Interventional

Endoscopic Ultrasound

Douglas G. Adler
Editor

1 23
Interventional Endoscopic Ultrasound
Douglas G. Adler
Editor

Interventional
Endoscopic Ultrasound
Editor
Douglas G. Adler
Department of Internal Medicine
Division of Gastroenterology and Hepatology
University of Utah School of Medicine
Salt Lake City, UT, USA

Additional material to this book can be downloaded from

https://link.springer.com/book/10.1007/978-3-319-97376-0.

ISBN 978-3-319-97375-3    ISBN 978-3-319-97376-0 (eBook)

https://doi.org/10.1007/978-3-319-97376-0

Library of Congress Control Number: 2018956728

© Springer Nature Switzerland AG 2019

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, express or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
For Harriet and Stanley & Karen and Joel
Preface

Technology progresses at an uneven rate. Since I first started performing

endoscopy almost 20 years ago, some procedures have remained fairly static,
while others have changed dramatically. ERCP, the first therapeutic proce-
dure I ever fell in love with, despite significant advances in endoscopes, cath-
eters, and wires, is still very similar today, i.e., we still use catheters and wires
to access the biliary tree and pancreatic ducts, we still perform sphincterot-
omy much the way we did back then, and we still use largely the same tools
(balloons and baskets) to remove stones. While much of the practice of ERCP
has changed, including which patients we select for ERCP, how we perform
the ERCP, and what steps we take to prevent pancreatitis, much of the
mechanics of ERCP today would look very familiar to someone who per-
formed the procedure in the 1980s.
The evolving practice of EUS, however, represents quite a different story.
EUS has undergone what can only be considered a radical transformation
over the past few years. From its inception and widespread dissemination in
the early 1990s until just a few years ago, EUS was comprised almost entirely
of a set of diagnostic procedures, with the vast majority of examinations
being used to look at and sample lesions or organs of concern. The idea of
EUS being used for therapeutic interventions was slow in coming. Concerns
about the mechanical limitations of echoendoscopes, fear of adverse events,
and a lack of commercially available accessories to perform these maneuvers
significantly hampered progress and development.
Only in the last few years has the idea of using EUS to perform interven-
tional procedures been embraced on a wide scale, and the pace of develop-
ment has been rapid. Centers around the globe are now actively working to
both develop new procedures and devices and to modify old procedures here-
tofore performed by surgeons or interventional radiologists to be performed
by interventional endosonographers.
While much of interventional EUS is still performed with ERCP accessories
in an off-label manner, the development and introduction of lumen-­apposing
metal stents (LAMS) that are supplied on catheters specifically designed to be
used with echoendoscopes represents the first true interventional EUS accessory
that was not simply a modified needle. LAMS have seen a rapid and widespread
dissemination into clinical practice. Although intended for, and widely used, to
drain pancreatic fluid collections, the development of LAMS has also led to the
development of a plethora of interventional EUS procedures including transmu-
ral gallbladder drainage, gastrojejunostomy creation, conduit creation in patients

vii
viii Preface

who have undergone Roux-en-Y gastric bypass to facilitate ERCP, and a host of
other procedures.
Beyond LAMS and their applications, interventional EUS has shown the
power of using needle-based technologies to do more than sample tissue or
fluid from target lesions. Modified needle devices can be used to measure
portal pressures, deliver therapeutic agents to treat solid and cystic tumors,
implant fiducials to facilitate targeted radiation therapy, and deliver analgesic
medications to treat benign and malignant conditions.
The time seems ripe for a single, comprehensive text on interventional
EUS and its myriad applications. This book contains 17 chapters that cover
the entire depth and breadth of interventional EUS, both with regard to how
it is currently practiced and with an eye toward future areas of investigation
and development. Each chapter is lavishly illustrated with endoscopic and
ultrasonographic images. In addition, each chapter is also accompanied by
one or more narrated video segments to allow readers to see how these proce-
dures are performed in real time by experts in the field.
I perform interventional EUS procedures of all manner in my daily thera-
peutic endoscopy practice and truly enjoy the work. It is my hope that readers
use the knowledge contained in this text to expand the range of therapeutic
and interventional EUS procedures that they feel comfortable adding to their
daily practice. In addition, I hope that readers will someday contribute to the
growing body of knowledge on these topics to promote the care of our patients
and the development of interventional EUS as a whole in the years to come.

Salt Lake City, UT, USA Douglas G. Adler

Contents

1 Endoscopic Ultrasound-Guided Drainage of Pancreatic

Fluid Collections����������������������������������������������������������������������������    1
Jeffrey S. Bank and Douglas G. Adler
2 Endoscopic Ultrasound-Guided Bile Duct Access
and Drainage: Antegrade Approaches ����������������������������������������   17
Nan Ge and Siyu Sun
3 Endoscopic Ultrasound-Guided Biliary Drainage:
Retrograde Approaches ����������������������������������������������������������������   25
Constantine Melitas and Douglas G. Adler
4 EUS-Guided Gallbladder Drainage ��������������������������������������������   35
Sunil Amin and Amrita Sethi
5 Endoscopic Ultrasound-Guided Pancreatic Duct
Drainage (EUS-PD)������������������������������������������������������������������������   45
Shawn L. Shah and Amy Tyberg
6 EUS-Guided Treatment of Gastrointestinal Bleeding����������������   55
Larissa L. Fujii-Lau, Louis M. Wong Kee Song,
and Michael J. Levy
7 EUS-Guided Celiac Plexus Block and Celiac
Plexus Neurolysis���������������������������������������������������������������������������   65
Truptesh H. Kothari, Shivangi Kothari, and Vivek Kaul
8 EUS-Guided Core Biopsy��������������������������������������������������������������   73
Ali Siddiqui
9 Endoscopic Ultrasound-Guided Liver Biopsy����������������������������   83
David L. Diehl
10 EUS-Guided Fiducial Placement��������������������������������������������������   95
Aamir N. Dam and Jason B. Klapman
11 EUS-Guided Therapies for Solid Pancreatic Tumors
Including Drug Delivery and Brachytherapy������������������������������ 109
Gursimran Singh Kochhar and Michael Wallace

ix
x Contents

12 EUS-Guided Enhanced Imaging and Sampling

of Neoplastic Pancreatic Cysts������������������������������������������������������ 119
Shivangi Kothari, Enqiang Linghu, Truptesh H. Kothari,
and Vivek Kaul
13 EUS-Guided Pancreatic Cyst Ablation���������������������������������������� 135
Kristopher Philogene and William R. Brugge
14 Endoscopic Ultrasound-Guided Access to the Stomach
in Patients with Prior Gastric Bypass to Facilitate Endoscopic
Retrograde Cholangiopancreatography�������������������������������������� 147
Christine Boumitri, Bhupinder Romana, and Michel Kahaleh
15 Endoscopic Ultrasound-Guided Gastroenterostomy
(EUS-GE)���������������������������������������������������������������������������������������� 159
Steven P. Shamah and Uzma D. Siddiqui
16 Endoscopic Ultrasound-Guided Portal Pressure
Measurement���������������������������������������������������������������������������������� 169
Jason B. Samarasena, Allen R. Yu, and Kenneth J. Chang
17 Endoscopic Ultrasound-Guided Drainage of Pelvic,
Intra-­abdominal, and Mediastinal Abscesses������������������������������ 177
Enad Dawod and Jose M. Nieto

Index�������������������������������������������������������������������������������������������������������� 189
Contributors

Douglas G. Adler Department of Internal Medicine, Division of

Gastroenterology and Hepatology, University of Utah School of Medicine,
Salt Lake City, UT, USA
Sunil Amin Division of Gastroenterology and Hepatology, Virginia Mason
Medical Center, Seattle, WA, USA
Jeffrey S. Bank Department of Internal Medicine, Division of
Gastroenterology and Hepatology, University of Utah School of Medicine,
Salt Lake City, UT, USA
Christine Boumitri Division of Gastroenterology and Hepatology,
University of Missouri, Columbia, MO, USA
William R. Brugge Department of Gastroenterology, Mt. Auburn Hospital,
Cambridge, MA, USA
Kenneth J. Chang Department of Gastroenterology, H. H. Chao
Comprehensive Digestive Disease Center, University of California, Irvine
Medical Center, Orange, CA, USA
Aamir N. Dam Department of Gastrointestinal Oncology, Moffitt Cancer
Center, Tampa, FL, USA
Enad Dawod Department of Gastroenterology and Hepatology, Weill
Cornell Medicine, New York, NY, USA
David L. Diehl Geisinger Commonwealth School of Medicine, Scranton,
PA, USA
Department of Gastroenterology and Nutrition, Geisinger Medical Center,
Danville, PA, USA
Larissa L. Fujii-Lau Division of Gastroenterology, Queens Medical Center,
Honolulu, HI, USA
University of Hawaii, Honolulu, HI, USA
Nan Ge Endoscopy Center, Shengjing Hospital of China Medical University,
Shenyang, Liaoning, China

xi
xii Contributors

Michel Kahaleh Division of Gastroenterology and Hepatology, Rutgers

Robert Wood Johnson, New Brunswick, NJ, USA
Vivek Kaul Division of Gastroenterology and Hepatology, Department of
Medicine, University of Rochester Medical Center, Rochester, NY, USA
Jason B. Klapman Department of Gastrointestinal Oncology, Moffitt
Cancer Center, Tampa, FL, USA
Gursimran Singh Kochhar Department of Gastroenterology and
Hepatology, The Mayo Clinic, Jacksonville, FL, USA
Shivangi Kothari Division of Gastroenterology and Hepatology, Department
of Medicine, University of Rochester Medical Center, Rochester, NY, USA
Truptesh H. Kothari Division of Gastroenterology and Hepatology,
Department of Medicine, University of Rochester Medical Center, Rochester,
NY, USA
Michael J. Levy Division of Gastroenterology and Hepatology, Mayo
Clinic, Rochester, MN, USA
Enqiang Linghu Department of Gastroenterology, Chinese People’s
Liberation Army General Hospital, Beijing, China
Constantine Melitas Michigan State University College of Human
Medicine/Providence-Providence Park Hospitals, Southfield, MI, USA
Jose M. Nieto Department of Gastroenterology and Hepatology, Borland
Groover Clinic, Advanced Therapeutic Endoscopy Center, Jacksonville, FL,
USA
Kristopher Philogene Department of Medicine, Mt. Auburn Hospital,
Cambridge, MA, USA
Bhupinder Romana Division of Gastroenterology and Hepatology,
University of Missouri, Columbia, MO, USA
Jason B. Samarasena Department of Gastroenterology, H. H. Chao
Comprehensive Digestive Disease Center, University of California, Irvine
Medical Center, Orange, CA, USA
Amrita Sethi Division of Digestive and Liver Diseases, Columbia University
Medical Center, New York, NY, USA
Shawn L. Shah Department of Gastroenterology and Hepatology, New
York-Presbyterian/Weill Cornell Medical Center, New York, NY, USA
Steven P. Shamah Center for Endoscopic Research and Therapeutics
(CERT), University of Chicago, Chicago, IL, USA
Ali Siddiqui Department of Gastroenterology, Fish Memorial Hospital,
Orange City, FL, USA
Uzma D. Siddiqui Center for Endoscopic Research and Therapeutics
(CERT), University of Chicago, Chicago, IL, USA
Contributors xiii

Siyu Sun Endoscopy Center, Shengjing Hospital of China Medical

University, Shenyang, Liaoning, China
Amy Tyberg Division of Gastroenterology, Department of Medicine,
Rutgers University Medical School, Robert Wood Johnson University
Medical Center, New Brunswick, NJ, USA
Michael Wallace Department of Gastroenterology and Hepatology, The
Mayo Clinic, Jacksonville, FL, USA
Louis M. Wong Kee Song Division of Gastroenterology and Hepatology,
Mayo Clinic, Rochester, MN, USA
Allen R. Yu Department of Gastroenterology, H. H. Chao Comprehensive
Digestive Disease Center, University of California, Irvine Medical Center,
Orange, CA, USA
About the Editor

Douglas G. Adler, MD, FACG, AGAF, FASGE attended SUNY Binghamton

as an undergraduate and received his medical degree from Cornell University
Medical College. He completed his residency in internal medicine at Beth
Israel Deaconess Medical Center/Harvard Medical School. Dr. Adler com-
pleted both a general gastrointestinal fellowship and a therapeutic endoscopy/
ERCP fellowship at Mayo Clinic in Rochester, MN. He then returned to the
Beth Israel Deaconess Medical Center for a fellowship in endoscopic ultra-
sound. Dr. Adler is currently a tenured Professor of Medicine and Director of
Therapeutic Endoscopy at the University of Utah School of Medicine in Salt
Lake City, UT. Dr. Adler is also the GI Fellowship Program Director at the
University of Utah School of Medicine. Working primarily at the University
of Utah School of Medicine’s Huntsman Cancer Institute, Dr. Adler focuses
his clinical, educational, and research efforts on the diagnosis and manage-
ment of patients with gastrointestinal cancers and complex gastrointestinal
disease, with an emphasis on therapeutic endoscopy. He is the author of more
than 300 scientific publications, magazine articles, and book chapters. This is
Dr. Adler’s seventh textbook on gastroenterology.

xv
 Endoscopic Ultrasound-Guided
Drainage of Pancreatic Fluid 1
Collections

Jeffrey S. Bank and Douglas G. Adler

Introduction Background

In the modern era, pancreatic fluid collections  efinitions of Pancreatic Pseudocysts

D
(PFCs) are most commonly diagnosed and treated (PPs) and Walled-Off Necrosis (WON)
by gastroenterologists. PFCs occur in the setting
of pancreatic ductal injury after episodes of acute Pancreatic fluid collections (PFCs) have been
pancreatitis and are also seen in patients with generally classified as being pancreatic pseudo-
chronic pancreatitis, iatrogenic causes (e.g., pan- cysts (PPs) and walled-off necrosis (WON). Per
creatic injury during surgery), trauma, or in the 2012 Atlanta classification criteria, a PP is an
patients with disconnected duct syndrome [1, 2]. “encapsulated collection of fluid with a well-­
They are divided into pancreatic pseudocysts (PP) defined inflammatory wall usually outside the
or walled-off necrosis (WON). This review will pancreas with minimal or no necrosis.” They usu-
discuss the diagnosis and management of PFCs ally form more than 4 weeks after the onset of
with an emphasis on endoscopic drainage utilizing edematous pancreatitis. They have the following
double pigtail plastic stents (DPPSs), fully cov- contrast-enhanced computerized tomography
ered self-expanding metal stents (FCSEMS), and (CECT) criteria: (1) well circumscribed, usually
lumen-apposing metal stents (LAMS). In addition, round or oval, (2) homogeneous fluid density, (3)
we cover the technique used for placement of each no non-liquid component, and (4) well-defined
stent, compare the advantages/disadvantages, effi- wall aka completely encapsulated. PPs most com-
cacy, and appropriate indications for each stent, monly occur as a result of disruption of the main
and discuss adverse event rates. pancreatic duct or its intrapancreatic branches.
Another less common etiology is disconnected
duct syndrome, where pancreatic parenchymal
necrosis of the neck or body of the gland damages
Electronic supplementary material: The online version
of this chapter (https://doi.org/10.1007/978-3-319-97376- a viable distal pancreatic remnant [3, 4].
0_1) contains supplementary material, which is available A WON is a “mature, encapsulated collection
to authorized users. of pancreatic and/or peripancreatic necrosis that
J. S. Bank · D. G. Adler (*) has developed a well-defined inflammatory wall.”
Department of Internal Medicine, Division of WON usually form more than 4 weeks after the
Gastroenterology and Hepatology, University of Utah onset of necrotizing pancreatitis. They have the
School of Medicine, Salt Lake City, UT, USA
e-mail: [email protected]; following CECT criteria: (1) heterogeneous with
[email protected] liquid and non-liquid density with varying

© Springer Nature Switzerland AG 2019 1

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_1
2 J. S. Bank and D. G. Adler

degrees of loculations, (2) well-defined wall aka PFCs” with a clarifying comment describing the
completely encapsulated, and (3) location— presence or absence of solid material within the
intrapancreatic and/or extrapancreatic. WON collection [9].
occurs as a result of necrotic pancreatic paren-
chyma and/or necrotic peripancreatic tissues and
may be infected, may be multiple, and may be  rief Overview of Endoscopic
B
present at distant sites from the pancreas [2]. Approaches to PFC

Endoscopic ultrasound (EUS)-guided drainage is

Differences Between PPs and WON currently the most commonly used endoscopic
method of drainage of PFCs, and at many centers
PPs are typically homogeneous in appearance has largely replaced surgical or interventional
and composed entirely of liquid components, radiology approaches in these patients. It has a
whereas WON are heterogeneous in appearance high clinical success rate, similar to surgical and
with at least some solid component. PPs occur percutaneous approaches, but with decreased
due to disruption of the main pancreatic duct morbidity and costs [10, 11]. EUS allows the
without pancreatic parenchymal necrosis; WON endoscopist to identify and drain PFCs without
arise from necrotic pancreatic parenchyma. endoluminal bulging, as compared to non-EUS-­
Despite the Atlanta criteria, it remains a chal- guided approaches. Transmural drainage alone
lenge to definitively distinguish PPs from WON. allows for resolution of PFCs in the majority of
In reality, many lesions thought to be PP on CT patients. Direct endoscopic necrosectomy (DEN)
scans are found to contain solid debris on mag- is sometimes necessary for WON. In the past,
netic resonance imaging (MRI) or EUS. For dis- double pigtail plastic stents (DPPSs) were used
tinguishing which type of PFC is present, MRI for management of PFCs. Recently, fully covered
has been shown to be better at detecting solid self-expandable metal stents (FCSEMS) and
debris than either CT or ultrasound [5, 6]. In a lumen-apposing metal stents (LAMS) are
series of 47 patients who developed PFCs, CT increasingly utilized, especially with WON,
scans performed within 3 days of onset of acute owing to their large diameter, which allows for
necrotizing pancreatitis (ANP) demonstrated evi- direction insertion of the endoscope into the
dence of greater than 50% necrosis in 57% of the WON for DEN [12].
patients. Upon repeat evaluation with EUS at
6 weeks, 87% had evidence of solid debris in
their PFCs, which slowly decreased on follow-up EUS-Guided Access
EUS exams at 3 and 6 months. Over half of PFCs
had no evidence of solid debris at 6 months. This Standard Approaches
was felt to be due to breakdown of the solid
debris over time [7]. There is no universally agreed upon technique of
Under the current Atlanta classification crite- PFC drainage, but some general approaches are
ria, both pseudocysts and WON can connect with widely utilized. EUS-guided drainage of PFCs
the pancreatic duct, which also makes it difficult can be performed with or without fluoroscopy, of
to distinguish between the two. Patients with note, depending on operator desire.
PFCs that contain solid debris tend to have poorer Once a PFC is visualized endosonographi-
outcomes and more adverse events compared to cally, a 19-gauge fine needle aspirate (FNA)
those with PFCs with only fluid [8]. In clinical needle is used for initial transmural puncture.
practice, many PFCs do not nicely fit into either The aspirated fluid is sometimes sent for cell
the PP or WON categories. Due to these gray count, gram stain, culture, and cytology—at
areas with the Atlanta criteria, it has been pro- some centers this step is obviated if the lesion is
posed that PP and WON should be called “mature obviously a PFC. If desired, contrast can then be
1 Endoscopic Ultrasound-Guided Drainage of Pancreatic Fluid Collections 3

injected inside the PFC to ensure the needle is in Most DPPSs do not migrate in or out of the cyst
the correct position. A 0.025″ or 0.035″ guide- as they are coiled at their proximal and distal
wire is advanced through the lumen of the needle ends. Seicean et al. used DPPSs over a guidewire
until it coils in the PFC. This forms extra wire in a prospective study of 24 patients (9 with
loops in the PFC cavity and helps solidify the abscesses, 15 with PPs) and achieved technical
endoscope/needle/guidewire position. With the success in 20 patients (83.3%) with complete
wire left in place, the needle is removed over the resolution after a median follow-up time period
wire, and a cystostome, needle knife, or dilation of 18 months. The four patients (16.7%) in whom
balloon can be threaded over the wire. The fis- failure occurred had diameter <6 cm and wall
tula tract is dilated either by balloon dilation or thickness >2 mm [17]. Seicean et al. felt that the
via a combination of a cystostome and/or dia- lack of fluoroscopy likely contributed to the tech-
thermy needle/needle knife. Once the tract is felt nical failure of drainage of PFCs with a diameter
to be sufficiently dilated, a stent can be placed <6 cm. All technical failures in their study were
over the guidewire. Freely flowing fluid through due to instability of a cystostome on the wall of
the stent and into the gastrointestinal cavity indi- a small pseudocyst, in which the absence of
cates successful stent placement [13]. fluoroscopy played a role.
A retrospective review of 93 patients with
symptomatic PFCs reported clinical success rates
Double Pigtail Plastic Stents (DPPSs) of 93.6% using a single plastic stent and 97.4%
using multiple plastic stents (P = 0.309). The
Typically, multiple DPPSs were used for trans- authors found that the secondary infection rate
mural drainage of PFCs (Fig. 1.1). When multi- for drainage utilizing a single stent was 18.4%
ple DPPSs are placed into a cyst cavity, several versus 5.3% for multiple stent drainage
guidewires can be placed into the cyst prior to (P = 0.134). Surprisingly, the secondary infection
insertion of the first to allow for easier stent rate for smaller diameter stents (8.5 Fr or less)
placement or a single wire can be used serially was less than that for larger diameter stents (10 Fr
with each new stent placement. The stents main- or larger), 3.4% versus 17.2%, respectively
tain the fistula tract between the gastric or duode- (P = 0.138) [18]. It should be emphasized, as
nal wall and the PFC, allowing for continued mentioned above, that drainage through the stents
drainage of the PFC. Of note, when multiple (when multiple stents are used) contributes to the
stents are placed the PFC can drain through and lower rate of infection.
between the stents. Placement of multiple stents Two studies have demonstrated high technical
also decreases the risk of stent dislodgement and success rates (93–94%) utilizing DPPSs for the
migration. Stent occlusion rates increase with drainage of PFCs. Clinical success rates were
smaller diameter stents, such as 7 Fr, so larger 74.2% and 82%. PFC recurrence with DPPSs typi-
sizes are typically used, but it is not wrong to use cally occurs as a result of stent clogging and/or
a 7 Fr stent [11]. migration and has been reported at a rate of 12–25%.
EUS has been used for drainage of PFCs since Procedure-related complication rates were 5.4–15%
1996, with several case series reporting technical and included perforation, bleeding, obstruction,
success rates of 83–100% [14–16]. Most endos- migration, recurrence, ­ secondary infection, and
copists use fluoroscopy as it optimizes visualiza- asymptomatic pneumoperitoneum [19, 20].
tion and access into PFCs as well as maintenance
of the position of various devices used, although
fluoroscopy is not mandatory or required for suc- Metal Biliary Stents
cessful PFC drainage. As most previous studies
had utilized fluoroscopy to aid with drainage of Another option for transmural drainage of PFCs
PFCs, Seicean et al. evaluated the safety of EUS-­ are fully covered self-expanding metal stents
guided drainage of PFCs without fluoroscopy. (FCSEMS) (Fig. 1.2). These devices have larger
4 J. S. Bank and D. G. Adler

Fig. 1.1 Use of double pigtail stents to drain a pancre- using an over-­the-­wire biliary dilation balloon catheter.
atic fluid collection (PFC). (a) CT image of a large PFC (f) A double pigtail stent is placed across the cystgastros-
abutting the stomach. (b) Linear EUS image of the PFC tomy into the PFC. (g) A second double pigtail stent is
showing largely fluid contents. (c) A 19-gauge needle has placed across the cystgastrostomy next to the first stent.
been passed into the PFC. (d) A biliary guidewire is (h) Coronal CT image of double pigtail stents in the PFC.
advanced into the PFC through the needle and the needle (i) Sagittal CT image of the double pigtail stents in the
is removed over the wire providing guidewire access to fully collapsed PFC shortly before they were removed.
the PFC. (e) The cystgastrostomy is dilated to 10 mm The PFC did not recur
1 Endoscopic Ultrasound-Guided Drainage of Pancreatic Fluid Collections 5

Fig. 1.1 (continued)

Fig. 1.2 Fully covered metal biliary stent used to drain a been placed through the metal stent to reduce the risk of
PFC. (a) Endoscopic image of fully covered metal biliary migration. (b) CT image of the same patient showing the
stent across a cystgastrostomy. A double pigtail stent has position of the stents across the cystgastrostomy
6 J. S. Bank and D. G. Adler

diameters (6, 8, or 10 mm) and placement of a The adverse event rate was higher (44%) in
single stent can provide a wide drainage opening, Talreja et al. as opposed to Penn et al. (15%),
as opposed to DPPSs, which typically require likely in part due to the inclusion of patients
placement of multiple stents. Owing to their with higher-risk WON. The most common
larger diameter, they have a decreased risk of adverse events between the two studies were
occlusion, especially for PFCs that contain a sig- superinfection, bleeding, stent migration, and
nificant amount of solid material [11]. fever. This rate of adverse events is similar to
It should be noted that FCSEMSs are designed Sharaiha et al., whose adverse event rate was
for drainage related to a luminal stricture and not 16% in those patients with PPs treated with
to a transluminal route and devices used in this FCSEMSs, although multivariate analysis dem-
manner should be considered off-label. When a onstrated patients with plastic stents were 2.9
biliary metal stent is utilized for drainage of a times more likely to experience adverse events
PFC, the ends of the stent protrude into both the than those treated with FCSEMSs [31]. A large
gastric or duodenal lumen and the PFC cavity, retrospective review of 211 patients treated with
which can cause contact ulceration and increase FCSEMSs for PFCs had an adverse event rate of
the risk of stent migration and bleeding. These 21%, which included infection (11%), bleeding
devices may not be the best option for PFCs that (7%), and stent migration and/or perforation
are not firmly adherent to the gastrointestinal (3%) [21].
wall as they do not have flanges to provide an
anchoring force; in addition, they may migrate
just like DPPSs. Many endoscopists place DPPSs  umen-Apposing Metal Stents
L
through a FCSEMS to decrease the risk of migra- (LAMS)
tion and help maintain stent patency [11].
Overall, the technical and clinical success rates Lumen-apposing metal stents (LAMS) have a
of FCSEMSs have been reported at 78–100% and saddle-shaped design and larger inner lumen
81–94%, respectively [21–25]. Biliary FCESMSs diameter than either plastic or metal biliary
have been used for drainage of symptomatic PFCs stents, which theoretically decreases the risk of
in multiple case series in addition to two prospec- migration and allows for an endoscope to pass
tive, single-center studies [26–28]. Talreja et al. into PFCs as well as the ability to perform DEN
utilized FCSEMSs in 18 patients for drainage of (Fig. 1.3 and Video 1.1). There are three different
PFCs and placed DPPSs alongside (n = 4) or LAMSs available at this time around the globe
through the FCSEMS (n = 14) to further promote (AXIOS, NAGI, and Niti-S Spaxus). The AXIOS
drainage and to prevent migration. At a mean fol- stent (Xlumena Inc., Mountain View, CA, USA)
low-up time period of 77 days, 14 (78%) patients consists of double-walled flanges perpendicular to
had complete resolution of their PFC [29]. Penn the lumen that hold the tissue walls in apposition
et al. used FCSEMSs in 20 patients for drainage [32]. The NAGI stent (Taewoong-Medical Co.)
of PPs and found that 14 (70%) patients had resolu- comes in 3 different lengths, 4 diameters and has
tion of their PPs after stent placement without flared ends of 20 mm [33]. The Niti-S Spaxus
known recurrence, adverse events, or the need for stent (Taewoong Medical Co., Ltd., Ilsan, Korea)
surgical intervention. One (5%) patient required consists of nitinol wire and is fully covered with
surgery for primary failure of endoscopic drain- a silicone membrane [34]. Currently available
age and another two (10%) patients developed LAMSs range in diameter from 8 to 20 mm.
pseudocyst infection, requiring surgery. Three Advances in endoscopy over the last
(15%) patients initially had resolution of their PP, 10 years with endoscopic ultrasound (EUS)-
but had recurrence after stent removal [30]. guided drainage of PFCs via transmural stent
Fig. 1.3 Use of a 15 mm diameter lumen-apposing metal in position across the cystgastrostomy. (g) Contents of
stents (LAMS) to treat a patient with infected walled-off pan- WON seen during endoscopic necrosectomy 1 week later.
creatic necrosis (WON). (a) CT scan image of a large WON (h) Solid necrosis being grabbed with an endoscopic net.
occupying much of the pancreatic bed. (b) EUS image of the (i) A large piece of necrotic tissue has been grasped and is
WON with copious turbid solid contents. (c) The electrocau- being pulled through the LAMS to be deposited in the
tery-enhanced LAMS has been advanced into the PFC and stomach. (j) Appearance of the WON cavity after total
half of the LAMS has been deployed. (d) Endoscopic debridement. Note the absence of any further necrotic tis-
image after the LAMS has been fully deployed showing sue. (k) Final appearance of fully collapsed WON cavity
WON contents spilling into stomach. (e) LAMS in good at 8 weeks. (l) The LAMS is removed using a rat-tooth
position after stomach aspirated. (f) CT scan showing LAMS forceps
8 J. S. Bank and D. G. Adler

