; PCR, polymerase chain reaction. (From Hoofnagle JH, DiBisceglie AM. Serologic diag-
nosis of acute and chronic renal hepatitis. Semin Liver Dis. 1991;11:78. Reprinted with permission.)
be vaccinated against HAV and HBV because of the high risk of severe liver disease
if superinfection occurs with these viruses.
1. Monotherapy. Treatment of chronic HCV infection is based on parenteral
interferon (IFN) therapy. Initially, the treatment consisted of IFN monotherapy
using 3 million units given subcutaneously three times per week for 6 months
and yielded about 10% end of treatment response (ETR). However, with this
therapy, there is a 90 % relapse rate 6 months after cessation of therapy, giv-
ing a sustained viral response (SVR) of about 3%. When this treatment was
extended to 12 months, the SVR increased to about 10%.
2. Combination therapy
a. Combination therapy using IFN- with ribavirin, an oral guanosine nucleoside
analog, has been shown to greatly improve sustained response rates in naive
and relapsing patients. In treatment of naive patients with IFN- at a dose of 3
million units given sc three times a week, in combination with oral ribavirin 800
to 1,200 mg per day for 12 months, achieves sustained virologic response rates
of approximately 30%. These rates are higher (up to 40%–50%) for patients
with low viral loads and with HCV genotypes II and III. In patients with HCV
genotypes II and III, combination therapy for 24 weeks achieves the same effi-
cacy as combination therapy administered for 48 weeks. The impact of combi-
nation therapy on interferon monotherapy nonresponders is less impressive.
b. The current standard of therapy for a patient with chronic HVC is a
combination of pegylated interferon and ribavirin.
The addition of polyethylene glycol (PEG) to interferon increases the
half-life of the drug in the patient, thereby allowing once-weekly adminis-
tration. When pegylated IFN therapy is combined with oral ribavirin, the
rate of SVR is greatly increased especially in patients with HCV genotype 1.
The optimal dose of PEG interferon--2b (PEG-IFN) is 1.5 g/kg per week
sc and for PEG interferon--2a (Pegasys) is 180 g per week sc. The pre-
ferred ribavirin dose is weight-based (10.6 mg/kg) for PEG-IFN and with
Pegasys it is 1,000 to 1,200 mg per day.
For naive patients with HCV genotype 1 infection, treatment is recom-
mended for 48 weeks. Early viral response is assessed by quantitative deter-
mination of HCV-RNA by PCR (viral load) at the end of the first month. In
patients who have undetectable viral load after one month of therapy, the
SVR is 90%.
For patients with undetectable viral load at 3 months of therapy or
those with a 2 log drop in the viral load, the SVR is in the range of 48%.
Patients with HCV genotypes 2 and 3 with the same treatment plan
have SVR 80% after treatment.
Patients who fail to respond to above therapy or those who relapse after
treatment may be offered therapies involving longer duration of therapy, higher
ribavirin and interferon doses, and/or treatment with consensus interferon
(Infergen) which is a synthetic interferon at a dose of 15 g sc daily and
ribovirim for 48 weeks. These treatments are still under study and are evolving.
c. There are numerous side effects of interferon and ribavirin therapy. Aside
from the flulike symptoms, fatigue, and depression, interferon causes
cytopenia and autoimmune thyroiditis. Ribavirin causes hemolytic anemia
and psychiatric and mood disorders and insomnia. It is paramount that
these side effects are treated promptly and effectively for patients’ comfort
and to ensure patient compliance. Growth factors, especially erythropoietin
is being used in many patients to alleviate anemia. Patients with mood dis-
orders or changes need to be treated with appropriate drugs and should be
followed by mental health professionals as needed.
Patients with HIV and HCV coinfection should be treated for HCV as
well as HIV.
Since ribavirin is teratogenic, patients who are taking ribavirin and are
sexually active need to use barrier contraception during the treatment and
6 months posttherapy.
79466_CH50.qxd 1/2/08 1:32 PM Page 385
VII. The hepatitis E virus (HEV) is a small (27–30 nm) RNA virus that causes acute epi-
demics of enterically transmitted hepatitis in underdeveloped areas of the world. HEV
is transmitted by the fecal-oral route and is associated with contamination of food or
water sources.
A. Diagnosis. HEV antigen can be detected in the liver, bile, and stool during the
incubation period and symptomatic phase of hepatitis E. Anti-HEV can be detected
by immune electron microscopy in serum during acute illness. Titers of anti-HEV
rise during convalescence; however, immunity to reinfection may not be absolute.
HEV-RNA has been cloned. Recombinant immunoassays are being developed for
both anti-HEV and HEV antigen. Diagnosis of acute HEV can be made in patients
presenting with acute hepatitis with a recent history of living in or travel to
endemic areas of the world (Central and South America, Mexico, Asia, Africa, and
the Middle East) and absence of serologic markers of acute hepatitis A, B, and C.
B. Clinical course. HEV causes an acute hepatitis in adults with varying severity. In
pregnant women, especially in the third trimester, it is associated with high
mortality—approximately 10% to 35% of instances. HEV infection does not lead
to chronic hepatitis or to the carrier state.
Selected Readings
Berg T, et al. Extended treatment duration for hepatitis C virus type 1: comparing 48 versus 72 weeks
of peginterferon-alpha-2a plus riboviron. Gastroenterology. 2006;130:1086–1097.
Chen CJ, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B DNA
level. JAMA. 2006;295(1):65–73.
Derenci P. Predictors of response to therapy for chronic hepatitis C. Semin Liver Dis. 2004;24(suppl
2):25–31.
Fartoux L, et al. Effect of prolonged interferon therapy on the outcome of hepatitis C virus-related
cirrhosis: A randomized trial. Clin Gastroenterol Hepatol. 2007;Apr;5:502–507.
Hoofnagle JH, et al. Management of hepatitis B: summary of a clinical research workshop. Hepatology.
2007;45:1056–1075.
Krawczynski K, et al. Hepatitis E virus: Epidemiology, clinical and pathological features, diagnosis and
experimental animal models. In: Thomas HC, Lemon S, Zuckerman AJ, eds. Viral Hepatitis. 3rd ed.
Malden, Mass.: Blackwell; 2005:624–634.
Lok ASF. Navigating hepatitis B treatments. Gastroenterol. 2007;132:1586–1594.
Lok ASF and McMahon BJ. AASLD practice guidelines: Chronic hepatitis B. Hepatology. 2007;
45:507–539.
McLean OR, et al. Acute Hepatitis C: Diagnosis and management. Pract Gastroenterol. 2007;(Sept)66–77.
Niro GA, et al. Peglyated interferon alpha-2b as monotherapy or in combination with ribovirin in
chronic hepatitis delta. Hepatology. 2006;44:713–720.
Oldfield EC, et al. The A’s and B’s of vaccine—Preventable hepatitis: Improving prevention in high risk
adults. Rev Gastroenterol Disord. 2007;7(1):1–21.
Pungpapong S, et al. Natural history of hepatitis B virus infection: an update for clinicians. Mayo Clin
Proc. 2007;82(8):967–975.
Re VL, et al. Prevalence, risk factors, and outcomes for occult hepatitis B virus infection among HIV-
infected patients. J Acquir Immune Defic Syndr. 2007;44:315–320.
Sarazin C, et al. A normal hepatitis C virus protense inhibitor, plus pegylated interferon alpha-2b for
genotype 1 nonresponders. Gastroenterology. 2007;132:1270–1278.
Shiffman ML, et al. Peginterferon Alpha-2a and Riborvirin for 16 or 24 weeks in HCV genetype 2 or 3.
N Eng J Med. 2007;357:124–134.
Shresthas MR, et al. Safety and efficacy of a recombinant hepatitis E vaccine. N Engl J Med.
2007;356:895–903.
Sorjano V, et al. Core of patients with HIV and Hepatitis C virus. AIDs. 2007;21(9):1073–1089.
Zeuzem S, et al. Efficacy of 24 week treatment with peginterferon alpha—2b plus ribovirin in patients
with chronic hepatitis C infected with genotype 1 and low pretreatment virimia. J Hepatol.
2006;44:97–103.
79466_CH51.qxd 1/2/08 1:32 PM Page 386
I. EPIDEMIOLOGY. The association of alcohol abuse and liver damage has been known
since the time of the ancient Greeks. The availability of alcoholic beverages, licensing
laws, and economic, cultural, and environmental conditions all influence both per
capita alcohol consumption and mortality from alcohol-related liver disease.
Alcoholism is, in part, inherited, and aberrant alcohol-drinking behavior is genetically
influenced. The risk factors that may affect the susceptibility to development of alco-
holic liver disease include genetic factors, malnutrition, female gender, and viral agents
(hepatitis virus B, C, and D).
Ninety percent to 95% of people with chronic alcohol consumption develop
fatty liver. In almost all instances, this lesion is thought to be reversible on cessation
of alcohol intake. Ten percent to 30% of individuals go on to perivenular sclerosis
(collagen deposition in and around central veins). Ten percent to 35% of chronic alco-
holics, however, have acute liver injury that may become recurrent or chronic. Some
of these patients recover, but 8% to 20% go on to sinusoidal, perivenular, and peri-
central fibrosis and cirrhosis. Even during the inflammatory stage without the pres-
ence of cirrhosis, patients may have portal hypertension, ascites, and esophageal
varices.
386
79466_CH51.qxd 1/2/08 1:32 PM Page 387
In an obese person, therefore, the same amount of alcohol per unit of weight gives
a higher blood alcohol concentration than in a thin person. The mean volume of
distribution of ethanol is less in women than in men, resulting in higher peak
blood concentrations and greater mean areas under the ethanol concentration–
time curves.
In humans, less than 1% is excreted in the urine, 1% to 3% via the lungs,
and 90% to 95% as carbon dioxide after it is oxidized in the liver.
B. Chemical metabolism
1. Alcohol dehydrogenase. Although most of the ingested ethanol is metabo-
lized by the liver, other tissues such as the stomach, intestines, kidney, and
bone marrow cells oxidize ethanol to a small extent. There is an alcohol
dehydrogenase (ADH) present in the mucosa of the stomach, jejunum, and
ileum, which results in a considerable first-pass metabolism of alcohol. The
gastric ADH activity is less in women than in men and decreases with chronic
alcoholism.
In the liver, the main pathway for ethanol metabolism is by its oxidation
to acetaldehyde by ADH. Alternative pathways of oxidation in other subcellu-
lar compartments are also present. Multiple molecular forms of ADH exist,
and at least three different classes have been described on the basis of structure
and function. Various ADH forms appear in different frequencies in different
racial populations. This polymorphism may explain, in part, individual varia-
tion in the rate of acetaldehyde production and first-pass elimination.
The hepatic metabolism of ethanol proceeds in three basic steps. First,
ethanol is oxidized within the hepatocyte cytosol to acetaldehyde. Second,
acetaldehyde is oxidized to acetate via catalysis mainly by aldehyde dehydroge-
nase (ALDH) in the mitochondria. Third, acetate ALDH is released into blood
and is oxidized by peripheral tissues to carbon dioxide and water.
79466_CH51.qxd 1/2/08 1:32 PM Page 388
Figure 51-1. Metabolic consequences of change in hepatic redox state secondary to oxidation
of ethanol.
79466_CH51.qxd 1/2/08 1:32 PM Page 389
In healthy people, almost all the acetaldehyde formed during ethanol oxi-
dation is effectively oxidized in the liver. However, detectable concentrations of
acetaldehyde have been found in the peripheral blood of chronic alcoholics and
in Asians, who experience alcohol-related flushing. The unusually high blood
acetaldehyde levels found in severely intoxicated, chronic alcoholics are thought
to result from an increased rate of ethanol oxidation of ADH and a decrease in
the liver ALDH activity associated with both liver injury and chronic ethanol
consumption.
V. DIAGNOSIS
A. Clinical presentation. Alcoholic hepatitis (AH) is an acute or chronic illness
associated with severe chronic alcoholism, involving extensive hepatocellular
necrosis, inflammation, and scarring. It is extremely variable in clinical presenta-
tion and may be superimposed on other forms of alcoholic liver disease such as
fatty liver and Laënnec’s cirrhosis.
Alcoholic liver disease may occur without any symptoms. The most common
complaints in symptomatic patients with alcoholic hepatitis are anorexia, nausea,
vomiting, and abdominal pain, especially in the right upper quadrant. Most
patients lose weight due to anorexia and nausea. One fourth of patients have fever.
Fever may be due to severe liver damage in the absence of an infection; however,
infections should be excluded in these immunocompromised patients. Jaundice,
when present, is usually mild, but in 20% to 35% of the patients who present with
a cholestatic picture, it is severe. Diarrhea and symptoms related to complications
of portal hypertension such as ascites, peripheral edema, and esophageal varices
are less common. Most patients with acute hepatitis have a tender and somewhat
enlarged liver. Jaundice, splenomegaly, ascites, peripheral edema, spider angiomata,
palmar erythema, parotid enlargement, gynecomastia, testicular atrophy, finger
clubbing, and Dupuytren’s contractures may also be present.
B. Diagnostic studies
1. Laboratory studies
a. The most characteristic laboratory findings in acute alcoholic hepatitis are
elevated serum transaminases and bilirubin. Serum glutamic-oxaloacetic
transaminase (SGOT, or aspartate aminotransferase [AST]) levels are usually
2 to 10 times normal. Values greater than 500 U/L are rare. Serum glutamic-
pyruvic transaminase (SGPT, or alanine aminotransferase [ALT]) levels are
also elevated, but they are increased less than SGOT levels. The reduced
SGPT levels may be related to mitochondrial injury in alcoholic hepatitis.
The SGOT/SGPT ratio in alcoholic hepatitis is usually greater than 1.
b. Serum alkaline phosphatase level is moderately elevated in at least one
half of the patients. In patients presenting with a cholestatic picture, the
elevation may be marked. -Glutamyltranspeptidase (GGTP) activity is
also often elevated in alcoholic hepatitis. This enzyme is quite sensitive for
alcoholic injury.
79466_CH51.qxd 1/2/08 1:32 PM Page 391
c. Serum globulins are often increased, and, interestingly, IgA levels are
disproportionately elevated compared to the other immunoglobulins. Serum
albumin levels may be normal initially in well-compensated and well-nourished
patients but are usually decreased.
d. Because the liver is the main site of synthesis of the coagulation factors, in
severely ill patients the prolongation of prothrombin time by greater than
7 to 10 seconds predicts a poor prognosis.
e. The hematologic parameters are usually abnormal in alcoholic hepatitis.
Leukocytosis with a shift to the left is common. Anemia and thrombocy-
topenia are usually present. Platelet count of 100 usually indicates the
presence of cirrhosis. Aside from the fact that alcohol is toxic to the bone
marrow, patients may have hypersplenism, disseminated intravascular coag-
ulation (DIC), and bone marrow suppression from their acute and chronic
disease state.
f. Other causes of liver injury should be ruled out by appropriate serologic
tests and a careful drug history. Acetaminophen even in therapeutic doses
may be toxic in alcoholic patients. Aspirin and acetaminophen levels may be
helpful in selected patients.
g. Hepatitis C infection increases the risk of developing cirrhosis in ALD.
This is also true in reverse. Thus, patients with ALD should be tested for
hepatitis C antibody and if the test is positive, treatment of hepatitis C
should be recommended.
2. Ultrasound. Even though the diagnosis may be easily established in most
patients from the history, physical examination, and blood work, it may be
helpful in some patients, especially in those presenting with the picture of
cholestasis, to obtain an abdominal ultrasound to rule out biliary tract disease,
ascites, and the presence of hepatocellular carcinoma. Some patients may
require computed tomography (CT) scan of the abdomen and endoscopic ret-
rograde cholangiopancreatography (ERCP).
3. Liver biopsy should be performed early in the course of the disease in most
patients to confirm the diagnosis and to stage the severity of the liver injury.
With deterioration of liver function (excessive prolongation of the prothrombin
time or the presence of ascites), liver biopsy may not be possible later in the
disease process.
Microscopic examination of the liver biopsy is essential for classification
of the alcoholic injury with regard to steatosis, early fibrosis, alcoholic hepati-
tis, or Laënnec’s cirrhosis. In patients with alcoholic hepatitis, pathologic fea-
tures may include the following:
a. The hepatocellular necrosis is mainly centrilobular but may be panlobu-
lar. Hepatocytes in various stages of degeneration are usually present. Many
hepatocytes are swollen, balloonlike, or vacuolated and often contain a fib-
rillar material or alcoholic hyalin called Mallory bodies. This cytoplasmic
material, even though commonly seen in patients with alcoholic hepatitis, is
not limited to this disease. It has also been reported in diseases such as
hepatoma, Indian childhood cirrhosis, and liver disease in patients with
intestinal bypass surgery, obesity, diabetes, and drug reaction. Occasionally
it is seen in primary biliary cirrhosis, primary sclerosing cholangitis, and
Wilson’s disease.
b. The inflammatory exudate consists predominantly of polymorphonuclear
(PMN) leukocytes in addition to some mononuclear cells. PMNs are seen in
portal tracts and sinusoids and are usually concentrated in areas of hepato-
cellular necrosis with or without alcoholic hyalin. In 50% to 75% of the
cases, rosettes of PMNs are seen around degenerating hepatocytes contain-
ing Mallory bodies. This finding constitutes the hallmark of the disease.
c. Fibrosis is usually extensive and seen early in the disease. It frequently sur-
rounds the central veins and extends into the perisinusoidal areas. The
stellate fibrosis often remains after inflammation resolves and leads to
micronodular cirrhosis.
79466_CH51.qxd 1/2/08 1:32 PM Page 392
VI. CLINICAL COURSE. In patients with steatosis only, abstinence from alcohol may lead
to resolution of the liver injury. However, if perivenular and sinusoidal fibrosis is pre-
sent, abstinence may halt further damage, but the scarring may remain.
Alcoholic hepatitis is a disease of varying severity. The overall mortality is 10%
to 15%. Healing is slow, and most patients require 2 to 8 months for recovery after
stopping consumption of alcohol. The disease usually worsens after the first several
days, regardless of the initial presentation. Patients may have encephalopathy and
other complications during this stage, such as gastrointestinal bleeding from
esophageal varices, gastritis or peptic ulcers, coagulopathy or DIC, infections espe-
cially of the urinary tract and lungs, spontaneous bacterial peritonitis, and hepatore-
nal syndrome. A combination of serious complications often leads to death of the
patient.
Many patients with a history of chronic alcoholism have episodes of acute alco-
holic hepatitis superimposed on existing cirrhosis. It is thought that repeated bouts of
alcoholic liver injury, necrosis, and fibrosis lead to extensive scarring and Laënnec’s
cirrhosis.
VII. TREATMENT. The most important component of therapy of alcoholic liver disease
is strict abstinence from alcohol. Alcohol is the most common cause of liver dis-
ease in the Western world. In many countries, the incidence of alcoholic liver disease
is increasing at a time when the incidences of other liver disorders are remaining
steady or declining.
A. Supportive therapy. The mainstay of therapy for alcoholic hepatitis is support-
ive. Most patients require bed rest. Seriously ill patients should be admitted to an
intensive care unit. If complications of portal hypertension exist, they should be
promptly treated. Infections and concomitant pancreatitis should be ruled out or
appropriately treated.
B. Diet. Alcohol intake should be strictly forbidden. Patients should be given
thiamine, folic acid, and multivitamins. The electrolyte, calcium, magnesium,
phosphate, and glucose levels should be strictly monitored and corrected.
Protein-calorie malnutrition is present in most patients. The energy expen-
diture and protein degradation are increased. Insulin resistance is a ubiquitous
finding. This causes fat to be preferred over carbohydrate as an energy source.
Frequent feedings with a high-carbohydrate, high-fiber, and low-fat diet may
decrease insulin resistance and improve nitrogen balance. In patients who cannot
eat, enteral tube feedings may be given with standard formulas. Special formula-
tions such as Hepatic-Aid may be reserved for the small minority of patients for
whom encephalopathy remains a major problem after a fair trial of standard ther-
apy. Patients should receive sufficient daily calories to provide basal energy expen-
diture (BEE) 1.2 to 1.4 total kcal or 25 to 30 kcal/ kg of ideal body weight.
Thirty percent to 35% of total calories should be given as fat and the remainder as
carbohydrate. Protein should not be restricted below 1 mg/kg unless encephalopa-
thy necessitates further restriction.
In patients in whom enteral therapy is not possible, parenteral nutrition
should be given promptly following guidelines similar to those outlined for enteral
nutrition. The use of solutions rich in branched-chain amino acids such as
HepatAmine is controversial and is not necessary except in cases of hepatic
encephalopathy and coma.
C. Alcohol withdrawal. Initially, symptoms of alcohol withdrawal may be present.
Alcoholic seizures and delirium tremens should be prevented by strict attention to
79466_CH51.qxd 1/2/08 1:32 PM Page 393
Selected Readings
Bridle K, et al. Hepcidin is down regulated in alcoholic liver injury: implications for the
pathogenesis of alcoholic liver disease. Alcohol Clin Exp Res. 2006;30:106–112.
Clouston AD, et al. Steatosis as cofactor in other liver diseases: Hepatitis C virus, alcohol,
hemochromatosis, and others. Clin Liver Dis. 2007;11(1):173–189.
Foody W, et al. Nutritional therapy for alcoholic hepatitis: New life for an old idea.
Gastroenterology. 2001;120:1053.
Kapadia C. Alcoholic steatosis: The Kupfer cell—A villain? Gastroenterology. 2001;120:581.
Purohit V, et al. Role of S-adenosylmethionine, folate, and betaine in the treatment of
alcoholic liver disease: summary of a symposium. Am J Clin Nutr. 2007;86(1):14–24.
Tilg H, et al. Cytokines in alcoholic and nonalcoholic steatohepatitis. N Engl J Med.
2000;343:1467.
Yipw W, et al. Alcoholic liver disease. Semin Diagn Pathol. 2007;23:149–160.