Fig. 1.3 (continued)

placement has become the first-line manage- scopic transmural drainage of PPs and WONs
ment of PFCs at most tertiary care centers [35]. [19–21, 36–38].
Over the last few years, LAMSs have been Due to their large diameter, AXIOS, Nagi, and
shown to be both safe and efficacious for endo- Spaxus LAMSs are preferable when DEN is
1 Endoscopic Ultrasound-Guided Drainage of Pancreatic Fluid Collections 9

required as it allows the endoscopist to pass the respectively) and WON (70% vs 78%, respec-
scope directly through the stent into the PFC to tively). In addition, there were no statistically
remove the necrotic material [23, 24]. significant differences in the rates of adverse
The graded dilation technique is the most events (e.g., bleeding, secondary infection, and
common technique utilized for the drainage of stent migration) with plastic versus metal
PFCs using cold LAMSs. A 19-gauge access stents (16% vs 23%, respectively) or recur-
needle is inserted into the PFC and a 0.035-in. rence (10% vs 9%, respectively). The overall
guidewire is then advanced through the needle high adverse event rate is likely due to the
into the collection and coiled under fluoro- high-risk nature of endoscopic PFC drainage
scopic guidance. The needle is then removed. [40].
The tract is subsequently enlarged by passing a A large retrospective study of 103 patients
dilating catheter, balloon dilator (4, 6, or with PFCs examined the efficacy of, and adverse
8 mm), or needle-­knife catheter over the guide- events from, LAMSs versus DPPSs. They
wire. In some centers, a larger caliber balloon reported 80 cases with PP (70 DPPSs, 10 LAMSs)
dilator (range, 8–15 mm) is used to further and 23 cases of WON (14 DPPSs, 9 LAMSs). In
expand the cystenterostomy tract, but this step patients for whom follow-up data was available,
is optional. After dilation, a LAMS is then clinical success rates were 67/70 (96%) with
deployed across the cystenterostomy. The DPPSs and 16/17 (94%) with LAMSs, as defined
endoscopist may then choose to perform DEN, by resolution of PFCs within 6 months (p = 0.78).
balloon dilation of the LAMS, or to place PFC recurrence occurred in four (3.9%) patients
DPPSs through the LAMS as needed and based (3 DPPSs, 1 LAMS). The overall adverse event
on patient and physician needs and prefer- rate was 18%. Adverse events occurred in nine
ences. If a DPPS (7F or 10F) is placed through (12%) patients treated with DPPSs and ten (53%)
the LAMS, it is placed in an over-the-­ wire patients treated with LAMSs (p = 0.0003). There
manner under endoscopic and/or fluoroscopic were more bleeding episodes in the LAMS group
guidance into the PFC with the internal pigtail than in the DPPS group (21% vs 1%, p = 0.0003).
inside the PFC and the external pigtail in the In addition, unplanned repeat endoscopy occurred
lumen of the stomach or duodenum. Some more frequently in the LAMS group (10% vs
endoscopists choose to place DPPSs through 26%, p = 0.07) [41].
the LAMS to decrease the risk of migration An additional retrospective review of 49
and to help break up solid contents inside a patients (31 with PP and 18 with WON) exam-
PFC through mechanical processes [39]. ined clinical success rates, cost, and adverse
events for drainage of PFCs utilizing LAMSs
(Nagi stent) versus DPPSs. Inadequate drainage
Comparison Between DPPSs occurred in 10 cases treated with DPPS com-
and Metal Stents (Both FCSEMSs pared to zero with LAMS. Clinical success was
and LAMSs) achieved in 25/38 (64.9%) for DPPS versus
11/12 (91.7%) for LAMS. Placement of DPPS
In a systematic review of 17 studies with 881 was associated with an increased frequency of
patients with PFCs who underwent EUS-guided repeat drainage (34.2% vs 6.3%, p = 0.032) com-
drainage with plastic versus metal (including pared to LAMS, which is different than the pre-
both FCSEMS and LAMS) stents, there were vious study. DPPSs were significantly cheaper
similar clinical success rates (defined as a for drainage of non-infected PPs; costs were
decrease in PFC size and/or resolution of similar for infected PP and WON. Overall
symptoms) for plastic stents (81%) and metal adverse event rate was 13.5% for DPPS and 0%
stents (82%) for both PPs (85% vs 83%, for LAMS [42].
10 J. S. Bank and D. G. Adler

Siddiqui et al. examined by the efficacy of Electrocautery-Enhanced LAMS

DPPSs, FCSEMSs, and LAMSs for the drainage (EC-LAMS)
of WON in 313 patients (106 with DPPSs, 121
with FCSEMSs, and 86 with LAMSs). There Recently, an electrocautery-enhanced LAMS
were no differences in technical success rates (EC-LAMS) has been made available for EUS-­
between the three stents, however, complete guided drainage of PFCs. EC-LAMS allows
resolution of WON at 6-month follow-up seen endoscopists to perform drainage of PFCs in a
in significantly fewer patients treated with more efficient manner by decreasing the number
DPPSs as opposed to FCSEMSs and LAMSs of steps needed, and in many cases completely
(81% vs 95% s 90%; P = 0.001). The average obviating the need for the use of a wire at all.
number of endoscopic sessions for WON reso- EC-LAMS procedures also should not require
lution was significantly lower in the LAMS the use of a dilating catheter as well, further sav-
group compared with the FCSEMS and DPPS ing time and the cost of additional supplies.
groups (2.2 vs 3 vs 3.6; P = 0.04). Adverse Three studies have examined the use of an
events occurred in 27 (8.6%) of 313 patients and electrocautery-enhanced LAMS for the drainage
included perforation (n = 6), bleeding (n = 8), of PFCs in a total of 131 patients. In a retrospec-
suprainfection (n = 9), and other (n = 7). Early tive study of 25 patients who underwent
adverse events occurred less often in the EC-LAMS drainage of PFCs (3 with PP and 22
FCSEMS group compared with the DPPS and with WON), technical success was achieved in
LAMS groups (1.6% vs 7.5% vs 9.3%; 25 (100%) patients and PFCs resolved in 24
P < 0.01). Overall, drainage of WON was felt to (96%) patients at an average follow-up of
be most efficacious with FCSEMSs or LAMSs 7.8 months. The resolution failure occurred in a
as opposed to DPPSs [25]. patient with WON. Adverse event rate was 8%
In a retrospective case-control study, Bang and included stent occlusion (n = 1), and sponta-
et al. compared the efficacy of LAMSs versus neous migration into enteral lumen after resolu-
DPPSs for PFC drainage in 21 patients undergo- tion (n = 1) [44]. In a large retrospective study of
ing PP drainage (7 via LAMS, 14 via conven- 93 patients (80% with complex collections) with
tional plastic double pigtail stents) and 39 patients PFCs at 13 European centers, Rinninella et al.
undergoing WON drainage (14 via LAMS, 26 via achieved a technical success rate of 99%. Overall
plastic stents). To be considered a treatment suc- clinical success occurred in 86/93 patients (92%)
cess, the pseudocyst or WON had to be ≤2 cm on without evidence of recurrence during average
CT/MRI in combination with resolution of the follow-up of 320 days. Treatment failure occurred
patient’s symptoms at 8 week follow-up. With in 6/93 patients (6%) due to persistent infection
regard to hospital costs, there was no difference requiring surgery (n = 3), perforation and signifi-
seen between WON treated with plastic stents vs cant bleeding caused by nasocystic drainage
LAMS, but hospital costs were significantly catheter (n = 2), and the need for a larger opening
decreased for pseudocysts drained with plastic to extract large necrotic tissue (n = 1). Major
stents ($18,996 vs $58,649, p = 0.03). The authors adverse events occurred in 5/93 patients (5%)
achieved treatment success in 80.9% of patients and included perforation and massive bleeding
(17/21); of the four patients who had treatment caused by the nasocystic drainage catheter
failure, three had WON and one had a PP. This (n = 2), pneumoperitoneum (n = 1), stent dis-
study did not demonstrate improved clinical out- lodgement during DEN (n = 1), and postdrainage
comes for LAMS over plastic stents for the treat- infection (n = 1) [45].
ment of pseudocysts or WON. The major In a recent retrospective, multicenter study of
advantage of LAMS was decreased procedure 13 patients with PFCs (69% with WON) who
time, while the major disadvantage of LAMS was underwent drainage with EC-LAMS, Adler
increased cost [43]. et al. had technical and clinical success rates of
1 Endoscopic Ultrasound-Guided Drainage of Pancreatic Fluid Collections 11

100%. Of note, DEN was carried out in all 9 treated with FCSEMSs (95% c­ onfidence interval,
patients with WON and there was no evidence 1.4–6.3) [29]. A large retrospective review of 211
of recurrence during an average follow-up time patients treated with FCSEMSs for PFCs had an
period of 2.5 months. There was one procedure- adverse event rate of 21%, which included infec-
related adverse event. In one patient, the LAMS tion (11%), bleeding (7%), and stent migration
was dislodged immediately after deployment, and/or perforation (3%) [21].
falling into the stomach where it was removed. The literature is conflicting on whether
A second electrocautery-enhanced LAMS was adverse events occur more frequently with metal
placed in this patient immediately afterward stents (LAMS and FCSEMS) versus DPPSs. A
[46]. Overall, EC-LAMS has been shown to be systematic review of 17 studies with 881 patients
safe and highly effective for drainage of PFCs with PFCs found there to be no statistically sig-
based on early data, although EC-LAMS has nificant differences in the rates of adverse events
now progressed to widespread adoption. (e.g., bleeding, secondary infection, and stent
migration) between metal vs plastic stents [38],
whereas another retrospective study of 103
Adverse Events patients with PFCs found that LAMSs have an
increased adverse event rate when compared with
 etal (LAMS and FCSEMS)
M DPPSs (53% versus 12%, P = 0.0003), specifi-
Versus DPPS cally with regard to more bleeding episodes and
unplanned repeat endoscopy [39]. A retrospec-
With regard to the safety of LAMSs versus tive review of 49 patients (31 with PP and 18 with
DPPSs, a recent small randomized trial for drain- WON) found that inadequate drainage occurred
age of PFCs via LAMSs versus DPPSs demon- in 10 cases treated with DPPSs compared to zero
strated stent-related adverse events in 50% (6/12) with LAMSs; in addition, the overall adverse
patients who received LAMSs and no adverse event rate was 13.5% for DPPSs and 0% for
events in patients who received DPPSs [47]. LAMSs [40]. A prospective study utilized DPPSs,
Similar results were seen in a previous study FCSEMSs, and LAMSs for the drainage of WON
using both LAMSs and DPPSs for drainage of in 313 patients had an overall adverse event rate
PFCs where stent-related adverse events occurred of 8.6%, which included perforation (n = 6),
in 10% (2/20) of patients who received LAMS bleeding (n = 8), suprainfection (n = 9), and other
and 2.5% (1/40) patients who received DPPSs (n = 7) [41].
[36]. Other studies using LAMSs for drainage of
PFCs with larger numbers of patients (n = 47 to
n = 124) have reported adverse event rates of EC-LAMS
5–20% [6, 43, 48–50]. Performing a CT scan
3 weeks post-procedure for all patients who Overall, adverse event rates for EC-LAMS are
received a LAMS followed by stent removal of low at 5–8%. The most common events reported
evidence of PFC resolution may be reasonable as include stent occlusion, stent migration, perfora-
proposed by Bang et al. [44]. The optimal time tion and bleeding, and infection [25, 43, 44].
frame to remove a LAMS after placement
remains unknown.
In 230 patients with PPs treated with DPPSs Infection
(n = 118) or FCSEMSs (n = 112), Sharaiha et al.
found a procedural adverse event rate of 31% in Infected pancreatic necrosis typically occurs
those treated with DPPSs and 16% with FCSEMSs within the first 2–3 months after the initial episode
(P = 0.006). Multivariate analysis demonstrated of necrotizing pancreatitis, but is rare during the
patients with plastic stents were 2.9 times more first week [51, 52]. The rates of secondary infec-
likely to experience adverse events than those tion after placement with each stent type is highly
12 J. S. Bank and D. G. Adler

variable in clinical studies, ranging 2.7–12% of Conclusion

DPPSs, 0–28% of FCSEMSs, and 0–15.2% in
LAMSs [16, 25, 28, 49, 51, 53, 54]. Interventional endoscopists can choose from
Guo et al. examined risk factors for infection DPPSs, FCSEMSs, and LAMSs for endoscopic
after EUS-guided drainage of PFCs in 83 patients drainage of PFCs. DPPSs have high technical and
and found evidence of infection in 7 (8.4%) clinical success rates and are less expensive than
patients. All 7 of these patients had a history of FCSEMSs and LAMSs, but are more prone to
acute pancreatitis and the cyst diameters were all stent occlusion, especially in the setting of
greater than 15 cm. They concluded that cyst WON. We recommend that they primarily be
diameter is an independent risk factor for infec- used for drainage of PPs as opposed to WON for
tion and that cysts with a diameter greater than this reason. FCSEMSs also have high technical
15 cm should be drained with either a large diam- and clinical success rates. They have a decreased
eter FCSEMS or multiple DPPSs to decrease the risk of occlusion compared to DPPSs, but have
risk of infection [53]. similar rates of stent migration. FCSEMSs should
One retrospective study of 93 patients found be utilized for drainage related to a luminal stric-
that the secondary infection rate for smaller diam- ture as opposed to LAMSs, which are designed
eter stents (8.5 Fr or less) was surprisingly less for a transluminal route. LAMSs have higher
than that for larger diameter stents (10 Fr or larger), technical and clinical success rates for drainage
3.4% versus 17.2%, respectively (P = 0.138) [16]. of WON as compared to both DPPSs and
The reasons for this are unclear and somewhat dif- FCSEMSs. This is likely due to their larger diam-
ficult to reconcile. eter, which allows for DEN. Endoscopists have
One case report described a case of infected many safe and effective stent options for EUS-­
pancreatic necrosis that developed 2 years after guided drainage of PFCs with high clinical suc-
necrotizing pancreatitis. The infected pancre- cess rates.
atic necrosis was treated using a LAMS with
subsequent endoscopic necrosectomy to
remove solid debris [54]. This case represents References
the longest time ever documented from the
onset of pancreatic necrosis to documented 1. Baillie J. Pancreatic pseudocysts (part I). Gastrointest
Endosc. 2004;59(7):873–9. Review.
infection. 2. Brun A, Agarwal N, Pitchumoni CS. Fluid collec-
tions in and around the pancreas in acute pancreatitis.
J Clin Gastroenterol. 2011;45(7):614–25.
Stent Migration and Occlusion 3. Pelaez-Luna M, Vege SS, Petersen BT, Chari ST, Clain
JE, Levy MJ, Pearson RK, Topazian MD, Farnell MB,
Kendrick ML, Baron TH. Disconnected pancreatic
There is a wide range of migration rates with duct syndrome in severe acute pancreatitis: clinical
each stent type reported in the literature, includ- and imaging characteristics and outcomes in a cohort
ing 0.7–18% with plastic stents, 0–10% with of 31 cases. Gastrointest Endosc. 2008;68(1):91–7.
4. Banks PA, Bollen TL, Dervenis C, Gooszen HG,
FCSEMSs, and 3–6.7% with LAMSs [25, 27, 28, Johnson CD, Sarr MG, Tsiotos GG, Vege SS,
55–60]. Acute Pancreatitis Classification Working Group.
A retrospective study of 31 patients with PPs Classification of acute pancreatitis—2012: revision of
treated with DPPSs had 6 (25%) symptomatic the Atlanta classification and definitions by interna-
tional consensus. Gut. 2013;62(1):102–11.
recurrences due to stent clogging or migration, 5. Morgan DE, Baron TH, Smith JK, Robbin ML,
with 2 secondary infections during a mean fol- Kenney PJ. Pancreatic fluid collections prior to inter-
low-­up time period of 12.6 months. Both were vention: evaluation with MR imaging compared with
treated with new DPPSs stents [17]. In DPPSs, CT and US. Radiology. 1997;203(3):773–8.
6. Dhaka N, Samanta J, Kochhar S, Kalra N, Appasani
occlusion rates increase with smaller diameter S, Manrai M, Kochhar R. Pancreatic fluid collec-
stents, such as 7 Fr, so larger sizes are typically tions: what is the ideal imaging technique? World
used [11]. J Gastroenterol. 2015;21(48):13403–10.
1 Endoscopic Ultrasound-Guided Drainage of Pancreatic Fluid Collections 13

7. Rana SS, Bhasin DK, Reddy YR, Sharma V, Rao C, t­ransmural drainage of pancreatic pseudocysts. Arq
Sharma RK, Gupta R. Morphological features of fluid Gastroenterol. 2008;45(1):17–21.
collections on endoscopic ultrasound in acute necro- 20. Antillon MR, Shah RJ, Stiegmann G, Chen
tizing pancreatitis: do they change over time? Ann YK. Single-step EUS-guided transmural drainage
Gastroenterol. 2014;27(3):258–61. of simple and complicated pancreatic pseudocysts.
8. Siddiqui AA, Adler DG, Nieto J, Shah JN, Binmoeller Gastrointest Endosc. 2006;63(6):797–803.
KF, Kane S, Yan L, Laique SN, Kowalski T, Loren 21. Singhal S, Rotman SR, Gaidhane M, Kahaleh
DE, Taylor LJ, Munigala S, Bhat YM. EUS-guided M. Pancreatic fluid collection drainage by endo-
drainage of peripancreatic fluid collections and necro- scopic ultrasound: an update. Clin Endosc. 2013;
sis by using a novel lumen-apposing stent: a large ret- 46(5):506–14.
rospective, multicenter U.S. experience (with videos). 22. Bang JY, Varadarajulu S. Metal versus plastic stent for
Gastrointest Endosc. 2016;83(4):699–707. transmural drainage of pancreatic fluid collections.
9. Adler DG, Siddiqui AA. What's in a name? Pancreatic Clin Endosc. 2013;46(5):500–2.
pseudocysts, walled-off necrosis, and pancreatic fluid 23. Vazquez-Sequeiros E, Baron TH, Pérez-Miranda M,
collections. Endosc Ultrasound. 2016;5(4):215–7. Sánchez-Yagüe A, Gornals J, Gonzalez-Huix F, de la
10. Seewald S, Ang TL, Kida M, Teng KY, Soehendra N, Serna C, Gonzalez Martin JA, Gimeno-Garcia AZ,
EUS 2008 Working Group. EUS 2008 Working Group Marra-Lopez C, Castellot A, Alberca F, Fernandez-­
document: evaluation of EUS-guided drainage of Urien I, Aparicio JR, Legaz ML, Sendino O, Loras C,
pancreatic-fluid collections (with video). Gastrointest Subtil JC, Nerin J, Perez-Carreras M, Diaz-Tasende
Endosc. 2009;69(2 Suppl):S13–21. J, Perez G, Repiso A, Vilella A, Dolz C, Alvarez A,
11. Teoh AY, Dhir V, Jin ZD, Kida M, Seo DW, Ho Rodriguez S, Esteban JM, Juzgado D, Albillos A,
KY. Systematic review comparing endoscopic, percu- Spanish Group for FCSEMS in Pancreas Collections.
taneous and surgical pancreatic pseudocyst drainage. Evaluation of the short- and long-term effectiveness
World J Gastrointest Endosc. 2016;8(6):310–8. and safety of fully covered self-expandable metal
12. Ang TL, Teoh AYB. Endoscopic ultrasonography-­ stents for drainage of pancreatic fluid collections:
guided drainage of pancreatic fluid collections. Dig results of a Spanish nationwide registry. Gastrointest
Endosc. 2017;29(4):463–71. Endosc. 2016;84(3):450–457.e2.
13. Mangiavillano B, Pagano N, Baron TH, Arena 24. Varadarajulu S, Wilcox CM, Latif S, Phadnis M,
M, Iabichino G, Consolo P, Opocher E, Luigiano Christein JD. Management of pancreatic fluid collec-
C. Biliary and pancreatic stenting: devices and inser- tions: a changing of the guard from surgery to endos-
tion techniques in therapeutic endoscopic retrograde copy. Am Surg. 2011;77(12):1650–5.
cholangiopancreatography and endoscopic ultra- 25. Siddiqui AA, Kowalski TE, Loren DE, Khalid A,
sonography. World J Gastrointest Endosc. 2016; Soomro A, Mazhar SM, Isby L, Kahaleh M, Karia K,
8(3):143–56. Yoo J, Ofosu A, Ng B, Sharaiha RZ. Fully covered
14. Wiersema MJ. Endosonography-guided cystoduode- self-expanding metal stents versus lumen-apposing
nostomy with a therapeutic ultrasound endoscope. fully covered self-expanding metal stent versus
Gastrointest Endosc. 1996;44(5):614–7. plastic stents for endoscopic drainage of pancreatic
15. Giovannini M, Bernardini D, Seitz JF. walled-off necrosis: clinical outcomes and success.
Cystogastrotomy entirely performed under endo- Gastrointest Endosc. 2017;85(4):758–65.
sonography guidance for pancreatic pseudocyst: 26. Adler DG. Drainage of a symptomatic pancreatic
results in six patients. Gastrointest Endosc. 1998; pseudocyst via a fully covered self expanding metal
48(2):200–3. stent (FCSEMS). Pract Gastroenterol. 2014;4:26–30.
16. Seifert H, Dietrich C, Schmitt T, Caspary W, 27. Berzosa M, Maheshwari S, Patel KK, Shaib
Wehrmann T. Endoscopic ultrasound-guided one-­ YH. Single-step endoscopic ultrasonography-guided
step transmural drainage of cystic abdominal lesions drainage of peripancreatic fluid collections with a
with a large-channel echo endoscope. Endoscopy. single self-expandable metal stent and standard linear
2000;32(3):255–9. echoendoscope. Endoscopy. 2012;44(5):543–7.
17. Seicean A, Stan-Iuga R, Badea R, Tantau M, Mocan 28. Weilert F, Binmoeller KF, Shah JN, Bhat YM, Kane
T, Seicean R, Iancu C, Pascu O. The safety of endo- S. Endoscopic ultrasound-guided drainage of pancre-
scopic ultrasonography-guided drainage of pancreatic atic fluid collections with indeterminate adherence
fluid collections without fluoroscopic control: a single using temporary covered metal stents. Endoscopy.
tertiary center experience. J Gastrointestin Liver Dis. 2012;44(8):780–3.
2011;20(1):39–45. 29. Talreja JP, Shami VM, Ku J, Morris TD, Ellen K,
18. Lin H, Zhan XB, Sun SY, Yang XJ, Jin ZD, Zou DW, Kahaleh M. Transenteric drainage of pancreatic-­
Li ZS. Stent selection for endoscopic ultrasound-­ fluid collections with fully covered self-expanding
guided drainage of pancreatic fluid collections: a metallic stents (with video). Gastrointest Endosc.
multicenter study in China. Gastroenterol Res Pract. 2008;68(6):1199–203.
2014;2014:193562. 30. Penn DE, Draganov PV, Wagh MS, Forsmark CE,
19. Lopes CV, Pesenti C, Bories E, Caillol F, Giovannini Gupte AR, Chauhan SS. Prospective evaluation of the
M. Endoscopic ultrasound-guided endoscopic use of fully covered self-expanding metal stents for
14 J. S. Bank and D. G. Adler

EUS-guided transmural drainage of pancreatic pseu- 43. Bang JY, Hasan MK, Navaneethan U, Sutton B,
docysts. Gastrointest Endosc. 2012;76(3):679–84. Frandah W, Siddique S, Hawes RH, Varadarajulu
31. Sharaiha RZ, DeFilippis EM, Kedia P, Gaidhane M, S. Lumen-apposing metal stents for drainage of pan-
Boumitri C, Lim HW, Han E, Singh H, Ghumman SS, creatic fluid collections: when and for whom? Dig
Kowalski T, Loren D, Kahaleh M, Siddiqui A. Metal Endosc. 2017;29(1):83–90.
versus plastic for pancreatic pseudocyst drainage: 44. Yoo J, Yan L, Hasan R, Somalya S, Nieto J, Siddiqui
clinical outcomes and success. Gastrointest Endosc. AA. Feasibility, safety, and outcomes of a single-­step
2015;82(5):822–7. endoscopic ultrasonography-guided drainage of pan-
32. Binmoeller KF, Shah J. A novel lumen-apposing stent creatic fluid collections without fluoroscopy using
for transluminal drainage of nonadherent extraintesti- a novel electrocautery-enhanced lumen-­ apposing,
nal fluid collections. Endoscopy. 2011;43(4):337–42. self-expanding metal stent. Endosc Ultrasound.
33. Itoi T, Nageshwar Reddy D, Yasuda I. New fully-­ 2017;6(2):131–5.
covered self-expandable metal stent for endo- 45. Rinninella E, Kunda R, Dollhopf M, Sanchez-Yague
scopic ultrasonography-guided intervention in A, Will U, Tarantino I, Gornals Soler J, Ullrich S,
infectious walled-off pancreatic necrosis (with video). Meining A, Esteban JM, Enz T, Vanbiervliet G,
J Hepatobiliary Pancreat Sci. 2013;20(3):403–6. Vleggaar F, Attili F, Larghi A. EUS-guided drainage
34. Moon JH, Choi HJ, Kim DC, Lee YN, Kim HK, Jeong of pancreatic fluid collections using a novel lumen-­
SA, et al. A newly designed fully covered metal stent apposing metal stent on an electrocautery-enhanced
for lumen apposition in EUS-guided drainage and delivery system: a large retrospective study (with
access: a feasibility study (with videos). Gastrointest video). Gastrointest Endosc. 2015;82(6):1039–46.
Endosc. 2014;79(6):990–5. 46. Adler DG, Taylor LJ, Hasan R, Siddiqui AA. A ret-
35. Baron TH, Harewood GC, Morgan DE, Yates rospective study evaluating endoscopic ultrasound-­
MR. Outcome differences after endoscopic drainage guided drainage of pancreatic fluid collections using
of pancreatic necrosis, acute pancreatic pseudocysts, a novel lumen-apposing metal stent on an electrocau-
and chronic pancreatic pseudocysts. Gastrointest tery enhanced delivery system. Endosc Ultrasound.
Endosc. 2002;56(1):7–17. 2017;6(6):389–93. https://doi.org/10.4103/eus.eus_
36. Lee HS, Chung MJ. Past, present, and future of gas- 4_17.
trointestinal stents: new endoscopic ultrasonography-­ 47. Bang JY, Hasan M, Navaneethan U, Hawes R,
guided metal stents and future developments. Clin Varadarajulu S. Lumen-apposing metal stents
Endosc. 2016;49(2):131–8. (LAMS) for pancreatic fluid collection (PFC) drain-
37. McVay T, Adler DG. EUS-guided drainage of pancre- age: may not be business as usual. Gut. 2017;66:2054.
atic fluid collections: double pigtails, metal biliary, 48. Walter D, Will U, Sanchez-Yague A, Brenke D,
or dedicated transluminal stents? Endosc Ultrasound. Hampe J, Wollny H, López-Jamar JM, Jechart G,
2015;4(1):1–3. Vilmann P, Gornals JB, Ullrich S, Fähndrich M, de
38. Bang JY, Hasan MK, Navaneethan U, Sutton B, Tejada AH, Junquera F, Gonzalez-Huix F, Siersema
Frandah W, Siddique S, et al. Lumen apposing metal PD, Vleggaar FP. A novel lumen-apposing metal stent
stents (LAMS) for drainage of pancreatic fluid collec- for endoscopic ultrasound-guided drainage of pan-
tions: when and for whom? Dig Endosc. 2017;29:83. creatic fluid collections: a prospective cohort study.
39. Varadarajulu S, Tamhane A, Blakely J. Graded dila- Endoscopy. 2015;47(1):63–7.
tion technique for EUS-guided drainage of peripan- 49. Chandran S, Efthymiou M, Kaffes A, Chen JW, Kwan
creatic fluid collections: an assessment of outcomes V, Murray M, Williams D, Nguyen NQ, Tam W,
and complications and technical proficiency (with Welch C, Chong A, Gupta S, Devereaux B, Tagkalidis
video). Gastrointest Endosc. 2008;68(4):656–66. P, Parker F, Vaughan R. Management of pancre-
40. Bang JY, Hawes R, Bartolucci A, Varadarajulu atic collections with a novel endoscopically placed
S. Efficacy of metal and plastic stents for transmural fully covered self-expandable metal stent: a national
drainage of pancreatic fluid collections: a systematic experience (with videos). Gastrointest Endosc.
review. Dig Endosc. 2015;27(4):486–98. 2015;81(1):127–35.
41. Lang GD, Fritz C, Bhat T, Das KK, Murad FM, 50. Sharaiha RZ, Tyberg A, Khashab MA, Kumta NA,
Early DS, Edmundowicz SA, Kushnir VM, Mullady Karia K, Nieto J, Siddiqui UD, Waxman I, Joshi
DK. EUS-guided drainage of peripancreatic fluid V, Benias PC, Darwin P, DiMaio CJ, Mulder CJ,
collections with lumen-apposing metal stents and Friedland S, Forcione DG, Sejpal DV, Gonda TA,
plastic double-pigtail stents: comparison of effi- Gress FG, Gaidhane M, Koons A, DeFilippis EM,
cacy and adverse event rates. Gastrointest Endosc. Salgado S, Weaver KR, Poneros JM, Sethi A, Ho S,
2018;87:150. Kumbhari V, Singh VK, Tieu AH, Para V, Likhitsup
42. Ang TL, Kongkam P, Kwek AB, Orkoonsawat P, A, Womeldorph C, Casey B, Jonnalagadda SS,
Rerknimitr R, Fock KM. A two-center comparative Desai AP, Carr-Locke DL, Kahaleh M, Siddiqui
study of plastic and lumen-apposing large diam- AA. Endoscopic therapy with lumen-apposing metal
eter self-expandable metallic stents in endoscopic stents is safe and effective for patients with pancre-
ultrasound-­guided drainage of pancreatic fluid collec- atic walled-off necrosis. Clin Gastroenterol Hepatol.
tions. Endosc Ultrasound. 2016;5(5):320–7. 2016;14:1797.
1 Endoscopic Ultrasound-Guided Drainage of Pancreatic Fluid Collections 15

51. Besselink MG, van Santvoort HC, Boermeester drainage of pancreatic pseudocyst: a prospective
MA, Nieuwenhuijs VB, van Goor H, Dejong CH, comparison with conventional endoscopic drainage.
Schaapherder AF, Gooszen HG, Dutch Acute Endoscopy. 2006;38(4):355–9.
Pancreatitis Study Group. Timing and impact of 56. Ahlawat SK, Charabaty-Pishvaian A, Jackson PG,
infections in acute pancreatitis. Br J Surg. 2009; Haddad NG. Single-step EUS-guided pancreatic
96(3):267–73. pseudocyst drainage using a large channel linear array
52. van Santvoort HC, Besselink MG, Bakker OJ, Hofker echoendoscope and cystotome: results in 11 patients.
HS, Boermeester MA, Dejong CH, van Goor H, JOP. 2006;7(6):616–24.
Schaapherder AF, van Eijck CH, Bollen TL, van 57. Itoi T, Binmoeller KF, Shah J, Sofuni A, Itokawa F,
Ramshorst B, Nieuwenhuijs VB, Timmer R, Laméris Kurihara T, Tsuchiya T, Ishii K, Tsuji S, Ikeuchi N,
JS, Kruyt PM, Manusama ER, van der Harst E, van der Moriyasu F. Clinical evaluation of a novel lumen-­
Schelling GP, Karsten T, Hesselink EJ, van Laarhoven apposing metal stent for endosonography-guided
CJ, Rosman C, Bosscha K, de Wit RJ, Houdijk AP, pancreatic pseudocyst and gallbladder drainage (with
van Leeuwen MS, Buskens E, Gooszen HG, Dutch videos). Gastrointest Endosc. 2012;75(4):870–6.
Pancreatitis Study Group. A step-up approach or open 58. Shah RJ, Shah JN, Waxman I, Kowalski TE, Sanchez-­
necrosectomy for necrotizing pancreatitis. N Engl Yague A, Nieto J, Brauer BC, Gaidhane M, Kahaleh
J Med. 2010;362(16):1491–502. M. Safety and efficacy of endoscopic ultrasound-­
53. Guo J, Feng L, Sun S, Ge N, Liu X, Wang S, Wang guided drainage of pancreatic fluid collections with
G, Sun B. Risk factors for infection after endoscopic lumen-apposing covered self-expanding metal stents.
ultrasonography-guided drainage of specific types of Clin Gastroenterol Hepatol. 2015;13(4):747–52.
pancreatic and peripancreatic fluid collections (with 59. Bapaye A, Itoi T, Kongkam P, Dubale N, Mukai
video). Surg Endosc. 2016;30(7):3114–20. S. New fully covered large-bore wide-flare remov-
54. Eliason K, Adler DG. Endoscopic ultrasound-guided able metal stent for drainage of pancreatic fluid col-
transmural drainage of infected pancreatic necrosis lections: results of a multicenter study. Dig Endosc.
developing 2 years after acute pancreatitis. Endosc 2015;27(4):499–504.
Ultrasound. 2015;4(3):260–5. 60. Varadarajulu S, Christein JD, Wilcox CM. Frequency
55. Kahaleh M, Shami VM, Conaway MR, Tokar J, of complications during EUS-guided drainage of pan-
Rockoff T, De La Rue SA, de Lange E, Bassignani M, creatic fluid collections in 148 consecutive patients.
Gay S, Adams RB, Yeaton P. Endoscopic ultrasound J Gastroenterol Hepatol. 2011;26(10):1504–8.
 Endoscopic Ultrasound-Guided
Bile Duct Access and Drainage: 2
Antegrade Approaches

Nan Ge and Siyu Sun

data are supported by a new prospective interna-

Introduction tional multicenter study citing comparable tech-
nical (95.7%) and clinical (95%) success rates in
EUS-BD an intention-to-treat analysis, with an adverse
event rate of 10.5% for patients with malignant
Endoscopic ultrasound (EUS), which couples distal biliary obstruction [6]. Compared with per-
endoscopy with ultrasound in examining the GI cutaneous transhepatic biliary drainage (PTBD),
tract and adjacent structures, has improved our EUS-BD is deemed superior in terms of possible
understanding of many disease states [1–3]. catheter dislodgement, recurrent infection, acute
EUS-guided fine needle aspiration (FNA) was cholangitis, pneumothorax, and cosmetic prob-
first reported in 1992. Its use has proved superior lems (due to external drainage) [9].
to EUS alone in accurately diagnosing and stag- EUS-BD by rendezvous technique and EUS-­
ing malignancies, thus aiding in treatment selec- guided choledochoduodenostomy (EUS-CDS);
tion. EUS is also a unique, effective, and hepaticogastrostomy (EUS-HGS) or hepaticodu-
minimally invasive therapeutic technique [4], odenostomy; and antegrade (EUS-AG) stenting
enabling bile duct drainage as well [5]. technique are the most common interventions in
Since the advent of EUS-guided biliary drain- this context.
age (EUS-BD) in 2001, [6, 7] the efficacy and
safety of this type intervention has been amply
demonstrated. A recent meta-analysis has shown EUS-AG
that EUS-BD has a pooled technical success rate
of 90–95% in therapeutic applications, with a EUS-AG stenting is a viable treatment option for
relatively low (14–17%) cumulative rate of patients with bile duct stones (BDS) or obstruc-
adverse events after failed endoscopic retrograde tion (BDO) in whom ERCP has failed, primarily
cholangiopancreatography (ERCP) [8]. These as a consequence of surgically altered anatomy
[10], duodenal obstruction, or inability to cannu-
Electronic supplementary material: The online version late the papilla.
of this chapter (https://doi.org/10.1007/978-3-319-97376- EUS-AG procedures may have several advan-
0_2) contains supplementary material, which is available tages over other EUS-guided techniques, espe-
to authorized users.
cially in patients whose surgically altered
N. Ge · S. Sun (*) anatomy precludes ready choledochoduodenos-
Endoscopy Center, Shengjing Hospital of China
Medical University, Shenyang, Liaoning, China tomy access. Likewise, EUS-AG may reduce the
e-mail: [email protected]; [email protected] occurrence of stent migration compared with

© Springer Nature Switzerland AG 2019 17

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_2
18 N. Ge and S. Sun

HGS. Long-term influx of intestinal content into help deter guide wire shearing during
the biliary system is then a potential risk. The interventions. In studies conducted by
EUS-AG approach confers clinical benefits of Godat et al. [11], both the 19G EchoTip
durable stent patency and fewer adverse events, Needle and 19G EchoTip Access Needle
such as bile peritonitis. EUS-AG holds promise were used for intrahepatic bile duct
as a safe and effective alternative in managing (IHBD) puncture, culminating in a higher
biliary disorders (Table 2.1). success rate for the 19G EchoTip Needle
(16/16, 100% vs. 5/7, 71%). No instances
of interventional guide wire shearing were
Indications recorded in this study.
(b) Boston 19G Flexible Needle (Boston
• Obstructive jaundice and failed ERCP or non-­ Scientific Corp, Marlborough, MA, USA)
access to ampulla of Vater (c) SonoTip Pro Control 19G Fine Needle
(a) Surgically altered anatomy (Medi-Globe GmbH, Rosenheim,
(b) Upper intestinal obstruction Germany; Medico’s Hirata Inc, Osaka,
(c) In situ enteral stents Japan)
(d) Problematic bile duct intubation (periam- (d) Dilators (hepaticogastric tract dilation)
pullary diverticular or infiltrative tumor (e) 6Fr cystotome (ENDO-FLEX GmbH ) for
impingement) dilation of hepaticogastric tract or biliary
• Common bile duct stones and failed ERCP or stenosis
non-access to ampulla of Vater (f) 6Fr, 8Fr Soehendra biliary dilation cathe-
(a) Surgically altered anatomy ter (Cook Medical)
(b) Upper intestinal obstruction (g) 6Fr diathermic dilator (Cysto-Gastro-Set;
(c) In situ enteral stents Endo-Flex, Voerde, Germany)
(d) Problematic bile duct intubation (periam- (h) Balloon for dilatation biliary stenosis
pullary diverticular or infiltrative tumor (i) 6Fr, 7Fr Bougie dilator (PD-SS6F180C;
impingement) Gadelius Medical, Tokyo, Japan)
• Guide wires
(a) 0.035-in (Tracer Metro Direct Wire Guide,
Contraindications Cook Medical or Radiofocus, Terumo,
Tokyo, Japan)
• Contraindication to EUS, including ERCP (b) Cyst guide wire (nitinol uncoated; Medi-­
• Normal coagulation studies and platelet count Globe GmbH)
• Severe organ failure (c) 0.025-in (VisiGlide, Olympus, Tokyo, Japan
• Failure to obtain consent or Revowave, Piolax Medical Devices,
Kanagawa, Japan)
(d) 0.018-in guide wire with 22G fine needle
Devices (Table 2.2) is also a choice for some operators

The following list included suggested devices for

EUS-AG biliary drainage procedures. Many dif- Approaches
ferent devices from multiple manufacturers could
potentially be used for these procedures. • EUS-AG bile duct stones removal (Fig. 2.1)
(a) EUS scan of left hepatic lobe for dilated
• Fine Needles IHBDs.
(a) 19G EchoTip Needle or EchoTip Access (b) Color Doppler identification of interven-
Needle (Cook Medical, Bloomington, IN, ing vessels to avoid during puncture.
USA). The EchoTip Access Needle may
 2

Table 2.1 Summary of EUS-AG stenting trials

Technical Clinical Puncture site
Study Author Treatment Cause success success Adverse event closure
1 Godat SEMS placement 20/20 Malignant biliary obstruction 1/20 20/20 17/20 Persistent obstructive N
[11] cholangitis 1/20
Benign biliodigestive anastomotic Infection 2/20
stenosis 19/20
2 Iwashita SEMS placement 1/6 Malignant biliary obstruction 1/6 6/6 6/6 Mild pancreatitis 1/6 N
[12] Balloon dilation (1–15 mm) and Choledocholithiasis 4/6 Mild abdominal pain 1/6
stone removal 5/6 Anastomotic stricture 1/6
3 Ogura SEMS placement Malignant biliary obstruction 42/49 40/49 Hyperamylasemia 4/49 N
[13] Bleeding 1/49
4 Iwashita SEMS placement Malignant 19/20 19/20 Mild pancreatitis 3/20 N
[14] Biliary obstruction Mild fever 1/20
Endoscopic Ultrasound-Guided Bile Duct Access and Drainage: Antegrade Approaches
19
20 N. Ge and S. Sun

Table 2.2 Device usage in EUS-AG stenting trials

Needles Guide wire (in) Dilator ENBD Stent
1 19G EchoTip (16/20) 0.035 Cystotome 6Fr Non-covered metal stents
19G EchoTip Access (7/20)
2 19G EchoTip (7/7) 0.025 Bougie dilator 7Fr 5/6 Non-covered metal
3 19G SonoTip Pro Control 0.025 Dilation catheter (23/49) Partly covered metal
19G EchoTip Ultra Balloon catheter (24/49)
Diathermic dilator (1/49)
4 19G SonoTip Pro Control 0.025 Bougie dilator 6Fr 3/19 Uncovered SEMS

(c) EchoTip Ultra Endoscopic Ultrasound (d) Endoscopic Ultrasound Needle intro-
Needle introduced via working channel of duced via working channel of echoendo-
echoendoscope, puncturing bile duct scope, puncturing bile duct under EUS
under EUS guidance. guidance.
(d) Sample aspirate obtained (after with- (e) Sample aspirate obtained (after with-
drawal of stylet) to confirm bile duct drawal of stylet) to confirm bile duct
puncture. puncture.
(e) Radiopaque contrast injected into bile (f) Radiopaque contrast injected into bile
duct for detecting common duct stones. duct to characterize stenosis.
(f) Guide wire inserted into bile duct. (g) Guide wire inserted into bile duct, maneu-
(g) Maneuvering of guide wire antegrade for vering past stenosis (minimizing with-
passage along common bile duct and drawal movements to avoid damaging
through papilla (minimizing withdrawal surface of guide wire).
movements to avoid damaging surface of (h) Once guide wire moves past stenosis,
guide wire). puncture needle is withdrawn, using cys-
(h) Once guide wire traverses papilla, punc- totome (6Fr) to dilate needle path.
ture needle is withdrawn, using cystotome (i) After successful dilation, SEMS place-
(6Fr) to dilate needle path (hepaticogas- ment proceeds (along guide wire) under
tric tract). fluoroscopic guidance, deployed at ste-
(i) Via newly made hepaticogastric tract, bal- notic site (preferably in transpapillary
loon dilator is placed in papilla under position).
X-ray surveillance (diameter of dilator (j) Nasobiliary drainage (NBD) tube placed
dependent on stone size). as warranted in fistula between stomach
(j) Stone balloon serves to push calculi along and liver, if necessary.
guide wire, through papilla, and into GI
lumen under fluoroscopic guidance.
(k) Nasobiliary drainage (NBD) tube placed Success Rate
as warranted in fistula between stomach
and liver; to be removed after 48 h, ensur- Although data is still limited, the technical and
ing proper functioning of biliary self-­ clinical success rates of EUS-AG procedures are
expandable metal stent (SEMS). relatively high (technical success: 85.7–100%;
• EUS-AG bile duct drainage (Fig. 2.2 and clinical success rate: 81.6–100%). Guide wire
Video 2.1) shearing may occur during manipulation and is one
(a) EUS scan of liver for dilated IHBDs. reason for failed attempts. Some interventionists
(b) Color Doppler identification of interven- may use small-caliber guide wires (0.025 in) or the
ing vessels to avoid during puncture. 19G EchoTip Access Needle to avoid this problem.
(c) Stylet withdrawn upon needle tip entry of Disadvantages of the 19G EchoTip Access Needle
bile duct. are the risk of submucosal parabiliary insertion and
2 Endoscopic Ultrasound-Guided Bile Duct Access and Drainage: Antegrade Approaches 21

Fig. 2.1 61-year-old man with common bile duct stones injection within bile duct. (c) Dilatation of papilla
and duodenal bulb stenosis, due to duodenal bulb ulcer (a) achieved via balloon. (d) Lithotomy balloon enabling
EUS scanning of dilated IHBD (left lobe), punctured by delivery of stone into duodenal cavity. (e) ENBD inserted
19G needle. (b) Guide wire entry of common bile duct into common bile duct (endoscopic view). (f) X-ray image
and passage through papilla, enabling contrast agent of NBD passage through papilla
22 N. Ge and S. Sun

Fig. 2.1 (continued)

a lesser rate of IHBD puncture, compared with the ensuring SEMS patency by maintaining a protec-
19G EchoTip Needle. In other instances, the extent tive nasobiliary catheter in 10% (2/20) of patients.
of biliary obstruction may prohibit guide wire In this particular trial, EUS-AG was deemed
advancement into the intestine. inadequate in 10% (2/20) of the patients treated,
although drainage was achieved by concurrent
EUS-HGS in one patient and by percutaneous
Complications drainage of right IHBD in another.
Bile peritonitis has not been reported in any of
Severe complications of EUS-AG stenting have the studies. Clinically, the patency of EUS-AG
not been reported. Other minor complications, stenting is durable, with little risk of adverse events,
including obstructive cholangitis, infection, pan- although there is a potential for acute pancreatitis
creatitis, mild abdominal pain, and fever, have due to obstruction of pancreatic duct orifice.
ranged from 10 to 33%. As mentioned, biliary EUS-AG has the potential to be an effective
SEMS should be placed in transpapillary position and safe alternative management method for bili-
to ease any resistance to bile flow. Certain ary disorders in patients with surgically altered
researchers [11] have extended the process, plac- anatomies or failed papilla cannulation. At pres-
ing a second non-covered SEMS through one-­ ent, it shows some possible advantages over other
step interventions in 15% (3/20) of patients and approach routes (ERCP, PTBD, or EUS-HGS),
2 Endoscopic Ultrasound-Guided Bile Duct Access and Drainage: Antegrade Approaches 23

Fig. 2.2 55-year-old patient with unresectable malignant mon bile duct with cystotome. (d) Guide wire passed
bile duct obstruction; 4 years ago, he had the surgery of through stenotic site, entering duodenal cavity. (e) Stent
pancreaticoduodenectomy; initial ERCP attempt unsuc- placement in common bile duct, past stenosis. (f)
cessful due to surgically altered anatomy (a) Guide wire Uncovered metal stent deployed at stenosis. (g) CT
entry of intrahepatic bile duct. (b) Needle path dilated by revealed the stent
cystotome (6Fr). (c) Guide wire maneuvered into com-

3. De Castro VL, Moura EG, Chaves DM, Bernardo WM,

however, there is not enough evidence to con- Matuguma SE, Artifon EL. Endoscopic ultrasound
clude that any one procedure is superior to versus magnetic resonance cholangiopancreatography
another, and further comparisons among the in suspected choledocholithiasis: a systematic review.
approaches are required. Endosc Ultrasound. 2016;5(2):118–28.
4. Martin A, Kistler CA, Wrobel P, Yang JF, Siddiqui
AA. Endoscopic ultrasound-guided pancreaticobili-
ary intervention in patients with surgically altered
References anatomy and inaccessible papillae: a review of current
literature. Endosc Ultrasound. 2016;5(3):149–56.
1. Saftoiu A, Vilmann P, Bhutani MS. The role of 5. Mukai S, Itoi T. How should we use endoscopic
contrast-enhanced endoscopic ultrasound in pan- ultrasonography-­ guided biliary drainage techniques
creatic adenocarcinoma. Endosc Ultrasound. 2016; separately? Endosc Ultrasound. 2016;5(2):65–8.
5(6):368–72. 6. Dhir V, Itoi T, Khashab MA, Park DH, Yuen Bun Teoh
2. Sharma M, Rameshbabu CS, Dietrich CF, Rai P, A, Attam R, et al. Multicenter comparative evaluation
Bansal R. Endoscopic ultrasound of the hepatoduo- of endoscopic placement of expandable metal stents
denal ligament and liver hilum. Endosc Ultrasound. for malignant distal common bile duct obstruction by
2016;7:168–74. ERCP or EUS-guided approach. Gastrointest Endosc.
2015;81(4):913–23.
24 N. Ge and S. Sun

7. Giovannini M, Moutardier V, Pesenti C, Bories 11. Godat S, Bories E, Caillol F, Pesenti C, Ratone JP, de
E, Lelong B, Delpero JR. Endoscopic ultra- Cassan C, et al. Efficacy and safety in case of technical
sound-guided bilioduodenal anastomosis: a success of endoscopic ultrasound-guided transhepatic
new technique for biliary drainage. Endoscopy. antegrade biliary drainage: a report of a monocentric
2001;33(10):898–900. study. Endosc Ultrasound. 2017;6(3):181–6.
8. Sharaiha RZ, Khan MA, Kamal F, Tyberg A, 12. Iwashita T, Yasuda I, Doi S, Uemura S, Mabuchi
Tombazzi CR, Ali B, et al. Efficacy and safety of M, Okuno M, et al. Endoscopic ultrasound-guided
EUS-guided biliary drainage in comparison with antegrade treatments for biliary disorders in patients
percutaneous biliary drainage when ERCP fails: a with surgically altered anatomy. Dig Dis Sci. 2013;
systematic review and meta-analysis. Gastrointest 58(8):2417–22.
Endosc. 2017;85(5):904–14. 13. Ogura T, Kitano M, Takenaka M, Okuda A, Minaga
9. Oh HC, Lee SK, Lee TY, Kwon S, Lee SS, Seo K, Yamao K, et al. A multicenter prospective evalua-
DW, et al. Analysis of percutaneous transhepatic tion study of endoscopic ultrasound-guided hepatico-
cholangioscopy-­related complications and the risk gastrostomy combined with antegrade stenting (with
factors for those complications. Endoscopy. 2007; video). Dig Endosc. 2017;30:252–9.
39(8):731–6. 14. Iwashita T, Yasuda I, Mukai T, Iwata K, Doi S,
10. Nakai Y, Isayama H, Koike K. Two-step EUS-guided Uemura S, et al. Endoscopic ultrasound-guided ante-
antegrade treatment of a difficult bile duct stone in grade biliary stenting for unresectable malignant
a surgically altered anatomy patient. Dig Endosc. biliary obstruction in patients with surgically altered
2017;30:125–7. anatomy: single-center prospective pilot study. Dig
Endosc. 2017;29(3):362–8.
 Endoscopic Ultrasound-Guided
Biliary Drainage: Retrograde 3
Approaches

Constantine Melitas and Douglas G. Adler

the bifurcation), then an EUS antegrade approach,

Introduction typically via hepaticogastrostomy, may be used
to gain access into the intrahepatic biliary system
Endoscopic ultrasound (EUS) has several advan- to assist with drainage. If the ampulla is accessi-
tages over luminal procedures as it allows for ble and standard ERCP cannulation techniques
access to structures which typically have no direct have failed or the intrahepatic bile ducts are not
communication. Furthermore, adjacent structures dilated, a retrograde EUS-guided approach is
may have altered anatomy based on disease infil- more favorable [4].
tration making typically simple luminal proce- Accessing the common bile duct (CBD)
dures difficult and sometimes impossible. Thus, directly has been shown to be a more favorable
in biliary obstruction, where endoscopic retro- approach when compared to accessing the intra-
grade cholangiopancreatography (ERCP) has hepatic bile ducts as the CBD has much thicker
failed or is not technically possible due to post- walls, closer apposition to the bowel lumen, and
surgical anatomy or malignant gastric outlet allows for greater stabilization of instrumenta-
obstruction, EUS can be used to directly access tion, and is a much more widely utilized target
the biliary system through the intestinal wall to for EUS-guided biliary drainage leading to
assist with biliary drainage [1–3]. higher success rates [5].
If the ampulla is not able to be reached endo-
scopically and/or the level of biliary obstruction
is very proximal (most commonly at the level of EUS-Guided Rendezvous

There are two main methods of retrograde biliary

Electronic supplementary material: The online version drainage via EUS. The first method is also known
of this chapter (https://doi.org/10.1007/978-3-319-97376-
0_3) contains supplementary material, which is available as a “rendezvous” procedure which has also been
to authorized users. performed with the assistance of interventional
C. Melitas radiologists in the past [6]. Endoscopically, this
Michigan State University College of Human procedure can be performed via intrahepatic
Medicine/Providence-Providence Park Hospitals, approach or extrahepatic approach [7]. The intra-
Southfield, MI, USA hepatic approach is a transgastric method
D. G. Adler (*) whereas the extrahepatic method is a transduode-
Department of Internal Medicine, Division of nal (either via the first or second portion of the
Gastroenterology and Hepatology, University of Utah
School of Medicine, Salt Lake City, UT, USA duodenum, known as D1 and D2). In the retro-
e-mail: [email protected] grade approach, the bile duct is accessed by using

© Springer Nature Switzerland AG 2019 25

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_3
26 C. Melitas and D. G. Adler

a 19-gauge FNA needle. To further ensure that or 2nd portion of the duodenum with an orienta-
the biliary system has been successfully pene- tion looking towards the common bile duct. The
trated, bile should be aspirated. It is at this point FNA needle in this position will exit the scope
that contrast should be injected to outline the and penetrate the CBD cephalad towards the
anatomy of the biliary system. A long guidewire liver hilum. Alternatively, the pull position is
can then be manipulated and passed down the similar to the short access positioning in
needle into the CBD and out through the ampulla ERCP. In the pull/short position, the tip of the
into the duodenum. Most commonly, the needle echoendoscope is flush with the duodenal wall
and wire are inserted retrograde via D1 or D2 and and the ultrasound transducer can be more easily
the wire is then manipulated in an antegrade oriented towards the distal common bile duct.
direction. The needle in this position will exit caudally and
Manipulation of the guidewire is by far the towards the ampulla.
most challenging aspect of the procedure as it When draining the biliary system using trans-
often may coil in the bile duct or begin to advance duodenal approaches, the push/long positioning
away from the desired direction. Getting the wire has more scope stability but may limit FNA nee-
to advance into the desired location can be a cum- dle maneuverability [4]. Needle maneuverability
bersome and sometimes a tedious process. This is is sometimes easier in the push/long position,
often the point of the procedure that takes the however, endoscopic stability is often compro-
most time. However, it is the most crucial aspect mised. Therefore, scope positioning should be
of the procedure [4, 8, 9]. based on individualized clinical and anatomical
Recent studies have shown that use of a hydro- conditions. When accessing the biliary system
philic guidewire and positioning as close to the through the intrahepatic bile duct, the endoscope
ampulla as possible, such as D2, allow for easier is in a straight position which allows both effec-
manipulation and minimize the challenges associ- tive needle maneuverability and a high degree of
ated with passing the guidewire [10]. This guidewire endoscope stability [4]. However, given the
can then be used to cannulate the biliary system in increased distance from the ampulla, an increase
retrograde fashion by means of ERCP. Cannulation in difficulty with guidewire maneuverability
can either be achieved by simply entering the papilla may be encountered with this endoscopic posi-
next to the previously passed wire, which acts as a tion [10, 11, 13].
guide for a 2nd wire to be passed retrograde, or by Many patients require biliary decompression
grasping the first wire and pulling it through a duo- due to the presence of biliary strictures. The
denoscope so a catheter can be advanced over it. cholangiogram is therefore a crucial step once
The decision on which of these two techniques to biliary access has been obtained in order to
choose is left to the operator. identify the presence, location, and extent of
these strictures. Gaining access through the left
intrahepatic bile duct is ideal as this access point
Extrahepatic Versus Intrahepatic allows for a more direct guidewire passage to
EUS-Guided Rendezvous the liver hilum and across a stricture [14]. The
EUS needle must be withdrawn while guidewire
The extrahepatic approach first involves proper manipulation is being performed as this may
positioning of the echoendoscope. The positions sheer or cut the guidewire [15]. After the guide-
for retrograde access have been termed the wire has been successfully manipulated through
“push” or “pull” positions [11, 12]. The push the stricture of interest, a 4 mm balloon may
position is similar to the long access in ERCP in be used to dilate the papilla and biliary system
which the echoendoscope travels along the to allow for advancement of a stent over the
greater curvature of the stomach. This often guidewire to traverse both ends of the stricture
places the tip of the echoendoscope in the bulb adequately [15].
3 Endoscopic Ultrasound-Guided Biliary Drainage: Retrograde Approaches 27

Choledochoduodenostomy as these stents have been used in most studies and

have been shown to decrease the risk of stent
The second method of performing EUS-guided migration [4, 24–29]. However, the success rates,
retrograde biliary dilation is by means of transmu- ease of the procedure, and procedure length have
ral stenting. Transmural stenting can be performed all been significantly improved with the introduc-
by four methods: choledochoduodenostomy, tion on lumen apposing metal stents (LAMS)
hepaticogastrostomy, choledochoantrostomy, and [30] (Fig. 3.2).
hepaticoduodenostomy, with the latter two being
very rarely performed [16–19]. Retrograde drain-
age is primarily done by means of choledocho- Success Rates
duodenostomy (Fig. 3.1). This method entails
transluminal stenting between the duodenal bulb Several meta-analyses have reported success
(D1) and the CBD. This method was first rates for EUS-guided biliary drainage procedures
described in 2001 and is most often used in in over 90% of patients [24–26, 31]. The overall
patients with distal bile obstruction and normal success rate with EUS-guided rendezvous proce-
gastrointestinal anatomy [20]. Patients with dures has been found to be about 81% and the
abnormal anatomy from surgery can undergo overall success rate of transluminal stenting has
other forms of retrograde access including cho- been shown to be higher, nearing 95% [25, 32]. It
ledochojejunostomy (Video 3.1). has been shown that success rates have improved
To perform the choledochoduodenostomy, the since the introduction of EUS-guided biliary
push/long position of the echoendoscope is pre- drainage and also have improved with the cumu-
ferred with the needle directed towards the liver lative experience of each operator over time,
hilum [21]. The echoendoscope should be demonstrating that these are complex, high risk
advanced to the first portion of the duodenum to procedures with their own learning curve [31].
adequately visualize the extrahepatic bile duct, The rates of procedure related deaths have also
which is usually quite dilated and easy to identify. significantly decreased due to these reasons as
Ideally, the penetration point into the bile duct well [33]. Success rates have also been impacted
should be at a point at which the CBD is dilated to by an experienced assistant as they become more
at least 5 mm in diameter and the length of the bile familiar with guidewire manipulation, stent
duct segment should be 1–3 cm to allow for suc- deployment, and other portions of the procedure
cessful stenting [21]. A 19-gauge FNA needle is which help increase the desired procedural out-
usually used to gain access into the biliary system comes [34].
in an angle which will allow for easy passage of There has yet to be a consensus as to which
the guidewire into the biliary system. To confirm biliary draining modality is superior when com-
adequate positioning, bile should be aspirated and paring EUS-guided rendezvous and transluminal
contrast can be injected to outline the biliary anat- stenting. Many operators are strongly influenced
omy. Some authors suggest that the tract should by their own experience with one or both tech-
then be dilated by means of balloon, bougie, cau- niques. However, in comparing EUS-guided
tery dilator, or fine-gauge balloon dilator, but this hepaticogastrostomy and choledochoduodenos-
maneuver is not performed universally [22, 23]. tomy, there have been no significant differences
The selected stent should then be deployed in success rates and adverse events [25]. Despite
within the dilated iatrogenic fistula which con- this, choledochoduodenostomy may be more
nects the duodenum to the CBD. Traditionally, beneficial as it is often technically more straight-
self-expanding metal stents (SEMS) which are forward, has been associated with lower rates of
greater than 4 cm in length and are either fully stent dysfunction, and greater 3-month stent
covered or partially covered are most often used patency rates [35].
28 C. Melitas and D. G. Adler

Fig. 3.1 EUS-guided choledochoduodenostomy and of a fully covered metal biliary stent after deployment
treatment of simultaneous malignant gastric outlet over the wire and across the choledochoduodenostomy.
obstruction. (a) Endoscopic image of gastric outlet (g) Endoscopic image of a fully covered metal biliary
obstruction at the level of the duodenal bulb. The ampulla stent after deployment over the wire and across the cho-
is thus not accessible for standard ERCP. (b) EUS image ledochoduodenostomy. (h) Using a biliary balloon cathe-
showing a large, solid pancreatic head mass obstructing ter, a guidewire is advanced across the gastric outlet
the CBD, which is markedly dilated. (c) Using a 19-gauge obstruction. (i) A 22 × 60 mm uncovered enteral stent is
EUS FNA needle, the CBD is accessed in a transduodenal advanced over the wire and across the gastric outlet
manner. (d) Injection of dye through the FNA needle obstruction. (j) Duodenal stent after deployment. The
results in a cholangiogram showing a distal CBD stric- patient has now had his biliary obstruction and gastric
ture. (e) A guidewire is passed through the FNA needle outlet obstruction relieved in a single outpatient
and into the proximal biliary tree. (f) Fluoroscopic image procedure
3 Endoscopic Ultrasound-Guided Biliary Drainage: Retrograde Approaches 29

Fig. 3.1 (continued)

Adverse Events toma, and pancreatitis (most often associated with

the ERCP portion of the procedure) [36].
Systematic reviews have shown complication When comparing approaches of EUS-guided
rates associated with EUS-guided biliary drainage biliary drainage, the extrahepatic approach has
as a whole to be between 16.5% and 23.3% [24– been found to be safer with adverse events occur-
26]. In a large systematic review consisting of ring in about 14% of cases as compared to 18%
over 1100 patients, the most common complica- of cases via the intrahepatic approach [37].
tions included: bleeding (4.03%), bile leakage However, in one compilation of almost 30 stud-
(4.03%), pneumoperitoneum (3.02%), stent ies, the complication rate seen with transmural
migration (2.68%), cholangitis (2.43%), abdomi- stenting was 24%. The most common adverse
nal pain (1.51%), and peritonitis (1.26%) [24–26]. events were stent migration (5.4%), pneumoperi-
Most complications were treated conservatively. toneum (3.4%), peritonitis (3%), cholangitis
The type of complication is often related to the (3%), bleeding (2.8%), and bile leakage (1.5%)
approach used, devices used, disease process, and [25]. Similar rates of these complications have
experience of the endoscopist [33, 34]. been seen in patients undergoing transduodenal
The reported rates of complication associated and transgastric transmural approaches.
with EUS-guided rendezvous specifically are When comparing the use of biliary stents,
approximately 15%. These complications include plastic stents have been shown to have a signifi-
bile leaks, pneumoperitoneum, subcapsular hema- cantly higher rate of bile leakage at 11% as
30 C. Melitas and D. G. Adler