79466_CH52.qxd 1/2/08 1:33 PM Page 395
II. ASSOCIATED CONDITIONS. Conditions associated with NAFLD include Type 2 dia-
betes mellitus, obesity, and dyslipidemia, all of which are related to metabolic syndrome.
A. Obesity. Obesity is the condition most often reported to be associated with
NAFLD. Obesity is described in 40% to 100% of patients with NASH. NASH has
been documented in 9% to 36% of obese patients. The prevalence and severity of
hepatic steatosis has been noted to be directly proportional to the grade of obesity,
and the severity of NASH has been noted to be proportional to the degree of hepatic
steatosis. The distribution of body fat seems to be important in the development of
hepatic steatosis. In one study, a significant correlation was found between the
degree of hepatic steatosis and waist-to-hip ratio, suggesting a relationship between
visceral and intraabdominal fat and accumulation of fat in the liver.
B. Hyperglycemia, insulin resistance, hyperinsulinemia, glucose intolerance,
and Type 2 diabetes mellitus have been described in 25% to 75% of adult patients
with NASH.
C. Hyperlipidemias, including hypertriglyceridemia and hypercholesterolemia or
both, have been found to be present in 20% to 80% of patients with NASH.
D. Most patients with NASH seem to have multiple risk factors including obesity,
Type 2 diabetes mellitus, and hyperlipidemia.
E. Other risk factors include female gender, rapid weight loss, acute starvation, total
parenteral nutrition, and small-bowel diverticulosis.
395
79466_CH52.qxd 1/2/08 1:33 PM Page 396
III. DIAGNOSIS
A. Clinical findings
1. Symptoms. Most patients with NAFLD are asymptomatic; however, fatigue,
malaise, or vague right upper quadrant pain or discomfort may cause patients
to seek medical attention.
2. Signs. Hepatomegaly may occur in up to 75% of patients. Splenomegaly may
be present in 25% of patients. Stigmata of portal hypertension and cirrhosis
occur rarely. In a subset of patients with NAFLD, especially in association with
certain drugs and toxins (i.e., certain nucleoside analogs, antimitotic agents, or
tetracycline) fulminant hepatic failure may develop rapidly. Cirrhosis may
develop rapidly in patients with steatosis and inborn errors of metabolism (i.e.,
tyrosinemia).
3. Laboratory findings. Laboratory features of NAFLD are nondiagnostic. Mild-
to-moderate increases in serum aminotransferases (alanine [ALT] and aspartate
[AST]) are the predominant laboratory abnormalties reported in patients with
NASH. Typically, the liver test abnormalities are noted during routine testing or
when patients seek medical attention for other complaints. However, liver
enzymes are not always sensitive markers for NASH because some patients
present with normal transaminases in the presence of severe liver disease. Also,
there is no significant correlation between the degree of serum aminotrans-
ferase elevation and the histologic findings and severity of the hepatic inflam-
mation or fibrosis. Serum ALT levels are often higher then AST levels, in
contrast to the pattern seen in alcoholic hepatitis in which the AST/ALT ratio is
higher and serum alkaline phosphatase levels may be mildly elevated; serum
bilirubin and albumin levels are usually normal. Prolongation of prothrombin
time (PT) is suggestive of hepatic decompensation. A small percentage of
patients with NASH may have positive low-titer antinuclear antibody (ANA)
levels. However, antimitochondrial antibody (AMA), antibody to hepatitis C
virus, and hepatitis B surface antigen are negative, and serum ceruloplasmin
and antitrypsin levels are normal. Elevated serum ferritin and transferrin satu-
ration seem to occur commonly in patients with NASH. Men with NASH may
have higher iron stores than women. One third of patients with NASH may
have one or two copies of the Cys282Tyr mutation in the HFE gene (genetic
marker for hemochromatosis). Trends toward more severe hepatic fibrosis in
patients with NASH with this gene mutation have been reported.
4. Imaging modalities. Several noninvasive imaging techniques including abdom-
inal ultrasound (US), computerized tomography (CT), and magnetic resonance
imaging (MRI) can reveal hepatic steatosis. None of the tests are significantly
sensitive for the detection of hepatic inflammation or fibrosis. CT and MRI
may be helpful only when extrahepatic manifestations of cirrhosis and portal
hypertension (i.e., ascites, splenomegaly, varices) are present. Thus, none of
these imaging modalities is sensitive or specific enough to definitively establish
the diagnosis of NASH or reliably grade its severity.
5. Liver biopsy is the only diagnostic test for confirming the clinical suspicion of
NAFLD and NASH, staging the severity of the liver disease, and determining
79466_CH52.qxd 1/2/08 1:33 PM Page 397
VI. TREATMENT. Currently, there is no effective therapy for NAFLD. Available treatment
modalities attempt to eliminate the factors that are commonly associated with
NAFLD. Weight loss, treatment of hyperglycemia, hyperlipidemia, and discontinua-
tion of alcohol and potentially toxic drugs such as corticosteroids, estrogens, amio-
darone, and perhexiline are recommended. Treatment for morbid obesity (i.e.,
bariatric surgery) should be considered. In small, short-term studies, the use of
ursodeoxycholic acid, vitamin E, gemfibrozil, betaine (a metabolite of choline),
N-acetylcysteine, and metformin resulted in improvement in liver chemistry tests and
hepatic steatosis, but no significant change in histological grade of inflammation or
fibrosis has been observed.
Thiazolidinediones (TZDs) (i.e., pioglitazone [Actos] and rosiglitazone
[Avandia]) is a class are drugs known as “insulin sensitizers” since they allow the adi-
pose and muscle cells to become more sensitive to insulin and allow better uptake of
serum glucose into these cells. In ongoing studies, TZDs show promise in reversing
NASH, especially in the diabetic patients. (Avandia has been removed from the U.S.
market for increasing cardiac risk factors.)
Newer treatment modalities using anticytokine (e.g., TNFa antibodies, antioxi-
dants, glutathione prodrugs, and insulin sensitizers) are being evaluated. Patients with
NAFLD who develop hepatic decompensation should be evaluated for liver transplan-
tation. In a number of patients, NASH has been observed to recur after successful liver
transplantation.
Selected Readings
Day CP. From fat to inflammation. Gastroenterology. 2006;130:207–210.
Ekstedt M, et al. Long-term followup of patients with NAFLD and elevated liver enzymes.
Hepatology. 2006;44:865–873.
Farrell GC, et al. Nonalcoholic fatty liver disease: from steatosis to cirrhosis. Hepatol.
2006;43(2 suppl I):S99–S112.
Gholam PM, et al. Nonalcoholic fatty liver disease in severely obese subjects. Am J
Gastroenteraol. 2007 Feb;102:399–408.
Hubscher SG. Histological assessment of nonalcoholic fatty liver disease. Histopathology.
2006;49:450–465.
McClaine CJ, et al. Good fat/bad fat. Hepatology. 2007;45:1343–1346.
79466_CH53.qxd 1/2/08 1:33 PM Page 399
M ore than 1,100 drugs and herbal remedies have been reported to cause liver injury.
Drug-induced liver injury can mimic almost all patterns of liver injury seen in humans.
Among causes of acute liver failure, drug-induced liver injury accounts for more than 50%
of them. The injury can be hepatocellular necrosis, cholestatic disease, deposition of
microvesicular fat in hepatocytes, or mixed patterns. In most cases, the damage is caused by
toxic metabolites of the drug or immune response to the drug or its metabolites. It may result
in chronic hepatitis and cirrhosis. Antimicrobials, psychotropic drugs, lipid-lowering agents
and nonsteroidal antiinflammatory drugs (NSAIDS), phenytoin, propylthiouracil antituber-
culosis drugs, sulfa compounds, and disulfirams are implicated most often. Table 53-1 is a
partial list of drugs for which liver chemistry test monitoring is recommended. For the major-
ity of drugs the risk of drug-induced liver injury is in the range of 1:104–105 in the United
States. The risk is greater in women and the elderly, and the risk varies depending on the
nutritional and other susceptibility characteristics of the patients and the drug class.
399
79466_CH53.qxd 1/2/08 1:33 PM Page 400
III. MIXED-PATTERN LIVER INJURY. In most cases, drug-induced liver injury causes a
combination of cholestasis and hepatocyte necrosis. Patients usually have moderate
elevation of the serum transaminases, bilirubin, and alkaline phosphatase levels. Most
of these reactions are the result of hypersensitivity to the drug and affect only a few
susceptible individuals.
A. Phenytoin (Dilantin) toxicity resembles viral mononucleosis. Patients present
with fever, lymphadenopathy, and a tender liver. The liver biopsy shows portal
lymphocytic infiltration and spotty necrosis of parenchymal cells.
B. Drugs such as quinidine, allopurinol, nitrofurantoin, diltiazem, and many others
cause a granulomatous reaction with some hepatocytic necrosis.
C. For a detailed list of drugs and their liver toxicity see the article by Lewis in the
Selected Readings.
V. ASPIRIN HEPATOTOXICITY. Aspirin (ASA) and other salicylates have been noted to
cause liver injury in patients with rheumatic and collagen vascular diseases such as
juvenile and adult rheumatoid arthritis, rheumatic fever, and systemic lupus erythe-
matosus. Normal subjects and patients with nonrheumatic diseases such as orthopedic
problems may also be affected.
A. Pathogenesis. The blood level of the drug (5 mg/dL) and the duration of intake
(6 days to weeks) seem to play an important role in the production of liver dam-
age. The injury appears to be a cumulative phenomenon, appearing after many
days of intake of large therapeutic doses. A single, toxic overdose of aspirin pro-
duces little or no hepatic injury.
Patients with rheumatic and collagen vascular disorders may be more sus-
ceptible to liver injury with aspirin than others. This may be due to the presence
of hypoalbuminemia allowing higher serum levels of unbound aspirin, underlying
liver damage, and possibly altered metabolism of the salicylates. The mechanism
of hepatic injury seems to be intrinsic toxicity of the salicylate moiety rather than
host idiosyncrasy to the drug. Salicylate choline and sodium salicylate also can
induce hepatic injury. The liver disease is usually mild, acute, and reversible.
Lowering the dose of ASA without totally discontinuing its use may be sufficient
to reverse the injury. There is strong evidence to suggest that aspirin in the setting
of a viral infection may provoke the development of Reye’s syndrome in children.
1. Clinical presentation. Clinical manifestations of hepatic injury are not promi-
nent. Most patients remain asymptomatic. Some patients complain of anorexia,
nausea, and mild abdominal distress. Almost all patients are anicteric.
The liver disease is usually mild. However, encephalopathy, severe coagu-
lopathy, and fatal liver failure have been reported. There is no established
evidence that aspirin causes chronic liver injury.
2. Diagnostic studies. Serum transaminases are usually moderately elevated (2–10
times normal). In 10% of the individuals, SGOT levels are greater than 100
IU/mL. The SGOT (AST) levels are usually higher than the SGPT (ALT) levels.
The alkaline phosphatase levels are usually normal or only modestly elevated.
Serum bilirubin levels have been elevated only in about 3% of the reported cases.
3. Treatment is supportive. In most instances, it is not necessary to discontinue
the drug for reversal of liver injury. Decreasing the dosage to attain blood lev-
els less than 15 mg/dL seems to be sufficient.
Figure 53-1. Major metabolic pathways of acetaminophen. (From Nelson SD. Molecular mecha-
nisms of the hepatotoxicity caused by acetaminophen. Semin Liver Dis. 1990;10:268. Reprinted with
permission.)
79466_CH53.qxd 1/2/08 1:33 PM Page 403
Figure 53-2. Semilogarithmic plot of plasma acetaminophen levels versus time. (From Rumack
BH, Matthew H. Acetaminophen poisoning and toxicity. Pediatrics. 1975;55:871. Reprinted with
permission.)
VII. DRUG-INDUCED CHRONIC HEPATITIS. Drugs that have been documented to cause
chronic hepatitis (CH) include oxyphenisatin, alpha-methyldopa, nitrofurantoin,
dantrolene, isoniazid, propylthiouracil, sulfonamides, and halothane. The incidence of
CH with any one drug is low, and the total number of cases is small. However, a drug
history should be obtained from each patient suspected to have CH. In most cases,
if drug-induced CH has not developed into cirrhosis, it improves or resolves with
discontinuation of the drug.
A. Methyldopa (Aldomet). The incidence of hepatitis with methyldopa is very low;
however, if the hepatitis is not noted in early stages, the disease may progress to
chronic active hepatitis (CAH). Hepatitis occurs within weeks after starting the
drug, suggesting a role for hypersensitivity. If the lesion is recognized early on, the
injury and inflammation regress with discontinuation of the drug.
B. Oxyphenisatin. This laxative, although taken off the market in the United States,
is still in popular use, especially among women in Europe and South America.
Oxyphenisatin has been shown to cause acute and chronic hepatitis resembling
“lupoid hepatitis.” The disease may progress to cirrhosis if the drug is continued
after the onset of liver injury. In most instances, however, after the withdrawal of
the drug, the disease is halted and even reversed.
C. Isoniazid (INH). Isoniazid produces asymptomatic elevation of serum transami-
nases with mild liver injury in up to 20% of the patients within the first 2 to
3 months of therapy. However, approximately 1% of the individuals develop
severe hepatic injury, even fulminant hepatic failure, associated with a high mor-
tality. Chronic hepatitis generally does not develop if the drug is stopped within the
first few weeks of recognition of the hepatitis. However, severe CAH accompanied
by cirrhosis develops with persistent administration of INH.
1. Pathogenesis. INH-induced liver injury is thought to result from hepatotoxic
reactive intermediates of INH metabolism. INH is first acetylated and then con-
verted to acetylphenylhydrazine, which is a potent hepatotoxin. There are some
79466_CH53.qxd 1/2/08 1:33 PM Page 406
data to suggest that rapid acetylators (e.g., most East Asians) are more suscep-
tible to INH-induced liver injury.
2. Clinical presentation. The clinical features of hepatitis induced by INH are
nonspecific and resemble those of viral hepatitis. Fatigue, malaise, anorexia,
nausea, vomiting, and abdominal discomfort are commonly noted. Jaundice is
the presenting complaint in 10% of the patients. Signs and symptoms of hyper-
sensitivity, rash, lymphadenopathy, arthralgia, and arthritis are rare.
Older patients, especially older women, have a higher susceptibility to
INH-induced hepatitis. Hepatitis is rare in patients younger than 20 years of
age. The risk increases to 0.5% in patients 20 to 35, 1.5% for patients 35 to 50,
and 3% for those over 50 years old. Alcohol and drugs that induce the hepatic
P-450 enzyme system such as rifampin seem to increase susceptibility to INH
injury. Continuation of INH after the onset of prodromal symptoms increases
the severity of the hepatitis. Thus it is crucial to stop INH therapy in sympto-
matic patients immediately during the first 1 to 2 weeks.
3. Treatment. There is no specific therapy for INH hepatitis other than prompt
withdrawal of the drug. Patients should be given supportive care. There is no
role for corticosteroid therapy.
Selected Readings
Buckley NA, et al. Oral or intravenous N-acetylcysteine: Which is the treatment of choice
for acetaminophen (paracetamol) poisoning? Toxicol Clin Toxicol. 1999;37:759.
Hanje AJ, et al. Thalidomide-induced severe hepatotoxicity. Pharmacotherapy. 2006;26:
1018–1022.
Kafrouni MI. Hepatotoxicity associated with dietary supplements containing anabolic
steroids. Clin Gastroeterol Hepatol. 2007;5(7)809–812.
Lewis JH. Drug induced liver disease. Med Clin North Am. 2000;84:1275.
Lewis JH, et al. Drug and chemical induced cholestasis. Clin Liver Dis. 1999;3:433.
Marino G, et al. Management of drug induced liver disease. Curr Gastroenterol Rep.
2001;3:38.
Woo OF, et al. Shorter duration of oral N-acetylcysteine therapy for acute acetaminophen
overdose. Ann Emerg Med. 2000;35:363.
79466_CH54.qxd 1/2/08 1:34 PM Page 407
CHRONIC IMMUNE-MEDIATED 54
LIVER DISEASE
407
79466_CH54.qxd 1/2/08 1:34 PM Page 408
SM, smooth muscle; KLM, kidney-liver-microsomal; SLA, soluble liver antigens; ANA, antinuclear
antibody; AMA, antimitochondrial antibody; LEZ, lupus erythematosus.
79466_CH54.qxd 1/2/08 1:34 PM Page 409
symptom. Darkening of the skin, hirsutism, anorexia, diarrhea, and weight loss
may also be present. Less commonly, patients may present with a complication
of portal hypertension, such as variceal bleeding or ascites, or have the disease
identified during the workup of an accompanying connective tissue disease
such as Sjögren’s syndrome, scleroderma or CREST syndrome, systemic lupus
erythematosus (SLE), or thyroiditis, or during routine blood screening. In fact,
up to 50% of the patients are asymptomatic when initially diagnosed. Physical
findings are variable and depend on the extent of the disease. Hepatomegaly,
splenomegaly, spider angiomata, palmar erythema, hyperpigmentation, hirsutism,
and xanthomata may be present. Complications of malabsorption, especially of
fat-soluble vitamins and calcium, and those patients with Sjögren’s also may
have pancreatic insufficiency.
Renal tubular acidosis with defective urinary acidification after an
acid load occurs frequently in PBC, but it is usually subclinical. Deposition
of copper in the kidneys may cause the renal dysfunction. An increased sus-
ceptibility to urinary tract infection has been observed in women with PBC;
the cause of this susceptibility is unknown. Also, there appears to be an
increased prevalence of hepatocellular and breast cancer among patients
with PBC.
2. Laboratory studies
a. Serum tests. Serum alkaline phosphatase level is usually elevated
markedly (2–20 times normal). The values for 5-nucleotidase and
-glutamyltranspeptidase (GGTP) parallel those for alkaline phosphatase.
Serum transaminase levels are slightly elevated (1–5 times normal). The
degree of elevation of these chemistry tests does not have prognostic sig-
nificance. The serum bilirubin level usually rises with the progression of
the disease and is a prognostic indicator of the disease. Serum albumin
and the prothrombin time are normal early in the course of the disease. A
low serum albumin level and prolonged prothrombin time that are not
corrected by vitamin K therapy indicate advanced disease: Both findings
are poor prognostic signs. Serum lipid levels may be strikingly elevated.
Patients with early PBC tend to have slight elevations of low-density
lipoproteins (LDL) and very-low-density lipoproteins (VLDL) and strik-
ing elevations of high-density lipoproteins (HDL). Patients with advanced
disease have marked elevations of LDL and decreased HDL; lipoprotein
X is present in patients with chronic cholestasis. Serum ceruloplasmin is
normal or elevated in contrast to Wilson’s disease. Elevated thyroid stim-
ulating hormone (TSH) levels may be present.
b. Serology and immunologic abnormalities. Serum IgM levels are markedly
increased (4–5 times normal), whereas IgA and IgG levels are commonly
within normal limits. The hallmark of the disease is the presence of AMA,
which is present in more than 99% of the patients. It is usually present in
high titer and is predominantly IgG. AMA has no known inhibitory effect on
mitochondrial function and it does not affect the course of the disease. The
finding of a significant AMA titer (1:40) is strongly suggestive of PBC, even
in the absence of symptoms and the presence of normal levels of serum alka-
line phosphatase. The liver biopsy in these patients shows the characteristic
lesions of PBC. Demonstration of AMA by indirect immunofluorescence
lacks complete diagnostic specificity, however, because AMA may be found
in other disorders using this technique. Newer techniques now available are
more sensitive than immunofluorescence in detecting AMA. These include
radioimmunoassays (RIA), enzyme-linked immunosorbent assays (ELISA),
and immunoblotting technique. A PBC-specific mitochondrial antibody
termed M2 has been characterized. The M2 autoantibody reacts with four
antigens on the inner mitochondrial membrane. These are the E2 components
and protein X of the pyruvate dehydrogenase complex (PDC). PDC is one of
the three related multienzyme complexes (the 2-oxo-acid dehydrogenase com-
plexes) of the Krebs cycle. These complexes are loosely associated with the
79466_CH54.qxd 1/2/08 1:34 PM Page 413
a. Genetic factors. A close association has been found between the HLA-B8
phenotype and PSC in 60% to 80% of patients with or without UC. Also,
HLA-DR3 has been found in 70% of patients with PSC. It appears that a
patient with UC who possesses HLA-B8, DR3 haplotype has a 10-fold
increase in the relative risk of development of PSC. Of patients who do not
possess HLA-B8, DR3, an association with HLA-DR2 has been found in
70%. HLA-B8, DR3 and HLA-DR2 are equally distributed in patients with
PSC, with or without UC. It appears that HLA-DR3, HLA-DR2, and UC are
separate, independent risk factors for the development of PSC.
b. Humoral immune abnormalities. As in primary biliary cirrhosis, adult
patients with PSC usually have hypergammaglobulinemia and increased lev-
els of IgM. High levels of IgG are found in all children with PSC, and 50%
of these children also have elevated IgM levels. Smooth muscle antibody
(SMA) and antinuclear antibody (ANA) may also be present in patients with
PSC. However, there is no correlation between the presence of circulating
autoantibodies and any clinical parameter of the disease.
Circulating anticolonic antibodies have been detected in 60% to 65%
of patients with PSC and UC. Antineutrophil cytoplasmic antibody (ANCA)
is present in 80% of patients.
Elevated levels of circulating immune complexes also have been found
in both sera and bile of patients with PSC. Whether this finding is primary
or an epiphenomenon is not clear. There is also activation of the complement
system via the classic pathway, supporting the involvement of humoral
immune mechanisms in PSC.
c. Cellular immune abnormalities. A significant reduction in the total number
of circulating T cells, especially in the suppressor-cytotoxic (T8) cells, has
been noted in PSC, leading to an increased ratio of helper suppressor (T4/T8)
cells. There is also an increase in the number and percentage of B cells.
In the liver, the portal T4/T8 ratio has been found to be higher in the
portal tract than around proliferating bile ductules. Cytotoxic (T8) cells
tended to localize in areas of bile duct proliferation and infiltrate the ductal
epithelium.
d. Bile duct expression of HLA-DR. Intra- and extrahepatic bile ducts in
normal subjects do not express HLA class II (HLA-DR) antigens. It has been
shown that most intrahepatic bile ducts in patients with PSC express HLA-
DR antigens on the surface cell membrane of the bile duct epithelial cells.
This aberrant expression may play a role in the pathogenesis of bile duct
damage in PSC.
e. Conclusion. There is strong evidence that genetic and immunologic factors
are important in the pathogenesis of PSC. The immunologic destruction of
bile ducts seems to be triggered in genetically predisposed individuals, possi-
bly by viruses, bacteria, or toxins. The association with IBD and UC may be
explained by the passage of these microorganisms or toxins across the dam-
aged colonic epithelial barrier into the portal circulation.
2. The prevalence of PSC in the United States is approximately 5 cases per
100,000 persons. It affects predominantly young males with an age range of
15 to 75 years. At the time of diagnosis, most patients are younger than
45 years. PSC has been reported in children 4 to 14 years old.
3. Associated disorders. PSC may be associated with a variety of autoimmune
disorders. These include Sjögren’s syndrome, retroperitoneal and mediastinal
fibrosis, thyroiditis, myasthenia gravis, Type 1 diabetes mellitus, celiac sprue,
sarcoidosis, cystic fibrosis, Peyronie’s disease, idiopathic chronic pancreatitis,
lupus, rheumatoid arthritis, systemic sclerosis, immune thrombocytopenic
purpura, and autoimmune hemolytic anemia.
4. Cholangiocarcinoma. Clinical experience and pathologic evidence strongly sup-
port an association between PSC and cholangiocarcinoma. Cholangiocarcinoma
arises in 5% to 10% of patients with preexisting PSC and can present in a syn-
chronous fashion with PSC. The most frequent location is the bifurcation of the
79466_CH54.qxd 1/2/08 1:34 PM Page 417
common right and left hepatic ducts (Klatskin tumor); however, it can be multifo-
cal. It is usually heralded by rapid clinical deterioration with progressive jaundice,
weight loss, and abdominal discomfort. Regardless of therapy, including trans-
plantation, the prognosis for patients with cholangiocarcinoma complicating PSC
has been poor. Active smoking and alcohol abuse may increase the risk of devel-
oping cholangiocarcinoma in patients with PSC. PSC may also be an independent
risk for colorectal carcinoma.
C. Diagnosis. The diagnosis of PSC is based on clinical, biochemical, radiologic, and
histologic criteria. The disease is suspected in patients with a cholestatic biochem-
ical profile for longer than 3 to 6 months.
1. Clinical presentation
a. History. Most patients are initially asymptomatic but have a chronic indo-
lent form of the disease, which may remain quiescent for many years. When
fatigue, pruritus, and jaundice develop, patients typically have advanced dis-
ease. Weight loss, abdominal pain, and sometimes complications of cirrhosis
or portal hypertension such as variceal bleeding, ascites, or encephalopathy
may be the presenting findings.
Approximately 15% of patients with PSC present with an acute relaps-
ing type of disease with fever, chills, night sweats, recurrent upper quadrant
pain, and intermittent episodes of jaundice. This “cholangitic” group of
patients may have self-limited, recurrent episodes of bacterial cholangitis
caused by biliary sludge, which may intermittently obstruct strictured bile
ducts.
The distinction between these two presentations of PSC may merge
with progression of disease. In both types, intrahepatic bile duct obstruction
may develop with sludge or pigment stones and recurrent symptoms of bac-
terial cholangitis.
b. Physical examination. Many patients, especially asymptomatic ones, have
a normal physical examination. However, with advanced disease, patients
may have jaundice and hepatosplenomegaly, xanthomas, signs of portal
hypertension, and stigmata of chronic liver disease.
2. The differential diagnosis should include extrahepatic bile duct obstruction;
drug-induced cholestasis; primary biliary cirrhosis; chronic active hepatitis,
autoimmune hepatitis, and alcoholic hepatitis with a cholestatic profile; cholan-
giocarcinoma and secondary sclerosing cholangitis arising in patients with
choledocholithiasis; and congenital or postsurgical abnormalities of the biliary
tract such as Caroli’s disease, choledochal cyst, or strictures. In immunodefi-
cient patients, infections with HIV, cytomegalovirus, and Cryptosporidium
should be considered. Vascular damage to the hepatic arterial tree by cytotoxic
drugs, chemotherapy, and graft-versus-host disease after liver transplantation
may mimic PSC.
3. Diagnostic studies
a. Laboratory tests. All patients with PSC have an elevated serum alkaline
phosphatase level (2 times normal). Serum transaminases are also usually
elevated (2–5 times normal). Serum bilirubin level is variable. Serum albu-
min and prothrombin time are dependent on the severity of the liver dys-
function. CA-19-9 levels greater than 100 U/mL may raise the suspicion for
the presence of cholangiocarcinoma.
Hepatic copper concentration is elevated in most of these patients due
to cholestasis; however, serum ceruloplasmin levels are also elevated. Serum
gamma globulin and IgM levels are elevated in half of the patients with PSC.
b. Radiologic visualization of the biliary tree is necessary for the diagnosis of
PSC. Endoscopic retrograde cholangiopancreatography (ERCP) is the preferred
technique over percutaneous endoscopic cholangiography (PTC) since the
intrahepatic ducts are not dilated. Cholangiography typically shows the pres-
ence of multifocal stricturing and irregularity of intra- or extrahepatic biliary
ducts, or both, giving a “beaded” appearance. There are bandlike short stric-
tures and diverticulumlike outpouchings in the biliary tree in one fifth of
79466_CH54.qxd 1/2/08 1:34 PM Page 418
itching does not diminish after 2 to 4 days, the dosage may be increased to
8 g three times daily. Additional increases in dosage have not been shown
to be effective.
In patients who cannot tolerate cholestyramine, colestipol hydrochlo-
ride may be substituted. If constipation occurs with either drug, laxatives
and fiber supplements may be added. Activated charcoal capsules also have
been used for pruritus. Phenobarbital 60 to 90 mg at bedtime may increase
bile flow and act as a sedative. Ultraviolet B light and large-volume plasma-
pheresis may be helpful in intractable cases. The use of ursodeoxycholic
acid is discussed in the next section. Intractable pruritus suggests intensive
bile duct scarring and obstruction and is an indication for liver transplan-
tation. Also see the recommendation mentioned for PBC-related pruritus in
section II.E.1.a.
iii. Recurrent cholangitis. Antibiotics may be used to treat episodes of
ascending cholangitis. Prophylactic antibiotics such as daily doses of
ciprofloxacin hydrochloride, amoxicillin, or double-strength trimetho-
prim/sulfamethoxazole may decrease the frequency and severity of such
episodes.
b. Treatment of the underlying disease process. A variety of immunosup-
pressive, antifibrotic, and antiinflammatory drugs have been tried in the
treatment of PSC. The slow progression of PSC and spontaneous fluctuation
in bilirubin levels make it difficult to evaluate treatment. However, no drug
has been shown to improve its natural history.
i. Cyclosporin, azathioprine, penicillamine, colchicine, and pred-
nisone have been studied and found ineffective.
ii. Ursodeoxycholic acid (UDCA) at 20 mg/kg of body weight per day has
given promising results in the long-term treatment of PSC in several stud-
ies. UDCA is postulated to decrease serum aminotransferase by stabiliz-
ing hepatocyte membranes rather than by decreasing levels of other bile
acids. It also alkalinizes bile and increases the bile flow. Theoretically,
the result should be a decrease in the formation of pigment stones in the
biliary tract.
iii. Methotrexate. In a small number of patients with PSC, methotrexate
15 mg/kg of body weight per week has resulted in dramatic improvement in
symptoms and biochemistry. In several patients, cholangiographic
improvement also has been noted with no worsening of liver histologic
findings. However, it is not certain that methotrexate is effective in most
patients with PSC.
iv. Tacrolimus decreased the serum bilirubin by 75%, alkaline phosphatase
by 70%, and aminotransferase by 83% after 1 year when given to 10
patients with PSC. The dose of tacrolimus was that which kept trough
levels between 0.6 and 1.0 mg/mL. There are no data on liver biopsy or
ERCP findings.
v. Combination therapy with prednisolone, 1 mg/kg of body weight per
day, azathioprine 1.0 to 1.5 mg/kg of body weight per day, and urso-
diol 500 to 750 mg per day in 15 patients has shown promising results
with significant improvement in liver enzyme levels and liver histology.
Studies using tumor necrosis factor (TNF) inhibitors (etanercept and
remicade) are under way.
vi. Summary. Because damaged or destroyed bile ducts either lack the
capacity to regenerate or do so slowly and ineffectively, diseases such as
PBC and PSC should be treated at early stages before the loss of much of
the biliary system and while there are still adequate numbers of func-
tioning bile ducts.
2. Surgical therapy
a. Biliary drainage procedures, either percutaneously, endoscopically, or surgi-
cally performed, provide only temporary benefit and are riddled with compli-
cations such as infection, obstruction, perforation, leakage, and hemorrhage.
79466_CH54.qxd 1/2/08 1:34 PM Page 420
Selected Readings
Alvarez F, et al. International autoimmune hepatitis group report: Review of criteria for diagnosis of
autoimmune hepatitis. J Hepatol. 1999;31:929.
Angulo P, et al. Primary biliary cirrhosis and primary sclerosing cholangitis. Clin Liver Dis. 1999;3:529.
Charatcharoenwitthaya, et al. Long-term survival and impact of ursodeoxycholic acid treatment for recurrent
PBC after liver transplantation. Liver Transpl. 2007;131:1236–1245.
Corpechot, et al. Assessment of biliary fibrosis by transient elastography in patients with PBC and PSC.
Hepatology. 2006;43:1118–1124.
Czaja AJ. Ursodeoxycholic acid in autoimmune hepatitis. Hepatology. 1999;30:138.
Czaja AJ. Autoantibodies in liver disease. Gastroenterology. 2001;120:239.
Donaldson PT, et al. HLA class II alleles, genotypes, haplotypes, and aminoacids in primary biliary cirrhosis:
A large scale study. Hepatology. 2006;44:667–674.
Ghali, P, Marotta PJ, et al. Liver transplantation for incidental cholangiocarcinoma: analysis of the Canadian
experience. Liver Transpl. 2005;11:1412–1416.
Gong T, et al. Ursodeoxycholic acid for patients with primary biliary cirrhosis: an updated systematic review
and meta analysis of randomized clinical trials using Bayesian approach as sensitivity analysis. Am J
Gastroenterol. 2007;102:1799–1780.
Graziadei IW, et al. Long-term results of patients undergoing liver transplantation for primary sclerosing
cholangitis. Hepatology. 1999;30:1121
Kaplan MM, et al. Primary biliary cirrhosis. N Engl J Med. 2005;353:1261–1273.
Kessler WR, et al. Fulminant hepatic failure as the initial presentation of acute autoimmune hepatitis. Clin
Gastroenterol Hepatol. 2004;2:625–631.
Lee J, et al. Transplantation trends in primary biliary cirrhosis. Clin Gastroenterol Hepatol. 2007; 5(11):
1313–1315.
Lempinen M, et al. Enhanced detection of cholangiocarcinoma with serum trypsinogen-2 I patients with
severe bile duct strictures. Journal of Hepatology. 2007;47(5):677–683.
Lindor KD. Ursodeoxycholic Acid for the treatment of primary biliary cirrhosis. New Engl J Med.
2007;357:1524–1529.
Manns MP, et al. Autoimmune hepatitis: Clinical challenges. Gastroenterology. 2001;120:1502.
Montano-Loza AJ, et al. Improving the end point of corticosteroid therapy in type 1 autoimmune hepatitis to
reduce the frequency of relapse. Am J Gastroenterol. 2007;May;102:1005–1012.
Panjala C, et al. Risk of lymphoma in primary biliary cirrhosis. Clin Gastroenterol Hepatol. 2007;5(6):761–764.
Pompon RE, et al. Quality of life in patients with primary biliary cirrhosis. Hepatology. 2004;40:489–494.
Rudolph G, et al. The incidence of cholangiocarcinoma in primary sclerosing cholangitis after long-time
treatment with ursodeoxycholic acid. Eur J Gastro & Hep. 2007;19(6):487–491.
Suzuki A, et al. Clinical predictors for hepatocellular carcinoma in patients with primary biliary cirrhosis. Clin
Gastroenterol Hepatol. 2007;Feb;5:259–264.
Tischendorf JJ, et al. Transpapillary intraductal ultrasound in the evaluation of dominant bile duct stenoses in
patients with primary sclerosing cholangitis. Scand J Gastro. 2007;42:1001–1017.
Ueno Y, et al. Primary biliary cirrhosis: what we know and what we want to know about human PBC and
spontaneous PBC Mouse models. J Gastroenterol. 2007;42:189–195.
79466_CH55.qxd 1/2/08 1:35 PM Page 421
I. WILSON’S DISEASE
A. Pathogenesis. Wilson’s disease is a treatable, genetic disorder. The metabolic
defect leads to progressive accumulation of copper in the liver, brain (particularly
in the basal ganglia), cornea, and kidneys, causing severe functional impairment
leading to irreversible damage. If not treated, this disease is invariably fatal, but
with early diagnosis and treatment, the clinical manifestations can be prevented
and reversed.
Wilson’s disease is an autosomal recessive disorder. The abnormal gene is dis-
tributed worldwide with a prevalence of heterozygotes of 1 in 200 and homozy-
gotes of 1 in 30,000. The genetic defect is on chromosome 13 near the red-cell
esterase locus. In 95% of patients, there is also an absence or deficiency of serum
ceruloplasmin, the main copper-transporting protein in blood. This deficiency
is caused by a decrease in transcription of the ceruloplasmin gene located on
chromosome 3.
B. Copper metabolism
1. Copper concentration in the liver. The liver of a human newborn contains
six to eight times the copper concentration of an adult liver. Within the first
6 months of life, this diminishes to a concentration of 30 mg/g of dry tissue.
Thereafter, throughout life, the liver concentration of copper is maintained at
this steady state by careful regulation of intestinal absorption and transport and
of the liver stores through the synthesis of plasma and tissue copper proteins
and excretion of copper from the body in the bile.
2. Absorption and excretion. Approximately 50% of the average dietary intake
of 2 to 5 mg of copper is absorbed from the proximal small intestine and loosely
binds to albumin. It is promptly cleared by the liver, where it is incorporated
into specific copper proteins such as cytochrome oxidase and ceruloplasmin or
is taken up by lysosomes before being excreted in bile. There are two main
routes by which copper is mobilized from the liver.
a. Synthesis of copper-containing ceruloplasmin and its release into the
circulation.
b. Biliary excretion amounting to 1.5 mg of copper per day. This is the princi-
pal route of elimination of copper from the body.
3. Genetics. The copper excess seen in Wilson’s disease has been shown to be the
result of decreased biliary excretion and not an increase in the absorption of
copper. The defect is caused by mutations in the Wilson’s gene (ATP7B) on
chromosome 13 gene. The gene ATP7b encodes a cation-transporting P-type
adenosine triphosphatase that is expressed in the liver, kidney, and placenta.
Mutations in ATP7b result in disordered export of copper from the liver into
bile with resultant accumulation of copper cation in hepatocytes. The ATP7b
protein is present primarily in the trans-Golgi where it is critical for excretion
of copper into bile, as well as providing appropriate copper for binding to ceru-
loplasmin. Lack of functional ATP76 limits the availability of copper for
ultimate incorporation into ceruloplasmin. When copper is not available for
binding, an apoprotein is secreted from the hepatocytes that is rapidly degraded
in the plasma, resulting in the hallmark of Wilson’s disease, diminished
circulating ceruloplasmin levels.
421
79466_CH55.qxd 1/2/08 1:35 PM Page 422
4. Family screening. More than 200 mutations have been identified in the
Wilson’s gene. Once a proband has been identified as having Wilson’s disease,
it may be possible to screen siblings on the basis of genetic analysis. As an auto-
somal recessive disorder, 1 in 4 siblings may be expected to be homozygous for
the gene defect. Genetic testing of siblings requires the sequencing of both alle-
les of the ATP7b gene in the proband and then subsequent comparison of those
alleles to the ATP7b alleles in the siblings.
C. Copper toxicity
1. Acute toxicity. Ingestion of gram quantities of copper causes serious gastroin-
testinal and systemic injuries and occasionally hepatic necrosis. Generally, how-
ever, the vomiting and diarrhea that follow the ingestion of copper salts protect
the patient from serious toxic effects.
2. Chronic toxicity. Hepatic copper overload may occur in disorders other than
Wilson’s disease. These include primary biliary cirrhosis, extrahepatic biliary
artesian, Indian childhood cirrhosis, and other chronic cholestatic disorders.
The excess hepatic copper may aggravate the underlying pathologic process by
direct damage to the organelles or through promotion of fibrosis.
D. Diagnosis
1. Clinical presentation. Wilson’s disease has many modes of presentation. It
may simulate several different neurologic and psychiatric disorders. It may pre-
sent as asymptomatic elevation of the transaminases, chronic active hepatitis,
fulminant hepatitis, cirrhosis of the liver, acquired hemolytic anemia, renal
disease, or eye abnormalities such as sunflower cataracts and Kayser-Fleischer
(K-F) rings.
a. Liver disease is the most common presentation of Wilson’s disease in
childhood. About 40% of all patients with Wilson’s disease come to med-
ical attention with evidence of liver disease. Because an increase of 30 to
50 times the normal hepatic concentration of copper can occur without
any clinical manifestations, symptoms of liver disease do not appear
before 6 years of age. However, one half of the patients have symptoms
by 15 years of age. Thus, overt Wilson’s disease is encountered predomi-
nantly in older children, adolescents, young adults, and rarely in older
adults.
i. Forms. The hepatic disease may take several different forms.
a) Commonly it begins insidiously and runs a chronic course character-
ized by weakness, malaise, anorexia, mild jaundice, splenomegaly, and
abnormal liver chemistry tests. The disease may mimic acute viral
hepatitis, mononucleosis, or chronic active hepatitis.
b) Fulminant hepatitis may occur suddenly, characterized by progres-
sive jaundice, ascites, and hepatic failure. The outcome is usually
fatal, particularly when the disorder is accompanied by hemolytic
anemia.
c) Some patients present with the typical picture of postnecrotic cirrho-
sis with spider angiomata, splenomegaly, portal hypertension, ascites,
bleeding esophageal varices, or thrombocytopenia mimicking idio-
pathic thrombocytopenic purpura (ITP). The liver enzymes may be
normal. The diagnosis of Wilson’s disease should always be consid-
ered in patients younger than 30 years with negative serology for viral
hepatitis; with a history of chronic active hepatitis; or with juvenile,
cryptogenic, or familial cirrhosis. Although fewer than 5% of such
patients have Wilson’s disease, it is one of the few forms of liver dis-
ease for which specific and effective therapy is available.
ii. Histology. There is no one specific histologic profile to identify Wilson’s
disease in liver biopsy specimens. In the early stages of copper accumula-
tion, when copper is diffusely distributed in the cytoplasm, it is unde-
tectable by rhodanine or rubeanic acid stains. At this stage, lipid droplets
are seen in the hepatocytes with ballooned, vacuolated nuclei containing
glycogen. This initial steatosis progresses to fibrosis, then ultimately to
79466_CH55.qxd 1/2/08 1:35 PM Page 423
cirrhosis. With time and progression of the liver disease, the hepatocyte
lysosomes seem to sequester the excess copper, which is detectable
throughout some nodules by routine histochemical staining. Because of
the variable stainability and the irregular distribution of copper among
adjacent nodules, absence of a positive rhodanine or rubeanic acid stain
on a histologic slide does not exclude the diagnosis of Wilson’s disease.
The parenchyma usually is infiltrated with mononuclear cells. There may
be cholestasis, focal necrosis, and Mallory’s hyalin. In other cases, the
histology may resemble that of acute or chronic active hepatitis.
Once macronodular cirrhosis develops, the microscopic findings are
nonspecific. Hepatocytes may contain some cytoplasmic lipid, vacuolated
glycogen-containing nuclei, and cytoplasmic inclusions containing
copper-rich lipofuscin granules.
b. Neurologic disease is the most common presentation of Wilson’s disease.
The usual age of onset is 12 to 32 years. The most common symptoms are
as follows:
i. Incoordination particularly involving fine movements such as handwrit-
ing, typing, and piano playing.
ii. Tremor is usually at rest but intensifies with voluntary movement
and emotion. It ranges from a fine tremor of one hand to generalized
tremor of the arms, tongue, and head. It may be slow, coarse, or choreoa-
thetoid. Dystonia, ataxic gait, spasticity, and rigidity are late neurologic
manifestations.
iii. Dysarthria begins with difficulty in enunciating words and progresses to
slurring of speech, microphonia, and aphasia.
iv. Excessive salivation occurs early in the course of the disease.
v. Dysphagia is progressive and oropharyngeal; patients have difficulty ini-
tiating swallowing, leading to regurgitation and aspiration.
c. K-F rings are corneal copper deposits laid in the Descemet’s membrane in
layers appearing as granular brown pigment around the periphery of the iris.
They may be absent in early stages but are present in all patients in the neu-
rologic stage of Wilson’s disease. Most K-F rings can be visualized by the
naked eye, but some require slit-lamp examination.
d. Psychiatric disease. Almost all of the patients demonstrate some form of psy-
chiatric disturbance, which may appear as teenage adjustment behavior, anxi-
ety, hysteria, or a manic-depressive or schizoaffective disorder. Psychotropic
drugs may accentuate the neurologic manifestations of Wilson’s disease and
increase the patient’s problems.
e. Hematologic disease. In a few patients, Wilson’s disease presents as a
Coombs-negative hemolytic anemia with transient jaundice. It may be inter-
mittent and benign, or it may occur with fulminant hepatitis. The hemolysis
occurs during phases of hepatocellular necrosis with sudden release of cop-
per from necrotic hepatocytes into the circulation. This effect is indicated by
a marked rise in the concentration of nonceruloplasmin copper in the blood
and in the amount of copper excreted through the urine.