Fig. 3.2 Use of a lumen apposing metal stent (LAMS) to deployment across the choledochoduodenostomy. (c)
create a choledochoduodenostomy. (a) After guidewire Axios stent in good position after deployment draining
access to the CBD has been obtained in a manner similar copious bile. (d) Fluoroscopic image demonstrating the
to that used in Fig. 3.1, a 10 mm Axios stent (Boston patient’s cholangiogram as well as the Axios stent in good
Scientific, Natick, MA) is advanced across the choledo- position with functional choledochoduodenostomy
choduodenostomy. (b) Axios stent immediately after

opposed to 4% in cases that utilized covered metal continue to decrease the procedural complications
stents [35]. Studies on newer equipment specific even further.
to these procedures such as LAMS, hybrid metal
stents (which consist of uncovered proximal por-
tions and covered distal portions), hook stents  ther Biliary Drainage Methods
O
(metal stents with anchoring hooks), and plastic Versus EUS-Guided Biliary Drainage
pigtail stents with four flanges have shown prom-
ise in further decreasing stent related complica- A commonly used alternative method for drain-
tions [30, 38–43]. Given that most of the tools ing the biliary system when ERCP has failed or is
and devices used are adapted from ERCP, there not possible is the percutaneous approach. Many
are very few devices specific to EUS-guided bili- studies and systematic reviews have compared
ary drainage. The development of more devices the percutaneous approach to the EUS-guided
specific to these procedures may very well likely approach and have shown similar success rates
3 Endoscopic Ultrasound-Guided Biliary Drainage: Retrograde Approaches 31

and adverse event rates [44]. However, the percu- even further. Furthermore, the decrease in proce-
taneous route has been found to require frequent dural times, especially seen with transluminal
reinterventions and higher costs [45]. Also, when stenting, and the decrease in overall cost, when
considering all reinterventions, a higher rate of compared to percutaneous biliary drainage, may
complications has been appreciated with the per- further assist in these procedures becoming routine
cutaneous approach [45]. Despite higher overall methods of biliary decompression.
costs, the percutaneous method has several other
disadvantages such as its affect on lifestyle. This
is an external drain which can interfere with References
daily activities, intimacy, physical appearance,
difficulty with sleeping, and can even interfere 1. Ekkelenkamp VE, de Man RA, Ter Borg F, et al.
Prospective evaluation of ERCP performance: results
with getting dressed. of a nationwide quality registry. Endoscopy. 2015;
Other studies have compared EUS-guided bili- 47:503–7.
ary drainage to ERCP as primary interventions for 2. Williams EJ, Ogollah R, Thomas P, et al. What predicts
biliary decompression. Success rates for primary failed cannulation and therapy at ERCP? Results of
a large-scale multicenter analysis. Endoscopy. 2012;
biliary interventions in the setting of malignant bili- 44:674–83.
ary obstructions between ERCP and EUS-guided 3. Widmer J, Sharaiha RZ, Kahaleh M. Endoscopic
approaches are similar, however, the EUS-guided ultrasonography-­ guided drainage of the pancre-
approach, especially transmural stenting, is also atic duct. Gastrointest Endosc Clin N Am. 2013;
23:847–61.
associated with decreased procedure times [46, 47]. 4. Iwashita T, Doi S, Yasuda I. Endoscopic ultrasound-­
Furthermore, EUS-guided biliary drainage after guided biliary drainage: a review. Clin J Gastroenterol.
failed ERCP has been compared to EUS-guided 2014;7(2):94–102.
biliary drainage without failed ERCP and has been 5. Paik WH, Park DH. Endoscopic ultrasound-guided
biliary access, with focus on technique and practical
found to have similar success rates and adverse tips. Clin Endosc. 2017;50(2):104–11.
events as well [48]. 6. De Cobelli F, Marra P, Diana P, Brembilla G, Venturini
M, Therapeutic EUS. Biliary drainage—the interven-
tional radiologist’s perspective. Endosc Ultrasound.
2017;6(Suppl S3):127–31.
Conclusion 7. Tarantino I, Barresi L, Repici A, Traina M. EUS-­
guided biliary drainage (EUS-BD): a case series.
EUS-guided retrograde approaches for biliary Endoscopy. 2008;40(4):336–9.
drainage are exciting revolutionary methods for 8. Isayama H, Nakai Y, Kawakubo K, et al. The endo-
scopic ultrasonography-guided rendezvous tech-
accessing the biliary tree, especially in patients nique for biliary cannulation: a technical review.
with failed ERCP, altered anatomy, and in J Hepatobiliary Pancreat Sci. 2013;20:413–20.
patients in whom the ampulla is inaccessible. 9. Kahaleh M, Hernandez AJ, Tokar J, Adams RB,
Both methods of EUS-guided retrograde biliary Shami VM, Yeaton P. Interventional EUS-guided
cholangiography: evaluation of a technique in evolu-
drainage have been found to be very safe and effi- tion. Gastrointest Endosc. 2006;64:52–9.
cacious with high success rates and acceptably 10. Dhir V, Bhandari S, Bapat M, Maydeo A. Comparison
low rates of complication. Studies evaluating of EUS-guided rendezvous and precut papillot-
these procedures as first-line biliary decompres- omy techniques for biliary access (with videos).
Gastrointest Endosc. 2012;75:354–9.
sive strategies have shown promising results. 11. Iwashita T, Lee JG, Shinoura S, Nakai Y, Park DH,
However, further prospective investigations will Muthusamy VR, et al. Endoscopic ultrasound-guided
need to be performed. As the majority of the tools rendezvous for biliary access after failed cannulation.
used for these procedures have been originally Endoscopy. 2012;44(1):60–5.
12. Kawakubo K, Isayama H, Sasahira N, Nakai Y, Kogure
developed for ERCP, the development of more H, Hamada T, et al. Clinical utility of an endoscopic
devices specific to these procedures as well as ultrasound-guided rendezvous technique via various
having these procedures performed by endosco- approach routes. Surg Endosc. 2013;27:3437–43.
pists experienced in these procedures may very 13. Iwashita T, Yasuda I, Mukai T, et al. EUS-guided
rendezvous for difficult biliary cannulation using
well likely decrease the procedural complications
32 C. Melitas and D. G. Adler

a standardized algorithm: a multicenter prospec- endoscopic retrograde cholangiopancreatography.

tive pilot study (with videos). Gastrointest Endosc. Endosc Int Open. 2016;4:E487–96.
2016;83:394–400. 28. Hara K, Yamao K, Mizuno N, et al. Endoscopic
14. Park DB. Endoscopic ultrasonography-guided hepati- ultrasonography-guided biliary drainage: who,
cogastrostomy. Gastrointest Endosc Clin N Am. when, which, and how? World J Gastroenterol. 2016;
2012;22:271–80. ix. 22:1297–303.
15. Park DH, Jeong SU, Lee BU, et al. Prospective evalu- 29. Bruno MJ. Interventional endoscopic ultrasonog-
ation of a treatment algorithm with enhanced guide- raphy: where are we headed? Dig Endosc. 2017;
wire manipulation protocol for EUS-guided biliary 29:503–11.
drainage after failed ERCP (with video). Gastrointest 30. Kunda R, Pérez-Miranda M, Will U, et al. EUS-­
Endosc. 2013;78:91–101. guided choledochoduodenostomy for malignant dis-
16. Itoi T, Itokawa F, Tsuchiya T, Tsuji S, Tonozuka tal biliary obstruction using a lumen-apposing fully
R. Endoscopic ultrasound-guided choledochoantros- covered metal stent after failed ERCP. Surg Endosc.
tomy as an alternative extrahepatic bile duct drainage 2016;30:5002–8.
method in pancreatic cancer with duodenal invasion. 31. Attasaranya S, Netinasunton N, Jongboonyanuparp T,
Dig Endosc. 2013;25(Suppl 2):142–5. et al. The spectrum of endoscopic ultrasound interven-
17. Cho MK, So H, Jung K, Lee JH, Park DH. Percutaneous tion in biliary diseases: a single center’s experience in
transhepatic biliary drainage-assisted, endoscopic 31 cases. Gastroenterol Res Pract. 2012;2012:680753.
ultrasound-guided hepaticoduodenostomy for iso- 32. Iwashita T, Lee JG. Endoscopic ultrasonography-­
lated complete right intrahepatic duct obstruction. guided biliary drainage: rendezvous technique.
Endoscopy. 2016;48:E317–8. Gastrointest Endosc Clin N Am. 2012;22(2):249–58.
18. Park do H. Endoscopic ultrasound-guided biliary viii–ix.
drainage of hilar biliary obstruction. J Hepatobiliary 33. Poincloux L, Rouquette O, Buc E, et al. Endoscopic
Pancreat Sci. 2015;22:664–8. ultrasound-guided biliary drainage after failed ERCP:
19. Park SJ, Choi JH, Park DH, et al. Expanding indica- cumulative experience of 101 procedures at a single
tion: EUS-guided hepaticoduodenostomy for iso- center. Endoscopy. 2015;47:794–801.
lated right intrahepatic duct obstruction (with video). 34. Vila JJ, Pérez-Miranda M, Vazquez-Sequeiros E,
Gastrointest Endosc. 2013;78:374–80. et al. Initial experience with EUS-guided cholangio-
20. Giovannini M, Moutardier V, Pesenti C, Bories E, pancreatography for biliary and pancreatic duct drain-
Lelong B, Delpero JR. Endoscopic ultrasound-guided age: a Spanish national survey. Gastrointest Endosc.
bilioduodenal anastomosis: a new technique for bili- 2012;76:1133–41.
ary drainage. Endoscopy. 2001;33:898–900. 35. Kawakubo K, Isayama H, Kato H, et al. Multicenter
21. Oh D, Park DH, Song TJ, et al. Optimal biliary access retrospective study of endoscopic ultrasound-
point and learning curve for endoscopic ultrasound-­ guided biliary drainage for malignant biliary
guided hepaticogastrostomy with transmural stenting. obstruction in Japan. J Hepatobiliary Pancreat Sci.
Therap Adv Gastroenterol. 2017;10:42–53. 2014;21:328–34.
22. Amano M, Ogura T, Onda S, et al. Prospective clinical 36. Weilert F, Binmoeller KF. Endoscopic ultrasound-­
study of EUS-guided biliary drainage using novel bal- guided access to the bile duct: a new frontier.
loon catheter (with video). J Gastroenterol Hepatol. Gastrointest Interv. 2012;1(1):11–8.
2017;32:716–20. 37. Alvarez-Sánchez MV, Jenssen C, Faiss S, Napoléon
23. Ogura T, Takagi W, Onda S, et al. Endoscopic B. Interventional endoscopic ultrasonography: an
ultrasound-­ guided biliary drainage with a novel overview of safety and complications. Surg Endosc.
­fine-­gauge balloon catheter: simplified technique using 2014;28:712–34.
a coaxial guidewire. Endoscopy. 2015;47(Suppl 1): 38. French JB, Coe AW, Pawa R. Endoscopic ultrasound-­
E573–4. guided choledochoduodenostomy with a lumen-­
24. Khan MA, Akbar A, Baron TH, et al. Endoscopic apposing, self-expandable fully covered metal stent
ultrasound-­guided biliary drainage: a systematic review for palliative biliary drainage. Clin J Gastroenterol.
and meta-analysis. Dig Dis Sci. 2016;61:684–703. 2016;9:79–85.
25. Wang K, Zhu J, Xing L, Wang Y, Jin Z, Li 39. Sejpal DV, Trindade AJ, Vamadevan AS. Lumen-­
Z. Assessment of efficacy and safety of EUS-guided apposing metal stent placement for biliary drainage
biliary drainage: a systematic review. Gastrointest may be preferable in the setting of duodenal obstruc-
Endosc. 2016;83:1218–27. tion when dual stenting is performed. Gastrointest
26. Moole H, Bechtold ML, Forcione D, Puli SR. A meta-­ Endosc. 2015;82:170.
analysis and systematic review: success of endoscopic 40. Brückner S, Arlt A, Hampe J. Endoscopic ultrasound-­
ultrasound guided biliary stenting in patients with guided biliary drainage using a lumen-apposing self-­
inoperable malignant biliary strictures and a failed expanding metal stent: a case series. Endoscopy.
ERCP. Medicine. 2017;96:e5154. 2015;47:858–61.
27. Khashab MA, Van derMerwe S, Kunda R, et al. 41. Cho DH, Lee SS, Oh D, et al. Long-term outcomes of
Prospective international multicenter study on endo- a newly developed hybrid metal stent for EUS-guided
scopic ultrasound-guided biliary drainage for patients biliary drainage (with videos). Gastrointest Endosc.
with malignant distal biliary obstruction after failed 2017;85:1067–75.
3 Endoscopic Ultrasound-Guided Biliary Drainage: Retrograde Approaches 33

42. Song TJ, Lee SS, Park DH, Seo DW, Lee SK, Kim age and percutaneous drainage in patients with distal
MH. Preliminary report on a new hybrid metal stent malignant biliary obstruction and failed ERCP. Dig
for EUS-guided biliary drainage (with videos). Dis Sci. 2015;60:557–65.
Gastrointest Endosc. 2014;80:707–11. 46. Hara K, Yamao K, Niwa Y, et al. Prospective clini-
43. Umeda J, Itoi T, Tsuchiya T, et al. A newly designed cal study of EUS-guided choledochoduodenostomy
plastic stent for EUS-guided hepaticogastrostomy: a for malignant lower biliary tract obstruction. Am
prospective preliminary feasibility study (with vid- J Gastroenterol. 2011;106:1239–45.
eos). Gastrointest Endosc. 2015;82:390–6. 47. Kawakubo K, Kawakami H, Kuwatani M, et al.
44. Artifon EL, Aparicio D, Paione JB, et al. Biliary Endoscopic ultrasound-guided choledochoduode-
drainage in patients with unresectable, malig- nostomy vs. transpapillary stenting for distal biliary
nant obstruction where ERCP fails: endoscopic obstruction. Endoscopy. 2016;48:164–9.
ultrasonography-­guided choledochoduodenostomy 48. Nakai Y, Isayama H, Yamamoto N, et al. Indications
versus percutaneous drainage. J Clin Gastroenterol. for endoscopic ultrasonography (EUS)-guided bili-
2012;46:768–74. ary intervention: does EUS always come after failed
45. Khashab MA, Valeshabad AK, Afghani E, et al. A endoscopic retrograde cholangiopancreatography?
comparative evaluation of EUS-guided biliary drain- Dig Endosc. 2017;29:218–25.
 EUS-Guided Gallbladder Drainage
4
Sunil Amin and Amrita Sethi

ERCP. Early adopters of EUS-GBD used plastic

Introduction or fully covered metal self-expandable metal bili-
ary stents (FCSEMS), but more recently the use
For most patients, acute cholecystitis is a surgical of a lumen-apposing metal stent (LAMS) has
disease. Laparoscopic cholecystectomy is the become widespread [5–8]. The most recent 2018
standard of care for operative candidates, while Tokyo Guidelines on the management of acute
poor surgical candidates typically receive percu- cholecystitis recognize EUS-GBD as an appro-
taneous cholecystostomy tube placement [1, 2]. priate treatment for high-risk surgical patients
Endoscopic transpapillary cystic duct stenting, when performed in high-volume tertiary care
while feasible, requires a high degree of technical centers by skilled endoscopists [9]. This inclu-
expertise and necessitates that an endoscopic ret- sion in the treatment algorithm is a paradigm
rograde cholangiopancreatography (ERCP) be shift as EUS-GBD was not recommended in ear-
performed, thereby exposing the patient to addi- lier versions of the text [10]. This review will
tional procedural-related risks. As such, cystic address the rationale, indications, technique,
duct stenting is often reserved for patients in complications, outcomes, and controversies
which the transhepatic approach is contraindi- regarding EUS-GBD.
cated or anatomically impossible [3, 4]. In recent
years, however, transmural gallbladder drainage
via an endoscopic ultrasound-guided approach Indications/Contraindications
(EUS-GBD) has emerged as a feasible and poten-
tial equally efficacious option that obviates the Several requirements must be met prior to con-
need for an external drainage catheter or sideration of EUS-GBD. First, the patient should
have a diagnosis of acute cholecystitis and either
Electronic supplementary material: The online version should be a high-risk surgical candidate (with an
of this chapter (https://doi.org/10.1007/978-3-319-97376- estimated mortality >10%) or should have
0_4) contains supplementary material, which is available refused surgical treatment. The authors of the
to authorized users.
2018 Tokyo Guidelines recommend percutane-
S. Amin ous transhepatic gallbladder drainage (PT-GBD)
Division of Gastroenterology and Hepatology,
as the first alternative to cholecystectomy in
Virginia Mason Medical Center, Seattle, WA, USA
high-risk surgical patients, but state that endo-
A. Sethi (*)
scopic drainage including EUS-GBD may be
Division of Digestive and Liver Diseases, Columbia
University Medical Center, New York, NY, USA considered in high-volume centers when
e-mail: [email protected] ­performed by skilled endoscopists [9]. Grade III

© Springer Nature Switzerland AG 2019 35

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_4
36 S. Amin and A. Sethi

disease, defined by the 2013 Tokyo Guidelines the cautery-enhanced catheter is placed over a
(TG13) as acute cholecystitis associated with pre-inserted guidewire), in general, it is felt that
organ dysfunction, is particularly suited for EUS-­ the placement of the wire can potentially push the
GBD [10]. In these patients, PT-GBD is associ- GB wall away from the gastric or duodenal wall,
ated with higher mortality, higher readmission subsequently increasing the risk of misdeploy-
rates, and prolonged hospital stay [10, 11]. ment or bile leak. Therefore, the free-hand tech-
Although not absolute contraindications, the niques appear to be the favored approach.
presence of ascites or severe coagulopathy should In the two-step procedure, a 19-gauge needle
be met with caution. In these circumstances, is first used to puncture the gallbladder, bile is
endoscopic transpapillary drainage either with aspirated, and a 0.035-in guidewire is advanced
placement of a cystic duct stent or naso-biliary through the needle and coiled in the gallbladder
catheter may be more appropriate. lumen. A long-wire exchange is then performed
under endosonographic guidance and flash dila-
tion of the fistulous tract is performed with a
Technique/Procedure 6 mm dilating balloon. This dilation, though nec-
essary, must not be too aggressive in order to
EUS-GBD should, in general, only be performed minimize the risk of peritoneal leakage during
in tertiary care centers where the full cadre of sur- the long-wire exchange [12, 13]. The 10.8F
gical and interventional radiological expertise is LAMS delivery catheter is then advanced over
available to the endoscopist should complica- the wire into the gallbladder lumen, where the
tions occur. Although slight variations in proce- distal flange is deployed in the gallbladder and
dural technique are described, EUS-GBD is the proximal flange is deployed in the gastroin-
sequentially performed as follows. Prior to the testinal lumen.
start of the procedure, the patient is sedated (often With both methods, the LAMS can be dilated
via general anesthesia), placed in a supine or left using a standard through-the-scope balloon dila-
lateral decubitus position, and given intravenous tor if so desired. At this point, certain endosco-
antibiotics. pists place a short double pig stent through the
A therapeutic linear echoendoscope is then LAMS prior to concluding the procedure or bring
advanced into the pre-pyloric antrum or the duo- the patient back several days later to perform this
denal bulb, and the gallbladder is identified endo- step. Again, there is variability among endosco-
sonographically. A suitable access point must be pists in procedural specifics such as patient posi-
chosen so that the distance between the gastroin- tioning, whether to dilate the LAMS
testinal lumen and the gallbladder wall is equal to post-deployment (and if so, to what size), and the
or less than the saddle length of the available need for and timing of the placement of a double
LAMS and no intervening vessels are present. pigtail stent.
Either a one-step or two-step procedure is then
performed in order to create a cholecysto-­
gastrostomy or cholecysto-duodenostomy and Complications
subsequently deploy a metal stent
trans-luminally. Pooled analyses report adverse events rates of
In the one-step procedure, an electrocautery EUS-GBD between 8% and 17% [6, 14, 15].
tipped 10.8F catheter containing a 15 mm inner-­ Potentially serious procedure-related complica-
diameter lumen-apposing metal stent (LAMS) tions may include bile leakage, perforation, stent
(AXIOS-EC, Boston Scientific, Natick, MA) is migration, and recurrent cholecystitis due to stent
used to simultaneously puncture the body or neck occlusion. When stratified by stent type,
of the gallbladder and deploy the LAMS (Fig. 4.1, Anderloni et al. report an adverse rate of 18.2%
Video 4.1). Whereas this method may be per- for plastic stents, 12.3% for SEMS, and 9.9% for
formed in a wire-guided fashion as well (in which LAMS [14]. Not surprisingly, their data suggest
4 EUS-Guided Gallbladder Drainage 37

Fig. 4.1 EUS-guided transgastric gallbladder drainage in Post-deployment dilation of lumen of deployed stent with
a 64-year-old female with metastatic breast cancer. (a) CRE balloon. (e) View of GB wall through deployed and
EUS image of GB with thickened wall and sludge-filled. dilated AXIOS stent. (f) Endoscopic view of GB wall
(b) EUS image of deployed distal flange within gallblad- after passing diagnostic endoscope through deployed and
der during step 3 of AXIOS deployment (pulling back on dilated AXIOS stent. (g) Gastric view of deployed AXIOS
catheter to pull flange up to GB wall and create apposi- stent with 7F double pigtail placed through lumen of
tion). (c) Post-deployment of proximal flange (gastric LAMS. Courtesy of Doug Adler, MD, Amrita Sethi, MD,
side) of AXIOS stent with drainage of GB contents. (d) and Reem Z. Sharaiha, MD
38 S. Amin and A. Sethi

pros and cons to each approach. Bile leakage was case reports) [6]. Most recently, a meta-analysis
only associated with the use of plastic stents (in looking exclusively at the use of LAMS in 181
which that GB wall can easily separate from the EUS-GBD procedures computed a pooled tech-
gastric or duodenal wall), while stent migration nical success rate of 95% (CI 91–99%) and clini-
and stent occlusion only occurred with cal success rate of 93% (CI 90–97%) [15].
SEMS. LAMS was associated with higher rates Several studies have favorably compared out-
of bleeding, infection, and pain. In a separate comes for EUS-GBD versus percutaneous gall-
meta-analyses looking at LAMS only, the most bladder drainage (PC-GBD) [5, 16–19]. Tyberg
common early adverse events were bleeding et al. retrospectively analyzed 155 patients who
(3.9%, 7/181), stent migration (1.1%, 2/181), and either underwent EUS-GBD (n = 42) or PC-GBD
recurrent cholecystitis due to stent occlusion (n = 113) across seven international tertiary cen-
(1.7%, 3/181) [15]. ters [16]. Whereas technical success was slightly
higher in the PC-GBD group (99% vs. 95%),
clinical success was higher in the EUS-GBD
Peri-Procedural Care group (95% vs. 86%). Nevertheless, neither of
these differences reached statistical significance
There is no standardized approach to the peri-­ (p = 0.179 and p = 0.157, respectively).
procedural management of patients undergoing Interestingly, significantly more patients in the
EUS-GBD. The majority of these patients are PC-GBD group required repeat interventions
high-risk surgical patients that may not be candi- (n = 28, 24%) compared to the EUS-GBD group
dates for percutaneous transhepatic drainage (n = 4, 10%). There was no difference in adverse
either due to ascites or coagulopathy. As such, the events between the two groups. The authors con-
nuances of managing these critically ill patients clude therefore that EUS-GBD and PC-GBD
is individualized. In general, however, patients have similar safety and efficacy; however, EUS-­
should remain NPO and intravenous antibiotics GBD results in significant cost savings.
initiated once the diagnosis of acute cholecystitis A follow-up study by Irani et al. compared 45
is made. Post-procedurally, patients are often patients with EUS-GB to 45 patients with
kept hospitalized, kept NPO until clinical resolu- PC-GBD but looked exclusively at LAMS [5].
tion of symptoms begins, and maintained on The authors similarly found no difference in
broad-spectrum antibiotics for at least several technical or clinical success between the two
days. Some practitioners place patients on a tra- groups. However, EUS-GBD was associated with
ditional enteral stent (low residue) diet to reduce a lower mean post-procedural pain score (2.5 vs.
the risk of stent occlusion but others allow 6.5, p < 0.05), shorter average length of stay in
patients to consume a full diet. the hospital (3 days vs. 9 days, p < 0.05), and
fewer repeat interventions per patient (0.2 vs. 2.5,
p < 0.05). Furthermore, there was a trend towards
 utcomes
O fewer adverse events in the EUS-GBD group
(11% vs. 32%, p = 0.065).
The technical and clinical success rates of EUS-­ In a slightly larger study, Teoh et al. looked
GBD are consistently reported at greater than specifically at adverse events between EUS-GBD
90%. In their systematic review and pooled anal- and PT-GBD [17]. Among 118 patients with
ysis of 166 high-risk surgical patients with acute acute cholecystitis deemed unfit for surgery, the
cholecystitis who underwent EUS-GBD, 59 patients who received EUS-GBD suffered sig-
Anderloni et al. report a pooled technical success nificantly fewer serious (23.7 % vs. 74.6 %,
rate of 95.8% and clinical success rate of 93.4% p < 0.001) and overall (32.2 % vs. 74.6 %,
[14]. Peñas-Herrero et al. report even higher p < 0.001) adverse events than the 59 patients
numbers (97% and 99%, respectively) in their who underwent percutaneous cholecystostomy.
pooled analysis of 155 patients (8 series and 12
4 EUS-Guided Gallbladder Drainage 39

Current Controversies/Future EUS-GBD and had their SEMS removed at

Considerations 4 weeks after resolution of symptoms [20]. The
SEMS was removed without replacement in 8
Although most experts performing EUS-GBD patients, and replaced with a 7F pigtail stent in 4
achieve high clinical and technical success rates patients. Recurrence of cholecystitis was seen in
(>90%), EUS-GBD remains in its infancy and only 1 patient who did not receive a pigtail stent.
optimal technique and best practices are still Thus, the authors conclude that 4 weeks is a rea-
being defined. sonable interval to remove the stent to prevent
At present, there is no current consensus migration and recurrence due to food impaction.
regarding optimal type of drainage stent (plastic, A third point of controversy, as in manage-
FCSEMS, and LAMS) or location of puncture ment after PT-GBD [10], is whether patients
site. Both transgastric and transduodenal routes should be referred for cholecystectomy upon
appear to be technically viable and safe (Fig. 4.2). clinical resolution of symptoms after EUS-GBD
Theoretically, LAMS results in better lumen and the timing of such an intervention. While one
apposition and thus might reduce risk of bile leak study reported significantly lower rates of com-
and stent migration. Furthermore, the larger cali- pletion cholecystectomy for patients undergoing
ber of the drainage stent may result in quicker EUS-GBD vs. PT-GBD (5% vs. 27%, p = 0.003)
symptom resolution. In a recent systematic [16], no substantial data is available comparing
review and pooled analysis of 21 studies and 166 outcomes for the two groups. Furthermore, there
patients, Anderloni et al. report the technical suc- are no long-term data regarding whether ulti-
cess rate of 100% with plastic stents, 98.6% with mately undergoing cholecystectomy results in
SEMS, and 91.5% with LAMS [14]. Clinical improved clinical outcomes in this sick popula-
success was 100%, 94.4%, and 91.5%, respec- tion of patients or whether EUS-GBD affects out-
tively. In their sub-group analysis of 42 patients comes of cholecystectomy.
who underwent EUS-GBD, Tyberg et al. found A final concern associated with EUS-GB is
no difference in adverse events (p = 0.895) or gastric reflux into the gallbladder lumen that may
clinical failure (p = 0.978) between plastic, result in delayed pain, stent occlusion, or even
FSCEMS, or LAMS [16]. There was also no dif- gallbladder perforation requiring surgical revi-
ference in adverse events (p = 0.289) or clinical sion. Kim et al. describe two such patients [21].
failure (p = 0.432) based on the location of the Although both patients experienced initial clini-
stent (transgastric vs. transduodenal vs. transjeju- cal success, the first patient was re-admitted with
nal). Regardless, a head-to-head trial of plastic intractable right upper quadrant pain with cross-­
vs. FCSEMS vs. LAMS with regard to technical sectional imaging confirming a large amount of
success, clinical success, and adverse events gastric reflux into the gallbladder. This patient
would certainly help guide future efforts. was managed with repeat outpatient endoscopic
A second consideration for which there is no drainage. The second patient developed a stent
clear agreement is the optimal time for stent occlusion that required a repeat endoscopic pro-
removal, or whether the stent should be removed cedure, an infected biloma, and, ultimately, intra-­
at all. Given the frailty of many patients undergo- peritoneal free air requiring surgical exploration.
ing EUS-GBD, certain experts argue against With regard to future considerations, the use
planned removed of the stent, thus mitigating the of a large bore LAMS for EUS-GBD provides
risk and cost associated with further procedures repeated access to the gallbladder for per-oral
[5]. This strategy makes EUS-guided gallbladder cholecystoscopy and other advanced interven-
drainage so-called destination therapy. The risk tions [22–24]. In their feasibility study of 27 cho-
associated with this strategy, however, is stent lecystoscopies following EUS-GBD, Chan et al.
occlusion and recurrent cholecystitis. As an alter- demonstrate the safety and efficacy of a number
native approach, Kamata et al. performed a retro- of practical applications [23]. First, the endosco-
spective study of 12 patients who received pist is able to confirm clearance of stones prior to
40 S. Amin and A. Sethi

Fig. 4.2 EUS-guided transduodenal gallbladder drainage into the gallbladder 1 week after initial deployment. (g)
in a male with inoperable IPMN. (a) EUS image of a Endoscopic image of a guidewire being passed through
large, distended gallbladder with a very thickened wall. the LAMS into the gallbladder. (h) Endoscopic image of a
(b) EUS image of the electrocautery-enhanced LAMS double pigtail stent through the lumen of the LAMS. Note
catheter during passage through the duodenal wall into the descending duodenum on the left side of image. (i)
gallbladder. (c) EUS image of deployment of the distal Fluoroscopic image of LAMS after deployment. Note
end of the LAMS within the gallbladder lumen. (d) EUS patient has a metal biliary stent as well as the LAMS. (j)
image of the fully deployed LAMS across the duodenal Fluoroscopic image of guidewire being passed into the
bulb into the gallbladder. (e) Endoscopic image of the gallbladder through the LAMS. (k) Fluoroscopic image of
proximal end of the LAMS after deployment. Note puru- double pigtail stent through LAMS. Courtesy of Douglas
lent drainage. (f) Endoscopic view through the LAMS G. Adler, MD
4 EUS-Guided Gallbladder Drainage 41

Fig. 4.2 (continued)

42 S. Amin and A. Sethi

removal of the LAMS. Should stones remain, stent type, location for drainage, and a standard-
they can be extracted with biliary baskets and ized management strategy of both the stent and
lithotripsy can be performed if required. Second, gallbladder itself, after resolution of symptoms.
should choledocholithiasis be suspected, the
endoscopist can perform a cholecystogram by Disclosure Statement
injecting contrast through the cystic duct opening Dr. A. Sethi serves as a paid consultant for Boston
under fluoroscopy. Finally, narrow band imaging Scientific Corp. and Olympus America.
(NBI) and confocal endomicroscopy (nCLE) can
be used to examine potential mucosal irregulari-
ties in the GB wall. Although not demonstrated to References
date, EUS-GBD could also provide a portal for
removal of large gallbladder polyps as well 1. Glenn F. Cholecystostomy in the high risk patient with
biliary tract disease. Ann Surg. 1977;185(2):185–91.
(>1 cm) that would otherwise require 2. Elyaderani MK, McDowell DE, Gabriele OF. A pre-
cholecystectomy. liminary report of percutaneous cholecystostomy
under ultrasonography and fluoroscopy guidance.
J Clin Gastroenterol. 1983;5(3):277–81.
3. Kozarek RA. Selective cannulation of the cystic duct at
Summary time of ERCP. J Clin Gastroenterol. 1984;6(1):37–40.
4. Itoi T, Sofuni A, Itokawa F, Tsuchiya T, Kurihara T,
Endoscopic ultrasound-guided gallbladder drain- Ishii K, et al. Endoscopic transpapillary gallbladder
age (EUS-GBD) is now an accepted treatment drainage in patients with acute cholecystitis in whom
percutaneous transhepatic approach is contraindicated
option for the high-risk surgical patient with or anatomically impossible (with video). Gastrointest
acute cholecystitis. Consensus guidelines suggest Endosc. 2008;68(3):455–60.
that this emerging procedure may be particularly 5. Irani S, Ngamruengphong S, Teoh A, Will U, Nieto
useful among patients with contraindications to J, Abu Dayyeh BK, et al. Similar efficacies of endo-
scopic ultrasound gallbladder drainage with a lumen-­
percutaneous transhepatic gallbladder drainage, apposing metal stent versus percutaneous transhepatic
or those with severe Type 3 cholecystitis [9, 10]. gallbladder drainage for acute cholecystitis. Clin
Adverse event rates are low (8–17%) but not Gastroenterol Hepatol. 2017;15(5):738–45.
insignificant, even in expert hands. Potential 6. Peñas-Herrero I, de la Serna-Higuera C, Perez-­
Miranda M. Endoscopic ultrasound-guided gall-
complications include recurrent cholecystitis due bladder drainage for the management of acute
to stent occlusion, stent migration, biliary leak- cholecystitis (with video). J Hepatobiliary Pancreat
age, or bleeding. As such, EUS-GBD must be Sci. 2015;22(1):35–43.
performed only at tertiary care centers with surgi- 7. Dollhopf M, Larghi A, Will U, Rimbaş M, Anderloni
A, Sanchez-Yague A, et al. EUS-guided gallbladder
cal and radiological backup. Early comparative drainage in patients with acute cholecystitis and high
studies suggest that compared to PT-GBD, EUS-­ surgical risk using an electrocautery-enhanced lumen-­
GBD provides a comparably safe, less invasive, apposing metal stent device. Gastrointest Endosc.
and more cost-effective option for high-risk sur- 2017;86(4):636–43.
8. Itoi T, Binmoeller KF, Shah J, Sofuni A, Itokawa F,
gical patients. Further study is needed to deter- Kurihara T, et al. Clinical evaluation of a novel lumen-­
mine additional roles for EUS-GBD in the apposing metal stent for endosonography-guided pan-
treatment algorithm for acute cholecystitis. While creatic pseudocyst and gallbladder drainage (with
early efforts used plastic and fully covered metal videos). Gastrointest Endosc. 2012;75(4):870–6.
9. Mori Y, Itoi T, Baron TH, Takada T, Strasberg SM,
stents, most experts now prefer the LAMS, which Pitt HA, et al. TG18 management strategies for gall-
provides better luminal apposition, is quicker to bladder drainage in patients with acute cholecysti-
deploy, and minimizes the risk of biliary leakage tis: updated Tokyo guidelines 2018 (with videos).
and stent migration. Furthermore, LAMS pro- J Hepatobiliary Pancreat Sci. 2017;25:87.
10. Takada T, Strasberg SM, Solomkin JS, Pitt HA,
vides continued access to the gallbladder lumen Gomi H, Yoshida M, et al. TG13: updated Tokyo
for cholecystoscopy and other advanced inter- guidelines for the management of acute cholangi-
ventions. Future prospective studies with ade- tis and cholecystitis. J Hepatobiliary Pancreat Sci.
quate follow-up are needed to clarify optimal 2013;20(1):1–7.
4 EUS-Guided Gallbladder Drainage 43