With portal hypertension and splenomegaly, hypersplenism may result
in thrombocytopenia and pancytopenia. Progressive liver disease also gives
rise to clotting factor deficiencies and bleeding.
f. Kidney disease. Renal abnormalities result from accumulation of copper
within the renal parenchyma. These abnormalities range from renal insuffi-
ciency with decreased glomerular filtration rate to proximal tubular defects
resembling Fanconi’s syndrome, renal tubular acidosis, proteinuria, and
microscopic hematuria.
g. In disorders involving “inflammation” and metabolic syndrome iron over-
loading may occur due to reduced iron egress. Iron is essential for many
bacterial and viral pathogens. Hepciden plays a key role in protecting the
body’s precious iron from these pathogens. In response to inflammatory
cytokins, hepcidin degrades ferroprotin, thus preventing iron from entering
79466_CH55.qxd 1/2/08 1:35 PM Page 424
also interferes with absorption of iron, alters immune responses, and affects the
serum lipoprotein profile.
Oral zinc therapy may serve as an adjunct to standard chelation therapy
with D-penicillamine or trientine; however, there are reports that have raised con-
cerns regarding the formation of zinc-penicillamine complexes that may diminish
or abolish the therapeutic effectiveness of both drugs when used in combination.
Zinc is not recommended as the sole agent for initial therapy of sympto-
matic patients, but it is recommended as maintenance therapy at 150 mg per
day in three divided doses to keep patients at negative copper balance. Zinc
acetate is better tolerated than zinc chloride or sulfate.
5. Tetrathiomolybdate, an agent that appears to block the absorption of copper
by holding the metal in a tight, metabolically inert bond, has been used in some
patients intolerant to penicillamine. It is not commercially available for use in
North America. Although the drug is generally well tolerated, at least two cases
of bone marrow suppression have been documented. Further clinical trials are
needed before it may be used as a primary therapy for Wilson’s disease.
6. Monitoring. Periodic physical examinations, slit-lamp examinations of the
cornea for documentation of the disappearance of K-F rings, and measurements
of 24-hour urinary copper excretion and serum free copper should be per-
formed to assess the effectiveness of therapy.
7. Significant clinical improvement may occur only after 6 to 12 months of
uninterrupted treatment.
8. Fulminant hepatic failure may develop in a number of patients with Wilson’s
disease, either as an initial manifestation of the disease or as a consequence of non-
compliance with medical therapy. A smaller subset of patients will have cirrhosis
and hepatic decompression unresponsive to medical interventions outlined above.
9. Orthotopic liver transplantation (OLT) is a lifesaving procedure for patients with
fulminant hepatitis or irreversible hepatic insufficiency due to Wilson’s disease. The
metabolic abnormality is reversed and the disease is cured. One-year survival follow-
ing liver transplantation is now approximately 80%. The replacement of the affected
liver expressing the mutant A7P7b gene protein product with a donor organ that
expresses the normal gene protein product is expected to correct the defect in hepatic
copper metabolism. Thus, the allograft is not susceptible to copper accumulation.
However, the resolution of the extra hepatic manifestations of Wilson’s
disease after OLT has been less than universal. Thus, OLT in the absence of
decompensated liver disease and solely for the management of extrahepatic dis-
ease such as neurologic defects is not routinely recommended. Liver cell trans-
plantation is currently under study as an alternative to liver transplantation.
2. The circulating forms of iron that lead to tissue iron overload are not tightly asso-
ciated with plasma transferring and are referred to as non-transferrin bound iron
(NTBI). NTBI increases whenever the capacity of transferring to incorporate incom-
ing iron from the gut or reticulo endothelian cells becomes a limiting factor. A frac-
tion of NTBI, called Labile plasma iron (LPI), is translocated across cell membranes
in a non-regulated manner and leads to excessive iron accumulation in various
organs. The extent of organ damage depends on the rate and magnitude of plasma
iron overload. In transfusion dependant iron overload and juvenile forms of
hemochromatosis early damage of the heart and endocrine glands predominate. In
milder forms of iron overload late onset liver disease is more common.
3. The control center that keeps blood levels of iron within the narrow physiolog-
ical range is in the hepatocyte. “Special sensors” respond to the iron and stimu-
late the synthesis and release of the iron-hormone hepcidin, which is encoded by
HAMP gene. Hepcidin circulates through the body and interacts with the iron-
exporter ferroprotin expressed in the surfaces of iron rich macrophages and
intestinal cells. As a result of this interaction, ferroportin is internalized and
degraded. The unneeded iron remain in the cells where it is saved for future use
in the form of ferritin. The diminished release of iron restores blood levels to the
non-toxic range, thus reinsuring the stimulus for further hepcidin synthesis and
ferroportin gradually resumes its iron-exporting activity.
4. The mechanisms underlying hepatocyte iron sensing are still being investigated.
However, as a result of their involvement in transferring and non-transferrin
mediated iron uptake, HFE (the hemochromatosis gene) and transferrin receptor 2
might play a role in conveying iron signals to the hepatocyte control center. The
details of this process are still unclear. However, both are important for hepcidin
expression. Their functional loss causes hepcidin insufficiency and iron overload.
5. Hepcidin production can be impaired by genetic as well as acquired factors.
Hepcidin deficiency has been associated with loss of HAMP, HJV, HFE, or
TFR2. Even though there is genetic heterogeneity of HH, all forms identified so
far originate from the presence of unneeded iron in the circulatory pool caused
by insufficient hepcidin.
Nongenetic factors which affect hepcidin output and lead to iron overload
include ethanol abuse, toxic and viral insults, e.g., hepatitis C virus, diminished
functional hepatic virus, e.g., in acute hepatic failure, end-stage liver disease or
immune mediated liver disease.
6. “Hepcidin resistance or insensitivity” may also result from genetic or acquired
factors that impair hepcidin-ferroprotein interaction. Mutations in ferroprotin
gene have been noted to cause disease similar to HFE-hemochromatosis.
7. All forms of iron overload above the same basic features. Iron overload initially
involves the plasma compartment signified biochemically by increased saturation
of the iron transporter, parenchymal cells of key organs is reflected by increasing
serum ferritin levels.
8. Expansion of the pool of non-transferrin-bound iron and its preferential uptake
by parenchymal cells as the iron level and transferrin saturation increase in
plasma. The result is total body iron overload with predominant deposition in
parenchymal liver cells, leading to hepatic fibrosis and cirrhosis. A growing body
of experimental evidence suggests that iron induces membrane lipid peroxidation,
possibly as a consequence of free radical formation. This process damages lysoso-
mal, microsomal, and other cellular membranes, resulting in cell death. Because
iron is a cofactor for proline and lysine hydroxylase, two critical enzymes involved
in the synthesis of collagen, some investigators have postulated that elevated tis-
sue iron levels may promote an increased deposition of collagen and hepatic fibro-
sis. In addition, iron overload may activate and enhance the expression of some
target genes in the liver, such as genes for ferritin and procollagen.
Parenchymal cell deposition of iron also occurs in other tissues and may
result in cardiac failure, diabetes mellitus, gonadal insufficiency, and arthritis.
E. Incidence. It has been estimated that the prevalence of homozygous and het-
erozygous HFE associated HH in populations of Northern European descent is
79466_CH55.qxd 1/2/08 1:35 PM Page 429
about one in 250 and 1 in 8 to 10, respectively. The HFE associated HH disease is
rarely identified in Africans or Asians. It is more common in men. The male-female
ratio is 5 to 10:1. Nearly 70% of the patients have their first symptoms between
the ages of 40 and 60 years. It is rarely clinically evident below the age of 20.
F. Genetics. In 1966 the HFE gene was identified in the short arm of chromosome 6.
HFE-linked HH is responsible for 85% to 90% of cases of HH. HFE-linked HH is
inherited as an autosomal recessive disorder. Its phenotypic expression is depen-
dent on diet, gender, and other factors. An affected individual is often the offspring
of two heterozygotes.
At least 30 missense mutations have been identified in HFE. One results in a
change of cysteine to tyrosine at amino acid (AA) position 282(C282Y); a second
mutation results in a change in histidine to aspartate at AA position 63(H63D); and a
third mutation results in a change of serine to cysteine at AA 65(S65C). Homozygosity
for the C282Y mutation has been identified in approximately 90% of individuals who
have typical phenotypic HH. The clinical impact of H63D and S65C mutations is
small. Approximately 83% of patients with HH are homozygous for C282Y muta-
tion; an additional 4% are compound heterozygotes (C282Y/H63D).
Heterozygosity for C2824 is approximately 1 in 10 in the United States.
Approximately one third of the males and one sixth of the females sharing one
haplotype (heterozygotes) exhibit partial biochemical expression; however, these
individuals rarely develop clinical manifestations of the disease.
Phenotypic expression of the inherited abnormality is modified by a variety of
factors, including dietary iron intake, iron supplementation, chronic hemodialysis,
alcohol consumption, menstrual blood loss, multiple pregnancies, and accelerated
erythropoiesis.
Ten percent to 15% of patients have a clinical syndrome similar to HH, but
do not have C282Y mutation.
The other inherited forms of iron overload, which do not involve mutations
of the HFE, include iron overload resulting from mutations in transferrin receptor-2.
Juvenile HH, which is caused by either mutation in hemojuvelin (HJV) or in the
hepcidin gene, or HAMP. Neonatal iron overload is a rare disorder that is
thought to be caused by an intrauterine hepatic viral infection resulting in an exces-
sive uptake of iron into the fetal liver.
Decreased cell iron efflux may also cause HH. Iron egress from mammalian
cells depends on the activity of a specialized membrane associated iron exporter,
ferroportin and a circulating protein, ceruloplasmin, which oxidizes Fe2+ to
Fe3+ and helps loading iron onto circulating transferring. Genetic or acquired fac-
tors that impair the function of these proteins lead to iron overload due to impaired
iron egress. In such disorders, the reduced ability to deliver iron into circulating
transferring causes low saturation of transferring with iron. Inefficient iron exit
out of the cells causes tissue iron overload and organ injury.
Ferroportin disease is associated with the A77D mutation of ferroportin as an
autosomal dominant disorder which causes progressive iron retention predominantly
in reticuloendothelial cells of the spleen and liver. It is characterized by steadily increas-
ing serum ferritin, marginal anemia and mild organ disease. The disorder appears to be
spread worldwide in the different ethnic groups. So far 32 pedigrees of ferroportin
mutations have been reported. Mutations of ferroportin cause impairment of iron recy-
cling, particularly by reticuloendothelial macrophages which normally process and
release a large quantity of iron derived from the lysis of senescent erythrocytes.
In hypo or aceruloplasminemia, an autosomal recessive neurodegenerative
disease characterized by iron accumulation with brain as well as visceral organs
such as the liver and pancreas is caused by mutations in the ceruloplasmin gene.
Cerulopasmin plays a role in brain iron traffic. Patients due to diminished exit of
iron from cells often present with iron deficiency anemia, neurologic disorder, reti-
nal degeneration and diabetes velitus.
G. Pathophysiology of HFE. Since the discovery of HFE as the gene responsible for
HH, its role in the dysregulated iron absorption seen in HH has been elucidated.
In addition to leading to diminished hepcidin expression, HFE protein is also found
79466_CH55.qxd 1/2/08 1:35 PM Page 430
in the crypt cells of the duodenum, associated with B2-microglobulin and transfer-
rin receptor. It is thought that HFE protein may facilitate transferrin receptor-
dependent iron uptake into crypt cells and that mutant HFE protein may lose this
ability, leading to a “relative” iron deficiency in the duodenal crypt cells. In turn,
this may result in an increase in the expression of an iron transport protein called
divalent metal ion transporter 1 (DMI-1) that is responsible for dietary iron
absorption in the villous cells of the duodenum.
H. Diagnosis
1. Clinical presentation. Early in the course of IHC, patients may have the fol-
lowing signs or symptoms: lethargy, weight loss, and change in skin color, con-
gestive heart failure, loss of libido, abdominal pain, joint pain, or symptoms
related to diabetes mellitus. Hepatomegaly, skin pigmentation, testicular atro-
phy, loss of body hair, and arthropathy are the most prominent physical signs.
Patients with hemochromatosis secondary to transfusion therapy for
chronic anemia present with clinical symptoms at a young age. The typical
patient with thalassemia, having received more than 100 blood transfusions,
experiences failure of normal growth and sexual development in adolescence
and hepatic fibrosis. Many patients die of cardiac disease by early adulthood.
a. Liver disease. The liver is the first organ affected in IHC, and
hepatomegaly is present in 95% of the symptomatic patients. Hepatomegaly
may exist in the absence of symptoms and abnormal liver tests. In fact, serum
aminotransferases are frequently normal or only slightly elevated in patients
with IHC, even in the presence of cirrhosis. This finding reflects the relative
preservation of the hepatocyte integrity, which usually persists throughout
the course of the disease.
Palmar erythema, spider angiomata, loss of body hair, and gynecomas-
tia are often seen. Manifestations of portal hypertension may occur but are
less common than in alcohol-related cirrhosis. Hepatocellular carcinoma
(HCC) develops in approximately 30% of the patients with cirrhosis. This
increased incidence of HCC may be due to chronic iron-overload-induced
damage to hepatic DNA.
b. Skin pigmentation, which may be absent early in the course of the disease, is
present in a large percentage of symptomatic patients. The dark metallic hue is
largely due to melanin deposition in the dermis. There is also some iron depo-
sition in the skin, especially around the sweat glands. Pigmentation is deeper
on the face, neck, exterior surfaces of the lower arms, dorsa of hands, lower
legs, and genital regions, and in scars. Ten percent to 15% of the patients have
pigmentation of the oral mucosa. Skin is usually atrophic and dry.
c. Endocrine disorders
i. Diabetes mellitus develops in 30% to 60% of the patients with
advanced disease. The presence of a family history of diabetes mellitus
and the presence of liver disease and direct damage to the beta cells of the
pancreas by deposition of iron all probably contribute to the develop-
ment of diabetes mellitus in IHC. Complications of diabetes mellitus
such as retinopathy, nephropathy, and neuropathy may occur. IHC spares
the exocrine pancreas.
ii. Loss of libido and testicular atrophy. Hypogonadism is common with
symptomatic IHC and is most likely due to hypothalamic or pituitary
failure with impairment of gonadotropin secretion. Liver damage, alco-
hol intake, and other factors may contribute to sexual hypofunction.
iii. Other endocrine disorders. Addison’s disease, hypothyroidism, and
hypoparathyroidism are less common in IHC.
d. Arthropathy is present in about one fifth of the patients with IHC. It is more
common in patients over 40 years of age and occasionally may be the pre-
senting symptom.
i. Osteoarthritis involving the metacarpophalangeal and proximal inter-
phalangeal joints of the hands, and later of the knees, hips, wrists, and
shoulders, is most commonly seen.
79466_CH55.qxd 1/2/08 1:35 PM Page 431
I. The treatment of IHC involves the removal of excess iron and therapy of func-
tional insufficiency of the organs involved, such as congestive heart failure, liver
failure, and diabetes mellitus.
1. Phlebotomy. Iron is best removed from the body by phlebotomy. There are
250 mg of iron in 500 mL of blood. Because the body burden of iron in IHC
may be in excess of 20 g, 2 to 3 years of weekly phlebotomy of 500 mL of
blood may be necessary to achieve hemoglobin of 11% and serum ferritin
level of 10 to 20 mg/L. Thereafter, the frequency of phlebotomy may be
decreased to 1 unit of blood every 3 months for the rest of the patient’s life.
2. Chelating agents such as deferoxamine remove only 10 to 20 mg of iron per
day. In patients with anemia, hypoproteinemia, or severe cardiac disease preclud-
ing phlebotomy, this technique may be used. However, it is difficult to achieve a
negative iron balance by this means. In patients with refractory anemia, if it is ini-
tiated early in the course of iron loading, this approach can lower the risk of
cardiac disease, promote sexual maturation, and generally improve the prognosis.
Nightly subcutaneous infusion of deferoxamine induces excretion of chelat-
able iron into urine and, probably via the bile, into stool. The recommended
dosage is approximately 40 to 80 mg/kg per day. At daily dosages of more than
50 mg/kg, the potential increases for hypersensitivity and ocular and otologic
complications, including night blindness, visual field changes, irreversible retinal
pigmentation, optic neuropathy, deafness, and other adverse reactions. In addi-
tion, the iron chelator deferoxamine can promote infections, including gram-
negative sepsis and abscesses, by functioning as a siderophore, delivering iron to
bacteria that use it in growth. Chelating compounds that can be taken orally,
most notably including -hydroxypyridine, are being developed.
Despite the potential of vitamin C deficiency to exacerbate iron over-
load, vitamin C supplementation is contraindicated in patients with
hemochromatosis. Sudden cardiac deaths have occurred in patients receiving
chelation therapy and vitamin C supplements. The cause of these events may
involve sudden shifts of iron from reticuloendothelial to parenchymal cells of
the myocardium or increased cellular injury from increased lipid peroxidation.
79466_CH55.qxd 1/2/08 1:35 PM Page 433
and live to adulthood. The remaining 25% appear to recover from their initial
insult and return to normal liver function with minimal evidence of residual
liver fibrosis.
5. In adults, as in children, men are twice as likely to have liver disease as women.
In studies done with patients with PI ZZ phenotype from Sweden or of Northern
European ancestry, the relative risk for cirrhosis was noted to be 37% to 47%,
and of hepatoma, 15% to 29%. In studies done in heterozygotes of PI MZ or PI
SZ phenotypes, the relative risk for cirrhosis was noted to be 1.8 and for
hepatoma 5.7, compared with other patients investigated for liver disease.
6. Diagnosis
a. Diagnosis of AAT deficiency should be considered in any chronic liver disease
of uncertain cause in children and adults of white and particularly Northern
European populations. The probability of AAT deficiency as a cause of the
disease increases in patients with a family history of liver or obstructive lung
disease and in patients in whom alcohol or viral hepatitis may be excluded.
Children with neonatal hepatitis, giant cell hepatitis, juvenile cirrhosis, or
chronically elevated liver chemistry tests should be investigated for AAT phe-
notype. Adults with chronic active hepatitis lacking serologic markers, or with
cryptogenic cirrhosis, with or without hepatoma, also should be evaluated.
b. The initial discovery of AAT deficiency is usually made by the absence of the
alpha-1 peak on serum protein electrophoresis. This method is not very sen-
sitive, and quantitative determination of serum AAT level and PI typing is
usually necessary for accurate diagnosis. Liver biopsy further supports the
diagnosis and helps to stage the extent of liver damage.
c. Biochemical studies. Plasma AAT levels may be determined in most clini-
cal laboratories by electroimmunoassay. Plasma levels should be related to
the normal reference range for each laboratory. Functional analysis may also
be performed by determination of total trypsin inhibitory capacity, 90% of
which is due to AAT activity. Discrepancies may occur between immunologic
and functional analyses due to the presence of dysfunctional or inactive
species. Subnormal levels suggest a genetic AAT variant and are only rarely
secondary to a disease process. Low levels have been described, however, in
the infant respiratory distress syndrome, in protein-losing states, and in ter-
minal liver failure. Levels below 20% of normal suggest the homozygous
deficiencies PI Z, PI M malton, PI M Duarte, and PI null or heterozygous
combinations of these alleles. Levels in the range 40% to 70% are compati-
ble with heterozygous deficiency (PI MO, SZ, MZ, etc.). Heterozygotes with
liver disease or active inflammation may well exhibit normal levels.
d. Phenotype determination. Isoelectric focusing is the method of choice for
phenotyping. A monoclonal antibody specific for PI Z AAT has been developed
and is useful in detecting the presence of the PI Z allele. It has been used for
mass population screening in an enzyme-linked immunosorbent assay (ELISA).
Such antibodies also may be used for specific staining of histologic material.
Specific diagnosis of most phenotypes also may be performed at the DNA
level. Some genetic mutations may be identified by Southern blotting due to
their fortuitous localization at a restriction endonuclease site (restriction frag-
ment length polymorphism). This method may be performed on DNA purified
from very few cells and therefore is ideally suited to prenatal diagnosis.
e. Liver biopsy and histopathology. 1-antitrypsin globular inclusions are
localized predominantly in periportal hepatocytes, are weakly acidophilic,
and may be overlooked easily on routine hematoxylin-eosin sections. After
diastase treatment to remove glycogen, the remaining immature glycoprotein
is strongly PAS-positive, reflecting high mannose content. In general, the num-
ber and size of globules increase with age and disease activity. Globules may
be missed entirely in liver biopsies from heterozygotes due to sampling error.
Immunofluorescence and immunoperoxidase staining with nonspecific
antisera against AAT are more sensitive than PAS-D staining for the detection
of intrahepatocellular aggregates. These techniques can be applied to both
79466_CH55.qxd 1/2/08 1:35 PM Page 436
frozen and formalin-fixed tissue. Electron microscopy can reveal AAT, in the
endoplasmic reticulum (ER), with dilatation in other periportal hepatocytes
in which biosynthesis generally predominates, in both hereto- and homozy-
gotes. Adult liver disease in AAT deficiency is usually characterized by rela-
tively low-grade inflammation radiating from portal tracts. Inflammatory
cells (primarily lymphocytes) are distributed in close proximity to areas of
abundant PAS-D globules. There may be piecemeal necrosis. As the disease
progresses, the liver becomes more fibrotic with the development of macron-
odular cirrhosis. Hepatoma of a hepatocellular or cholangiocellular type may
accompany the cirrhosis.
C. Treatment
1. Prevention. As in all genetic disease, the diagnosis of 1-antitrypsin deficiency
in a patient, regardless of the mode of clinical presentation (lung, liver, or other
symptoms), should lead to investigations of the family. It is desirable to identify
homozygotes in an asymptomatic stage. Patients should be advised against
smoking. Genetic counseling should be provided to families with one or more
children affected by liver disease.