11. Dimou FM, Adhikari D, Mehta HB, Riall 18. Choi JH, Kim HW, Lee J-C, Paik K-H, Seong NJ,
TS. Outcomes in older patients with grade III Yoon CJ, et al. Percutaneous transhepatic versus EUS-­
cholecystitis and cholecystostomy tube place- guided gallbladder drainage for malignant cystic duct
ment: a propensity score analysis. J Am Coll Surg. obstruction. Gastrointest Endosc. 2017;85(2):357–64.
2017;224(4):502–511.e1. 19. Kedia P, Sharaiha RZ, Kumta NA, Widmer J, Jamal-­
12. Itoi T, Baron TH, Mouen K, Tsuchiya T, Irani S, Dhir Kabani A, Weaver K, et al. Endoscopic gallblad-
V, et al. Technical review of EUS-guided gastroenter- der drainage compared with percutaneous drainage.
ostomy in 2017. Dig Endosc. 2017;29(4):495–502. Gastrointest Endosc. 2015;82(6):1031–6.
13. Itoi T, Ishii K, Ikeuchi N, Sofuni A, Gotoda T, 20. Kamata K, Takenaka M, Kitano M, Omoto S, Miyata
Moriyasu F, et al. Prospective evaluation of endoscopic T, Minaga K, et al. Endoscopic ultrasound-guided
ultrasonography-­guided double-balloon-occluded gallbladder drainage for acute cholecystitis: long-­
gastrojejunostomy bypass (EPASS) for malignant term outcomes after removal of a self-expandable
gastric outlet obstruction. Gut. 2016;65(2):193–5. metal stent. World J Gastroenterol. 2017;23(4):661–7.
14. Anderloni A, Buda A, Vieceli F, Khashab MA, Hassan 21. Kim JJ, Hiotis SP, Sur MD. Gastric reflux into the
C, Repici A. Endoscopic ultrasound-guided transmural gallbladder after EUS-guided stenting-letter to the
stenting for gallbladder drainage in high-risk patients editor regarding “EUS-guided versus percutaneous
with acute cholecystitis: a systematic review and gallbladder drainage: isn’t it time to convert?” J Clin
pooled analysis. Surg Endosc. 2016;30(12):5200–8. Gastroenterol. 2017. July 10.
15. Han D, Inamdar S, Lee CW, Miller LS, Trindade AJ, 22. Yeung B, Teoh AYB. Endoscopic management of
Sejpal DV. Lumen apposing metal stents (LAMSs) for gallbladder stones: can we eliminate cholecystec-
drainage of pancreatic and gallbladder collections: a tomy? Curr Gastroenterol Rep. 2016;18(8):42.
meta-analysis. J Clin Gastroenterol. 2017. Oct 9. 23. Chan SM, Teoh AYB, Yip HC, Wong VWY, Chiu
16. Tyberg A, Saumoy M, Sequeiros EV, Giovannini M, PWY, Ng EKW. Feasibility of per-oral cholecys-
Artifon E, Teoh A, et al. EUS-guided versus percuta- toscopy and advanced gallbladder interventions
neous gallbladder drainage: isn’t it time to convert? after EUS-guided gallbladder stenting (with video).
J Clin Gastroenterol. 2016;52(1):79–84. Gastrointest Endosc. 2017;85(6):1225–32.
17. Teoh AYB, Serna C, Penas I, Chong CCN, Perez-­ 24. Teoh AYB, Chan AWH, Chiu PWY, Lau JYW. In
Miranda M, Ng EKW, et al. Endoscopic ultrasound-­ vivo appearances of gallbladder carcinoma under
guided gallbladder drainage reduces adverse events magnifying endoscopy and probe-based confo-
compared with percutaneous cholecystostomy cal laser endomicroscopy after endosonographic
in patients who are unfit for cholecystectomy. gallbladder drainage. Endoscopy. 2014;46(Suppl 1
Endoscopy. 2017;49(2):130–8. UCTN):E13–4.
 Endoscopic Ultrasound-Guided
Pancreatic Duct Drainage (EUS-PD) 5
Shawn L. Shah and Amy Tyberg

procedure of EUS-PD remains among the most

Introduction technically challenging of all therapeutic EUS
procedures, the ability to provide successful drain-
Over the past 25 years, endoscopic ultrasound age without percutaneous or surgical intervention
(EUS) has evolved from a diagnostic endoscopic makes the procedure an attractive option when
tool to a versatile therapeutic one [1–3]. One conventional approaches fail.
emerging indication is drainage of the pancreatic Pancreatic duct opacification using EUS guid-
duct (PD) in patients with symptomatic PD ance was first described in 1995 by Harada et al.
obstruction who fail conventional endoscopic ret- in a patient status post pancreaticoduodenectomy
rograde cholangiopancreatography (ERCP). Until [5]. Since then, there have been multiple case
recently, percutaneous or surgical drainage was series describing the EUS-PD experience of ther-
the primary option for these patients (and these apeutic endoscopists around the world, with
were only rarely attempted); however, EUS-guided improvements in technique allowing for thera-
pancreatic duct drainage (EUS-PD) has risen as a pies such as stricture dilation, calculi extraction,
minimally invasive, efficacious, and safe alterna- tissue sampling, and long-term drainage [2, 6–8].
tive [4]. The procedure involves visualizing and In this chapter, we will describe the EUS-PD
accessing the PD through EUS, advancing a wire technique and highlight the current literature on
into the duct, creating a fistulous tract, and finally EUS as a modality for pancreatic duct access and
deploying a stent for decompression. While the drainage, evaluating the strengths and limitations
of prior studies, and propose recommendations
therapeutic endoscopists can consider when can-
Electronic supplementary material: The online version
of this chapter (https://doi.org/10.1007/978-3-319-97376-
nulation of the PD fails with conventional ERCP.
0_5) contains supplementary material, which is available
to authorized users.
S. L. Shah Procedural Technique
Department of Gastroenterology and Hepatology,
NewYork-Presbyterian/Weill Cornell Medical Center, Patient Selection
New York, NY, USA
A. Tyberg (*) The most common indications for EUS-PD are
Division of Gastroenterology, Department of inaccessible ampullas, PD strictures, obstructing
Medicine, Rutgers University Medical School, Robert
Wood Johnson University Medical Center, calculi, chronic pancreatitis, disconnected PD syn-
New Brunswick, NJ, USA drome, and complex postsurgical PD anatomy

© Springer Nature Switzerland AG 2019 45

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_5
46 S. L. Shah and A. Tyberg

(e.g., post-pancreaticoduodenectomy) [2]. Common two main approaches: transgastrically or transen-

contraindications to EUS-PD include severe terically. The optimal access site is often deter-
uncorrectable coagulopathy or thrombocytope- mined by the shortest distance between the bowel
nia, inability to visualize the PD on EUS or find a lumen and the main PD without interposed vas-
safe and appropriate access window without culature, as well as on the targeted portion of the
interposed vessels, multifocal PD strictures, or duct and patient anatomy (Fig. 5.1). In most
any condition preventing endoscopy. The most reported studies, the transgastric approach
important aspect of deciding to convert to appears to have the highest technical success [2].
EUS-PD after a failed conventional approach is After achieving relative endoscope stability, a
the experience of the therapeutic endoscopists as puncture is made with a 19 gauge FNA needle
well as appropriate patient selection. and contrast is gently injected to obtain a pan-
creatogram (Figs. 5.2, 5.3, and 5.4). A guidewire
is then advanced through the needle and coiled
Procedural Prerequisites into the duct (Figs. 5.5 and 5.6). Cautery (Fig. 5.7)
of the needle tract followed by balloon dilation
In our practice, all EUS-PD procedures are per- (Figs. 5.8 and 5.9) is used to create a fistulous
formed under general anesthesia with use of car- tract to allow for stent placement. Cautery is
bon dioxide insufflation and fluoroscopy although often applied via needle knife or a cystotome. To
MAC may be a viable alternative from a sedation avoid the risk of a pancreatic fluid leak, the tract
point of view. Intravenous antibiotics are given should be ballooned to the smallest diameter to
periprocedurally along with indomethacin per facilitate stent deployment.
rectum to prevent and decrease the severity of Plastic stents are utilized in the majority of
potential post-procedure pancreatitis [9]. Pre-­ cases. Occasionally, metal stents are deployed
procedure imaging with computed tomography when the PD is markedly dilated (Figs. 5.10 and
or magnetic resonance imaging is necessary for 5.11). Stents can be deployed in an antegrade
optimizing procedural approach and interven- fashion (towards the head of the pancreas) or ret-
tion. A therapeutic channel linear echoendoscope rograde (towards the tail of the pancreas)
is preferred as it allows for the passage of a vari- (Fig. 5.12). Crossing the papilla or anastomosis is
ety of accessories and stents. Equipment includ- prioritized whenever possible (Figs. 5.13 and
ing fine-needle aspiration (FNA) needles for 5.14). When not feasible, transluminal stent
access, guidewires, PD tract dilators and cautery placement is performed with the distal end of the
agents, and stents should be available to the stent in the pancreatic duct and the proximal end
endoscopist as well as a variety of standard ERCP in the gastrointestinal lumen (Figs. 5.15, 5.16,
tools and accessories to help manage unforeseen and 5.17). However, successful crossing can
difficulties [10, 11]. In addition, a multidisci- often be achieved on repeat intervention after the
plinary team including interventional radiolo- duct has been successfully decompressed.
gists and surgeons is required in conjunction with In cases where the papilla/anastomosis is
the endoscopy staff to provide full comprehen- crossed and accessible endoscopically, a rendez-
sive care, although they are not needed to be pres- vous technique can be used and conventional
ent during the EUS-PD procedure. ERCP can be performed: the guidewire is left in
place and coiled into the small bowel as the echo-
endoscope is removed over the wire. A duodeno-
Performing the Procedure (Video 5.1) scope is then inserted to the ampulla/anastomosis,
where the wire is grasped with a biopsy forceps
Once the decision is made to perform EUS-PD, or snare and pulled through the working channel
the patient’s individual anatomy dictates the of the scope, followed by performance of a con-
approach and guides technique and route of ventional ERCP. Alternatively, the wire can be
access selected. A therapeutic channel linear left in place as a guide and cannulation can be
echoendoscope is used to access the main PD via performed with a second wire next to the first
5 Endoscopic Ultrasound-Guided Pancreatic Duct Drainage (EUS-PD) 47

Fig. 5.1
Endosonographic image
of the pancreatic duct
visualized from the
gastric lumen with no
interposing vessels

Fig. 5.2 Fluoroscopic image of a pancreatogram obtained

via transgastric FNA of the tail of the pancreas

Fig. 5.3 Fluorosocpic image of a pancreatogram obtained

via transgastric FNA from the body of the pancreas with
filling of the small bowel

wire, often in short order. The defect at the access most centers performing this procedure have
site is then closed with a hemostatic clip. operators trained in both modalities [12].
In cases where the ampulla/anastomosis is Procedure times vary widely among the
crossed but not reachable endoscopically, trans- EUS-PD studies with a trend towards shorter
papillary/anastomotic transluminal stent deploy- procedure times upon completion of a greater
ment is preferred in which the distal end of the number of cases [13]. In Tyberg et al.’s multi-
stent crosses the ampulla/anastomosis, the mid- center retrospective study of 80 patients, the pro-
dle portion traverses the pancreatic duct, and the cedure success rate was higher than several
proximal end terminates in the gastrointestinal previously reported studies given the experience
lumen (Fig. 5.18). While there are reports of of the endoscopists involved. This highlights the
increased technical success with two endosco- learning curve needed prior to developing a pro-
pists (one trained in ERCP and the other in EUS), ficiency for performing EUS-PD.
48 S. L. Shah and A. Tyberg

Fig. 5.6 Fluoroscopic image of wire advanced into the

pancreatic duct, retrograde orientation
Fig. 5.4 Fluoroscopic image of a pancreatogram obtained
via contrast injection through an FNA needle, transduode-
nal access

Fig. 5.7 Fluoroscopic image of a needle-knife catheter

being advanced over the wire using cautery

Fig. 5.5 Fluoroscopic image of wire advanced into the

pancreatic duct, antegrade orientation

studies, there remains no standardized approach

Post-Procedure Management to aftercare for patients having undergone
EUS-PD and it is unclear if all patients undergo-
Patients undergoing EUS-PD are typically hospi- ing this procedure warrant hospital admission.
talized for observation post-procedure, and While some experts recommend revising stents
provided with analgesics and anti-emetics as
­ contingent on patient symptoms, others use
needed as well as a short course of oral antibiot- cross-­sectional imaging to determine the need for
ics. However, given the overall small number of revision.
5 Endoscopic Ultrasound-Guided Pancreatic Duct Drainage (EUS-PD) 49

Fig. 5.10 Fluoroscopic image of a metal pancreaticogas-

trostomy stent, antegrade orientation, with the distal end
in the pancreatic duct and proximal end in the gastric
lumen
Fig. 5.8 Endoscopic image of the fistulous tract being
dilated using a dilating balloon

Fig. 5.11 Endoscopic image of a metal pancreaticogas-

trostomy stent, antegrade orientation, with the distal end
Fig. 5.9 Fluoroscopic image of the fistulous tract being in the pancreatic duct and proximal end in the gastric
dilated using a dilating balloon lumen

 nalysis of Available Literature

A and EUS-guided pancreatograms [2]. The authors
on EUS-PD included 14 retrospective studies with a total of
222 patients with both native and surgically
While there have been numerous publications altered anatomy, and reported an encouraging
describing EUS-PD outcomes, the overall data is overall technical success rate of 76.6% (n = 170)
limited with no prospective studies to date and a clinical success rate of 70% [7, 10, 12, 14–
(Table 5.1). In 2014, Fujii-Lau et al. published a 24]. However, technical success ranged from as
review of EUS-PD studies excluding case reports low as 36% to as high as 100%.
50 S. L. Shah and A. Tyberg

Fig. 5.14 Fluoroscopic image of wire advanced trans-

gastrically into the pancreatic duct, antegrade orientation,
and across the ampulla into the small bowel

Fig. 5.12 Fluoroscopic image of a plastic pancreatico-

duodenostomy stent, retrograde orientation, with the dis-
tal end in pancreatic duct and the proximal end in the
duodenum

Fig. 5.15 Fluoroscopic image of wire advanced trans-

gastrically into the pancreatic duct, antegrade orientation,
and coiled in the proximal portion of the duct

Fig. 5.13 Fluoroscopic image of wire advanced trans-

gastrically into the pancreatic duct, antegrade orientation,
and across the ampulla into the small bowel

Fig. 5.16 Fluoroscopic image of a plastic pancreatico-

gastrostomy stent, antegrade orientation, with the distal
end in the pancreatic duct and the proximal end in the gas-
tric lumen
5 Endoscopic Ultrasound-Guided Pancreatic Duct Drainage (EUS-PD) 51

Fig. 5.18 Fluoroscopic image of a plastic pancreatico-

Fig. 5.17 Endoscopic image of a plastic pancreaticogas- gastrostomy stent, antegrade orientation, with the distal
trostomy stent, antegrade orientation, with the distal end end in the small bowel, the middle portion in the pancre-
in the pancreatic duct and the proximal end in the gastric atic duct, and the proximal end in the gastric lumen
lumen

Table 5.1 Outcomes of EUS-guided pancreatic duct drainage (studies with 30 or more patients)
Author, year Number of patients Technical success (%) Clinical success (%) Complications
Tessier et al., 2007 [22] 36 33/36 (91.7%) 25/36 (69.4%) 5/36 (13.9%)
Fujii-Lau et al., 2013 [16] 43 32/43 (74.4%) 20/29a (69.0%) 16/43 (37.2%)
Will et al., 2015 [23] 94 94/94 (100%) 68/83b (81.9%) 24/111c (21.6%)
Tyberg et al., 2017 [12] 80 71/80 (88.8%) 65/80 (81.3%) 16/80 (20%)
Total 253 230/253 (90.9%) 178/253 (70.4%) 61/253 (24.1%)
a
Patients who followed up to death or at least 12 months
b
83 patients required drainage
c
111 procedures performed on 94 patients; EUS endoscopic ultrasound

In 2013, Fujii-Lau and his colleagues at the the 11 patients with failed EUS-PD required
Mayo Clinic published the then largest single pancreatic surgery while 2 patients remained
center Unites States experience of EUS-PD in asymptomatic, 2 had recurrent pancreatitis, and
43 patients who failed a conventional ERCP or 2 were lost to follow-up. Interestingly, the
had surgically altered anatomy [17]. The overall authors reported a statistically greater chance of
technical success rate was 74% (n = 32) with an unsuccessful EUS-PD when the procedure
symptom resolution in 68.9% of patients seen in was performed on the same day as a failed
follow-up beyond 1 year. Of the technical suc- ERCP. Shorter stent length and an indication
cesses, stent insertion was antegrade in 18 other than benign anastomotic structure were
patients and retrograde in 14 patients. In those associated with greater clinical success on uni-
with failed stent placement, reasons included variate analysis; prior pancreatic surgery
inability to advance the guidewire through the trended towards a lower l­ikelihood of clinical
papilla/anastomosis (n = 8) or main PD (n = 1), success, possibly due to a larger pancreatic duc-
difficulty dilating the tract (n = 1), and loss of tal diameter. Limitations similar to other prior
the guidewire (n = 1). This emphasizes the need studies included its retrospective nature, small
to decompress the MPD regardless of crossing sample size, and lack of a standardized EUS-PD
the papilla or anastomosis. Furthermore, 5 of technique.
52 S. L. Shah and A. Tyberg

In 2015, Will and his colleagues from Germany technical successes, stents were deployed in an
published the largest international series of antegrade approach in 51 patients and retrograde
patients who underwent EUS-PD after a failed in 20 patients. The method of approach, even when
ERCP over a 12-year period [25]. The authors controlling for indication, altered anatomy and
performed 111 procedures on 94 patients, and prior failed ERCP, did not predict technical suc-
reported a 100% rate of achieving a successful cess (p = 0.23). That said, there appeared to be a
pancreatogram. The overall technical success rate trend towards greater clinical success in patients
was 56.6% (n = 47) with a clinical success rate of who underwent retrograde stent placement (95%
81.9% (n = 68). Of those with successful stent vs. 76%, p = 0.67). Mean follow-up post-EUS-­PD
deployment, 26 patients underwent transgastric or was 24 months with only 1 patient ultimately
transenteric stent insertion while 21 patients requiring surgical intervention. Limitations
underwent transpapillary stent insertion using the included the lack of a standardized approach and
rendezvous technique. On follow-up, 3 patients an adverse event rate of 20%.
underwent a total of 6 reinterventions as a result There were no comparative studies evaluating
of stent migration (n = 2), stent occlusion (n = 2), ERCP and EUS-PD until recently when Chen
and unsuccessful positioning of the PD drain et al. retrospectively compared the two modali-
(n = 2). Interestingly, 12 patients of the 36 with ties in patients with prior pancreaticoduodenecto-
failed EUS-PD experienced clinical improvement mies [26]. The authors evaluated 66 patients who
after additional manipulation at the access site underwent 75 procedures (40 EUS-PD and 35
with use of an endoscopic knife and/or balloon enteroscopy-assisted ERCP [e-ERCP]) at 7 ter-
dilator. Additionally, 15 patients of the 36 had tiary care centers across the world. The overall
continued complaints after EUS-PD failure with 1 technical success rate of EUS-PD was 92.5% as
patient requiring urgent surgical ­ intervention compared to 20% in the e-ERCP cohort (odds
from resultant perforation and 4 requiring surgery ratio [OR] 49.3, p < 0.001). Transluminal stent-
on follow-up. The authors reported that the most ing occurred in 52.5% of cases, followed by ante-
frequent cause for failed PD was because of dif- grade stenting in 40% and rendezvous stenting in
ficulty obtaining PD access. However, after intro- the remaining 7.5% of cases. Clinical success
duction of the Will’s high-frequency ring knife to was achieved in 87.5% of procedures in the
create access to the PD (used in 48 cases), the EUS-PD cohort as compared to 23.1% in the
endoscopist’s rate of clinical success rose from e-ERCP group (OR 23.3, p < 0.001). The authors
71.4% to 89.6%, while the rate of reinterventions reported no severe adverse events in the EUS-PD
dropped from 31.4% to 12.5%. Additionally, the group as well as no significant difference in pro-
total number of adverse events decreased from 14 cedure time or length of stay. While this is the
to 11. The authors reported no intervention-related first study to compare EUS-PD and e-ERCP
deaths. Limitations of the study included a single directly, limitations included lack of randomiza-
experienced endoscopist performing all of the tion and likely selection bias. Furthermore, it is
procedures and use of an endoscopic tool not somewhat difficult to compare these two proce-
widely available, raising concern about the lim- dures directly as e-ERCP is heavily contingent on
ited applicability of this data. even reaching the pancreaticojejunostomy, let
In 2017, Tyberg et al. published the largest mul- alone accessing it.
ticenter experience to date on EUS-PD. [13] The
authors evaluated 80 patients who failed conven-
tional ERCP at 4 academic centers in 3 countries. Adverse Events
Endoscopic retrograde cholangiopancreatography
failure was attributed primarily to either an anasto- One of the most significant barriers to widespread
motic or benign main PD stricture. Overall techni- adoption of EUS-PD is the high rate of adverse
cal success was achieved in 89% of patients events in published studies. Even when performed
(n = 71) with improvement in clinical symptoms by experienced endoscopists, adverse events occur
or imaging in 92% of these patients. Of the 71 in a significant number of patients. In Fujii-Lau
5 Endoscopic Ultrasound-Guided Pancreatic Duct Drainage (EUS-PD) 53

et al.’s review of 222 EUS-PD cases, the authors understanding the long-term outcomes of
reported 42 adverse events (18.9%) including pan- EUS-PD and improved standardization of the
creatitis (n = 7), bleeding (n = 4), perforation (n = 2), technique still need to be cultivated and evaluated
peripancreatic abscess (n = 2), shearing of the guide- in prospective studies. With improved technology
wire (n = 2), perigastric fluid collection (n = 1), both in accessories and endoscopic instruments,
pneumoperitoneum (n = 1), pancreatic pseuodocyst an overall reduction in the rate of adverse events
(n = 1), and pancreatic aneurysm (n = 1) [2]. In Fujii- will hopefully be achieved with increased success
Lau’s large single center study of 43 patients, mod- rate. At this stage, patients who fail ERCP for PD
erate or severe complications occurred in 3 patients intervention should be considered for EUS-PD at
(5.8%) including acute pancreatitis, peripancreatic an experienced center prior to referral to interven-
abscess requiring EUS-guided transmural drainage, tional radiology or surgical intervention.
and shearing of the guidewire into the retroperito-
neum [17]. Of note, 13 patients (31.0%) required
hospitalization post-procedure for abdominal pain References
(median hospitalization of 2 days).
Similarly, in the largest EUS-PD study to date 1. Erickson RA. EUS-guided pancreaticogastrostomy:
invasive endosonography coming of age. Gastrointest
with 80 patients by Tyberg et al., there were 12 Endosc. 2007;65(2):231–2.
serious immediate adverse events (15%) [13]. 2. Fujii-Lau LL, Levy MJ. Endoscopic ultrasound-­
This included 6 patients with post-ERCP pancre- guided pancreatic duct drainage. J Hepatobiliary
atitis, 4 with pancreatic fluid collections, 1 with a Pancreat Sci. 2015;22(1):51–7.
3. Giovannini M. EUS-guided pancreatic duct drain-
main PD leak, and 1 with a luminal perforation. age: ready for prime time? Gastrointest Endosc.
In addition, the authors reported delayed adverse 2013;78(6):865–7.
events in 10% of patients (n = 8) with 4 patients 4. Oh HC, Lee SK, Lee TY, et al. Analysis of percuta-
developing pancreatic abscesses requiring antibi- neous transhepatic cholangioscopy-related compli-
cations and the risk factors for those complications.
otics, 2 with pancreatitis, 1 with a luminal perfo- Endoscopy. 2007;39(8):731–6.
ration requiring surgical repair, and 1 with a main 5. Harada N, Kouzu T, Arima M, Asano T, Kikuchi T,
PD leak. No deaths were reported from any of the Isono K. Endoscopic ultrasound-guided pancreatog-
abovementioned complications. Still, it must be raphy: a case report. Endoscopy. 1995;27(8):612–5.
6. Dhir V, Isayama H, Itoi T, et al. Endoscopic
recognized that EUS-PD is a complex and rela- ultrasonography-­guided biliary and pancreatic duct
tively high risk pancreatic intervention. This interventions. Dig Endosc. 2017;29(4):472–85.
needs to be looked at in light of the invasive 7. Kahaleh M, Hernandez AJ, Tokar J, Adams RB, Shami
nature of alternative treatments, including pan- VM, Yeaton P. EUS-guided pancreaticogastrostomy:
analysis of its efficacy to drain inaccessible pancreatic
creatic surgery, recognizing that those interven- ducts. Gastrointest Endosc. 2007;65(2):224–30.
tions are at least as high risk, if not higher. 8. Chapman CG, Waxman I, Siddiqui UD. Endoscopic
ultrasound (EUS)-guided pancreatic duct drainage:
the basics of when and how to perform E ­ US-­guided
pancreatic duct interventions. Clin Endosc.
Discussion 2016;49(2):161–7.
9. Sotoudehmanesh R, Eloubeidi MA, Asgari AA,
Recent advances in therapeutic EUS have allowed Farsinejad M, Khatibian M. A randomized trial of
endoscopists to provide patients with efficacious rectal indomethacin and sublingual nitrates to pre-
vent post-ERCP pancreatitis. Am J Gastroenterol.
and minimally invasive options. Whereas in the 2014;109(6):903–9.
past, patients who failed conventional ERCP were 10. François E, Kahaleh M, Giovannini M, Matos C,
limited only to percutaneous or surgical interven- Devière J. EUS-guided pancreaticogastrostomy.
tion, EUS-PD now offers therapeutic endosco- Gastrointest Endosc. 2002;56(1):128–33.
11. Kahaleh M, Hernandez AJ, Tokar J, Adams RB,
pists a feasible alternative for PD intervention. Shami VM, Yeaton P. Interventional EUS-guided
Currently, EUS-PD is performed only at highly cholangiography: evaluation of a technique in evolu-
specialized centers. Certainly, efforts at better tion. Gastrointest Endosc. 2006;64(1):52–9.
54 S. L. Shah and A. Tyberg

12. Vila JJ, Pérez-Miranda M, Vazquez-Sequeiros E, 20. Kurihara T, Itoi T, Sofuni A, Itokawa F, Moriyasu
et al. Initial experience with EUS-guided cholangio- F. Endoscopic ultrasonography-guided pancreatic
pancreatography for biliary and pancreatic duct drain- duct drainage after failed endoscopic retrograde chol-
age: a Spanish national survey. Gastrointest Endosc. angiopancreatography in patients with malignant and
2012;76(6):1133–41. benign pancreatic duct obstructions. Dig Endosc.
13. Tyberg A, Sharaiha RZ, Kedia P, et al. EUS-guided 2013;25(Suppl 2):109–16.
pancreatic drainage for pancreatic strictures after 21. Mallery S, Matlock J, Freeman ML. EUS-guided
failed ERCP: a multicenter international collaborative rendezvous drainage of obstructed biliary and pan-
study. Gastrointest Endosc. 2017;85(1):164–9. creatic ducts: report of 6 cases. Gastrointest Endosc.
14. Barkay O, Sherman S, McHenry L, et al. Therapeutic 2004;59(1):100–7.
EUS-assisted endoscopic retrograde pancreatog- 22. Shah JN, Marson F, Weilert F, et al. Single-operator,
raphy after failed pancreatic duct cannulation at single-session EUS-guided antegrade cholangiopan-
ERCP. Gastrointest Endosc. 2010;71(7):1166–73. creatography in failed ERCP or inaccessible papilla.
15. Brauer BC, Chen YK, Fukami N, Shah RJ. Single-­ Gastrointest Endosc. 2012;75(1):56–64.
operator EUS-guided cholangiopancreatography 23. Tessier G, Bories E, Arvanitakis M, et al. EUS-guided
for difficult pancreaticobiliary access (with video). pancreatogastrostomy and pancreatobulbostomy
Gastrointest Endosc. 2009;70(3):471–9. for the treatment of pain in patients with pancreatic
16. Ergun M, Aouattah T, Gillain C, Gigot JF, Hubert C, ductal dilatation inaccessible for transpapillary endo-
Deprez PH. Endoscopic ultrasound-guided translumi- scopic therapy. Gastrointest Endosc. 2007;65(2):
nal drainage of pancreatic duct obstruction: long-term 233–41.
outcome. Endoscopy. 2011;43(6):518–25. 24. Will U, Fueldner F, Thieme AK, et al. Transgastric
17. Fujii LL, Topazian MD, Abu Dayyeh BK, et al. pancreatography and EUS-guided drainage of the
EUS-guided pancreatic duct intervention: outcomes pancreatic duct. J Hepato-Biliary-Pancreat Surg.
of a single tertiary-care referral center experience. 2007;14(4):377–82.
Gastrointest Endosc. 2013;78(6):854–64. e851. 25. Will U, Reichel A, Fueldner F, Meyer F. Endoscopic
18. Kahaleh M, Yoshida C, Yeaton P. EUS antegrade ultrasonography-guided drainage for patients
pancreatography with gastropancreatic duct stent with symptomatic obstruction and enlargement
placement: review of two cases. Gastrointest Endosc. of the pancreatic duct. World J Gastroenterol.
2003;58(6):919–23. 2015;21(46):13140–51.
19. Kinney TP, Freeman ML. The role of endoscopic 26. Chen YI, Levy MJ, Moreels TG, et al. An international
retrograde cholangiopancreatography and endo- multicenter study comparing EUS-guided pancreatic
scopic ultrasound in diagnosis and treatment of duct drainage with enteroscopy-assisted endoscopic
acute pancreatitis. Minerva Gastroenterol Dietol. retrograde pancreatography after Whipple surgery.
2005;51(4):265–88. Gastrointest Endosc. 2017;85(1):170–7.
 EUS-Guided Treatment
of Gastrointestinal Bleeding 6
Larissa L. Fujii-Lau, Louis M. Wong Kee Song,
and Michael J. Levy