2. Medical therapy. No medical therapy is beneficial in patients with liver
disease.
3. Liver transplantation. Liver transplantation provides a new source of plasma
AAT. In these patients, the PI typing converts to that of the donor. The 1-year
survival is 90% and the 5-year survival of juvenile patients in most transplant
centers is 80% to 85%. The experience of liver transplantation for cirrhosis in
PI Z adults is limited. In the absence of serious pulmonary manifestations, indi-
cations for liver transplantation are essentially those for decompensation due to
chronic liver disease of any cause. Preoperative evaluation of pulmonary func-
tion and the exclusion of hepatoma are essential.
Selected Readings
Wilson’s Disease
Brewer GI, et al. Treatment of Wilson’s disease with zinc, XVII: Treatment during
pregnancy. Hepatology. 2000;31:304.
Eghtesad B, et al. Liver transplantation for Wilson’s disease: A single center experience.
Liver Transplant Surg. 1999;5:467.
Ferenci P, et al. Late onset Wilson’s disease. Gastroenterol. 2007;132:1294–1298.
Mufti AR, et al. Is a copperbinding protein deregulated in Wilson’s disease and other
copper toxic disorders? Mol Cell. 2006;21:775–785.
Schilsky ML. Treatment of Wilson’s disease: What are the roles of penicillamine, trientine,
and zinc supplementation? Curr Gastroenterol Rep. 2001;3:54.
Sternlieb I. Wilson’s disease and pregnancy. Hepatology. 2000;31:304.
Taly AB, et al. Wilson’s disease: description of 282 patients evaluated over 3 decades.
Medicine (Baltimore). 2007;86:112–121.
Hemochromatosis
Brunt EM, et al. Histologic evaluation of iron in liver biopsies: Relationship of HFE
mutations. Am J Gastroenterol. 2000;95:1788.
Harrison-Findik DD, et al. Alcohol metabolism mediated oxidative stress down porter
expression. J Biol Chem. 2006;281:22974–22982.
Papanikolanon G, et al. Mutations in HFE2 cause iron overload in chromosome 1g-linked
juvenile hemochromatosis. Nat Genet. 2004;36:77–82.
Pietrangelo A. Hereditary hemochromasis—a new look at an old disease. N Eng J Med.
2004;350:2382–2397.
Pietrangelo A. Hereditary hemochromatosis: Biochemia and biophysician. Acta. 2006;
1763:700–710.
1-Antitrypsin Deficiency
Arroyo M, et al. Hepatic inherited Metabolic disorders. Semin Diag Pathol. 2006;
23:182–189.
Perlmutter DH. The cellular basis of liver injury in alpha-1-antitrypsin deficiency.
Hepatology. 1990;13:172.
Perlmutter DH, et al. Molecular pathogenesis of alpha-1-antitrypsin deficiency associated
liver disease. A meeting review. Hepatology. 2007;45:1313–1323.
Pietrangelo A. The ferroportin disease. Blood Cells Mol D13. 2004;32:131–138.
Pietrangelo A, et al. Genetics in liver diseases. J Hepot. 2007;46:1143–1146.
Powell PF, et al. Steatosis is a cofactor in liver injury in hemochromatosis. Gastroenterol.
2005;129:1932–1943.
Sveger T, et al. The liver in adolescents with a1-antitrypsin deficiency. Hepatology.
1995;22:514.
Whiting PF. Fetal and infantile hemochromatosis. Hepatology. 2006;43:654–660.
Wong K, et al. The diversity of liver disease associated with an elevated ferratin. Con J
Gastro. 2007;20:467–470.
79466_CH56.qxd 1/2/08 1:36 PM Page 438
I. Cirrhosis is a disease state that is the consequence of a wide variety of chronic, pro-
gressive liver diseases. These result in diffuse destruction of hepatic parenchyma and
its replacement with collagenous scar tissue and regenerating nodules with disruption
of the normal hepatic lobular and vascular architecture. Regardless of etiology, the
triad of parenchymal necrosis, regeneration, and scarring is present in all cirrhotic
patients.
A. Classification
1. Morphologic. The pattern of scarring and gross appearance of the liver can be
used to classify cirrhosis into three groups:
a. Micronodular (Laënnec’s)
b. Macronodular
c. Mixed
Morphologic classification seldom permits the determination of the spe-
cific etiology. However, micronodular cirrhosis is most commonly seen as the
consequence of alcoholic liver disease, and macronodular and mixed cirrhosis
are the result of most other inflammatory or infiltrative diseases of the liver.
2. Etiologic
a. Alcohol
b. Viral hepatitis B, C, and D—most common causes of cirrhosis in the United
States
c. Drug- or toxin-induced
d. Hemochromatosis
e. Wilson’s disease
f. 1-Antitrypsin deficiency
g. Autoimmune hepatitis
h. Nonalcoholic steatohepatitis (NASH)
i.Biliary obstruction
i. Primary biliary cirrhosis (without extrahepatic bile duct obstruction)
ii. Secondary biliary cirrhosis (with extrahepatic bile duct obstruction)
j. Venous outflow obstruction
i. Budd-Chiari syndrome
ii. Venoocclusive disease
k. Cardiac failure
i. Chronic right-sided failure
ii. Tricuspid insufficiency
l. Malnutrition
i. Jejunoileal bypass surgery
ii. Gastroplasty
m. Miscellaneous
i. Schistosomiasis
ii. Congenital syphilis
iii. Cystic fibrosis
iv. Glycogen storage disease (type IV)
n. Idiopathic
B. Diagnosis. A specific diagnosis generally requires a combination of history, physical
findings, laboratory tests, and the identification of characteristic histologic features.
438
79466_CH56.qxd 1/2/08 1:36 PM Page 439
Intrahepatic
IV. VARICES. The thin-walled varices in the lower esophagus and upper stomach may
bleed extensively and constitute the major complication of portal hypertension.
Variceal bleeding occurs without an obvious precipitating cause and presents usually
as a painless massive hematemesis or melena.
Variceal bleeding primarily reflects portal hypertension. The role of acid reflux
and its contribution to initiation of variceal bleeding is not clear. Even though there is
no clear agreement as to whether bleeding correlates with the severity of portal hyper-
tension, it is generally accepted that hemorrhage usually is seen with a portal pressure
above 12 mm Hg and is more likely in patients with large varices.
A. Diagnosis. The presence of varices may be detected by barium swallow and upper
gastrointestinal (GI) series (40% sensitivity), angiography, and endoscopy. Upper
GI endoscopy is preferred; it not only shows the presence and size of the varices
but also reveals whether they are the sites of bleeding. Forty percent of the bleed-
ing in patients with cirrhosis with known varices has a nonvariceal source.
Congestive or portal hypertensive gastropathy is a major source of bleeding in
these patients.
B. Prognosis. Once esophageal varices are diagnosed, the risk of bleeding ranges
from 25% to 35% within 1 year of diagnosis of large varices. Risk factors for
variceal bleeding include size of varices, severity of liver disease, and presence of
active alcohol consumption. The overall mortality of variceal bleeding is 70% to
80% in patients with cirrhosis. The prognosis is dependent on the patient’s nutri-
tional status, presence or absence of ascites; encephalopathy, bilirubin level, albu-
min level, and prothrombin time (see modified Child’s criteria in Table 56-2).
C. Treatment. Prompt care of the patient with massive hematemesis or melena from
bleeding esophageal or gastric varices requires coordinated medical and surgical
efforts.
1. Transfusion. The first step is to ensure adequate circulation with transfusion
of blood, fresh-frozen plasma, and, if necessary, platelets. Because patients with
79466_CH56.qxd 1/2/08 1:36 PM Page 441
Scorea
Variable 1 2 3
liver disease often have deficiency of clotting factors, the infusion of fresh blood
or fresh-frozen plasma is important.
2. Endoscopy or angiography. After the vital signs are stabilized, the site
and cause of the bleeding should be established by endoscopy. If bleeding is
too brisk and endoscopic diagnosis is not possible, angiography may be per-
formed to determine the site of bleeding and the vascular anatomy of the portal
circulation.
3. Choice of therapeutic method. Once the diagnosis of active variceal bleeding
is made, there are several therapeutic options. The treatment of choice is endo-
scopic sclerotherapy or endoscopic variceal banding.
If these methods are not immediately available, medical drug therapy, bal-
loon tamponade, or transhepatic variceal obliteration may be used. Surgical
therapy with a portosystemic shunt (PSS) carries a very high mortality but may
be lifesaving. Transjugular introduction of an expandable stent (transjugular
intrahepatic portosystemic shunt [TIPS]) into the liver may create a PSS with
much less morbidity or mortality.
4. Endoscopic sclerotherapy, the direct injection of a sclerosing agent into the
esophageal varices, is effective in the immediate control of variceal bleeding.
This technique is preferentially used as an initial therapy before the infusion of
vasopressin or balloon tamponade. The sclerosants most commonly used are
tetradecyl, sodium morrhuate, and ethanolamine oleate. The sclerosing agent is
injected directly into the variceal wall or into the mucosa between the varices.
It causes clotting of the varices and a severe necrotizing inflammation of the
esophageal wall followed by a marked fibrotic reaction.
After control of the bleeding, the endoscopic sclerotherapy is repeated at
weekly or monthly intervals until the varices are totally obliterated, leaving a
scarred esophagus. Sclerotherapy of gastric varices has not been shown conclu-
sively to be effective and may result in gastric ulceration. The complications of
endoscopic sclerotherapy of esophageal varices include ulceration, hemorrhage,
perforation, stricture, and pleural effusion. Sclerotherapy controls acute variceal
hemorrhage in 80% to 90% of patients. Chronic sclerotherapy that obliterates
the esophagus and varices decreases the risk of rebleeding.
5. Endoscopic banding of esophageal varices has been shown to be as effec-
tive as or slightly more effective than injection sclerotherapy in the initial treat-
ment of bleeding esophageal varices. The technique requires expertise and a
79466_CH56.qxd 1/2/08 1:36 PM Page 442
V. Ascites refers to the accumulation of excessive volumes of fluid within the peritoneal
cavity. Cirrhosis is the most common cause of ascites. The other main causes are infec-
tion (acute and chronic, including tuberculosis), malignancy, pancreatitis, heart fail-
ure, hepatic venous obstruction, nephrotic syndrome, and myxedema.
A. Pathogenesis. Ascites forms because of an imbalance between the formation and
resorption of peritoneal fluid. Distribution of fluid between vascular and tissue
spaces is determined by the equilibrium of hydrostatic and oncotic pressures in the
two compartments. The accumulation of fluid in the peritoneal cavity of patients
with cirrhosis results from an interaction of a number of factors:
1. Portal hypertension with increased total splanchnic plasma volume.
2. Renal changes favoring increased sodium and water resorption and retention
include the following:
a. Stimulation of the renin-aldosterone system
b. Increased antidiuretic hormone release
c. Decreased release of “natriuretic” hormone or third factor
3. Imbalance in the formation and removal of hepatic and gut lymph.
Lymphatic drainage (removal) fails to compensate for the increased lymph leak-
age, mainly due to elevated hepatic sinusoidal pressure, leading to formation of
ascites.
4. Hypoalbuminemia from decreased hepatic synthesis, which results in decreased
intravascular oncotic pressure. The leakage of albumin through lymph into the
peritoneal cavity increases the intraperitoneal oncotic pressure, favoring ascites
formation.
5. Elevated plasma vasopressin and epinephrine levels. These hormonal responses
to a volume-depleted state further accentuate renal and vascular factors.
B. Diagnosis
1. Clinical presentation. The physical findings of ascites include the puddle sign,
bulging flanks, flank and shifting dullness, and fluid wave. Abdominal or umbil-
ical herniation, penile or scrotal edema, or right-sided pleural effusion may be
present.
Detection of more than 2 L of ascites is not difficult, but less than this
amount is not always ascertained by physical examination. The percussion
maneuvers require the presence of at least 500 mL of fluid. The diagnostic accu-
racy of all the maneuvers described is only about 50%. The maneuver to deter-
mine flank and shifting dullness is probably the most reliable.
2. Radiologic studies
a. Plain abdominal x-ray films may reveal general haziness of the abdomen
with loss of the psoas shadow. Usually, there is centralization and separation
of bowel loops.
b. Ultrasonography can detect as little as 30 mL of ascitic fluid with the
patient in the right lateral decubitus position. Free and loculated collections
can be identified.
c. Computed tomography (CT) scans of the abdomen may detect small
amounts of ascites and at the same time evaluate intraabdominal anatomy,
giving important information on the size and state of the liver and spleen
and on the presence or absence of varices and tumors.
3. Characterization of ascitic fluid
a. Paracentesis. A diagnostic paracentesis should be performed for new onset
ascites or when a complication is suspected. The procedure is performed
under sterile conditions using a 20- to 23-gauge angiocatheter. The most
frequent site of puncture is on the linea alba slightly below the umbilicus.
The iliac fossa also may be used. Paracentesis rarely leads to serious compli-
cations (1%), which include bowel perforation, hemorrhage, and persis-
tent ascitic fluid leakage. It is much safer to perform the paracentesis under
79466_CH56.qxd 1/2/08 1:36 PM Page 445
VII. HYPERSPLENISM. Portal hypertension and obstruction to portal flow may cause
enlargement and engorgement of the spleen. The resulting hypersplenism causes
sequestration and destruction of all or some of the blood cells, leading to decreased
red cell survival, neutropenia, and thrombocytopenia. These hematologic abnormali-
ties are usually not severe and do not require specific therapy.
In patients with thrombocytopenia who require a surgical procedure, a normal
bleeding time is a good index of adequacy of hemostasis. If it is abnormal, cryopre-
cipitate and platelet transfusions should be administered prior to surgery.
Splenectomy is not routinely recommended for patients with hypersplenism. The
effect of surgical portal decompression on hypersplenism is unpredictable.
was directly related to the severity and duration of PSE. It is thought to be a meta-
bolic disorder, and in most patients the cerebral dysfunction is reversible.
However, in patients with hepatocerebral degeneration and spastic paraparesis, it
may be progressive. Reduced neuronal activity and brain energy metabolism and
decreased brain glucose utilization in PSE is due to neurotransmission failure,
rather than primary energy failure.
The shunting of blood from the gut directly to the systemic circulation,
bypassing the liver, due to portal hypertension and the poor function of the dis-
eased liver, allows the accumulation of various “toxins,” leading to deranged cere-
bral function. The various toxins and states implicated in the pathogenesis of
hepatic encephalopathy include ammonia, -aminobutyric acid (GABA), mercap-
tan, short-chain fatty acids, false neurotransmitters, an imbalance of excitatory
and inhibitory neurotransmitters, and the reversed ratio of serum aromatic amino
acids to branched-chain amino acids.
Several pathogenetic mechanisms involving neurotransmitter systems have
been proposed to explain hepatic encephalopathy, including the following:
1. Ammonia neurotoxicity (most commonly accepted theory)
a. Direct effects of ammonia on inhibitory and excitatory synaptic transmission
b. Glutamate synaptic dysregulation due to astrocytic changes
c. Increased serotonin turnover (especially in preclinical PSE)
2. The GABA-benzodiazepine system
a. Increased brain GABA uptake
b. Increased GABA-ergic tone
c. Presence of “endogenous” benzodiazepines
3. Other neurotransmitter systems (acetylcholine, histamine, taurine, opioid system,
other peptides).
Moreover, in these patients, other metabolic disturbances such as
azotemia, electrolyte abnormalities, acid-base disorders, such as hypokalemic
alkalosis, and infections may also contribute to the cerebral dysfunction. In
addition, in patients with cirrhosis and PSE, the central nervous system (CNS)
appears to be very sensitive to the sedative effects of drugs. Decreased hepatic
drug metabolism with subsequent accumulation of the drug, decreased binding
of the drug to decreased plasma proteins (making more free drug available to
penetrate the altered, “more porous” blood-brain barrier), and enhanced cere-
bral receptor sensitivity to sedative drugs such as benzodiazepines and barbitu-
rates probably combine to increase susceptibility of the patient to development
of encephalopathy and coma.
B. Diagnosis
1. Clinical presentation. The diagnosis of hepatic encephalopathy is clinical and
depends on documentation of the presence of mental status changes, fetor
hepaticus, and asterixis in a patient with parenchymal liver disease. Fetor
hepaticus refers to the feculent-fruity odor of the breath in these patients.
Asterixis is a flapping motion of the hands caused by intermittent loss of exten-
sor tone. It is best elicited by asking the patient to outstretch his or her arms,
with hands extended at the wrist and fingers separated. The flap is maximal
with sustained posture. The motor disturbance can also be detected in tightly
closed eyelids, pursed lips, and protruded tongue. Asterixis is not specific for
hepatic encephalopathy. It can also occur in patients with uremia, severe pul-
monary disease, and sedative overdose.
a. Stages. Hepatic encephalopathy can be subclinical or overt; the overt form
can be classified roughly into four stages of increasing severity as follows:
i. Stage I. Patients exhibit inappropriate behavior, altered sleep pattern,
loss of affect, depression, or euphoria. There is usually asterixis and dif-
ficulty with writing and other fine motor skills.
ii. Stage II. Patients are moderately obtunded, confused, and disoriented.
There is accompanying fetor hepaticus and asterixis.
iii. Stage III. Patients are stuporous with marked confusion. They are
barely responsive to painful stimuli. If it can be elicited, asterixis should
79466_CH56.qxd 1/2/08 1:36 PM Page 451
Since the cost of these formulations is very high, they may be reserved for
patients in whom other measures have failed.
h. Zinc. Zinc deficiency has been suggested to play a role in hepatic encephalopa-
thy, but few data are available to support this hypothesis. Zinc sulfate and zinc
acetate supplementation have been studied in limited trials. Zinc histidine holds
promise as a superior oral preparation to use in future trials.
i. Benzodiazepine receptor antagonists such as flumazenil (Romazicon)
have been used in several trials. Sixty to seventy percent of patients improved
after IV flumazenil administration. Larger trials are under way at this time.
j. Surgical considerations
i. Portosystemic shunts. Distal splenorenal shunts, portacaval H-graft
shunts, mesocaval calibrated shunts, and TIPS have been successful in
decompressing portal hypertension and ameliorating PSE. All of these
special shunt procedures do not completely decompress the preexisting
portal hypertension and allow hepatic perfusion.
ii. Liver transplantation. Intractable hepatic encephalopathy unmanage-
able by medical treatment is a clear indication for liver transplantation.
Nontransplant surgical procedures, with the exception of TIPS, should be
considered only for patients who are not transplant candidates.
iii. Other procedures. Patients with demonstrated large, spontaneous, or
surgically created PSSs and intractable hepatic encephalopathy some-
times benefit from the closure of the shunt. This operation should be
accompanied by additional procedures to prevent variceal bleeding.
Shunt occlusion by radiologic techniques has been described. Duplex
ultrasound may be used instead of angiography to identify patients
with large spontaneous PSSs. Portosystemic shunts, especially end-to-
side shunts and sometimes side-to-side shunts, can induce hepatic
encephalopathy. Colonic exclusion has been used in selected patients for
the management of severe post–PSS hepatic encephalopathy.
Selected Readings
Cirrhosis
Fellowfield JA, et al. Reversal of fibrosis: no longer a pipedream? Clin Liver Dis.
2006;10:481–497.
Friedman SL. Reversibility of hepatic fibrosis and cirrhosis. Is it all hype? Nat Clin Pract
Gastroenterol Hepatol. 2007;4:236–237.
Groszman R, et al. Measurement of portal pressure: when, how, and why to do it. Clin
Liver Dis. 2006;10:499–512.
Tripodi A, et al. Abnormalities of hemostasitis in chronic liver disease. Reappraisal of their
clinical significance and need for clinical and laboratory research. J Hepatol.
2007;46:727–733.
Variceal Bleeding
Sarin SK, et al. Comparison of endoscopic ligation and propranolol for the primary
prevention of variceal bleeding. N Engl J Med. 1999;340:988.
Hepatorenal Syndrome
Arroyo V, et al. Advances in the pathogenesis and treatment of type-1 and type-2
hepatorenal syndrome. J Hepatol. 2007;46(5):935–946.
Moreau R, et al. The use of vasoconstrictors in patients with cirrhosis: type 1 HRS.
Hepatology. 2006;43(3):385–394.
Salerno F, et al. Diagnosis prevention and treatment of hepatorenal syndrome in cirrhosis.
A consensus workshop of the international ascites club. Gut. 2007;56(9):1310–1318.
Wong F, et al. Midodrive, octreodde, albumin and TIPs in selected patients with cirrhosis
and type 2 hepatorenal syndrome. Hepatology. 2004;40(1):55–64.
Hepatic Encephalopathy
Nutrition
Hepatocellular Carcinoma
I. LIVER TUMORS. The liver filters both arterial and portal venous blood and thus is a
major site for the spread of metastatic cancers, particularly those that originate in the
abdomen. Metastatic liver tumors can develop after the primary tumor has been iden-
tified, or patients can present initially with the signs and symptoms of metastatic liver
disease. Common primary tumors that metastasize to the liver are colon, pancreas,
stomach, breast, lung, gallbladder, and bile duct tumors, and lymphoma.
The most common primary liver cancer is hepatoma, or hepatocellular carci-
noma (HCC). HCC often is a consequence of cirrhosis. Worldwide, chronic hepatitis
B and C are major causes of HCC. Other primary liver malignancies are fibrolamellar
carcinoma and cholangiohepatocellular carcinoma and sarcoma, including angiosar-
coma, leiomyosarcoma, fibrosarcoma, and mesenchymal sarcoma.
Benign tumors of the liver include hepatic hemangioma, adenoma, focal nodular
hyperplasia, and focal regenerative hyperplasia.
A. Clinical presentation
1. History. Primary hepatoma often occurs in the setting of established cirrhosis
of any cause. In fact, abrupt deterioration of a patient with known cirrhosis is
a signal to consider the possibility of HCC. Other pathogenic antecedents of
HCC include hepatitis C virus, chronic hepatitis B virus infection (HBV) with
or without cirrhosis, exposure to aflatoxins in food (implicated in parts of
Africa and Asia), and exposure in the distant past to thorium dioxide, a radi-
ologic contrast material, nonalcoholic steatohepatitis (NASH) associated with
obesity and diabetes mellitus. Uncommon consequences of HCC include fever
of unknown origin, portal vein thrombosis, hypoglycemia, polycythemia,
hypercalcemia, porphyria, and dysglobulinemia.
The incidence of HCC has doubled in the United States for the past
20 years and continues to increase due to HCV- and HBV-related complications
and NASH-related cirrhosis.
The mean age at the time of diagnosis of HCC in the United States is
65 years. Seventy-four percent of cases occur in men. In patients younger than
40 years of age who present with liver cancer, only a third are HCC; others
include fibrolamellar carcinoma, which has a much better prognosis, and
metastatic cancer.
In patients with metastatic liver cancer, the primary lesion may not be
known. Thus the initial presentation may be due to the metastatic disease to the
liver. About half of the patients who die from malignant disease have metas-
tases in the liver at postmortem examination.
Abdominal pain is a common complaint of patients with primary HCC or
metastatic liver disease. Some patients have nonspecific complaints, such as
anorexia, weight loss, and malaise.
2. Physical examination. The liver typically is enlarged and nodular and may be
tender. Ascites often has developed. A friction rub heard over the liver with respi-
ration indicates involvement of the liver capsule. Rarely, a bruit is heard, reflecting
the vascular nature of most HCC and some metastatic tumors. Jaundice usually
develops later in the course of both HCC and metastatic liver disease. If jaundice
is present initially, it means that preexisting liver disease is present, the tumor
involves much of the liver parenchyma, or a large bile duct is obstructed.
455
79466_CH57.qxd 1/2/08 1:36 PM Page 456
B. Diagnostic studies
1. Laboratory studies. Anemia is common in patients with liver cancer. It may be
the nonspecific anemia of chronic disease or the macrocytic anemia associated
with a chronic liver disorder. Bilirubin is elevated for the same reasons that the
patient may be jaundiced (see section I.A.2). Elevation of alkaline phosphatase
is common simply because obstruction of even the small biliary radicals causes
generation and release of this enzyme. If the source of an elevated alkaline
phosphatase level is in question, a 59-nucleotidase elevation will confirm its
origin in the liver. Often mild elevations of aspartate aminotransferase (AST)
and alanine aminotransferase (ALT) occur.
The serum alpha-fetoprotein concentration is elevated in about half of the
patients with hepatoma in the United States; thus, the measurement is useful in
helping make the diagnosis. However, some patients with gonadal malignancies
and metastatic disease to the liver also have elevated serum alpha-fetoprotein.
2. Radiologic studies include ultrasound, computed tomography (CT), magnetic
resonance imaging (MRI), dimethylphenylcarbamylmethyliminiodiacetic acid
(HIDA) scan, and sulfur colloid liver scan.
Many physicians proceed immediately to CT in the evaluation of liver
tumors after the initial blood chemistry studies have been done. The CT scan
has the advantage of not only providing accurate information about the liver
but also identifying enlarged lymph nodes and other abnormalities of the
abdominal organs. Furthermore, a CT-guided needle biopsy of a liver lesion or
other abdominal mass may provide important diagnostic information.
3. Liver biopsy. Percutaneous liver biopsy is diagnostic of liver cancer in about
80% of patients in whom alkaline phosphatase is elevated due to intrahepatic
cancer. Biopsy of the liver under direct laparoscopic vision is an alternative to
percutaneous liver biopsy. Laparoscopy also can assess the spread of tumor to
the peritoneum, lymph nodes, and other abdominal organs.
4. Angiography. Celiac axis angiography can determine operability in a patient
with a HCC or with a solitary metastatic lesion to the liver. If the CT scan sug-
gests tumor in both lobes of the liver, arteriography is not indicated. An arteri-
ogram is helpful in differentiating a benign hemangioma from a malignant
tumor when the CT scan suggests a vascular lesion.
C. Natural history and treatment. Survival of patients diagnosed with HCC remains
very poor. Survival is related to tumor size, liver function, and the receipt of effec-
tive and potentially curative treatment (resection or orthotopic liver transplanta-
tion [OLT]). In recent years, living-donor partial liver transplantation (LT) has
made transplantation more available for this indication. Five-year survival rates
after OLT may be as high as 70% for a Child class A patient with one tumor less
than 2 to 5 cm in diameter. Patients with intermediate to advanced HCC tend to
have uniformly poor prognosis.
1. Surgical resection is possible only for a small number of patients. Tumor
spread, poor liver function, and the presence of portal hypertension preclude
patients for surgery. Even with resectable tumors, there is a high recurrence
rate. This increases with large tumors, those with vascular invasion, and those
with poor differentiation. Well-differentiated and encapsulated HCC especially
less than 3 cm in size has been associated with lower recurrence rates after
resection.
2. Liver transplantation (OLT or living-donor partial LF) may be offered to
patients with HCC. The eligibility criteria include solitary tumors less than 5
cm or up to three tumors each less than 3 cm. Recurrence is much lower than
that of resection if these criteria are used.
3. Local ablation using US- or CT-guided percutaneous ethanol injection (PEI) or
radiofrequency ablation have become established modes of therapy for patients with
HCC with no significant coagulopathy or ascites. For patients with a single tumor
less than 3 cm, HCC recurrence-free survival is similar to that surgical resection.
4. Palliative therapy is used for patients not suitable for potentially curative
therapy. Unfortunately, several studies have confirmed no survival benefit with
79466_CH57.qxd 1/2/08 1:36 PM Page 457
II. BILIARY TUMORS. Biliary tumors include those that arise in the gallbladder and in the
intra- and extrahepatic biliary system. Benign tumors are rare and include papilloma,
leiomyoma, lipoma, myxoma, and fibroma. On the other hand, adenocarcinomas of the
gallbladder and bile ducts (cholangiocarcinomas) are not rare, accounting for about
5% of all malignancies.
A. Carcinoma of the gallbladder (CG)
1. Epidemiology. CG is extremely rare in the United States. However, CG is the
most common gastrointestinal malignancy in the Native Americans living in the
Southwest and in Mexican Americans. The incidence is highest in Chile and
Bolivia. The risk is higher in women and in the elderly. If it is diagnosed early or
incidentally on routine cholecystectomy for gallstone disease, the prognosis is
very good. Incidental CG is noted in 1% to 3% of cholecystectomy specimens.
2. Risk factors for CG include gallstones, chronic cholecystitis, and porcelain
gallbladder. Adenomatous polyps of the gallbladder may progress to cancer.
Polyps smaller than 1 cm seldom undergo malignant change. Other risk factors
for CG include congenital biliary cysts, anomalous drainage of the pancreatic
duct to the common bile duct (CBD), and chronic infection of the gallbladder
with Salmonella.
CG associated with gallstones are most commonly seen with large gall-
stones (2.5 cm) and in patients with long duration of gallstone disease. Use of
certain drugs (i.e. isoniazid, methyldopa, and oral contraceptive) may also be
associated with CG.
3. Clinical picture. Most patients with CG may present with symptoms of gallstone
disease (i.e., right upper quadrant pain, malaise, anorexia, nausea, vomiting, weight
loss, and jaundice). Unfortunately, most patients at the time of diagnosis have tumors
that have invaded adjacent organs and lymph nodes and may even have distant
metastases. The 5-year survival for such patients is usually less than 5%.
4. Diagnosis is usually obtained by imaging techniques including abdominal US,
CT, MRI (magnetic resonance imaging), and MRCP (magnetic resonance
cholangiopancreatography). Tumor markers such as CEA and CA-19-9 are not
helpful.
5. Prognosis and treatment. Prognosis depends on the stage of CG. If the CG is
thought to be a T1 lesion, surgical resection (i.e., simple cholecystectomy) may
be sufficient. Advanced cases may require radical resection. Postoperative radi-
ation and chemotherapy may reduce rates of recurrence.
B. Cholangiocarcinoma (CGC)
1. Epidemiology. CGC arises from the epithelium of the intra- or extrahepatic
biliary ductular system. It is less common than HCC and tends to occur at an
older age than HCC. The incidence is similar in both men and women.
2. Risk factors of CGC include primary sclerosing cholangitis (PSC), ulcerative
colitis, inflammatory bowel disease (IBD), choledochal cysts, biliary ductal
ectasia (Caroli’s disease), intrahepatic gallstones, biliary duct infections (chronic
cholangitis), infections with Clonorchis sinensis and Opisthorchis viverrini,
multiple biliary papillomatosis, bile duct adenoma, and exposure to thorotrast
(contrast agent used in the past for radiologic imaging).
3. Clinical features depend on whether CGC is central or peripheral.
a. Central CGC arises in major bile ducts and is often associated with chronic
inflammation of the bile ducts ( i.e., with PSC). Klatskin tumor arises in the
bifurcation of the CBC. Central CGC is often present with obstructive jaundice.
b. Peripheral CGC is rarely associated with PSC or cirrhosis. It often presents
with weight loss and abdominal pain.
c. Mixed HCC and CGC may be found in association with cirrhosis.
4. Diagnosis. Abnormal laboratory tests include elevated levels of alkaline phos-
phatase, bilirubin, 5nucleotidase, -glutamyltransferase, and serum CA19-9
above 100 mm /mL.
79466_CH57.qxd 1/2/08 1:36 PM Page 458
Selected Readings
Bruix J, et al. Management of hepatocellular carcinoma. Hepatology. 2006;42:1202–1236.
Case Records of the Massachusetts General Hospital (Case 13-2006). A man with bone
mass and lesions in the liver. N Engl J Med. 2006;354:1828–1837.
Colli A, et al. Accuracy of ultrasonography, spiral CT, magnetic resonance and alfa-etoprotein
diagnosing hepatocellular carcinoma. A systematic review. Am J Gastroenterol. 2006;
101:513–523.
deGroen PC. Cholangiocarcinoma: Making the diagnosis. Clin Perspect Gastroenterol.
2001;4:77–89.
DeVreede T, et al. Prolonged disease five-year survival after orthotopic liver transplantation
plus adjuvant chemo-irradiation for cholangiocarcinoma. Liver Transplant. 2000;6:399.
DiBisceylia AM. Liver tumors. Clin Liver Dis. 2001;5:1–286.
Geahigan TA. Primary hepatic lymphoma: A case report and discussion. Pract Gastroenterol.
2001;25:58.
Heimbach JK, et al. Liver transplantation for perihilar cholangiocarcinoma after aggressive
neoadjuvant therapy: a new paradigm for liver and biliary malignancies? Surgery.
2006;140:331–334.
Koslin DB. Hepatocellular carcinoma. Rev Gastroenterol Disord. 2001;1:58.
Larsson SC. Coffee consumption and risk of liver cancer: A metaanalysis. Gastroenterology.
2007 May;132:1740–1745.
Llover JM, et al. Resection and liver transplantation for hepatocellular carcinoma. Seminar
Liver Dis. 2005;25:181–200.
Llovert JM, et al. Hepatocellular carcinoma. Lancet. 2003;362:1907–1917.
Lopez PM, et al. Systematic review: evidence based management of hepatocelluar
carcinoma—an updated analysis of randomized controlled trials. Ailment Pharmacol
Ther. 2006;23:1537–1547.
Rea DJ, et al. Liver transplantation with neoadjuvant chemoradiation is more effective
than resection for hilar cholangiocarcinoma. Ann Surg. 2005;242:451–461.
Sheth S, et al. Primary gallbladder cancer: Recognition of risk factors and the role of
prophylactic cholecystectomy. Am J Gastroenterol. 2000;95(6):1402–1409.
79466_CH58.qxd 1/2/08 1:36 PM Page 459
LIVER TRANSPLANTATION 58
Figure 58-1. Anastomoses in auxiliary partial liver transplantation. VC, inferior vena cava; PV,
portal vein; AO, aorta; CJ, choledochojejunostomy. (From Terpstra OT, et al. Auxiliary partial liver
transplantation for end-stage chronic liver disease. N Engl J Med. 1988;319:1507. Reprinted with
permission. Copyright © 1988 Massachusetts Medical Society. All rights reserved.)
459
79466_CH58.qxd 1/2/08 1:36 PM Page 460
Figure 58-2. Orthotopic liver transplantation. Biliary reconstruction can be accomplished through
choledochojejunostomy or duct-to-duct anastomosis (inset). (From Starzl TE. Liver transplantation, II.
N Engl J Med. 1989;321:1092. Reprinted with permission. Copyright © 1989 Massachusetts Medical
Society. All rights reserved.)
Figure 58-3. Diagram of appearance after orthotopic liver transplantation. Sequence of anasto-
moses is as follows: upper vena caval anastomosis above the liver (1), lower vena caval anastomosis
beneath the liver (3), end-to-side hepatic artery anastomosis (4), portal vein anastomosis (2), and com-
mon bile duct anastomosis over a T tube (5). (From Krom RAF. Liver transplantation at the Mayo Clinic.
Mayo Clin Proc. 1986;61:278–282. Reprinted with permission.)
460
79466_CH58.qxd 1/2/08 1:36 PM Page 461
The second factor is due partly to inadequate identification of potential organ donors
by physicians. Members of the medical community, especially physicians dealing with
patients who may require transplantation (internists, nephrologists, gastroenterolo-
gists, cardiologists), must begin to increase their own and their patients’ awareness of
the need for donor organs.
In an attempt to increase the number of organ donors, 27 states have passed leg-
islation for required request. This legislation obligates hospitals to have policies in place
to offer organ donation as an option to families of patients dying in that hospital.
These laws are designed to relieve the attending physician of the burden of requesting
organ donation from grieving families by making the request for organ donation part
of the routine hospital policy.
Many patients do not receive a liver transplant because they are never referred to
a transplantation center. Other patients are referred only after their liver disease has
reached its terminal stage, and they often die before a suitable donor organ can be
found. To allow all appropriate patients the opportunity to undergo liver transplanta-
tion and survive the procedure, physicians must be aware of the criteria for candidacy
and for timing of the referral.
There are more than 30 active centers in which liver transplant operations are
performed in the United States. For the success of this procedure, centers must main-
tain high standards and achieve and maintain acceptable 1- and 5-year survival rates.
Centers performing fewer than 10 liver transplantations each year are unlikely to
maintain the necessary expertise for optimal management of patients with this
extremely complex disease. Currently 1-year survival rates range from 80% to 95%
and 5-year survival rates are 80% to 85%. The dramatic increase in the survival rates
results from the following factors:
A. Advances in standardization and refinement of the transplantation procedure.
B. Expertise of anesthesiologists in the prevention and treatment of the metabolic
abnormalities that occur in patients with end-stage liver disease during the trans-
plantation procedure.
C. Use of the venovenous bypass system, which ensures venous return to the heart
from both the portal and systemic venous systems during the hepatic phase of the
transplantation procedure, reduces blood loss, decreases the incidence of postop-
erative renal failure, and generally results in less hemodynamic instability during
the procedure.
D. Improved techniques for the identification and support of potential organ donors.
E. Refinement of operative techniques for the recovery and preservation of the donor
livers.
F. Use of more effective and less toxic immunosuppressive regimens. It is important to
remember that successful liver transplantation does not return a patient to normal.
Rather, a new disease, a “transplanted liver,” replaces the former disease. However,
this new state allows patients a chance for both long-term survival and a more nor-
mal lifestyle than were possible during the late stages of their liver disease. After
liver transplantation, patients must take immunosuppressive medications for the
remainder of their lives. Discontinuation of the prescribed medications may lead to
rejection and rapid deterioration in the patient’s condition.
II. CANDIDACY (PATIENT SELECTION). In the past, patients were referred for liver
transplantation only after their disease had reached its end stage. For best results,
however, earlier referral to a transplantation center is desirable for all appropriate
patients. Because of the variable course of many liver diseases, the determination of
the most appropriate time for referral of any patient for transplantation is difficult.
A. Criteria. There are three general criteria used in most transplantation centers.
These are as follows:
1. The unavailability of other surgical or medical therapies that offer the patient
an opportunity for long-term survival.
2. Absence of complications of chronic liver disease that may significantly increase
the patient’s operative risk or lead to the development of absolute or relative
contraindications to transplantation.
79466_CH58.qxd 1/2/08 1:36 PM Page 462
Adults Children
Absolute contraindications
Extrahepatic malignancy
Advanced cardiopulmonary disease
Acquired immunodeficiency syndrome (AIDS)
Active substance abuse
Other organ system failure not curable with hepatic transplantation
Relative contraindications
Active sepsis outside the hepatobiliary tract
Renal insufficiency/failure
Advanced protein-calorie malnutrition
Portal vein thrombosis
Operative procedures, such as end-to-side portacaval shunt or complex hepatobiliary surgery
Previous extrahepatic malignancies
III. EVALUATION
A. Goals. Once the attending physician identifies a patient as a potential candidate
for liver transplantation, he or she refers the patient to a transplant center where
the patient undergoes a thorough evaluation to satisfy four specific goals:
1. The establishment of a specific diagnosis
2. Documentation of the severity of the disease
79466_CH58.qxd 1/2/08 1:36 PM Page 465
Figure 58-4. Venovenous bypass system in place during the anhepatic phase of the transplant
procedure. Venous blood from the femoral and portal veins is returned to the heart via the axillary
vein. (From Griffith BP, et al. Venovenous bypass without systemic anticoagulation for transplantation
of the human liver. Surg Gynecol Obstet. 1985;160:270. Reprinted with permission.)
79466_CH58.qxd 1/2/08 1:36 PM Page 467
transplanted liver. Measurements of blood amino acid clearance and other prod-
ucts of intermediary metabolism have been used to distinguish between patients in
whom the new liver is expected to recover and those in whom it is not.
Host immune factors and hyperacute rejection may result in primary failure of
the liver graft. Compared to other transplanted organs such as kidney and heart, the
human liver seems to be more resistant to antibody-mediated injury. Because of this
resistance, liver transplantation is often performed in spite of major blood-group
incompatibilities. However, a progressive and severe coagulopathy that develops
shortly after hepatic revascularization should arouse suspicion of an accelerated
rejection.
C. Nontechnical complications
1. Hypertension is almost universally present in a transplant recipient, especially
in the early postoperative period, and requires aggressive treatment. The cause
of the hypertension is probably multifactorial and includes the use of
cyclosporine. Blood pressure control often requires at least two antihyperten-
sive medications, usually a vasodilator and a beta-blocker. In addition, patients
need diuretics both for blood pressure control and to relieve the ascites that
usually develops after transplantation. Hypertension is usually less severe after
the first 6 months after transplantation, and many patients require no antihy-
pertensive medications by 1 year after the procedure.
2. Infection. Serious bacterial, viral, and fungal infections may occur in “routine”
liver transplantation patients up to several years after the transplant operation.
However, infectious complications are much more common in patients who
require multiple reoperations or who have poor initial function of the trans-
planted liver. Therefore, fever in any posttransplant patient requires a thor-
ough investigation to rule out infection. The initial workup of fever should
begin with a detailed history and physical examination and appropriate cultures
of blood, urine, and sputum. Suspected bacterial infections should be treated
aggressively with broad-spectrum antibiotics, which are modified when culture
results become available.
Fever may be one of the earliest signs of rejection of the transplanted liver,
and patients experiencing an acute rejection episode may complain of flulike
symptoms. Therefore, the workup of any fever must include a close examina-
tion of the patient’s most recent liver function tests.
Liver transplantation patients do not appear to be more susceptible to
common viral illnesses than other people. However, if the physical examination
reveals oral or cutaneous lesions compatible with either herpes simplex or
herpes zoster, the lesions should be cultured and acyclovir therapy instituted
for at least 2 weeks. When the cause of the fever cannot be identified or the
patient fails to respond to initial treatment, repeat cultures and serum titers for
cytomegalovirus and Epstein-Barr virus should be obtained and compared
to those obtained before transplantation to identify patients with acute viral
infections. Biopsy specimens, especially of liver, skin, and lung, should be cul-
tured and histologically examined for evidence of virus. Any patient with evi-
dence of a severe viral infection should return to the transplantation center for
treatment. Appropriate management includes careful reduction in the patient’s
immunosuppression and treatment with antiviral agents.
Opportunistic infections may occur in the posttransplant patient and
require prompt diagnosis and treatment. Pneumocystis carinii pneumonitis,
rare in the early posttransplantation period, most commonly presents between 3
and 6 months after surgery. In most patients, the presenting symptom is tachyp-
nea. Physical examination and chest x-ray are usually normal initially, but
arterial blood gases reveal moderate-to-severe hypoxemia. All patients with
suspected pneumocystic pneumonia should be started on intravenous trimetho-
prim/sulfamethoxazole and undergo immediate bronchoscopy with bron-
choalveolar lavage to confirm the diagnosis. A dramatic worsening of the
pulmonary status commonly occurs after treatment is initiated, and the patient
may require intubation and mechanical ventilation. Early diagnosis and treatment
79466_CH58.qxd 1/2/08 1:36 PM Page 469
usually result in a relatively brief, limited illness, whereas failure to make the
diagnosis expediently may prove fatal.
Other opportunistic infections include Cryptococcal meningitis, coc-
cidiomycosis, Listeria meningitis, and tuberculosis.
Because a reduction in the dosage of immunosuppressive medications is
essential, a transplant patient with a serious infection should return to the
transplantation center for treatment of the infection. Too rapid or aggressive a
reduction in the immunosuppressive regimen may put the patient at risk for
acute rejection. Full immunosuppressive therapy should be reinstituted as soon
as possible. If full immunosuppression is reinstituted before complete recovery,
however, the infection may recur; on the other hand, failure to resume full
immunosuppressive therapy at the appropriate time may lead to an acute rejec-
tion episode, which could result in loss of the transplanted liver.