Introduction Rationale for and Limitations

of EUS-Guided Angiotherapy
Acute gastrointestinal (GI) bleeding is one of the
most common indications for inpatient GI con- There are several potential advantages to EUS-­
sultation. Although most causes of GI bleeding guided angiotherapy over standard therapy. EUS
can be successfully treated with standard endo- imaging can enhance detection of the culprit ves-
scopic techniques, refractory or recurrent bleed- sel or vessels and allows for precise vascular tar-
ing is encountered in up to 25% of patients [1]. In geting. The use of Doppler before and during
this situation, patients typically undergo salvage therapy allows immediate monitoring of treat-
surgery or interventional radiology (IR)-guided ment response and guides the need for additional
angiotherapy. Often, it is the location and size of therapy. EUS can also be used for surveillance of
the bleeding lesion that complicates source iden- the lesion to determine the long-term efficacy of
tification and delivery of effective hemostasis. treatment.
The efficacy and safety of endoscopic ultrasound EUS-guided angiotherapy has several limita-
(EUS)-guided angiotherapy has recently been tions. First, its performance is generally limited
demonstrated, and is a valuable alternative tech- to tertiary centers with skilled interventional
nique to IR or surgery in select cases presenting endoscopists. Second, the echoendoscope has a
with challenging GI hemorrhage. smaller caliber working channel relative to thera-
peutic upper endoscopes that can limit its ability
to suction blood and clots (which impair visual-
ization), as well as limited capability to perform
Electronic supplementary material: The online version retroflexion and circumferential torque that can
of this chapter (https://doi.org/10.1007/978-3-319-97376- be useful when delivering therapy in some loca-
0_6) contains supplementary material, which is available
to authorized users. tions, such as the gastric fundus and second por-
tion of the duodenum or colon. Third, EUS is
L. L. Fujii-Lau more time consuming than standard endoscopic
Division of Gastroenterology, Queens Medical
Center, Honolulu, HI, USA therapy and may require the use of fluoroscopy,
which makes it difficult to perform in the inten-
University of Hawaii, Honolulu, HI, USA
sive care unit where portable fluoroscopy must be
L. M. Wong Kee Song · M. J. Levy (*) employed. Fourth, the cost of repairing an echo-
Division of Gastroenterology and Hepatology, Mayo
Clinic, Rochester, MN, USA endoscope must be considered given the risk of
e-mail: [email protected] damage to the channel, elevator or transducer

© Springer Nature Switzerland AG 2019 55

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_6
56 L. L. Fujii-Lau et al.

when cyanoacrylate (glue) compounds are used. response with the understanding that the full
The risk of instrument damage secondary to glue extent of hemostasis or complete vascular oblit-
is minimized by careful technique and vigilance eration may not be readily apparent. Nevertheless,
when cleaning the echoendoscope immediately this initial assessment of the treatment response is
following therapy. often used to determine the extent of therapy for
Given the complexity and limitations of EUS-­ that particular procedure.
guided angiotherapy, it is often reserved for Patients should be informed and consented to
patients in whom bleeding lesions are unsuitable the fact that EUS-guided angiotherapy is not a
or refractory to conventional therapy, including Food and Drug Administration (FDA)-approved
standard endoscopic and interventional radio- indication regarding use of various injectates, such
logic treatment. Most published data on EUS-­ as coils or glue. Moreover, the use of glues and
guided angiotherapy pertain to its use in treating coils is investigational and associated with specific
gastric varices (GV). Herein, current techniques risks that should be fully conveyed to the patient
and clinical applications for EUS-guided angio- during the consent process. Due to the duration
therapy for treating variceal as well as nonvari- and complexity of the procedure, we recommend
ceal bleeding lesions are highlighted. the use of general anesthesia. Preprocedure pro-
phylactic intravenous antibiotics should be admin-
istered. The use of postprocedure oral antibiotics is
Technique of EUS-Guided individualized.
Angiotherapy

The technique of EUS-guided vascular access is EUS-Guided Angiotherapy

the same irrespective of the injectate used. After of Varices
identification of the target lesion by curvilinear
echoendoscopy and Doppler imaging, the vascu- EUS-Guided Coil Injection
lar network should be thoroughly mapped to iden-
tify not only the specific site of bleeding, but also Technical Considerations
the network of communicating and feeding ves- The microcoils (Tornado or Nester Embolization
sels when possible. In the case of varices, the goal Coils, Cook Medical Inc., Bloomington, IN) used
is to target either the feeding vessel for a localized during EUS-guided angiotherapy are the same as
network of varices or the largest varix when vari- those deployed during IR angiotherapy. The
ces are extensive. The hemostatic agent of choice selected coil diameter depends on which needle is
(e.g., coil, cyanoacrylate, thrombin, or sclerosant) used, with 0.035-in. coils requiring the use of
should be preloaded into a fine needle aspiration 19-gauge needles while 0.018-in. coils may be
(FNA) needle. In general, we prefer using a deployed via 22-gauge needles. To minimize the
22-gauge FNA needle over a larger 19-gauge nee- risk of the coil itself embolizing, we typically
dle due to ease of use and decreased likelihood of select a coil diameter that is 1.2–1.6 times the
bleeding at the needle puncture site. Larger FNA diameter of the target vessel. The FNA needle is
needles, typically 19 gauge, are warranted if preloaded with the appropriate coil by removing
EUS-guided intravascular coil placement is to be the needle stylet, which is then used to push the
performed. During the actual injection, both EUS coil from its original angiocatheter assembly and
and fluoroscopy are used to ensure proper place- then advanced through the FNA needle until it lies
ment and therapy. Although fluoroscopy is not just short of the needle tip. Although some employ
mandatory, we favor its use when available, espe- a guidewire to advance the coil, we recommend
cially for complex lesions and during the opera- using the stiffer stylet which has easier “pushabil-
tor’s initial experience with EUS-guided ity” and foregoes the need for additional equip-
angiotherapy. After each hemostatic injection, ment. To decrease the risk of coil migration, we
Doppler may be used to monitor the immediate often intentionally advance the needle through the
6 EUS-Guided Treatment of Gastrointestinal Bleeding 57

entire vessel and deposit the distal aspect of the perforating vessel in one treatment session.
coil into the deeper tissues to provide an anchor Although more patients in the cyanoacrylate
for the coil. The stylet is then advanced as the group required subsequent therapy as compared
needle is slowly withdrawn to deposit the major- to the coil group, endosonographers in the study
ity of the coil within the vessel lumen, while leav- perceived coil injection to be more technically
ing the most proximal portion of the coil within demanding. Due to the retrospective and non-
proximal structures, thereby providing an addi- randomized design of the study, the two groups
tional point of anchor (Figs. 6.1 and 6.2) should be cautiously compared.
Our reported experience with EUS-guided coil
Clinical Applications injection encompassed both esophago-­gastric and
Romero-Castro et al. first reported on EUS-­ ectopic varices [4]. Of the fourteen patients
guided coil injection in four patients with included in the study, 10 underwent coil injection
cirrhosis-­related GV, leading to eradication in only for esophago-gastric (n = 1), gastric (n = 2),
three (75%) patients [2]. Per the authors, the first duodenal (n = 2), and choledochal (n = 5) varices
patient had 13 coils inserted throughout the GV for a total of 18 procedures. The median size of
complex to theoretically reduce the risk of migra- the targeted varix was 6.5 mm (range 4.4–16 mm)
tion, followed by 9 coils placed into a 13 mm per- and a mean of 4.6 ± 1.8 coils were placed during
forating vessel. The subsequent three patients the index procedure. During a median follow-up
had 2–7 coils placed only within the 6–12 mm of 18 months (range 0–104 months), three patients
perforating vessels. There was no migration of died and four patients did not experience recur-
coils over a 5-month follow-up period. rent bleeding up to 8 years following their index
The same group published a multicenter, ret- procedures. One patient had successful coil injec-
rospective study comparing the use of coils to tion of GV, but re-bled 6 months later from esoph-
cyanoacrylate injection in managing GV [3]. In ageal varices which were treated with band
the coil only group, complete GV obliteration by ligation and sclerotherapy. The remaining two
injection into the perforating vein occurred in 10 patients had improvement in bleeding from chole-
of 11 patients (91%) with a mean of 5.8 ± 1.2 dochal varices after the initial EUS therapy, but
coils placed per patient. The majority of patient required additional EUS-guided coil placement or
(n = 9; 82%) had complete obliteration of the endoscopic retrograde cholangiopancreatography

Fig. 6.1 In a patient with multiple prior episodes of gas- sodes of clinically significant gastric variceal bleeding
tric variceal bleeding, EUS demonstrates a network of resulting from alcohol and hepatitis C induced cirrhosis
varices with power Doppler imaging. Image demonstrate (Courtesy of the Mayo Clinic)
EUS-guided angiotherapy in a patient with multiple epi-
58 L. L. Fujii-Lau et al.

Fig. 6.2 A needle

(orange arrow) is
advanced into the
varices and microcoils
(yellow arrow) are
inserted. Image
demonstrate EUS-­
guided angiotherapy in a
patient with multiple
episodes of clinically
significant gastric
variceal bleeding
resulting from alcohol
and hepatitis C induced
cirrhosis (Courtesy of
the Mayo Clinic)

cyanoacrylate once the target vessel has been iden-

tified. Preloading the needle avoids the need to
remove the stylet after needle puncture, resulting
in vacuum suction of blood into the needle that
could be reinserted along with air during glue
injection. If an oil-based contrast agent (e.g.,
Lipiodol, Guerbet LLC, Bloomington, IN) is
employed to allow for fluoroscopic guidance, we
typically use a 2:1 mixture of 2-octyl-­cyanoacrylate
(Dermabond; Ethicon Inc., Somerville, NJ) to lipi-
odol in 3 ml syringes when using a 22-gauge nee-
dle versus a 2.5:0.5 mixture when using a 19-gauge
needle. To decrease the risk of glue embolization,
Fig. 6.3 Despite having experienced multiple clinically the least amount of glue needed to adequately
significant bleeds, including a recent episode when blood occlude the vessel is used and the mixture is
was seen emanating from gastric varices, there was diffi-
injected at a rate approximating 1 ml per 15 s.
culty on routine endoscopy identifying the gastric varices.
Image demonstrate EUS-guided angiotherapy in a patient
with multiple episodes of clinically significant gastric Clinical Applications
variceal bleeding resulting from alcohol and hepatitis C The first study on EUS-guided cyanoacrylate
induced cirrhosis (Courtesy of the Mayo Clinic)
injection compared a historical group of patients
who underwent conventional endoscopic glue
(ERCP) with the placement of a fully covered injection during an acute episode of GV hemor-
metal stent (to treat bile duct related bleeding) to rhage with a group who underwent endoscopic
obtain long-­term hemostasis. glue injection for initial hemostasis followed by
EUS surveillance and further glue injection until
eradication [5]. Primary hemostasis during the
 US-Guided Cyanoacrylate Injection
E index procedure occurred in >95% of patients in
(Figs. 6.3, 6.4, 6.5, 6.6, 6.7, and 6.8 both groups. In 43 of 54 patients (80%) in the EUS
and Video 6.1) group, complete GV obliteration occurred after
2.2 ± 1.7 procedures. No adverse events were
Technical Considerations reported during the EUS-guided injection. Patients
Cyanoacrylate polymerizes upon contact with in the EUS surveillance group had significantly
blood, resulting in hemostasis and vascular occlu- fewer episodes of recurrent GV bleeding com-
sion. We recommend preloading the needle with pared to those who only received conventional
6 EUS-Guided Treatment of Gastrointestinal Bleeding 59

Fig. 6.4 (a, b) At EUS, grey

scale (Fig. 6.2) and power
Doppler (Fig. 6.3) imaging
demonstrated a network of
gastric varices. Image
demonstrate EUS-­guided
angiotherapy in a patient with
multiple episodes of clinically
significant gastric variceal
bleeding resulting from
alcohol and hepatitis C
induced cirrhosis (Courtesy
of the Mayo Clinic)

endoscopic glue injection (26% vs 57%, respec- of patients had successful treatment after one
tively, p = 0.002). Although the use of a historical session of EUS-guided glue injection. A mean
cohort is subject to bias, this study introduced the of 1.5 ± 0.1 ml of cyanoacrylate was injected
concept that patients with active GV hemorrhage per patient. Although 12 adverse events
may benefit from EUS surveillance with second- occurred in 11 patients in the cyanoacrylate
ary prophylaxis until eradication to decrease the group, only two patients were symptomatic,
risk of re-bleeding. including fever (n = 1) and chest pain (n = 1).
A case series of patients with cirrhosis- There were 9 pulmonary glue embolisms (47%)
related GV who underwent EUS-guided cyano- detected on routine chest CT performed in all
acrylate injection into perforating vessels patients in the EUS-guided glue injection
showed complete obliteration in all five patients group, which significantly lengthened their
[6]. A mean of 1.6 ml of glue was injected. hospital stay.
During a mean follow-up of 10 months, no
adverse events or recurrent bleeding was
observed. When the same group focused on EUS-Guided Combination Therapy
patients who underwent only EUS-guided glue
injection, all 19 patients had complete oblitera- The injection of coils prior to glue theoretically
tion of the feeding gastric vessels [3]. The five provides a scaffold to help anchor the glue, thereby
patients reported in the initial case series were decreasing the amount of glue required for oblit-
not included in the subsequent study. Only 42% eration and minimizing the risk of embolization,
60 L. L. Fujii-Lau et al.

Fig. 6.5 (a, b) As the patient had a large splenorenal adrenal vein. A fluoroscopic run was obtained with the
shunt, balloon-occluded retrograde transvenous oblitera- balloon inflated to confirm that it does occlude flow. The
tion (BRTO) was performed to minimize the risk of glue balloon was left inflated while the gastric varices were
embolization. The left inferior phrenic vein was selec- treated with endoscopic glue injection. Image demon-
tively catheterized and satisfactory position confirmed strate EUS-guided angiotherapy in a patient with multiple
with venography. This catheter was exchanged for a tem- episodes of clinically significant gastric variceal bleeding
porary balloon occlusion catheter, positioned in the infe- resulting from alcohol and hepatitis C induced cirrhosis
rior phrenic vein above the confluence with the left (Courtesy of the Mayo Clinic)

the most concerning adverse event during and a mean of 1.4 ml of 2-octyl-cyanoacrylate
EUS-guided angiotherapy [7]. Binmoeller et al. was injected. No immediate adverse events
described an ex-vivo experiment where 1 ml of occurred, particularly regarding glue emboliza-
cyanoacrylate was injected into heparinized blood tion. In patients who underwent subsequent
that contained a previously placed coil. The glue surveillance endoscopy (n = 24), 96% had com-
adhered to the coil fibers, allowing all of the glue plete obliteration of the feeding vessels and no
to be removed with the coil in a single piece. They evidence of flow on color Doppler within the
hypothesized that EUS-guided coil insertion fol- variceal complex. One patient had recurrent
lowed by cyanoacrylate injection improves vari- GV bleeding 21 days after the initial procedure,
ceal obliteration while decreasing the risk of glue which was treated with repeat EUS-guided
embolization. Prospective trials are needed to con- combined coil and glue injection. At follow-up
firm the theoretical benefit of using coils to anchor endoscopy, the glue and coils had spontane-
the glue. ously extruded into the stomach and eventually
The same group retrospectively analyzed 30 formed a scar.
patients with acute (n = 2), recent (defined as In the largest study to date, the same group
<1 week, n = 18), or remote (n = 10) bleeding reported on 152 patients with GV treated with a
from GV who underwent EUS-guided coil and mean of 1.4 coils and a mean of 2 ml cyanoac-
glue embolization of feeding vessels [8]. rylate for active hemorrhage (n = 7), stigmata
Technical success of coil and glue injection of recent bleeding (n = 105), and primary pro-
occurred in all patients, while immediate hemo- phylaxis (n = 40) [9]. In one patient, technical
stasis was achieved in both patients with overt failure occurred with the inability to control the
bleeding at the time of the endoscopy. The bleed despite injection of a coil and 6 ml of
majority of cases (93%) only had 1 coil placed glue, requiring urgent transjugular intrahepatic
6 EUS-Guided Treatment of Gastrointestinal Bleeding 61

portosystemic shunt (TIPS). Of the 100 patients

with follow-up EUS, 93% had complete oblit-
eration of the GV by Doppler evaluation after
one (n = 79), two (n = 10), three (n = 2), or four
(n = 2) procedures. Of these 93 patients, three
had recurrent bleeding after a median of
324 days and were treated with additional coil
and glue therapy. There were nine procedure-
related adverse events, including self-limited
pain (n = 4), embolization (n = 1), and minor
bleeding from coil/glue extrusion (n = 4). The
one patient with embolization presented 1 week
after discharge with shortness of breath, hemop-
tysis, and fever. CT identified an acute pulmo-
nary embolism and associated pneumonia.
Although this study included patients who
Fig. 6.6 Glue and lipiodol were injected under EUS
underwent primary prophylaxis, additional
guidance, resulting in complete filling of the variceal net- studies are required in this patient population
work. The glue migrated into the feeding vessel with pro- before advocating routine EUS-guided
tection provided by the occlusion balloon to prohibit angiotherapy.
shunt embolization. Image demonstrate EUS-guided
angiotherapy in a patient with multiple episodes of clini-
In our experience, four patients (3 with gas-
cally significant gastric variceal bleeding resulting from tric and 1 with duodenal varices) underwent a
alcohol and hepatitis C induced cirrhosis (Courtesy of the combination of coil and glue injection [4]. In
Mayo Clinic) these patients, the varices were deemed to be

Fig. 6.7 (a, b) Standard endoscopy was then performed. Image demonstrate EUS-guided angiotherapy in a patient
A closed biopsy forceps was placed at the site of pre- with multiple episodes of clinically significant gastric
ciously identified bleeding demonstrated the site of glue variceal bleeding resulting from alcohol and hepatitis C
injection. When applying gentle pressure, firmness was induced cirrhosis (Courtesy of the Mayo Clinic)
appreciated as a result of the injected glue and thrombus.
62 L. L. Fujii-Lau et al.

Fig. 6.8 (a, b) Following therapy, the IR-inserted balloon in a patient with multiple episodes of clinically significant
was deflated demonstrating no further spread or emboliza- gastric variceal bleeding resulting from alcohol and hepa-
tion of glue. Image demonstrate EUS-guided angiotherapy titis C induced cirrhosis (Courtesy of the Mayo Clinic)

too large to be treated either with standard glue case report described placement of 1 coil and
injection or with EUS-guided coil injection 1 ml of cyanoacrylate into a large rectal varix,
only. A mean of 7.5 coils and 3 ml of glue were with repeat sigmoidoscopy confirming the lack
injected in these patients. There was no further of large feeding vessels and collapse of the rectal
evidence of bleeding at a mean follow-up of varix [13]. One case report described the success-
4 months. ful use of sclerotherapy for the treatment of
recurrent bleeding from rectal varices with 2 ml
of sodium tetradecyl sulfate [14].
 US-Guided Angiotherapy of Rectal
E
Varices
EUS-Guided Angiotherapy
EUS can more accurately characterize the extent of Nonvariceal Bleeding
of rectal varices, define the hemodynamics, and
target the perforating veins with therapeutic Since most studies on EUS-guided angiotherapy
intent [10, 11]. Coil and glue injection has been are limited to the treatment of esophagogastric
reported in the treatment of recurrent bleeding varices, data pertaining to the use of EUS for
rectal varices. One patient underwent 4 punctures treating nonvariceal bleeding are scant. One of
along a large >30 mm varix with deployment of the first applications of EUS-guided angiother-
either 1 or 2 coils and 1 ml of cyanoacrylate into apy was in eight patients with suspected
each puncture site and no recurrent bleeding after Dieulafoy lesions [15]. In these patients, the
12 months of follow-up [12]. These authors pos- stomach was filled with 200–400 ml of water and
tulated the advantages of EUS over standard radial echoendoscopy identified potential culprit
endoscopic treatment of these large varices, 2–3 mm vessels penetrating the muscularis pro-
including its ability to deliver precise treatment pria and coursing through the submucosa for
without being obscured by luminal contents, 2–4 cm. Four patients underwent sclerotherapy, 3
visualize deeper collaterals, and confirm the of which were EUS-guided. During a median
absence of flow by Doppler after therapy. Another follow-up of 10 months, two patients re-bled at 3
6 EUS-Guided Treatment of Gastrointestinal Bleeding 63

and 5 months post-therapy. One patient who was ligation with alcohol injection (n = 1), and com-
receiving nonsteroidal anti-inflammatory medi- bined epinephrine injection, endoloop ligature
cations re-bled from a duodenal ulcer that con- and polypectomy (n = 1). Doppler ultrasound at
tained a visible vessel, while the other patient had the conclusion of the EUS confirmed complete
a lesion located at 1.5 cm from the prior sclero- cessation of flow in 11 patients (84.6%) and
therapy scar; both patients had repeat sclerother- marked decrease in 1 patient. Only 1 patient
apy performed. In three patients, surgical (7.7%) had recurrent bleeding attributed to the
pathology confirmed the presence of a submuco- treated lesion. This patient had a gastric
sal vessel consistent with Dieulafoy lesions. Dieulafoy lesion that was managed with
Similarly, Ribeiro et al. described a case of EUS-­ EUS-guided India ink marking and band liga-
guided bipolar coagulation followed by sclero- tion. This patient experienced re-­ bleeding
therapy of a Dieulafoy lesion located 4 cm distal 38 months later, which was again treated with
to the gastroesophageal junction [16]. EUS-guided India ink marking and band ligation
EUS-guided thrombin injection has been using multiple bands. This patient subsequently
described in the treatment of pseudoaneurysms experienced complete cessation of Doppler flow
[17–20]. Thrombin promotes the conversion of to the lesion and had no further evidence of re-
fibrinogen to fibrin, resulting in clot production bleeding at 5 months follow-up. No adverse
[21]. In one study of four cases of patients who events were encountered from these procedures.
developed pseudoaneurysm formation secondary
to pancreatitis, a 22-gauge needle was used to
inject the thrombin solution. Within a minute, Summary
blood flow to the aneurysm ceased in each case
and a thrombus had formed within the aneurys- Difficult-to-treat GI bleeding remains a common
mal sac. This thrombus persisted at 6- to 42-week clinical challenge for gastroenterologists. EUS
follow-up. Partial aneurysmal recanalization was detection and EUS-guided angiotherapy using a
noted on CT angiography at 12 weeks post-­ variety of agents, including coils, glue, scle-
thrombin injection with associated reports of rosant, or a combination thereof, has increasingly
melena; however, this did not persist and the been used in this situation. However, additional
aneurysm had spontaneously thrombosed on fol- studies are needed to determine the optimal use
low-­up CT scans at 28 and 42 weeks [17]. of EUS-guided angiotherapy, including the ideal
The largest case series to date from our insti- target lesions, injectate(s), dose, and follow-up in
tution included 13 patients who had either failed these complex patients. In the interim, manage-
prior therapy or were deemed unsuitable candi- ment decisions should be based on local expertise
dates for other endoscopic, interventional radio- and in a multidisciplinary manner that incorpo-
logic or surgical procedures [22]. Indications rates standard endoscopy, EUS, interventional
for EUS-guided angiotherapy included gastric radiology, and surgery.
gastrointestinal stromal tumors (n = 4),
Dieulafoy lesions (n = 2), duodenal metastases Disclosures None.
(n = 2), esophageal cancer (n = 1), intractable
marginal ulcer after gastric bypass (n = 1), duo-
denal ulcer (n = 1), duodenal Brunner’s gland
References
hamartoma (n = 1), and pancreatic pseudoaneu-
rysm (n = 1). All but one patient failed prior 1. Sarin SK, Lahoti D, Saxena SP, Murthy NS, Makwana
endoscopic and/or IR-guided therapy. Fifteen UK. Prevalence, classification and natural his-
EUS-guided procedures were performed in tory of gastric varices: a long-term follow-up study
in 568 portal hypertension patients. Hepatology.
total, including cyanoacrylate injection (n = 5), 1992;16(6):1343–9.
hyaluronate injection (n = 3), alcohol ablation 2. Romero-Castro R, Pellicer-Bautista F, Giovannini M,
(n = 3), band ligation (n = 2), combined band Marcos-Sanchez F, Caparros-Escudero C, Jimenez-­
64 L. L. Fujii-Lau et al.

Saenz M, et al. Endoscopic ultrasound (EUS)-guided 12. Weilert F, Shah JN, Marson FP, Binmoeller KF. EUS-­
coil embolization therapy in gastric varices. Endoscopy. guided coil and glue for bleeding rectal varix.
2010;42(Suppl 2):E35–6. Gastrointest Endosc. 2012;76(4):915–6.
3. Romero-Castro R, Ellrichmann M, Ortiz-Moyano C, 13. Philips CA, Augustine P. Endoscopic ultrasound-­
Subtil-Inigo JC, Junquera-Florez F, Gornals JB, et al. guided management of bleeding rectal varices. ACG
EUS-guided coil versus cyanoacrylate therapy for the Case Rep J. 2017;4:e101.
treatment of gastric varices: a multicenter study (with 14. Connor EK, Duran-Castro OL, Attam R. Therapy for
videos). Gastrointest Endosc. 2013;78(5):711–21. recurrent bleeding from rectal varices by EUS-guided
4. Fujii-Lau LL, Law R, Wong Kee Song LM, sclerosis. Gastrointest Endosc. 2015;81(5):1280–1.
Gostout CJ, Kamath PS, Levy MJ. Endoscopic 15. Fockens P, Meenan J, van Dullemen HM, Bolwerk
ultrasound (EUS)-guided coil injection therapy of CJ, Tytgat GN. Dieulafoy’s disease: endosonographic
esophagogastric and ectopic varices. Surg Endosc. detection and endosonography-guided treatment.
2016;30:1396. Gastrointest Endosc. 1996;44(4):437–42.
5. Lee YT, Chan FK, Ng EK, Leung VK, Law KB, Yung 16. Ribeiro A, Vazquez-Sequeiros E, Wiersema MJ.
MY, et al. EUS-guided injection of cyanoacrylate Doppler EUS-guided treatment of gastric Dieulafoy’s
for bleeding gastric varices. Gastrointest Endosc. lesion. Gastrointest Endosc. 2001;53(7):807–9.
2000;52(2):168–74. 17. Roach H, Roberts SA, Salter R, Williams IM, Wood
6. Romero-Castro R, Pellicer-Bautista FJ, Jimenez-­ AM. Endoscopic ultrasound-guided thrombin injec-
Saenz M, Marcos-Sanchez F, Caunedo-Alvarez A, tion for the treatment of pancreatic pseudoaneurysm.
Ortiz-Moyano C, et al. EUS-guided injection of Endoscopy. 2005;37(9):876–8.
cyanoacrylate in perforating feeding veins in gas- 18. Robinson M, Richards D, Carr N. Treatment of
tric varices: results in 5 cases. Gastrointest Endosc. a splenic artery pseudoaneurysm by endoscopic
2007;66(2):402–7. ultrasound-guided thrombin injection. Cardiovasc
7. Weilert F, Binmoeller KF. EUS-guided vascular Intervent Radiol. 2007;30(3):515–7.
access and therapy. Gastrointest Endosc Clin N Am. 19. Lameris R, du Plessis J, Nieuwoudt M, Scheepers
2012;22(2):303–14, x. A, van der Merwe SW. A visceral pseudoaneurysm:
8. Binmoeller KF, Weilert F, Shah JN, Kim J. EUS-­ management by EUS-guided thrombin injection.
guided transesophageal treatment of gastric fundal Gastrointest Endosc. 2011;73(2):392–5.
varices with combined coiling and cyanoacrylate glue 20. Chaves DM, Costa FF, Matuguma S, Lera Dos Santos
injection (with videos). Gastrointest Endosc. 2011; ME, de Moura EG, Maluf Filho F, et al. Splenic
74(5):1019–25. artery pseudoaneurysm treated with thrombin injec-
9. Bhat YM, Weilert F, Fredrick RT, Kane SD, Shah JN, tion guided by endoscopic ultrasound. Endoscopy.
Hamerski CM, et al. EUS-guided treatment of gastric 2012;44(Suppl 2 UCTN):E99–100.
fundal varices with combined injection of coils and cya- 21. Bhat YM, Banerjee S, Barth BA, Chauhan SS,
noacrylate glue: a large U.S. experience over 6 years Gottlieb KT, Konda V, et al. Tissue adhesives: cyano-
(with video). Gastrointest Endosc. 2016;83(6):1164–72. acrylate glue and fibrin sealant. Gastrointest Endosc.
10. Sato T, Yamazaki K, Akaike J. Evaluation of the 2013;78(2):209–15.
hemodynamics of rectal varices by endoscopic ultra- 22. Law R, Fujii-Lau L, Wong Kee Song LM, Gostout CJ,
sonography. J Gastroenterol. 2006;41(6):588–92. Kamath PS, Abu Dayyeh BK, et al. Efficacy of endo-
11. Sharma M, Rai P, Bansal R. EUS-assisted evaluation scopic ultrasound-guided hemostatic interventions
of rectal varices before banding. Gastroenterol Res for resistant nonvariceal bleeding. Clin Gastroenterol
Pract. 2013;2013:619187. Hepatol. 2015;13(4):808–12.e1.
 EUS-Guided Celiac Plexus Block
and Celiac Plexus Neurolysis 7
Truptesh H. Kothari, Shivangi Kothari,
and Vivek Kaul

the agents used for treatment and indications

Introduction [7]. Celiac plexus block (CPB) refers to a tem-
porary disruption of pain transmission from
Abdominal pain can be a very common debilitat- the afferent nerves to the spinal cord via the
ing symptom in patients with upper abdominal celiac plexus and is accomplished by injecting
malignancy or a benign disease process such as corticosteroids (e.g., triamcinolone) and local
chronic pancreatitis. Cancer related pain could anesthetics. CPB is generally used for pain
dramatically affect the quality of life of these relief in benign disease processes such as
patients [1–3]. The pain originating from upper chronic pancreatitis [7]. In contrast, celiac
abdominal viscera is carried by special visceral plexus neurolysis (CPN) refers to the perma-
afferent fibers that relay through the splanchnic nent destruction of the celiac plexus with a
nerves and the celiac plexus [4, 5]. The celiac neurolytic agent along with local anesthetics.
plexus is a network of nerve fibers located in the Ethanol is generally used as a neurolytic with
retroperitoneum, adjacent to the anterolateral bupivacaine as a local anesthetic for CPN. CPN
wall of the aorta [6]. The principle of relieving is indicated for alleviating pain from abdomi-
the pain associated with upper abdominal nal malignancy [7].
malignancy is to disrupt nociceptive impulses at The CPN technique was initially reported in
the level of the celiac plexus or splanchnic 1914 and was performed as an intraoperative
nerves [6]. procedure [8]. CPN has been carried out under
The terms celiac plexus block and neuroly- radiographic, fluoroscopic, computed tomogra-
sis are sometimes used interchangeably, but phy (CT), or ultrasonographic imaging guidance.
they differ in terms of duration of action and Faigel et al. [9] and Wiersema and Wiersema [10]
introduced endoscopic ultrasound (EUS)-guided
CPN in 1996. This technique is considered safer,
precise, and more convenient than other previ-
Electronic supplementary material: The online version ously used techniques since EUS-guided CPN is
of this chapter (https://doi.org/10.1007/978-3-319-97376- carried out under real time ultrasound imaging
0_7) contains supplementary material, which is available
to authorized users.
and color Doppler assessment, which helps avoid
injury to interposing vessels. In a randomized
T. H. Kothari (*) · S. Kothari · V. Kaul controlled trail (RCT), Gress et al. showed that
Division of Gastroenterology and Hepatology,
the EUS-guided CPB technique provided more
Department of Medicine, University of Rochester
Medical Center, Rochester, NY, USA long-term pain relief than the CT-guided
e-mail: [email protected] approach [11].