3. Rejection. Despite advances in immunosuppression, rejection of the grafted
liver is one of the most common indications for retransplantation. Acute rejec-
tion episodes may occur at any time after transplantation but become less
frequent with time. Severe rejection episodes should be treated at the liver
transplant center, but mild rejection episodes usually can be handled by the
patient’s physician in consultation with the transplant surgeons.
After transplantation, the follow-up of the patient may continue at the
transplantation center or may be done by the patient’s physician. Initially, rou-
tine blood work is performed three times per week, including a complete blood
count with platelet count, electrolytes, blood urea nitrogen, creatinine, aspar-
tate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl
transpeptidase (GGTP), alkaline phosphatase, total and direct bilirubin, and a
cyclosporine trough level. If the patient remains stable, the frequency of routine
blood work may be decreased to once or twice a month by 6 months after
transplantation.
Most acute rejection episodes that occur in the late postoperative period are
due to either too rapid a reduction in immunosuppressive drugs or a decreased
cyclosporine level. The cyclosporine level depends on absorption from the small
intestine and is affected by vomiting, diarrhea, and interaction with other drugs
taken by the patient (e.g., phenytoin [Dilantin], ketoconazole, and rifampin).
Gastroenteritis and diarrheal illnesses may be associated with a rapid fall in the
patient’s cyclosporine level and may result in a severe rejection episode. Therefore,
these patients may require hospitalization for the administration of intravenous
cyclosporine and corticosteroids until the gastrointestinal function returns to nor-
mal. Drug interactions that may affect cyclosporine levels also should be closely
monitored. Patients with mild rejection episodes may be entirely asymptomatic or
may have flulike symptoms. Such episodes are generally diagnosed by a slight ele-
vation in any of the liver chemistry tests; after consultation with the transplant
surgeons, they are usually treated with a single 500- to 1,000-mg bolus of methyl-
prednisolone or with a short course (usually 5 days) of an increased dosage of oral
prednisone. If the liver chemistry tests improve, nothing else may be required
except a temporary increase in the frequency of laboratory tests.
Patients who fail to respond to steroid therapy or respond only tem-
porarily should be returned to the transplant center for additional investiga-
tion and treatment of the suspected rejection episode. The workup includes a
liver biopsy and an ultrasound study of the biliary tract to rule out obstruc-
tion, and Doppler ultrasound to ascertain the patency of the hepatic artery
and portal vein. The liver biopsy may differentiate rejection from cholangitis,
hepatitis, and ischemic injury, all of which may clinically mimic rejection.
Because therapy for most of these disorders is vastly different from treatment
of acute rejection, making an accurate and prompt diagnosis is essential.
Treatment of biopsy-confirmed rejection is initiated with a short course of
high-dose steroids. Patients who fail to respond to this therapy are usually given
a 10- to 14-day course of either antilymphocyte globulin or the monoclonal
antibody OKT3. Repeat liver biopsy to confirm continued rejection and to rule
79466_CH58.qxd 1/2/08 1:36 PM Page 470
out the other diagnoses should be performed if liver chemistry tests do not
improve after a course of either of these drugs. Patients in whom rejection is
confirmed should be treated with whichever therapy they have not yet received.
Cytomegalovirus hepatitis may develop during the antirejection therapy. A liver
biopsy is necessary to make this diagnosis so that appropriate antiviral therapy
may be initiated.
If repeat biopsy reveals a severely damaged liver secondary to rejection,
additional antirejection therapy is not administered, and the patient is considered
at high priority for retransplantation. Retransplantation should be considered for
any patient who continues to demonstrate severe liver dysfunction despite
adequate therapy for rejection.
Chronic rejection, in contrast to acute rejection, often fails to respond to
any form of antirejection therapy. It is a more indolent process, characterized
by a gradual, progressive, and unrelenting rise in the patient’s liver chemistry
tests. This form of rejection may occur at any time after transplantation, and
most of these patients eventually require retransplantation. Approximately
20% of all liver transplantation patients require retransplantation. The 1-year
survival rate after retransplantation is approximately 50%.
D. Immunosuppression. Most transplant centers use similar immunosuppression
protocols. The primary immunosuppressive agents currently used are cyclosporine
and corticosteroids.
1. Corticosteroids. Adults receive 1,000 mg of methylprednisolone intravenously
after revascularization of the donor liver. The dosage of methylprednisolone is
tapered rapidly over the first 6 days by 40 mg/day until a baseline of 20 mg/day
is reached. By day 6, most patients are able to tolerate oral medications, and
prednisone 20 mg/day is begun. Most adult patients are discharged with this
dosage of prednisone; depending on the frequency and severity of rejection
episodes, the dosage is reduced in 2.5 mg increments in the first year until a
baseline of 5 to 10 mg is reached in adult patients.
2. Cyclosporine and tacrolimus are the two agents used in combination with
prednisone.
a. Cyclosporine is given as a single dose (2 mg / kg IV) before the start of the
transplantation procedure, and the next dose is given immediately postop-
eratively when the patient is in the intensive care unit. If urinary output is
adequate, a dose of 2 mg / kg is administered q8h. The frequency of the
cyclosporine dose of 2 mg / kg may be reduced to twice daily in patients with
marginal urinary output or with prior evidence of renal dysfunction. Oral
cyclosporine is initiated at a dosage of 10 mg / kg twice daily as soon as ade-
quate gastrointestinal function has returned. When oral cyclosporine ther-
apy is begun, the intravenous dose of cyclosporine is gradually reduced and
eventually eliminated. All additional adjustments in dosage are based on
daily cyclosporine trough levels. A cyclosporine level of 1,000 ng /dL by
radioimmunoassay (RIA) is accepted as ideal. The intestinal absorption of
cyclosporine tends to be erratic, and the maintenance of stable blood levels
is sometimes difficult. The cyclosporine level is maintained at a level that is
proportionately lower than 1,000 ng/dL if toxicity develops.
In patients in whom cyclosporine is not well tolerated, it may be nec-
essary to add azathioprine 1.5 to 2.5 mg / kg daily to ensure adequate
immunosuppression. After a stable course of several months, the dosage of
cyclosporine may be reduced gradually to obtain a blood level of 500 to 800
ng /dL by 1 year after transplantation. During the late postoperative period,
it is difficult to predict how much the dosages of immunosuppressive med-
ications can be lowered without causing an acute rejection episode.
However, doses should be maintained at as low a level as possible because
most side effects of immunosuppressive drugs are dose related. Except for
the immunosuppressive regimen, which is usually managed indefinitely by
the transplant surgeon, care is resumed by the referring physician when the
patient returns home.
79466_CH58.qxd 1/2/08 1:36 PM Page 471
b. Side effects. The most common side effects directly related to cyclosporine
are hypertension, nephrotoxicity, hepatotoxicity, hirsutism, gum hyperpla-
sia, and a fine motor tremor.
i. Hypertension. The hypertension may be quite severe initially but
becomes less severe during the first year after transplantation. Often the
referring physician is left with the task of gradually reducing the patient’s
antihypertensive medications during the first year after transplantation.
ii. Nephrotoxicity. Some degree of nephrotoxicity is experienced by most
patients taking cyclosporine, but in most cases it is not clinically signifi-
cant. During the early postoperative period, some degree of nephrotoxi-
city is tolerated because the patient is most vulnerable to severe rejection
episodes during this time. This nephrotoxicity is usually reversible when
the dosage of cyclosporine is reduced. However, patients who require
continued high dosages of cyclosporine because of repeated rejection
episodes might have a significant deterioration in renal function after an
otherwise successful liver transplantation. Azathioprine may be added as
a third immunosuppressive drug in patients in whom severe nephrotoxi-
city develops so that the dosage of cyclosporine may be reduced. In most
liver transplant patients, however, close monitoring of the patient’s renal
function and cyclosporine trough levels allows the cyclosporine dosage
to be adjusted appropriately both to prevent rejection and to avoid sig-
nificant toxicity without the addition of a third drug.
iii. Hepatotoxicity is common when the patient’s cyclosporine level is
greater than 1,200 ng/dL; therefore, it is rarely seen in patients who take
a lower dosage of cyclosporine.
iv. Hirsutism, tremor, and gum hyperplasia. These side effects are less
serious but may be indicative of overimmunosuppression. If clinically
possible, the transplant surgeon may reduce the dose of the immunosup-
pressants. To prevent serious gum disease, all liver transplant patients
should receive frequent dental examinations. Excessive hair growth may
require the use of a depilatory, especially in young female patients with
body and facial hair.
Tacrolimus has similar pharmacologic properties as cyclosporine
and gives similar patient and graft survival rates. It is available as 0.5 mg,
1 mg, and 5 mg capsules for oral use and as a solution 5 mg/mL for par-
enteral use. The principal side effects include nephrotoxicity and neuro-
toxicity, hypertension, hyperbulemia, hyperglycemia, and nausea, vomit-
ing, and diarrhea.
Patients receiving cyclosporine and tacrolimus may develop lym-
phoprolerative disorders.
3. Carcinogenesis. The incidence of cancer is significantly increased in immuno-
suppressed patients. The most common cancers are carcinoma of the cervix,
vulva, perineum, shin, and lip; Kaposi’s sarcoma; and non-Hodgkin’s lym-
phoma. Most lymphomas appear to be related to infection with or reactivation
of the Epstein-Barr virus. In most patients, the presenting problem is fever,
lymphadenopathy, gastrointestinal perforation, obstruction, or hemorrhage.
The treatment of these lymphomas is controversial. Some authorities believe
that reduction of immunosuppression is sufficient, and others recommend
chemotherapy or radiation therapy or both. All physicians involved in the long-
term care of transplantation patients must be aware that these patients are at
increased risk of the development of some cancers (especially lymphomas). Any
patient suspected of having a lymphoma should be evaluated promptly and
referred back to the transplant center.
E. Long-term management of the liver transplant patient. Adult patients’ hospi-
tal stay for OLT is 7 to 14 days and children’s 10 to 20 days. After discharge,
patients are asked to remain near the transplant center for 2 to 6 weeks for inten-
sive monitoring of liver function and immunosuppression. They are also monitored
for the need for nutritional support, antibiotics, and/or antiviral therapy. Initially,
79466_CH58.qxd 1/2/08 1:36 PM Page 472
outpatient visits occur twice weekly, then weekly for the first month, then less
frequently. Laboratory tests are initially performed weekly, then monthly indefi-
nitely. In the outpatient setting, patient’s primary caregivers need to follow patient’s
symptoms, assess and encourage strict compliance with medications, perform
physical exams and appropriate laboratory testing, including a chemistry panel,
complete blood count, trough levels for cyclosporine, and tacrolimus at the indi-
cated intervals. Patients should be referred to the transplant center if at any time
rejection is suspected.
V. QUALITY OF LIFE. Even though a successful liver transplantation does not return a
patient to “normal” but to a life of indefinite immunosuppression, the operation
allows patients a chance for long-term survival and a much more productive lifestyle
than they experienced during the late stages of their liver disease. Most patients are
able to return to work in 3 to 6 months after OLT. Active physical fitness programs
and psychotherapeutic support facilitate patients’ return to normal daily activities.
Selected Readings
Alexander JW, et al. The influence of immunomodulatory diets on transplant success and
complications. Transplantation. 2005;79:460–465.
Clavien PA, et al. Strategies for safer liver surgery and partial liver transplantation. N Eng
J Med. 2007;356:1545–1550.
Di Bisceglie A. Pretransplant treatments for hepatocellular carcinoma: Do they improve
outcomes? Liver Transplantation. 2005;11:510–513.
Fausto N, et al. Liver regeneration. Hepatology. 2006;43(suppl 1):S45–S53.
Lee J, et al. Transplantation trends in primary biliary cirrhosis. Clin Gastroenterol Hepatol.
2007;5(11):1313–1315.
Levy M, et al. The elderly liver transplant recipient: A call for caution. Ann Surg. 2001;
233:107.
Mantel HT, Heimbach JK, et al. Vascular complications after orthotopic liver transplantation
following neoadjuvant therapy for hilar cholangiocarcinoma. Liver Transpl. 2006;82.
Middleton PF, et al. Living donor liver transplantation—adjust donor outcomes: a
systematic review. Liver Transpl. 2006;12:24–30.
Pomfret EA, et al. Liver and intestine transplantation in the United States. Am J Transpl.
2007;7:1376–1389.
Tan HP, et al. Adult living donor liver transplantation: Who are the ideal donor and
recipient: J Hepatol. 2005;43:13–17.
Tillman HL. Successful orthotopic liver transplantation. Gastroenterology. 2001;120:1561.
VanThiel D, et al. Liver transplantation: What the non-hepatologist should know. Pract
Gastroenterol. 2001;25(4):46.
79466_CH59.qxd 1/2/08 6:15 PM Page 473
A norexia nervosa and bulimia nervosa are eating disorders that are distinct but
related syndromes that have in common an intense preoccupation with food. Patients with
anorexia nervosa are characterized by a fear of becoming obese, disturbance in body
image, anorexia, extreme weight loss, and amenorrhea. On the other hand, bulimia is
characterized by periods of binge eating, alternating with fasting; self-induced vomiting;
and the use of diuretics and cathartics. Bulimic behavior is seen in some patients with
anorexia.
II. ETIOLOGY. The cause of these disorders has been attributed in part to the great
emphasis our society places on thinness. However, the disorders are caused by a com-
bination of genetic, neurochemical, psychological, and sociocultural factors.
Despite profound weight loss, patients with anorexia nervosa usually regard
themselves as being too fat. They may be reluctant to admit this and, in fact, often
agree to increase their food intake when prompted by relatives and friends. However,
they typically continue to avoid food, indulge in excessive physical exercise, and con-
sume medications to inhibit appetite and promote diuresis and catharsis.
Emotionally, the anorectic patient tends to be isolated and shuns relationships,
whereas the bulimic patient is likely to be outgoing and sociable. Weight may
fluctuate in bulimics because of the alternate food binging and purging, but the low
points in weight usually do not reach the point of dangerous weight loss that some-
times occurs in anorectics.
473
79466_CH59.qxd 1/2/08 6:15 PM Page 474
General Dry skin and hair, hair loss, altered temperature regulations
Cardiovascular Hypotension, abnormal electrocardiogram: bradycardia,
low voltage, prolonged Q interval, arrhythmias, mitral
valve prolapse, cardiomyopathy
Renal Edema, decreased glomerular filtration rate, increased
blood urea nitrogen level, renal calculi
Endocrine/Metabolic Electrolyte abnormalities, hypoglycemia, hypomagnesemia,
hyperphosphatemia, euthyroid sick syndrome, delay in
puberty, menorrhea, infertility, decreased estradiol and
testosterone, lipid abnormalities, osteoporosis
Gastrointestinal Constipation, rectal prolapse, decreased intestinal motility,
esophagitis, Mallory-Weiss tear, gastric dilatation,
delayed emptying, elevated liver enzymes, elevated
amylase level
Hematologic Anemia, leukopenia, thrombocytopenia, elevated
prothrombin time
Pulmonary Aspiration pneumonia
Neurologic Peripheral neuropathy
V. DIFFERENTIAL DIAGNOSIS. Clearly, other disorders that cause anorexia, weight loss,
or vomiting must be considered. Crohn’s disease, which typically begins in adolescence
or young adulthood, can present with many of the signs and symptoms of anorexia
nervosa. Malabsorption syndromes (see Chapter 31) also can cause profound weight
loss and diarrhea (steatorrhea). Neoplasia, unusual in the age group under considera-
tion, may also be considered. Finally, metabolic, endocrinologic, cardiovascular, renal,
pulmonary, and hematologic disorders must be considered when appropriate. The
complications of anorexia nervosa or bulimia may mimic those conditions.
VI. Diagnostic studies are performed in patients suspected of having anorexia nervosa
or bulimia for two reasons: to document the severity and complications of the disease
and to exclude other disorders. Thus a complete blood count, serum electrolytes,
serum proteins, liver tests, and renal function tests should be obtained. Examination
of the entire gastrointestinal tract by sigmoidoscopy, barium enema or colonoscopy,
and barium swallow–upper gastrointestinal– small-bowel x-ray series may be done. If
vomiting persists, upper gastrointestinal endoscopy and esophageal motility testing
are appropriate in most patients because of the high frequency of Crohn’s disease, gas-
trointestinal motility disorders, and other conditions in patients originally thought to
have anorexia nervosa. Studies for the evaluation of diarrhea or malabsorption may
be indicated in some patients.
VII. TREATMENT. Typically, anorectic patients deny the severity of their illness and evade
adequate psychiatric and medical care or fail to comply with the treatment regimen.
Bulimic patients are more motivated to seek treatment but do not tolerate therapeutic
regimens that fail to give immediate relief of symptoms.
Treatment includes improvement of medical and nutritional status, reestablish-
ment of healthful patterns of eating, and identification and resolution of psychosocial
precipitants of the disorder. Most patients are treated on an outpatient basis; however,
some patients may require inpatient medical and psychiatric care. Rapid weight loss,
intractable purging, severe electrolyte imbalances, cardiac disturbances, and a high
suicide risk are some of the indications for inpatient care.
79466_CH59.qxd 1/2/08 6:15 PM Page 475
Selected Readings
Becker AE, et al. Eating disorders. N Engl J Med. 1999;340:1092.
Gard ME, et al. The dismantling of a myth: A review of eating disorder and socioeconomic
status. Int J Eat Disord. 1996;20:1.
Kohn MR, et al. Cardiac arrest and delirium: Presentations of the refeeding syndrome in
severely malnourished adolescents with anorexia nervosa. J Adolesc Health. 1998;22:239.
Pike KM, et al. Ethnicity and eating disorders. Psychopharmacol Bull. 1996;32:265.
Vitiello B, et al. Research on eating disorders: Current status and future of prospects. Biol
Psychiatry. 2000;47:777.
Walsh BT, et al. Medication and psychotherapy in the treatment of bulimia nervosa. Am J
Psychiatry. 1997;154:523.
79466_CH60.qxd 1/2/08 6:15 PM Page 476
60 OBESITY
I. EPIDEMIOLOGY. Obesity has become an epidemic throughout the world, even in the
Third World countries. In the United States it is estimated that three of five American
adults are overweight or obese and the cost of obesity is in excess of $100 billion
annually. In the last 35 years, the prevalence of obesity has more than doubled in the
United States. The prevalence of obesity is particularly high in many ethnic minority
women (e.g., African American, Mexican American, Native American, Pacific Islander
American, Puerto Rican, and Cuban American). Obesity, in fact, is equal to tobacco
use as a public health hazard, contributing to more than 500,000 premature deaths
annually and is associated with a twofold increase in mortality. Obesity is a major
health problem in young adults and children. In minority populations, up to 30% to
40% of the children and adolescents are overweight.
476
TABLE 60-1 Body Mass Index
79466_CH60.qxd
BMI 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
1/2/08
58 91 96 100 105 110 115 119 124 129 134 138 143 148 153 158 162 167
59 94 99 104 109 114 119 124 128 133 138 143 148 153 158 163 168 173
6:15 PM
60 97 102 107 112 118 123 128 133 138 143 148 153 158 163 168 174 179
61 100 106 111 116 122 127 132 137 143 148 153 158 164 169 174 180 185
62 104 109 115 120 126 131 136 142 147 153 158 164 169 175 180 186 191
63 107 113 118 124 130 135 141 146 152 158 163 169 175 180 186 191 197
64 110 116 122 128 134 140 145 151 157 163 169 174 180 186 192 197 204
Page 477
65 114 120 126 132 138 144 150 156 162 168 174 180 186 192 198 204 210
66 118 124 130 136 142 148 155 161 167 173 179 186 192 198 204 210 216
67 121 127 134 140 146 153 159 166 172 178 185 191 198 204 211 217 223
68 125 131 138 144 151 158 164 170 177 184 190 197 203 210 216 223 230
69 128 135 142 149 155 162 169 176 182 189 196 203 209 216 223 230 236
70 132 139 146 153 160 167 174 181 188 195 202 209 216 222 229 236 243
71 136 143 150 157 165 169 179 186 193 200 208 215 222 229 236 243 250
72 140 147 154 162 169 174 184 191 199 206 213 221 228 235 242 250 258
73 144 151 159 166 174 179 189 197 204 212 219 227 235 242 250 257 265
74 148 155 163 171 179 184 194 202 210 218 225 233 241 249 256 264 272
75 152 160 168 176 184 189 200 208 216 224 232 240 248 256 264 272 279
76 156 164 172 180 189 193 205 213 221 230 238 246 254 263 271 279 287
77 160 168 177 185 193 202 210 219 227 235 244 252 261 269 278 286 294
78 164 173 181 190 198 207 216 224 233 232 250 259 268 276 285 293 302
To obtain the BMI, locate height on the left, then move across the line to the weight; the BMI will be on the top.
BMI, body mass index.
477
79466_CH60.qxd 1/2/08 6:15 PM Page 478
Waist circumference
women 88 cm or
Body mass 35 in. Men 102 cm
Weight class index (kg/m2) or 40 in.
III. PATHOGENESIS. Both genetic and environmental factors contribute to body size.
Genetic background can explain up to 40% of the variance in body mass in humans.
However, the marked increase in the prevalence of obesity in the last 20 years cannot
be attributed to genetic change and may be caused by alterations in the environment,
such as sleep deprivation, increased stress at work and home, lack of regular meal times,
and choices of foods, i.e., fast food versus mediterranian-type diet.
Simplistically, obesity originates from ingesting more energy and calories than is
expended over a long period of time. The excess ingested calories are stored as fat. Even
small, but persistent differences between energy intake and energy expenditure can lead
to large increases in body fat. For example, ingestion of only 5% more calories than
expended could result in the accumulation of about 5 kg of adipose tissue in one year.
Ingestion of 7 kcal/day more than expended over 30 years can lead to an increase of 10
kg body weight, which is the average amount of weight gained by American adults from
25 to 55 years of age.