© Springer Nature Switzerland AG 2019 65

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_7
66 T. H. Kothari et al.

Indications 3. Gastric and/or esophageal varices.

4. Use of anticoagulation—medication hold
EUS-guided CPN is often used for persistent and required prior to procedure.
intractable abdominal pain due to gastric, pancre- 5. Severe alcohol intolerance.
atic, esophageal, and biliary malignancy as well
as retroperitoneal lymph node metastasis or pain
due to metastatic liver cancer [12]. CPB is usu- Techniques of EUS-CPN/CPB
ally performed for abdominal pain associated
with chronic pancreatitis [12]. It has been Central CPN
reported to also be beneficial in cases of nausea
and vomiting due to pancreatic cancer. CPB This technique is also known as the single injec-
results in unopposed parasympathetic activity, tion technique. The abdominal aorta is located
which helps with increased peristalsis relieving through the posterior gastric wall using the linear
nausea and vomiting symptoms. The unopposed echoendoscope, at the level just below the gastro-­
parasympathetic activity is due to the sympa- esophageal junction. Once the abdominal aorta is
thetic denervation of the gastrointestinal tract. visualized, the celiac artery (CA) is identified as
the first vessel originating from the aorta (Fig. 7.1).
In this technique for CPN, an FNA needle (19 or
Contraindications to CPN/CPB 22 gauge) is advanced to the level just above the
origination of the CA from the abdominal aorta
1. Supratherapeutic INR (International normal- (Figs. 7.2 and 7.3). A normal saline syringe is
ized ratio >1.5). attached to the needle and negative suction is
2. Thrombocytopenia (platelets <50,000). applied by aspirating the syringe to ensure a ves-

Fig. 7.1 EUS image showing the origin of the celiac artery and the expected location of the celiac plexus in most
patients
7 EUS-Guided Celiac Plexus Block and Celiac Plexus Neurolysis 67

Fig. 7.2 EUS image of a central celiac plexus injection (in this case a neurolysis) into the celiac ganglia just anterior
to the origin of the celiac artery

Fig. 7.3 EUS image during a celiac plexus neurolysis showing the cloudy distortion that accompanies ethanol
injection
68 T. H. Kothari et al.

Fig. 7.4 EUS image showing persistent haziness following removal of the needle after ethanol injection

sel has not been inadvertently entered. Once the withdrawn and the echoendoscope is rotated
absence of blood in the syringe is confirmed, counterclockwise until the CA and SMA are no
0.25% × 20 ml bupivacaine is administered longer visible (Fig. 7.6). Now the needle is
through the needle, followed by 10 ml absolute advanced again up to the level of the SMA origin
alcohol. An echogenic cloud is evident at the site from the aorta and the injection agents are admin-
of injection (Figs. 7.4 and 7.5). The needle is istered. The steps and the volume of agents
flushed with normal saline at the injection site fol- injected for bilateral technique remain the same
lowed by withdrawal of the needle (Video 7.1). as in central technique [13, 14].
In CPB, bupivacaine is administered followed by
3 ml (40 mg) triamcinolone [7].
Central Versus Bilateral Technique

Bilateral Several studies have been published comparing

the two CPN techniques (central vs bilateral) in
After the identification of the CA origin from the terms of pain relief but it is still a matter of
abdominal aorta (Fig. 7.1), the echoendoscope debate as to which technique is superior to the
is rotated clockwise until the CA and superior other. Puli et al. compared both the techniques
mesenteric artery (SMA) are not in the field of in terms of pain relief in the pancreatic cancer
vision. The FNA needle is then advanced via the group [7], concluding that the bilateral tech-
trans-­gastric approach up to the point of origin of nique (84.54%) was a superior modality com-
the SMA from the aorta. The injection agents are pared to the central technique (45.99%) (95%
administered into this region. The needle is then CI = 37.33–54.78).
7 EUS-Guided Celiac Plexus Block and Celiac Plexus Neurolysis 69

Fig. 7.5 EUS with schematic arrow showing needle path for a central celiac plexus injection

Fig. 7.6 EUS image with schematic arrows showing the needle path for bilateral celiac plexus injections
70 T. H. Kothari et al.

In 2009, Sahai et al. compared the short-term and chronic pancreatitis [19]. During this retro-
efficacy of pain relief (pain score reduction) with spective trial, the volume of injection and num-
both techniques in EUS-guided CPN/CPB in 160 ber of celiac ganglia injected were not
patients [15]. Bilateral technique was found to be consistently reported, perhaps since it was a new
more effective than the central CPB/CPN tech- technique then. However, in general the mean
nique with a mean pain reduction of 70.4% vs. number of ganglia injected were 2.7 and 2.3 in
45.9% (P = 0.0016). Bilateral CPB/CPN was patients with pancreatic cancer and chronic pan-
reported to be the only predictor of a >50% pain creatitis, respectively. Depending on the size of
reduction (OR 3.55, 1.72–7.34). ganglion—if smaller than 1.0 cm as measured
However, LeBlanc et al. [16] performed a within the axis of the needle plane, the needle tip
randomized controlled trial of 50 patients, which was positioned within the central point of the
concluded no significant difference between the ganglia. If the ganglion was 1.0 cm or greater in
two groups in terms of pain relief. It is to be noted the needle plane axis, the needle tip was inserted
though that for the bilateral technique in the to the deepest point within the ganglion and
Leblanc study, the needle was advanced laterally intraganglion injection was performed on slow
to the CA with injection performed on both sides withdrawal [19].
of the CA without any distal advancement of the
needle towards the base of the CA, whereas in the
study by Sahai et al. the needle was advanced lat- CPN Versus CGN
eral to the CA and further advanced to the region
lateral to the base of SMA. However, self-limited EUS-guided CPN is a well-established interven-
bleeding in an anticoagulated patient has been tion for the relief of cancer pain. With CPN, neu-
reported due to laceration of the adrenal artery rolytic agents are injected around the celiac
with the bilateral technique and thus caution trunk where the ganglion is thought to reside.
should be taken while performing this technique However, injection directly into the ganglion (if
[15, 17]. visualized on EUS) may provide more effective
response in the treatment of pain. In 2013 Doi
et al. reported a multicenter, randomized study
 US-Guided Direct Celiac Ganglion
E of 68 patients comparing EUS CPN vs. EUS
Neurolysis (CGN) and Celiac CGN for pain relief in patients with upper
Ganglion Block (CGB) abdominal malignancy. In this study, CPN was
performed using the “central method” with
In 2006, Levy et al. demonstrated that the celiac injection performed just above the origin of the
ganglia can be visualized with the aid of celiac artery. EUS CGN was performed in 30 of
EUS. Levy et al. suggested that EUS-guided 34 patients (88%); the celiac ganglia could not
direct CGN was highly effective in pancreatic be visualized in 4 patients. In patients with gan-
cancer pain relief with a 94% success rate when glia <1 cm in size the injection was performed in
alcohol was injected (16/17 patients) and 0% the center whereas in patients with ganglia
success rate (0/1 patient) when steroid was >1 cm in size the needle was advanced deep into
injected [18, 19]. In patients with chronic pan- the ganglion and injection performed as the nee-
creatitis, 80% (4/5 patients) pain relief was dle was slowly withdrawn so as to distribute the
reported with alcohol injection versus 38% (5/13 injection throughout the ganglion. Significantly
patients) who received steroids. Initial experience higher pain response rate (decrease in pain score
by Levy et al. in 2008 suggested that EUS-guided to ≤3) was seen in the EUS-CGN group (73.5%)
CGN or CGB was safe. It was also noted that compared to the EUS- CPN group (45.4%,
alcohol injection directly into the ganglia p = 0.026). Also 50% of patients in the EUS-
appeared to be effective in patients with cancer CGN group had complete response to treatment,
7 EUS-Guided Celiac Plexus Block and Celiac Plexus Neurolysis 71

compared to 18.2% in the EUS-CPN group the commonly reported adverse symptoms [7,
(p = 0.010). No difference was seen in the dura- 10–12]. Alvarez-Sanchez et al. reported a review
tion of pain relief or in the complications of 1142 patients, which showed complications in
between the two groups [20]. 7% of 481 EUS CPB procedures and 21% of 661
EUS-CPN procedures [22]. Most frequent com-
plications experienced were transient diarrhea
EUS-Guided CPB (7% of patients) which spontaneously resolved
Versus CT-Guided CPB and hypotension (4% of patients). Transient
increase in pain occurred in 2% of EUS-CPB and
Gress et al. [21] conducted a prospective random- in 4% of EUS-CPN cases.
ized comparison of EUS-guided CPB and Several major adverse events have also been
CT-guided CPB for management of chronic pan- reported which include infectious complica-
creatitis pain. Twenty-two consecutive patients tions in patients with chronic pancreatitis.
were enrolled in the study from 7/1/1995 to Therefore, antibiotic prophylaxis is recom-
12/30/1995. Ten patients underwent EUS-guided mended before EUS-CPB when steroids are used
CPB and 8 patients underwent CT-guided [23]. Irreversible paraparesis is one of the major
CPB. Four patients were excluded due to proto- adverse events reported with EUS-guided CPB
col violations. EUS-guided CPB was performed with posterior approach [24]. In an RCT of
with a 22-gauge FNA needle. Ten milliliter of EUS-­CPN and EUS-CGN, the overall complica-
bupivacaine (0.75%) and 3 ml (40 mg) of triam- tion rates were similar in both groups but the
cinolone was injected on both sides of the celiac overall volume of ethanol injected was signifi-
artery under real time imaging with linear array cantly less in EUS-CGN [24]. Also the target was
endosonography. clearly visualized in EUS-CGN reducing the
CT-guided CPB was performed in the radiol- ischemic complication rates in that group.
ogy department using a trans posterior approach
with a 22-gauge, 15 cm spinal needle. The needle
was inserted anterior to the aorta under CT guid- Discussion/Conclusion
ance and a similar volume of bupivacaine and
triamcinolone was injected. Abdominal pain is the most frequent presenting
In the group receiving EUS-guided CPB, 50% symptom among patients with chronic pancreati-
(5/10) of patients experienced decreased pain with tis and/or pancreatic malignancy, often leading to
a mean post procedure follow-up of 15 weeks. a significant impairment of quality of life.
In the group who received CT-guided CPB, Standard management of these patients involves
25% (2/8) experienced reduction in pain score the use of opioid analgesic medications, though
with a mean post procedure follow-up of 4 weeks. its long-term efficacy is limited. CPN (ethanol)
Only 12% patients had some pain relief at and CPB (steroid) serve as an alternative inter-
12 weeks follow-up. In summary, 75% (6/8) vention for pain management in pancreatic can-
returned to baseline or pretreatment pain score cer and chronic pancreatitis patients respectively,
within 6 weeks after the CT block. by disrupting nociceptive impulses at the level of
the celiac plexus. CPN has been carried out under
radiographic, fluoroscopic, CT, or EUS guidance.
Complications There is increasing evidence that EUS-guided
CPN and CPB are more effective than other
EUS-CPN related common complications are methods in providing persistent pain relief (for
due to the unopposed parasympathetic activity cancer and CP patients, respectively), with a sim-
that develops as a result of the sympathetic block- ilar safety profile to the other procedures (namely,
age of the celiac plexus. Transient diarrhea, pain CT-guided CPB). Durability of effective pain
exacerbation, hypotension, and inebriation are relief has been demonstrated for up to a year fol-
72 T. H. Kothari et al.

lowing EUS-guided CPB, although typically 10. Wiersema MJ, Wiersema LM. Endosonography-­
guided celiac plexus neurolysis. Gastrointest Endosc.
lasts only 2–3 months [11]. The majority of 1996;44:655–62.
reported complications are mild (diarrhea, pos- 11. Gress F, Schmitt C, Sherman S, Ciaccia D, Ikenberry
tural hypotension, and pain exacerbation), S, Lehman G. Endoscopic ultrasound-guided coeliac
although severe complications can occur rarely. A plexus block for managing abdominal pain associated
with chronic pancreatitis: a prospective single-center
preferred injection technique (single vs. bilateral) experience. Am J Gastroenterol. 2001;96:409–16.
is still widely debated, as some studies have 12. Wyse JM, Battat R, Sun S, et al. Practice guidelines
suggested higher pain relief for patients receiving for endoscopic ultrasound-guided celiac plexus
a bilateral injection, while others indicate there is neurolysis. Endosc Ultrasound. 2017;6(6):369–75.
13. Erdine S. Celiac ganglion block. Agri. 2005;17(1):
no difference in the outcomes between the two 14–22.
techniques. In conclusion, EUS-guided CPN and 14. Yasuda I, Wang HP. Endoscopic ultrasound-guided
CPB offer a safe and effective technique to control celiac plexus block and neurolysis. Dig Endosc.
abdominal pain in patients with pancreatic/upper 2017;29(4):455–62.
15. Sahai AV, Lemelin V, Lam E, et al. Central vs. bilateral
abdominal malignancy and chronic pancreatitis, endoscopic ultrasound-guided celiac plexus block or
respectively. neurolysis: a comparative study of short-term effec-
tiveness. Am J Gastroenterol. 2009;104:326–9.
16. LeBlanc JK, Al-Haddad M, McHenry L, et al. A
prospective, randomized study of EUS-guided celiac
References plexus neurolysis for pancreatic cancer: one injection
or two? Gastrointest Endosc. 2011;74:1300–7.
1. de Oliveira R, dos Reis MP, Prado WA. The effects of 17. Sakamoto H, Kitano M, Kamata K, et al. EUS-guided
early or late neurolytic sympathetic plexus block on broad plexus neurolysis over the superior mesenteric
the management of abdominal or pelvic cancer pain. artery using a 25-gauge needle. Am J Gastroenterol.
Pain. 2004;110(1-2):400–8. 2010;105:2599–606.
2. Staats PS, Hekmat H, Sauter P, Lillemoe K. The 18. Levy M, Rajan E, Keeney G, et al. Neural ganglia visu-
effects of alcohol celiac plexus block, pain, and mood alized by endoscopic ultrasound. Am J Gastroenterol.
on longevity in patients with unresectable pancre- 2006;101:1787–91.
atic cancer: a double blind, randomized, placebo-­ 19. Levy MJ, Topazian MD, Wiersema MJ, et al. Initial
controlled study. Pain Med. 2001;2(1):28–34. evaluation of the efficacy and safety of endoscopic
3. Wong GY, Schroeder DR, Carns PE, et al. Effect of ultrasound-guided direct ganglia neurolysis and
neurolytic celiac plexus block on pain relief, quality block. Am J Gastroenterol. 2008;103:98–103.
of life, and survival in patients with unresectable 20. Doi S, Yasuda I, Kawakami H, et al. Endoscopic
pancreatic cancer: a randomized controlled trial. ultrasound-guided celiac ganglia neurolysis vs. celiac
JAMA. 2004;291(9):1092–9. plexus neurolysis: a randomized multicenter trial.
4. De Cicco M, Matovic M, Balestreri L, Fracasso A, Endoscopy. 2013;45:362–9.
Morassut S, Testa V. Single-needle celiac plexus 21. Gress F, Schmitt C, Sherman S, Ikenberry S, Lehman
block: is needle tip position critical in patients with G. A prospective randomized comparison of endo-
no regional anatomic distortions? Anesthesiology. scopic ultrasound- and computed tomography-guided
1997;87(6):1301–8. celiac plexus block for managing chronic pancreatitis
5. Loukas M, Klaassen Z, Merbs W, Tubbs RS, Gielecki pain. Am J Gastroenterol. 1999;94(4):900–5.
J, Zurada A. A review of the thoracic splanchnic nerves 22. Alvarez-Sánchez MV, Jenssen C, Faiss S, et al.
and celiac ganglia. Clin Anat. 2010;23(5):512–22. Interventional endoscopic ultrasonography: an over-
6. Mercadante S, Nicosia F. Celiac plexus block: a reap- view of safety and complications. Surg Endosc. 2014;
praisal. Reg Anesth Pain Med. 1998;23(1):37–48. 28:712–34.
7. Puli SR, Reddy JB, Bechtold ML, Antillon MR, 23. Fusaroli P, Jenssen C, Hocke M, Burmester E,
Brugge WR. EUS-guided celiac plexus neurolysis for Buscarini E, Havre RF, Ignee A, et al. EFSUMB
pain due to chronic pancreatitis or pancreatic cancer guidelines on interventional ultrasound (INVUS),
pain: a meta-analysis and systematic review. Dig Dis part V—EUS-guided therapeutic interventions (short
Sci. 2009;54(11):2330–7. version). Ultraschall Med. 2016;37:412–20.
8. Kappis M. Erfahrungen mit Lokalanasthesie bei 24. Gress F, Ciaccia D, Kiel J, Sherman S, Lehman
Bauchoperationen. Verh Dtsch Gesellsch Chir. 1914; G. Endoscopic ultrasound (EUS) guided celiac plexus
43:87–9. block (CB) for management of pain due to chronic
9. Faigel DO, Veloso KM, Long WB, et al. Endosonography pancreatitis (CP): a large single center experience.
guided celiac plexus injection for abdominal pain due to Gastrointest Endosc. 1997;45:AB173.
chronic pancreatitis. Am J Gastroenterol. 1996;91:1675.
 EUS-Guided Core Biopsy
8
Ali Siddiqui

3. Enhanced ability to perform immunostaining

Introduction or advanced molecular diagnostic testing,
thus allowing for targeted therapies for indi-
While endoscopic ultrasound-guided fine needle vidualized treatment of patients with malig-
aspiration (EUS-FNA) has been for many years nancies [7–9].
the procedure of choice to obtain samples from 4. Lack of clear need for on-site cytology if core
lesions of the gastrointestinal (GI) tract and of tissue is simply placed into formalin.
adjacent organs, its sensitivity is highly influ-
enced upon the availability of rapid on-site cyto-
pathology, which significantly influences the Fine Needle Biopsy Needles
overall diagnostic accuracy [1–4]. However,
access to on-site cytopathology and an expert The first needle used to perform EUS-guided biop-
cytopathologist is limited, thus [5] creating a bar- sies was the Quick-Core (Cook Medical Inc.,
rier to the widespread use of EUS-FNA [6]. Bloomington, in, United States), a 19-gauge Tru-­
The ability of a tissue biopsy specimen for Cut biopsy needle (TCB) with a spring-loaded fir-
histologic examination by EUS-guided fine nee- ing mechanism that collected an 18-mm tissue
dle biopsy (EUS-FNB) in many ways overcomes specimen for histologic analysis [10, 11].
the limitations of EUS-FNA. The primary advan- Unfortunately, performance of EUS-TCB was dis-
tages of a histological core specimen are: appointing with a significant variation in diagnostic
accuracy (52–100%) [12, 13] and tissue yield (50–
1. Improved tissue architecture interpretation 100%) [14, 15]. There was also no obvious advan-
and improves diagnosis of both malignant and tage for EUS-TCB over EUS-FNA [16]. The
benign lesions. Tru-Cut needle was also challenging to operate and
2. Improved diagnostic accuracy. cumbersome to use, thus making its widespread
use limited. An additional drawback to this device
was difficulty in operation when the echoendo-
scope was in a flexed position, making biopsies of
Electronic supplementary material: The online version the pancreatic head and uncinated process difficult
of this chapter (https://doi.org/10.1007/978-3-319-97376-
0_8) contains supplementary material, which is available and, in some cases, unworkable.
to authorized users. The newer EchoTip Procore FNB (Cook
Medical) needles are available in a wide range
A. Siddiqui (*)
Department of Gastroenterology, Fish Memorial of sizes (19, 20, 22, 25-G) and have a unique
Hospital, Orange City, FL, USA lateral opening in the distal needle shaft, presenting

© Springer Nature Switzerland AG 2019 73

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_8
74 A. Siddiqui

suspicious for autoimmune pancreatitis (AIP),

adequate specimens for histologic analysis were
succesfully obtained in 93% of patients. However,
the diagnostic accuracy to diagnose AIP was only
43%; this was probably accounted for by the
patchy distribution of histologic changes of AIP,
making tissue obtained with EUS-FNB insuffi-
cient due to sample variability. There were no
patients in this subset in whom a malignant etiol-
ogy was not accurately excluded, making the
19-G needle 100% specific to exclude a cancer
[19]. Varadarajulu et al. [20] recently looked at
the new flexible 19-gauge needle (Expect™ 19
Fig. 8.1 ProCore needle tip (Permission for use granted
by Cook Medical, Bloomington, Indiana) Flex, Boston Scientific) which is made of nitinol,
giving it more flexibility for transduodenal punc-
ture. The procedure was successful in all 32
a reverse bevel to hook and cut a core tissue patients with pancreatic lesions approached from
(Fig. 8.1). Other FNB needles use the novel nee- the duodenum. Histologic core tissue to make an
dle tip design (Sharkcore; Medtronic, accurate diagnosis of cancer was obtained in 94%
Minneapolis, MN) with multiple parallel cutting patients.
surfaces at the tip and a Franseen needle tip
design (Acquire, Boston Scientific, Natick, MA) ProCore™ Needle
(Fig. 8.2). There are now multiple high quality clinical
studies that have evaluated the role of the
Procore needle to obtain EUS-FNB from solid
 US-Guided Fine Needle Biopsy
E pancreatic lesions. Iglesias-Garcia et al. used
of Solid Pancreatic Lesions the 19-gauge ProCore™ needle to perform
EUS-FNB of solid lesions in a large multicenter
Solid pancreatic lesions are the most common study, majority of which were pancreatic
target for EUS-FNB, so much research has been masses [21]. EUS-FNB using the ProCore™
performed to evaluate these devices in this con- needle was technically successful in 98% with
text (Fig. 8.3). no adverse events. Tissue deemed to be ade-
quate for histologic examination was obtained
 tandard 19 Gauge Needle Aspiration
S in 90% of cases with a diagnostic accuracy was
Devices 86% for all lesions and 93% for malignant
In 2005 Itoi et al. reported using the standard lesions. In another study evaluating the
19-gauge needle to obtain histologic biopsy spec- 22-gauge ProCore™ needle in 61 patients with
imens in patients with solid pancreatic lesions pancreatic masses, [22] adequate tissue for his-
[17]. They reported overall diagnostic accuracy tologic diagnosis (diagnostic accuracy) was
of only 69%. The reason for the low diagnostic obtained in 88.5%. Interestingly, a randomized
accuracy was because transduodenal biopsy of trial comparing a standard 22-gauge FNA nee-
pancreatic head and uncinate process masses was dle to the 22-gauge ProCore™ in pancreatic
unsuccessful as a result of stiffness of the 19-G mass patients [23] showed no significant differ-
needle. If the study from Itoi et al. [17] was ence in the median number of passes required
excluded, the overall technical success and diag- for diagnosis, rates of diagnostic accuracy, or
nostic yield to obtain histologic core tissue is technical failure. Histologic core to make a
above 90% in other studies [18, 19]. In a study diagnosis was present in 66.7% of FNA speci-
[18] looking at patients with a pancreatic mass mens and 80% of FNB specimens (P = 0.66).
8 EUS-Guided Core Biopsy 75

Fig. 8.2 (a) Handle and catheter of Acquire core biopsy needle. (b) Magnified image of an Acquire core biopsy needle
tip (Courtesy of Boston Scientific Corporation)

Fig. 8.3 (a) EUS image of a 22 gauge Acquire needle in the SharkCore needle, showing long tissue cores and
a pancreatic mass. (b) Histologic specimen of a core of malignant adenocarcinoma. (d) Photo of a tissue core
tissue in the patient from (a), showing adenocarcinoma. extruded onto a glass slide from a patient with a pancre-
(c) Histologic specimen of a core biopsy obtained using atic mass (Courtesy of Douglas G. Adler MD)
76 A. Siddiqui

[23] Iwashita et al. [24] evaluated the use of the Nayar et al. compared the diagnostic perfor-
25-gauge ProCore™ needle for EUS-FNB of mance and yield for tissue acquisition from solid
50 patients with solid pancreatic lesions. While pancreatic lesions of the ProCore and SharkCore
the sensitivity to obtain adequate tissue for needles. In this single-center study, the
cytologic diagnosis was high (96%), the pres- SharkCore™ afforded substantially superior tis-
ence of a histologic core was found in only 32% sue yield and diagnostic performance compared
of the patients. This study suggests that while with ProCore™ [28].
the 25-gauge ProCore™ needle is excellent to All these studies suggest that the SharkCore™
diagnostic cytologic specimen, its use to obtain needle allows for adequate specimen collection
a tissue core biopsy specimen to make the diag- in order to obtain a diagnosis in solid pancreatic
nosis may be limited. Overall results with the lesions. It is also useful to obtain a tissue core
ProCore needle are mixed with some studies biopsy specimen to make the diagnosis in
showing very positive results and others with 50–90% of cases.
more mixed results.
Acquire™ Needle
 harkCore™ Needle (Video 8.1)
S Mitri et al. assessed the safety, histological sam-
DiMaio et al. looked at the ability to obtain suf- ple procurement yield, and diagnostic accuracy
ficient tissue for pathologic evaluation by using of a newly available Acquire™ (Boston
the 22-G and 25-G SharkCore needle in 136 solid Scientific) histology needle for pancreatic
pancreatic lesions (Fig. 8.2) [25]. The diagnostic lesions [29] A mean of 2.8 needle passes per
yield to obtain adequate histological core to make lesion site were performed, without any major
a diagnosis was 85% when using the 25-G needle complication. A tissue core biopsy sample for
and 86% when using the 22-G needle. Adverse histological evaluation was obtained in 93%
events included post-procedure pain in 5 patients, cases. Considering malignant versus nonmalig-
mild acute pancreatitis in 4 patients, and fever/ nant disease, sensitivity and specificity were
cholangitis in one patient. 12 days after com- 98.2% and 100%, respectively. EUS-FNB using
bined EUS/ERCP a pancreatic head cancer. the 22-gauge Acquire™ needle was able to
Kandel et al. compared the histology yield of reach a very high procurement yield and diag-
EUS-FNB sampling using the SharkCore needle nostic accuracy.
(19, 22, or 25-G) to EUS-FNA in patients who had
solid pancreatic lesions. Ninety-five percent of the
specimens obtained from the SharkCore needle  US-Guided Fine Needle Biopsy
E
group were of sufficient size for histologic screen- of Gastrointestinal Subepithelial
ing, compared with only 59% from the EUS-FNA Tumors
group (P = 0.01). The median number of passes
required to achieve a sample was significantly lower After pancreatic masses, subepithelial lesions are
in the SharkCore needle group compared with the among the most common targets for EUS-guided
EUS-FNA group (2 passes vs 4 passes) [26]. core needle biopsy (Fig. 8.4).
Another comparative study evaluating the Lee et al. looked at the tissue acquisition and
SharkCore™ needles with the standard EUS-­ diagnostic yield of EUS-FNB for gastric subep-
FNA needles by Jovani et al. showed that more itheal tumors greater than 2 cm in size. They used
histological specimens were obtained with the the ProCore™ 22-gauge needle in this study and
SharkCore™ needles compared to standard FNA performed EUS-FNB in 78 patients. The authors
needles (59 versus 5%; P < 0.001). However, found that EUS-FNB was diagnostic in 82% of
overall diagnostic test characteristics were not patients, and tissue of histologic evaluation was
significantly different (diagnostic accuracy: 92.2 obtained in 97% of cases. An important observa-
versus 85.4% for SharkCore™ versus standard tion found was that FNB specimens permitted
needles) [27]. immunostaining for the diagnosis of gastrointes-
8 EUS-Guided Core Biopsy 77

Fig. 8.4 EUS image of

a 22 gauge Acquire
needle being used to
obtain a core of tissue
from a large
subepithelial lesion. The
pathology revealed a
gastrointestinal stromal
tumor (Courtesy of
Douglas G. Adler MD)

tinal stromal tumors 48%, a capability that is  US-Guided Fine Needle Biopsy
E
rarely possible when tissue is obtained using of Lymphadenopathy
standard EUS-FNA needles. There was only a
single case of self-limited post-procedural bleed- Adequate tissue acquisition from lymph nodes
ing [30]. using standard EUS-FNA needles can be chal-
El Chafic et al. perfomed a large retrospective lenging. Lymphoproliferative disorders often
study [31] evaluating patients suspected GI stro- require histologic specimens in order to obtain
mal tumors greater than 2 cm that underwent architecture and allow performance of flow
EUS-FNA (n = 91) or EUS-FNB using the cytometry. Although FNA specimens have a high
SharkCore needle (n = 15). The needle size at yield for metastatic lesions, FNA is not ideal for
was used most often was 22 gauge in both hematologic malignancy. Core biopsies of nodes
groups. Adequate tissue was procured, allowing are often preferable (Fig. 8.5). Two studies in
immunohistochemical staining in 65% patients which FNB with Tru-cut sampling was per-
in the FNA group and 100% patients in the formed for enlarged lymph nodes produced diag-
SharkCore group. A diagnosis was reached by nostic yields ranging from 69% to 73% [15, 32].
immunohistochemical staining in 53% patients There continues to be limited data on the role
in the FNA group and 87% patients in the of the newer EUS-FNB needles for the diagnosis
SharkCore group. Tissue was insufficient to of lymphadenopathy of unknown etiology. In a
make a cytologic diagnosis in 24% patients in randomized study comparing conventional nee-
the FNA group compared with none in the FNB dle fine needle aspiration to ProCore biopsy nee-
group. There were no reported immediate dle. In patients with mediastinal lymphadenopathy
adverse events or technical difficulties in either [33], the diagnostic sensitivity of aspirated mate-
group. The authors concluded that EUS-FNB by rial obtained using EUS-FNA needle and ProCore
using a SharkCore needle for suspected GI stro- needle were comparable (69% vs 79% respec-
mal tumors is technically similar and equally tively; P > 0.05). In another multicenter, random-
safe as FNA, with better tissue acquisition, ized trial, Nagula et al. [33] compared EUS-FNA
which was achieved with fewer needle passes and EUS-FNB for tissue sampling of 135 solid
and an improved diagnostic yield by immuno- lesions, 46 of which were enlarged lymph nodes
histochemical staining. adjacent to the GI tract. This study found that
78 A. Siddiqui