Technological advances have led to changes in lifestyle that favor a positive
energy balance due to an increased availability and palatability of inexpensive energy-
dense foods, decreased daily physical activity because of labor-saving devices, changes
in job-work patterns, and accessibility to mechanical transportation. Persons with
certain genetic backgrounds are particularly predisposed to weight gain when they are
exposed to this “modern” lifestyle. For example, Pima Indians living in reservations
in Arizona have a much greater prevalence of obesity and diabetes mellitus than their
genetic counterparts who live in rural areas of Mexico.
The modern American diet of fast food and beverages are high in fat and calo-
ries and low in nutritional value. An estimated 60% to 90% of Americans are under-
nourished meaning that despite excessive caloric intake, they do not meet their daily
recommended dietary allowances (RDAs) in one or more food groups. In addition to
increased caloric intake, only about 9% of men and 3% of women exercise vigorously
on a regular basis as part of their leisure time activities.
V. THERAPY
A. General principles. Table 60-4 lists the key principles involved in the therapy of
obesity. Americans spend in excess of $70 billion a year on commercial weight-loss
products. Most persons lose weight on these diets, but unfortunately within 1 to
5 years the weight is gained back with extra pounds.
Obesity is a chronic illness and requires long-term management for long-
term success. Behavior modification is necessary for long-term lifestyle changes.
Dietary and nutritional education is also very important. Patients should be
encouraged to lose weight in a systemic and modest way through increased insulin
sensitivity, decreased blood pressure and blood lipid levels, and reduction of fatty
infiltration of the liver.
Caloric reduction needs to be individualized based on the individual’s age and
comorbid risk factors. A useful formula for losing about a pound a week is as follows:
Current weight in pounds 13 kcal 500 kcal daily caloric requirement
Reduction of fat intake is essential in a successful weight-reduction pro-
gram. Many patients will do well by reducing their total dietary fat intake to 10%
to 20% of their total caloric intake (about 20–30 g of total dietary fat daily). Most
commercial weight loss programs limit caloric intake to 800 to 1,200 calories a
day. When followed carefully, these programs will induce a weight loss of 0.5 to
2.0 lbs a week for up to 30 weeks.
Many “fad diets” have little merit. Some of these diets may be actually harm-
ful. For example, avoiding carbohydrates will induce ketosis; excessively high pro-
tein intake may adversely affect the kidneys and accelerate calcium loss from
bones, thereby promoting or enhancing osteoporosis. In addition, reduced caloric
intake may lead to micronutrient deficiencies that impair metabolic fitness.
The assistance of a dietitian is very important. The National Registry of
Dietitians may be contacted for recommendations of a dietitian in patients’ living
areas (telephone: 1-800-366-1655). Diet advice should include encouraging
patients to eat three meals a day, avoid snacking between meals, avoid energy-
dense and high-fat foods, and increase the intake of fruits and vegetables.
Physical activity is important for long-term weight management and
improved health. Physical activity should be increased slowly over time. Studies
suggest that about 80 minutes per day of moderate-intensity exercise (e.g.,
brisk walking or 35 minutes per day of vigorous activity such as fast bicycling
or aerobic exercise) is needed for long-term weight maintenance after initial
weight loss has been achieved. Physical activity does not necessarily have to be
part of a structured exercise program. Increasing daily lifestyle activities is just
as effective in maintaining weight loss as participating in an aerobic program
exercise.
Recent data indicate that weight-resistance training appears to be the most
beneficial form of exercise for successful weight management. By improving
the integrity of existing muscle or by developing muscle mass, this type of exercise
increases overall metabolism and enhances the oxidation of fat as fuel. These
effects make long-term weight control far more likely.
B. Very-low-calorie diets (VLCDs) or protein-sparing diets are proven, safe
alternatives to starvation for significant sustained and progressive weight loss.
These diets deliver a total daily caloric intake of 400 to 800 calories. Successful
and safe programs include a daily intake of 0.8 to 1.0 g of protein per kilogram of
desirable body weight or about 70 to 100 g of protein and at least 45 to 50 g of
carbohydrate to minimize nitrogen losses and ketosis, respectively. All VLCDs
require careful physician supervision and close patient monitoring. Generally,
weight loss is rapid and progressive for several weeks to months. After about
6 months, weight loss slows and plateaus and further weight loss becomes difficult
to achieve. Unfortunately, once the VLCD is discontinued, initial weight-loss main-
tenance is also difficult to sustain. Incorporation of regular exercise and lifestyle
changes improve the likelihood of sustaining the weight loss. Intermittent use of
VLCD products or meal substitutions along with restrained eating patterns offers
considerable promise.
C. Pharmacotherapy. It is extremely difficult to achieve and sustain significant
reductions in weight and body fat in obese patients without pharmacotherapy or
VLCDs. Pharmacotherapy can help selected patients maintain long-term weight
loss, but it should not be considered a short-term treatment approach. Obesity is a
chronic disease and patients who respond to drug therapy usually regain weight
when the drug therapy is stopped. Also, the effectiveness of pharmacotherapy may
diminish with time. It is of paramount importance that pharmacotherapy be cou-
pled with dietary, lifestyle, and behavioral changes.
1. Sibutramine hydrochloride maleate (Meridia) is a relatively new drug that
was approved by the U.S. Food and Drug Administration (FDA) for long-term
use in 1997. It is a monoamine reuptake inhibitor that was initially developed as
an antidepressant to prevent the reuptake of serotonin, dopamine, and norepi-
nephrine; thus, it synergistically promotes enhanced satiety and a reduction in
food intake. In most patients, it induces a dose-dependent reduction in weight.
The drug is available in capsule form in once-daily doses of 5, 10, and 15 mg. In
one study, at 1 year, 39% of patients randomized to sibutramine hydrochloride
maleate (15 mg per day) lost 70% of their initial body weight compared to 9%
of those randomized to receive placebo. Based on clinical trials, it appears to be
safe. With long-term use, parallel to the weight loss, successful reduction in obe-
sity associated comorbid conditions has also been observed.
79466_CH60.qxd 1/2/08 6:15 PM Page 481
distal gastric bypass is 150 cm. Weight loss occurs due to early satiety
from the small gastric size and induction of mild malabsorption. If weight
loss is not satisfactory the standard Roux gastric bypass can be revised to a
very long limb Roux-en-Y procedure.
b. Laparoscopic mini-gastric bypass (LMGB) is a modification of gastric
bypass with a longer lesser curvature tube.
c. Gastric banding or laparoscopic adjustable gastric banding (LABG) is usu-
ally performed laparoscopically. It involves placement of a “band” high on the
stomach creating a small pouch of 15-mL capacity without cutting the stomach.
Adjustments are made up to six times per year to limit gastric capacity.
LABG is reversible by the removal of the band, tubing and port or it
can be revised by removal of the device and performance of a restrictive-
malabsorptive or malabsorptive procedure.
d. Vertical banded gastroplasty ( VBG) is usually performed using an open
surgical approach and involves the creation of a small 15-to 25-mL linear
gastric pouch along the lesser curve of the stomach with an outlet of 0.75
to 1.25 cm in diameter. This procedure may be revised by the removal of
the ring or the band allowing the outlet to dilate. In patients with inade-
quate weight loss, VGB may be revised by conversion to gastric bypass or
duodenal switch.
3. Results
a. Benefits. Weight loss is greater with gastric bypass (about 70 kg vs. 30 kg
in 12 months) than with gastroplasty or gastric banding procedures.
Postoperatively, patients experience prolonged satiety and have improved
well-being, attractiveness, self-regard, and social or physical activity. There
is a significant reduction in the incidence of diabetes, hyperlipidemia, hype-
ruricemia, sleep apnea, hypertension, and mortality rate. In addition, there
are significant risk reductions for developing cardiovascular, malignant,
endocrine, infectious, psychiatric, and mental disorders.
b. Complications. Short-term complications are those expected as in any
postoperative period.
Long-term complications may include anastomotic ulceration, bleed-
ing, stenosis, and possible increase in certain GI symptoms such as diarrhea.
Deficiencies in certain vitamins and nutrients and neurologic and psychiatric
complications may occur. Patients should be monitored for vitamin and nutri-
ent deficiencies and should be placed on replacement therapies as needed.
Selected Readings
Balasekasan R, et al. Positive results in tests for steatorrhea in person consuming olestra
potato chips. Am Intern Med. 2000;132:279.
Bardia A, et al. Diagnosis of obesity by primary care physicians and impact on obesity
management. Mayo Clin Prac. 2007;82(8):927–932.
Bloomberg RD, et al. Nutritional deficiencies following bariatric surgery: What have we
learned? Obes Surg. 2005;15(2):145–154.
Buchwald H, et al. Bariatric surgery: A systematic review and meta-analysis. JAMA. 2004;
292(14):1724–1737.
Choban PS, Dickerson RN. Morbid obesity and nutrition support: is bigger different?
Nutr Clin Pract. 2005;20:480–487.
Christou NV, et al. Surgery decreases long-term mortality, morbidity and health care use in
morbidly obese patients. Ann Surg. 2004;240(3):416–423.
DeMaria EJ. Bariatric surgery for morbid obesity. N Eng J Med. 2007;356:2176–2183.
Dixon AF, et al. Laparoscopic adjustable gastric banding induces prolonged satiety:
A randomized blind crossover study. J Clin Endocrinol Metab. 2005;90(2):813–819.
Flum DR, et al. Impact of gastric bypass on survival: A population-based analysis. J Am
Coll Surg. 2004;199(4):543–551.
Inabnet WB, et al. Laparoscopic Roux-en-Y gastric bypass in patients with BMI 50:
A prospective randomized trial comparing short and long limb lengths. Obes Surg.
2005;15(1):51–57.
79466_CH60.qxd 1/2/08 6:15 PM Page 483
Koffman BM, et al. Neurologic complications after surgery for obesity. Muscle Nerve. June
22, 2005 (Epub ahead of print).
Kruger J, et al. Attempting to lose weight: specific practices among adults. Am J Prev Med.
2004;26:402–406.
Lee WJ, et al. Laparoscopic vertical banded gastroplasty and laparoscopic gastric bypass:
A comparison. Obes Surg. 2004;14(5):626–634.
Lujan JA, et al. Laparoscopic versus open gastric bypass in the treatment of morbid
obesity: A randomized study. Ann Surg. 2004;239(4):433–437.
Madan A. Laparoscopic bariatric surgery. US Gastroneterol Rev. 2007;1:29–331.
Maggard MA, et al. Meta-analysis: Surgical treatment of obesity. Ann Intern Med.
2005;142(7):547–559.
Malone AM. Permissive underfeeding: Its appropriateness in patients with obesity, patients
on parenateral nutrition, and non-obese patients receiving enteral nutrition. Curr
Gastroenterol Rep. 2007;9(4):317–322.
Padwal RS, et al. Drug treatments for obesity: orlistat, sibutramine, and rimonabant.
Lancet. 2007:369:71–77.
Perrin EM. Preventing and treating obesity: pediatricians’ self-efficacy, barriers, resources,
and advocacy. Ambul Pediatr. 2005;5:150–156.
Redinges RN. The pathophysiology of obesity and its clinical manifestations. Gastroenteric
Hepatol. 2007;3(11):856–874.
Sjostrom L, et al. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric
surgery. N Engl J Med. 2004;351(26):2683–2693.
Slater GH, et al. Serum fat-soluble vitamin deficiency and abnormal calcium metabolism
after malabsorptive bariatric surgery. J Gastrointest Surg. 2004;8(1):48–55.
Wang W, et al. Short-term results of laparoscopic mini-gastric bypass. Obes Surg. 2005;
15(5):648–654.
Yan LL, et al. Midlife body mass index and hospitalization and mortality in older age.
JAMA. 2006;295:190–1989.
79466_Index.qxd 1/2/08 6:16 PM Page 484
INDEX
Page numbers followed by “f” indicate figures; those followed by “t” indicate tables.
A Abdominal tenderness
AAs. See Amino acids in acute abdomen, 81
AAT (1-antitrypsin deficiency) examination for, 8
Abdomen Abetalipoproteinemia, and malabsorption,
acute. See Acute abdomen 219–220
areas of, 6, 7f Abscesses
auscultation of, 6–7 anorectal, 284
percussion and palpation of, 708 pancreatic, 321
Abdominal angina, 264–265 Absorbents in diarrhea treatment, 193
Abdominal blunt trauma, in pancreatitis, Acalculous cholecystitis, 301–302
308 Acetaldehyde, 389
Abdominal examination, 6–8, 7f Acetaminophen, in alcoholic liver
Abdominal films, 29, 39 disease, 391
of acute abdomen, 82t Acetaminophen hepatoxicity, 401–405
of acute pancreatitis, 312 Acetylcysteine, in fulminant hepatic
of ascites, 443 failure, 109
of diverticulitis, 238 Achalasia, 144, 151, 152f
of inflammatory bowel disease, 248, treatment of, 153
250f Acid clearance in GERD, 129
of jaundice, 345 Acid ingestion, 100t, 101
of malabsorption, 210, 210f AcipHex (rabeprazole sodium), 136
of small-intestinal neoplasms, 222 Acquired deficiency of glucuronyl
of toxic megacolon, 259 transferase, 342
Abdominal pain, 79–83 Acquired immunodeficiency syndrome
and bloating, 234 (AIDS). See also Hepatobiliary
clinical presentation of, disease in AIDS; Human
79–81 immunodeficiency virus
differential diagnosis of, 79, 80t and cytomegalovirus (CMV), 361–362
in diverticular disease, 239 cytomegalovirus esophagitis (CMV) in,
evaluation of, 81–83, 82t 124
indication for laparoscopy, 27 HSV esophagitis in, 123
in inflammatory bowel disease, 246 idiopathic esophageal ulcer (IEU) in, 125
and intestinal gas, 234 sexually transmitted enteric disorders,
in irritable bowel syndrome, 246 288
in pancreatic cancer, 327 Acticall (ursodeoxycholic acid), 114
in pancreatitis Actos (pioglitazone), 398
acute, 79 Acute abdomen, 79–83
chronic, 318–319 clinical presentation, 79, 81
referred, 79 history, 79, 81
somatic, 79 known disorders, 79, 81
treatment of, 83 pain, 79
in vascular disorders, 264 vomiting, 79
visceral, 79 physical examination, 81
484
79466_Index.qxd 1/2/08 6:16 PM Page 485
Index 485
486 Index
Index 487
488 Index
Index 489
490 Index
Index 491
492 Index
Index 493
494 Index
Index 495
496 Index
Index 497
498 Index
Index 499
500 Index
Index 501
502 Index
Index 503
psychological and emotional aspects in, for alcoholic liver disease (ALD), 344,
10, 229–230 390–391
synonyms for, 227, 227t for ascites, 444
treatment for, 229–231, 230t for autoimmune hepatitis, 408–409
Ischemic colitis, 246, 246t, 265–267 for choledocholithiasis, 356
clinical presentation of, 265 for gastrointestinal bleeding, 86–87
diagnostic studies and differential for hemochromatosis, 430
diagnosis of, 265, 266f for hepatic encephalopathy, 450
pathophysiology of, 265 for hepatitis B virus (HBV),
treatment of, 265, 267 367–368
Islet cell tumors of the pancreas, 326. for hepatocellular necrosis, 399
See also Pancreatic cancer for human immunodeficiency virus
Isoniazid (INH), 405–406 (HIV), 294
Itraconazole, 122, 294 for liver disease, 335–340, 344, 345t
IVP. See Intravenous pyelogram for liver tumors, 456
for nonalcoholic fatty liver disease, 396
J for pancreatic cancer, 327–328
Jaundice, 335–348 for pancreatitis, 311–312
and cancer, 344–345 for peptic ulcer disease, 163
diagnostic studies, 345–348, 348f for primary biliary cirrhosis, 412–413
invasive techniques, 346–347 for toxic megacolon, 258
noninvasive techniques, 345–346 for Wilson’s disease, 424
drug-induced, 343–344 Lactase deficiency, in malabsorption, 219
extrahepatic billiary obstruction, 344 Lactic acidosis, in fulminant hepatic
and hepatocellular disease, 344 failure, 108
hyperbilirubinemia, 340–345, 341t Lactic dehydrogenase (LDH), 336
conjugated, 340, 342–345, 343t, Lactose absorption tests, in malabsorption,
345t 213–214, 213f
hereditary unconjugated, 341–342 Lactose-free formulas with intact nutrients
unconjugated, 340–341 in enteral nutrition, 59, 57t
laboratory studies for liver disease, Lactose intolerance, and irritable bowel
335–340, 344, 345t syndrome, 228
cholestasis tests, 337–340, 338t Lactose tolerance test, for irritable bowel
liver cell injury and necrosis, syndrome, 229
335–336 Lactulose, for constipation, 242
synthetic function, 336–337, 336f Laënnec’s cirrhosis, 392, 437
postoperative, 344 Lamina propria, 24–25
in pregnancy, 343 Lamivudine, for HBV infection, 374
and sepsis, 344 Lanreotide, in metastatic carcinoid
Jejunostomies, 280. See also Ostomy care tumors, 226
Jevity, 57t, 59 Lansoprazole (Prevacid), 136
in peptic ulcer disease, 161, 166
Laparascopy (peritoneoscopy), for
K jaundice, 346–347
Kaopectate, 193 Laparoscopic adjustable gastric
Kaposi’s sarcoma, 289, 291, 301 banding (LAGB), 482. See also
and dysphagia, 142 Obesity
Keofeed feeding tube, 60 Laparoscopic cholecystectomy, 28–29
Ketoconazole, 121 Laparoscopic mini gastric bypass (LMGB),
Klatskin tumors, 457 482. See also Obesity
Klebsiella pneumoniae, 65 Laparoscopic Nissen fundoplication, 138
Kock pouches, 280. See also Ostomy care Laparoscopic surgery, 28–29
KUB (kidney, ureter, and bladder) film, 39 laparoscopic cholecystectomy, 28–29
indications and contraindications,
L 28, 28t
Laboratory studies results, 29
in acute abdomen, 81–82 technique, 28–29
for acute calculous cholecystitis, 354 other operations, 29
79466_Index.qxd 1/2/08 6:16 PM Page 504
504 Index
Index 505
506 Index
Index 507
508 Index
Index 509
510 Index
Index 511
512 Index
Radiotherapy S
for colorectal cancer, 275 Saliva in esophageal clearance, 129
in gastric adenocarcinoma, 178 Salmonella, 203–204
Ranitidine (Zantac), in peptic ulcer in hepatobiliary disease, 299
disease, 165 SAMe, for alcoholic liver disease, 393
Raynoud’s phenomenon, and reflux Savory dilators, 133
esophagitis, 145 Schilling test, for malabsorption, 214, 216
Rectal carcinoid tumors, 225 Scleroderma, and reflux esophagitis, 145
Rectal examination Sclerotherapy (banding), 19
in acute abdomen, 81 Secretin stimulation test
in gastrointestinal bleeding, 86 in pancreatitis, 320
Rectal prolapse, 285 in peptic ulcer disease, 163–164
Rectum, foreign bodies in, 98 Secretory diarrhea, 185, 186, 186t.
Recurrent alcoholic pancreatitis, 318 See also Diarrhea
Red blood cells, radiolabeled for scans, 43 Sedation, conscious, 18
REE. See Resting energy expenditure Selenium deficiency, and parenteral
Refeeding syndrome, 53–54 nutrition, 74
Referred pain in acute abdomen, 79 Sengstaken-Blakemore (SB) balloon
Reflux esophagitis, 126, 127–128. See also tamponade, 93
Gastroesophageal reflux disease for varices, 442
Regurgitation, in GERD, 130 Sepsis
Rehydralylate, 198 in fulminant hepatic failure, 109
Remicade (infliximab), for inflammatory parenteral nutrition in, 68
disease, 254, 256 Serous cyctadenomas, 331
Renal abnormalities, in fulminant hepatic Serum alkaline phosphatase level,
failure, 108 337–338, 338t
Renal failure, parenteral nutrition in, 68 in alcoholic liver disease, 390
Renal formulations for enteral nutrition, 59 Serum amylase levels
Respiratory disorders, in fulminant hepatic in irritable bowel syndrome, 229
failure, 108 in pancreatitis, 311–312, 320
Respiratory failure, parenteral nutrition Serum bilirubin, 339–340
in, 68, 74 Serum carcinoembryonic antigen (CEN),
Respiratory formulations for enteral in gastric neoplasms, 177
nutrition, 60 Serum enzymes as marker of
Respiratory quotient (RQ), 52–53 hepatocellular injury and
Resting energy expenditure (REE), 65 necrosis, 335–336
Restitution, in peptic ulcer disease, 160 Serum gastrin, in peptic ulcer disease,
Retained antrum syndrome, in peptic ulcer 163, 163t
disease, 168 Serum globulins, in alcoholic liver disease, 391
Rifampin, in primary biliary cirrhosis, 414 Serum glutamic-oxaloacetic transaminase
Rifaximin (Xifaxan), 110, 193, 195 (SGOT), 335
for small-bowel bacterial overgrowth, 216 Serum glutamic-pyruvic transaminase
Rigid sigmoidoscope, 21 (SGPT), 335
Ringed esophagus, 143 Serum iron studies, for colorectal cancer,
Rings and webs, in esophageal dysphagia, 142 274–275
Riopan, 135 Serum lipase levels, in pancreatitis, 312
Romazicon (flumazenil), for hepatic Serum proteins, 336–337, 336f
encephalopathy, 453 Serum tests, for primary biliary cirrhosis, 412
Rosiglitazone (Avandia), 398 Sessile polyps, 269. See also Colonic
Rosol, 198 polyps
Rotavirus, 197–198 Sexually transmitted enteric disorders,
Rotor’s syndrome, 343 286, 288–291
Rowasa enema anorectal disorders, 289, 291
for inflammatory bowel disease, 254 anorectal syphilis, 289
for proctitis, 285 Chalamydis trachomatis, 289
RQ. See Respiratory quotient condylomata acuminata (anal
Rubber-band ligation, for hemorrhoids, 283 warts), 291
Rubine tube, 23 gonococcal proctitis, 289
79466_Index.qxd 1/2/08 6:16 PM Page 513
Index 513
514 Index
Index 515
516 Index