Fig. 8.5 EUS image of

a 22 gauge Procore
needle being used to
obtain a core of tissue
from a mediastinal
lymph node (Courtesy
of Douglas G. Adler
MD)

there was no difference between FNA and FNB alternative means for safe and accurate liver tissue
when stratified by the presence of on-site cytopa- acquisition for focal and parenchymal disease
thology or by type of lesion sampled. A median (Fig. 8.6). It should be noted that due to anatomy,
of 1 needle pass was needed to obtain a diagnos- the left lobe of the liver is easily accessible for
tic sample for both needles. FNA and FNB EUS-LB from the stomach, while the right lobe
obtained a similar diagnostic yield with a compa- needs to be assessed from the duodenum.
rable number of needle passes. This studied
argued against routine use of FNB for lymph
node biopsy [34]. EUS-Guided Tru-Cut Biopsy
There continues to be controversy about when
EUS-FNB is superior to EUS-FNA for the diag- EUS-guided Tru-Cut biopsy uses the spring-­
nosis of malignancy in patients with lymphade- loaded Quick-Core needle (Cook Medical,
nopathy. In cases when a lymphoproliferative Bloomington, IN, USA) in order to obtain a tis-
disorder is suspected, this author would recom- sue sample (Quick-Core, Cook Medical,
mend EUS-FNB so as to allow acquisition of tis- Bloomington, IN, USA). This is a 19-gauge nee-
sue architecture and allow immunostaining to be dle capable of collecting an 18-mm tissue speci-
performed. men sufficient for histologic examination. Initial
trails using the Quick-Core needle demonstrated
its safety and efficiacy in acquiring liver tissue in
 US-Guided Fine Needle Biopsy
E a pig model [10] which then prompted its use in
of the Liver humans.
The initial human study evaluating the use of
Liver biopsy is not only used to determine the EUS-guided Tru-Cut biopsy in benign liver dis-
underlying etiology of liver disease but also to ease was performed by Dewitt et al where they
evaluate the extent of liver damage, both of which adequate liver tissue to make a histologic diagno-
are essential in determining how such patients are sis was obtained in 19/21 (90%) patients [35]. No
treated. adverse events occurred. While the specimen
There is now increasing data to demonstrate mean length was 9 mm, the size of the samples
that EUS-guided liver biopsy (EUS-LB) is an obtained was smaller than those usually consid-
8 EUS-Guided Core Biopsy 79

Fig. 8.6 (a) Diagram showing EUS-guided liver biopsy len/ballooned hepatocytes consistent with mild steatohep-
of the right lobe of the liver from the duodenal bulb. (b) atitis. There is also mild to moderate inflammatory cell
EUS image of a 22 gauge core biopsy needle in the left infiltrates (lymphocytes) in areas of scar. Portal tracts
lobe of liver after a transgastric passage (Courtesy of appear intact with surrounding inflammatory cells
Douglas G. Adler MD) (c) Liver transgastric biopsy on (Courtesy of Douglas G. Adler MD and Nicole Girard
H + E stain: High power view revealing occasional swol- MD)

ered adequate for histologic assessment. Gleeson EUS-LB and failure to obtain tissue on many
et al. utilized Tru-Cut EUS-LB to evaluate the occasions [35–37]. Hence, the Tru-Cut never
number of liver portal triads obtained in 9 patients reached widespread acceptance and use among
undergoing liver ibopsy [35]. This study obtained endosonographers, leading to the use of alterna-
adequate diagnostic tissue in all nine cases, with tive needle types to obtain EUS-LB.
a total of 63 portal triads.
While initial studies with Tru-Cut needle
appeared promising, this needle is technically  US-Guided Fine Needle Biopsy
E
more challenging to use compared to conven- with a 19-Gauge Needle to Obtain
tional EUS-FNA. This is especially true when Histological Tissue in Patients
EUS-LB is performed with the echoendoscope in with Benign Liver Disease
a long position, i.e. in the duodenum. These rea-
sons accounted for the significant variability seen The first study to evaluate the use of EUS-LB
in studies evaluating the Tru-Cut technique for using the standard 19-gauge FNA needle was
80 A. Siddiqui

performed by Stavropoulos et al. [38] The 22 biopsy [43]. Adequate tissue to obtain a histo-
patients in this study underwent EUS for elevated logical diagnosis of malignancy was obtained in
liver function tests of unknown etiology, and 19 patients (91%). The overall diagnostic accu-
EUS-LB of the left liver was done when no evi- racy for malignancy and specific tumor type were
dence of biliary obstruction was seen. Results of 90.5% and 85.7%, respectively. No complica-
EUS-LB using the 19-gauge FNA were the fol- tions were seen when the EUS core biopsy needle
lowing: median specimen length = 37 mm, nine was used. The authors concluded that EUS-FNB
complete portal triads, and diagnostic ade- with core biopsy needle for solid liver masses
quacy = 91%. No procedure related adverse may be helpful in the management of patients
events were seen. who are unable to be diagnosed using percutane-
In a large multi-center study, Diehl et al evalu- ous liver biopsy.
ated 110 patients with elevated liver enzymes EUS-LBs allows an effective and targeted
who underwent EUS-LB [39]. These investiga- approach for liver biospy, particularly for focal
tors used suction on the needle in most cases, and lesions. Use of the 19-G standard needle or the
then performed up to 10 to-and-fro needle move- newer EUS core-biopsy needs may also provide a
ments per pass to obtain adequate tissue. The higher yield as compared to the standard EUS-­
diagnostic yield to obtain adequate tissue for FNA needles. Advantages of EUS-LB include
diagnosis was 89% with a median core length of performing bilobar liver biopsy to increase diag-
38 mm a median of 14 complete portal triads. In nostic accuracy in parenchymal disease and the
this study, there was one bleeding adverse event ability to accurately target and biopsy focal liver
that led to a subcapsular hematoma; this patient masses.
was treated conservatively and did well.
The above studies confirmed that EUS-LB
using a 19-gauge FNA needle is safe, effective, Conclusion
and allows for a high diagnostic yield and speci-
men adequacy. The use of the 19-gauge FNA Overall, the ability of EUS to obtain core tissue
needle has been demonstrated to be easier to use specimens from primary tumors, lymph nodes,
and possibly yield better liver core tissue com- the liver, and metastases makes these devices
pared to the Tru-Cut needle. invaluable in the era of modern, interventional
EUS. Ongoing studies will further clarify ideal
needle types and sizes for different indications
 US-FNB for Malignant Liver
E and target locations.
Lesions

EUS-LB allows us to obtain a histological biopsy References

of malignant liver lesion(s) seen on cross-­
sectional imaging modalities (with the exception 1. Iglesias-Garcia J, Dominguez-Munoz JE, Abdulkader
I, et al. Influence of on-site cytopathology evalu-
of the right posterior segments), thus allowing ation on the diagnostic accuracy of endoscopic
confirmation of suspected malignant or meta- ultrasound-­guided fine needle aspiration (EUS-FNA)
static lesions [40]. While several large trials have of solid pancreatic masses. Am J Gastroenterol. 2011;
confirmed the efficacy and safety of using EUS-­ 106:1705–10.
2. Alsohaibani F, Girgis S, Sandha GS. Does onsite
FNA cytological aspiration to diagnose hepato- cytotechnology evaluation improve the accuracy of
cellular carcinoma and liver metastasis, [41, 42], endoscopic ultrasound-guided fine-needle aspiration
there is currently very limited data on the use of biopsy? Can J Gastroenterol. 2009;23:26–30.
EUS-LB in such lesions. 3. Hébert-Magee S, Bae S, Varadarajulu S, et al. The
presence of a cytopathologist increases the diagnostic
Lee et al. evaluated 21 patients that underwent accuracy of endoscopic ultrasound-guided fine needle
EUS-LB using a EUS 22-Gauge core biopsy nee- aspiration cytology for pancreatic adenocarcinoma: a
dle in patients who failed percutaneous liver meta-analysis. Cytopathology. 2013;24:159–71.
8 EUS-Guided Core Biopsy 81

4. Eloubeidi MA, Tamhane A, Jhala N, et al. Agreement 19. Iwashita T, Yasuda I, Doi S, et al. Use of samples from
between rapid onsite and final cytologic interpreta- endoscopic ultrasound-guided 19-gauge fine-needle
tions of EUS-guided FNA specimens: implications aspiration in diagnosis of autoimmune pancreatitis.
for the endosonographer and patient management. Clin Gastroenterol Hepatol. 2012;10:316–22.
Am J Gastroenterol. 2006;101:2841–7. 20. Varadarajulu S, Bang JY, Hebert-Magee S. Assessment
5. Jhala NC, Jhala DN, Chhieng DC, et al. Endoscopic of the technical performance of the flexible 19-gauge
ultrasound-guided fine-needle aspiration. A cyto- EUS-FNA needle. Gastrointest Endosc. 2012;76:
pathologist’s perspective. Am J Clin Pathol. 2003; 336–43.
120:351–67. 21. Iglesias-Garcia J, Poley JW, Larghi A, et al. Feasibility
6. Kalaitzakis E, Panos M, Sadik R, et al. Clinicians’ and yield of a new EUS histology needle: results
attitudes towards endoscopic ultrasound: a survey from a multicenter, pooled, cohort study. Gastrointest
of four European countries. Scand J Gastroenterol. Endosc. 2011;73:1189–96.
2009;44:100–7. 22. Larghi A, Iglesias-Garcia J, Poley JW, et al. Feasibility
7. Braat H, Bruno M, Kuipers EJ, et al. Pancreatic can- and yield of a novel 22-gauge histology EUS needle
cer: promise for personalised medicine? Cancer Lett. in patients with pancreatic masses: a multicenter pro-
2012;318:1–8. spective cohort study. Surg Endosc. 2013;27:3733–8.
8. Wakatsuki T, Irisawa A, Terashima M, et al. ATP 23. Bang JY, Hebert-Magee S, Trevino J, et al.
assay-guided chemosensitivity testing for gemcitabine Randomized trial comparing the 22-gauge aspiration
with biopsy specimens obtained from unresectable and 22-gauge biopsy needles for EUS-guided sam-
pancreatic cancer using endoscopic ultrasonography-­ pling of solid pancreatic mass lesions. Gastrointest
guided fine-needle aspiration. Int J Clin Oncol. Endosc. 2012;76:321–7.
2011;16:387–94. 24. Iwashita T, Nakai Y, Samarasena JB, et al. High single-­
9. Brais RJ, Davies SE, O’Donovan M, et al. Direct his- pass diagnostic yield of a new 25-gauge core biopsy
tological processing of EUS biopsies enables rapid needle for EUS-guided FNA biopsy in solid pancre-
molecular biomarker analysis for interventional pan- atic lesions. Gastrointest Endosc. 2013;77:909–15.
creatic cancer trials. Pancreatology. 2012;12:8–15. 25. DiMaio CJ, Kolb JM, Benias PC, et al. Initial
10. Wiersema MJ, Levy MJ, Harewood GC, et al. Initial experience with a novel EUS-guided core biopsy
experience with EUS-guided Trucut needle biop- needle (SharkCore): results of a large North
sies of perigastric organs. Gastrointest Endosc. American multicenter study. Endosc Int Open.
2002;56:275–8. 2016;4(9):E974–9.
11. Levy MJ, Jondal ML, Clain J, et al. Preliminary 26. Kandel P, Tranesh G, Nassar A, et al. EUS-guided
experience with an EUS-guided Trucut biopsy nee- fine needle biopsy sampling using a novel fork-tip
dle compared with EUS-guided FNA. Gastrointest needle: a case-control study. Gastrointest Endosc.
Endosc. 2003;57:101–6. 2016;84(6):1034–9.
12. Shah SM, Ribeiro A, Levi J, et al. EUS-guided fine 27. Jovani M, Abidi WM, Lee LS. Novel fork-tip nee-
needle aspiration with and without Trucut biopsy of dles versus standard needles for EUS-guided tis-
pancreatic masses. JOP. 2008;9:422–30. sue acquisition from solid masses of the upper GI
13. Wahnschaffe U, Ullrich R, Mayerle J, et al. EUS-­ tract: a matched cohort study. Scand J Gastroenterol.
guided Trucut needle biopsies as first-line diagnostic 2017;52(6-7):784–7.
method for patients with intestinal or extraintestinal 28. Nayar MK, Paranandi B, Dawwas MF, et al.
mass lesions. Surg Endosc. 2009;23:2351–5. Comparison of the diagnostic performance of 2 core
14. Sakamoto H, Kitano M, Komaki T, et al. Prospective biopsy needles for EUS-guided tissue acquisition
comparative study of the EUS guided 25-gauge FNA from solid pancreatic lesions. Gastrointest Endosc.
needle with the 19-gauge Trucut needle and 22-gauge 2017;85(5):1017–24.
FNA needle in patients with solid pancreatic masses. 29. Mitri RD, Rimbaş M, Attili F, et al. Performance of
J Gastroenterol Hepatol. 2009;24:384–90. a new needle for endoscopic ultrasound-guided fine-­
15. Ribeiro A, Pereira D, Escalón MP, et al. EUS-guided needle biopsy in patients with pancreatic solid lesions:
biopsy for the diagnosis and classification of lym- a retrospective multicenter study. Endosc Ultrasound.
phoma. Gastrointest Endosc. 2010;71:851–5. 2017. https://doi.org/10.4103/eus.eus_33_17.
16. Levy MJ, Wiersema MJ. EUS-guided Trucut biopsy. 30. Lee JH, Cho CJ, Park YS, et al. EUS-guided 22-gauge
Gastrointest Endosc. 2005;62:417–26. fine needle biopsy for the diagnosis of gastric subepi-
17. Itoi T, Itokawa F, Sofuni A, et al. Puncture of solid thelial tumors larger than 2 cm. Scand J Gastroenterol.
pancreatic tumors guided by endoscopic ultraso- 2016;51(4):486–93.
nography: a pilot study series comparing Trucut 31. El Chafic AH, Loren D, Siddiqui A, et al. Comparison
and 19-gauge and 22-gauge aspiration needles. of FNA and fine-needle biopsy for EUS-guided sam-
Endoscopy. 2005;37:362–6. pling of suspected GI stromal tumors. Gastrointest
18. Larghi A, Capurso G, Carnuccio A, et al. Ki-67 grad- Endosc. 2017;86(3):510–5.
ing of nonfunctioning pancreatic neuroendocrine 32. Storch I, Shah M, Thurer R, et al. Endoscopic
tumors on histologic samples obtained by EUS-­ ultrasound-guided fine needle aspiration and Trucut
guided fine-needle tissue acquisition: a prospective biopsy in thoracic lesions: when tissue is the issue.
study. Gastrointest Endosc. 2012;76:570–7. Surg Endosc. 2008;22:86–90.
82 A. Siddiqui

33. Pathrabe R, Udawat H, Dave D, et al. Randomized to exclude biliary obstruction. Gastrointest Endosc.
study comparing conventional needle fine needle aspi- 2012;75:310–8.
ration to procore biopsy needle in patients with medi- 39. Diehl DL, Johal AS, Khara HS, et al. Endoscopic
astinal lymphadenopathy and masses of unknown ultrasound-guided liver biopsy: a multicenter experi-
origin. Endosc Ultrasound. 2017;6(Suppl 1):S10. ence. Endosc Int Open. 2015;3:e1–6.
34. Nagula S, Pourmand K, Aslanian H, et al. Comparison 40. Nguyen P, Feng JC, Chang KJ. Endoscopic ultrasound
of endoscopic ultrasound-fine-needle aspiration (EUS) and EUS-guided fine-needle aspiration (FNA)
and endoscopic ultrasound-fine-needle biopsy for of liver lesions. Gastrointest Endosc. 1999;50:357–61.
solid lesions in a multicenter, randomized trial. Clin 41. tenBerge J, Hoffman BJ, Hawes RH, et al. EUS-­
Gastroenterol Hepatol. 2017:16(8):1307–13 guided fine needle aspiration of the liver: indications,
35. Dewitt J, McGreevy K, Cummings O, et al. Initial yield, and safety based on an international survey of
experience with EUS-guided Tru-cut biopsy of benign 167 cases. Gastrointest Endosc. 2002;55:859–62.
liver disease. Gastrointest Endosc. 2009;69:535–42. 42. Singh P, Erickson RA, Mukhopadhyay P, et al.
36. Gleeson FC, Clayton AC, Zhang L, et al. Adequacy of EUS for detection of the hepatocellular carcinoma:
endoscopic ultrasound core needle biopsy specimen results of a prospective study. Gastrointest Endosc.
of nonmalignant hepatic parenchymal disease. Clin 2007;66:265–73.
Gastroenterol Hepatol. 2008;6:1437–40. 43. Lee YN, Moon JH, Kim HK, et al. Usefulness of
37. Mathew A. EUS-guided routine liver biopsy in endoscopic ultrasound-guided sampling using core
selected patients. Am J Gastroenterol. 2007;102: biopsy needle as a percutaneous biopsy rescue for
2354–5. diagnosis of solid liver mass: combined histological-­
38. Stavropoulos SN, Im GY, Jlayer Z, et al. High cytological analysis. J Gastroenterol Hepatol. 2015;
yield of same-session EUS-guided liver biopsy by 30(7):1161–6.
19-gauge FNA needle in patients undergoing EUS
 Endoscopic Ultrasound-Guided
Liver Biopsy 9
David L. Diehl

There remains an important role for liver biopsy Other Methods for Liver Biopsy
in the current management of liver diseases
despite advancements in noninvasive hepatic Development of transjugular access to the hepatic
assessment [1, 2]. Historically, large gauge (typi- venous system and liver led to a safer option for
cally 15–16 gauge, G) biopsy needles were used liver biopsy in patients with coagulopathy or
to obtain a biopsy after localization of a target ascites [4–6]. Using the same approach, it became
site by percussion of the liver span [3]. Risk of possible to measure portal pressures (“portal
inadvertent puncture of the pleural space or gall- package”), and rapidly led to the development of
bladder led to increasing use of transcutaneous transjugular intrahepatic portosystemic shunt
ultrasound-guided biopsy site selection. Because (TIPS) for the management of the complications
on-site ultrasound machines may not be widely of portal hypertension [7].
available in the endoscopy unit or GI/Hepatology The development of endoscopic ultrasound
Clinic, much of the liver biopsy case load was quickly led to refinement of the technology to
moved to the general or interventional radiology allow real-time fine needle biopsy of various
department. With decreasing case volumes, most lesions around the esophagus, stomach, and duo-
hepatologists and gastroenterologists got out of denum. Fine needle aspiration of focal liver
the business of liver biopsy. GI fellowships also lesions was found to be safe [8, 9] but the use of
dropped the requirement of training in percutane- EUS to obtain core biopsy of liver parenchyma
ous liver biopsy, leading to even fewer non-­ occurred later, with the adaptation of a Tru-Cut
radiologists doing this procedure. needle (Fig. 9.1) that could be used through the
echoendoscope (QuickCore, Cook Medical,
Winston Salem, NC). The first report of use of
Electronic supplementary material: The online version the EUS-guided Tru-Cut needle for parenchymal
of this chapter (https://doi.org/10.1007/978-3-319-97376- liver biopsy was published in 2007 [10]. Several
0_9) contains supplementary material, which is available case series were subsequently reported with this
to authorized users.
needle [11–13]. However, the device was some-
D. L. Diehl (*) what technically difficult to use, and did not reli-
Geisinger Commonwealth School of Medicine,
Scranton, PA, USA ably deliver liver core biopsy samples. As a
result, this needle never saw widespread adop-
Department of Gastroenterology and Nutrition,
Geisinger Medical Center, Danville, PA, USA tion, and the use of this device was essentially
e-mail: [email protected] abandoned.

© Springer Nature Switzerland AG 2019 83

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_9
84 D. L. Diehl

specimen fragmentation and could adversely

affect measurements. For total aggregate speci-
men lengths, lengths of longest piece, and portal
counts, it was found that specimen yields were
comparable between the three methods, and in
fact trended to higher for EUS-LB due to ease of
performing multiple needle passes (Fig. 9.2).

Indications and Contraindications

Indications for EUS-LB are broadly any patient

who needs a liver biopsy and does not require a
transjugular approach, most commonly in patients
with abnormal liver function tests of unclear etiol-
ogy. If the patient requires an endoscopy (e.g.,
rule out varices, Barrett’s surveillance or detec-
tion, and evaluation of upper GI symptoms) or an
EUS (evaluation of the common bile duct, gall-
bladder, pancreas, or other EUS indication) in
addition to a liver biopsy, then EUS-LB is the
Fig. 9.1 This close-up view of the Tru-Cut needle shows most efficient means to accomplish the testing
a “tissue tray” (left); a cutting sheath cuts a core of tissue that is required. It has previously been shown that
upon deployment (right) (Image used with the permission
of Allan Darr, ProAct Ltd.) endoscopy and percutaneous liver biopsy done on
the same day is safe [17]. EUS-LB is safer for
liver sampling than percutaneous biopsy due to
The first published experience of use of a reg- continuous real-time imaging of the needle. Some
ular EUS-FNA needle for liver biopsy was patients require a sedated liver biopsy, for exam-
described by Stavropoulos et al. [14]. In this sem- ple, the patients with a great deal of anxiety
inal prospective case series, 22 patients under- regarding the procedure, or for some children
went liver biopsy of the left hepatic lobe through undergoing liver biopsy [18], and sedated
a transgastric route with a regular 19G EUS nee- EUS-LB an excellent approach in these cases.
dle (EchoTip, Cook Medical, Bloomington, IN). Situations where EUS-guided liver biopsy
Adequate tissue yield was obtained in 20/22, may not be ideal include significant coagulopa-
with median specimen length of 36.9 mm (range thy and/or the use of anticoagulants, while ascites
2–185 mm) and median portal triad counts of 9 is a relative contraindication based on its location
(range 1–73 CPTs). Extending this pilot data, [1, 3]. The transjugular approach is preferred in
Diehl et al. conducted a multicenter study of these situations.
EUS-LB in 110 patients at 8 centers. Tissue
yields were excellent, with adequate specimen
lengths and portal counts [15]. EUS-LB Technique (Video 9.1)
Comparison of tissue yields between percuta-
neous, transjugular, and EUS-guided routes was Identification of Liver Lobes
carried out by Pineda et al. [16]. In this study,
19G EUS-FNA needles were used by the endo- EUS-LB is performed with the curvilinear echo-
sonographer, and 18G or 20G needles used by endoscope, which allows real-time monitoring
interventional radiologist. Only non-cirrhotic of needle entry into the liver. A distinct advan-
patients were included from the transjugular tage of EUS-LB is the ability to target spatially
liver biopsy group, as cirrhosis leads to more distinct parts of the liver (left and right lobes)
9 Endoscopic Ultrasound-Guided Liver Biopsy 85

Fig. 9.2 When compared to percutaneous (left bars) and transjugular (right bars) liver biopsy, bilobar EUS-LB (center
bar) gives comparable or superior samples in terms of portal triad count and total specimen length

Fig. 9.3 Visualization

of the left hepatic lobe
from the proximal
stomach; the cursor
shows the expected
trajectory of the biopsy
needle when obtaining
the core specimen

[19]. The left lobe is found by identifying the the liver or tracing hepatic veins to the IVC. This
liver from the proximal stomach (Fig. 9.3). It is confirmation will allow avoidance of inadver-
important to positively identify liver and distin- tent splenic puncture.
guish it from the spleen, which is found in a The right hepatic lobe is found by placing the
similar location and may be enlarged in patients tip of the EUS scope in the duodenal bulb and
with portal hypertension. In some cases, the torqueing until the large mass of the right lobe is
echotexture of the liver can be very similar to identified (Fig. 9.4). Gallbladder (if present) may
the spleen (Video 9.1) which can lead to confu- be seen from this duodenal position.
sion. The two organs can be distinguished by The presence of larger vessels in the hepatic
either identification of portal vein branches in parenchyma is expected, and as a rule, direct
86 D. L. Diehl

Ex vivo liver biopsy studies have been done to

try to determine the optimal needle gauge and tip
characteristics to optimize tissue yield. In one
study, a core biopsy needle (SharkCore 19G,
Medtronic) was found to give the highest tissue
yields in terms of increased mean portal tracts
[22]. Preliminary experience with a 19G core nee-
dle (Acquire 19G, Boston Scientific) has sug-
gested that specimen lengths are longer with the
core type needle than regular beveled needle [23].
We have completed a prospective randomized
study of the standard 19G FNA needle that we
have used for EUS-LB (19G Expect Flexible,
Boston Scientific) to the 19G Acquire core biopsy
needle. In a pilot group of patients randomized to
Fig. 9.4 Right lobe biopsies are obtained with the echo- either needle, there were increased tissue yields
endoscope tip placed in the duodenal bulb (Image adapted and higher portal tract counts in the core needle
from Boston Scientific) grout [24]. Importantly, in this group there was
also an increased “length of longest piece,” a criti-
puncture of these vessels should be avoided. A cal metric for assessing quality of the liver biopsy.
reasonable trajectory length without intervening A natural question is whether a 22G core nee-
vessels is identified for needle puncture. This is dle could be used to obtain adequate liver biopsy
typically approximately 2.5–3 cm, although a cores. We recently conducted a prospective ran-
slightly longer trajectory can often be found. domized study of the 22G SharkCore FNB nee-
One of the distinct advantages of EUS-LB dle for EUS-LB compared to a standard 19G
compared to other liver biopsy techniques is the needle [25]. We found that adequate liver biopsy
ability to sample spatially distinct areas of liver. specimens were obtained in 90% of cases with
Some parenchymal liver diseases may have dif- the 19G needle compared to only 60% with the
ferent lobar distributions [20], and bilobar biopsy 22G SharkCore needle. This appeared to be due
can minimize sampling error. Bilobar biopsy, to increased tissue fragmentation during histo-
with 1 or 2 passes per lobe, is very safe, and does logic processing, even though before processing,
not lead to increased risk of complications. liver cores of reasonable length were obtained
(Fig. 9.6). We concluded that the 19G FNA nee-
dle was more reliable for EUS-LB. With the
Needle Selection promising results from use of 19G core needles,
excellent safety profile, and higher likelihood of
Most of the studies on EUS-LB have utilized delivering an adequate specimen, these needles
19G needles. A 19G spring-loaded EUS Tru-Cut are likely to become the standard needle for
needle (QuickCore, Cook) was previously used, EUS-LB.
but disappointing tissue yields and cumbersome
operation led to this needle falling out of favor
[21]. The seminal study of Stavropoulos [14]  eedle Preparation and Amount
N
proved that a regular 19G needle is adequate in of Suction
delivering liver cores. More recently, “core nee-
dles” with special cutting tips have become avail- Full or “slow-pull” suction (wherein the stylet is
able and appear to be preferred for EUS-LB slowly removed during needle actuations in the
(Fig. 9.5a, b). liver) can be used for EUS-LB. Most published
9 Endoscopic Ultrasound-Guided Liver Biopsy 87

Fig. 9.5 Close-up of the tips of the core needles used for EUS-LB: (a) Acquire needle, Boston Scientific (used with the
permission from Boston Scientific), and (b) SharkCore needle, Medtronic (used with the permission from Medtronic)

saline [27, 28]. In the past 2 years, we have

primed the needle with a heparin flush instead of
saline. This had led to good biopsy yields and no
ill effect on liver histology [29]. The heparin
leads to less clotting of blood in the needle, which
can lead to a less bloody tissue specimen. This in
turn leads to easier tissue handling by the pathol-
ogy laboratory. In addition, stylet reinsertion is
considerably easier for the endoscopy assistant.

Needle Technique

In the chosen lobe, a trajectory for needle pas-

sage is identified that does not include larger
hepatic or portal vein branches. The longest tra-
jectory of needle travel is sought, which is typi-
Fig. 9.6 Prefixation tissue cores obtained with a 22G
SharkCore needle cally about 3–4 cm (Fig. 9.7). We have found
advantage to having the endoscopy assistant hold
the echoendoscope at the level of the bite block,
cases utilize the high suction achieved with the which prevents “recoiling” of the echoendoscope
vacuum syringe included with the EUS needles backward during the needle throw, while still
(20 cc VacLok, Merit EndoTek, Salt Lake City, allowing torqueing of the scope to maintain
UT). For slow-pull technique, the stylet is pulled ­needle visualization. After a suitable trajectory is
slowly as the needle is advanced into the target defined, the luminal wall is punctured, and the
lesion or organ. This has been demonstrated to needle positioned in the liver parenchyma.
deliver about 2–3 mL of vacuum [26]. At this point, the stopcock on the suction
“Dry suction” and “wet suction” have both syringe is turned to “on.” To-and-fro actuations of
been used for EUS-LB. For wet suction, the sty- the needle are made into the liver, while “fanning”
let is removed and the needle lumen primed with the needle trajectory in the same manner as is done
88 D. L. Diehl

Fig. 9.7 EUS imaging

of the right hepatic lobe
with a long (46.9 mm)
needle trajectory

for FNA of masses. The fanning is accomplished

with both use of the echoendoscope elevator and
the up-down wheel of the scope handle. The num-
ber of passes can be 1–10 depending on endo-
sonographer preference; we are currently using
1–3 passes. The stopcock on the vacuum syringe is
turned off prior to removing the needle from the
parenchyma. The needle is then removed from the
echoendoscope. A variation is the use of wet suc-
tion while making a single long pass into the liver
[30]. If the slow-pull technique is being used, the
assistant pulls back on the stylet as the endosonog-
rapher makes a needle pass into the liver.

 pecimen Handling in the Endoscopy

S
Suite and the Pathology Laboratory

The contents of the needle are expressed directly Fig. 9.8 Liver tissue is captured on a microsieve and
into a formalin cup by either stylet reinsertion, or blood is washed away
flushing the contents with saline or the heparin
flush. Most if not all the specimen will be in the mation of blood clots in the needle, which can be
needle lumen, although if blood has entered the visualized as “blood noodles” in the formalin.
vacuum syringe, tissue can be found there as When using the tissue filter, the needle contents
well. It is important to avoid excessive handling are expressed first onto the sieve. Blood generally
of the specimen, including expressing the tissue does not clot with the heparin priming, and the
onto gauze or a telfa pad. specimen is washed off using a light rinse with
For the last 2 years, we have utilized a “tissue saline, which leaves only (or mainly) liver tissue
sieve” to separate tissue from blood (Fig. 9.8). on the sieve. This tissue is then floated off into the
Heparinization of the needle tends to prevent for- formalin (Fig. 9.9).
9 Endoscopic Ultrasound-Guided Liver Biopsy 89

The liver biopsy literature cites a wide range

of “minimal number” of complete portal triads
that should be obtained, typically from 6 to 10.
There is no rigorously demonstrated number.
However, interpretation of liver biopsies has an
important qualitative aspect, rather than purely
quantitative. A fewer number of portal structures
with a lot of liver parenchyma around them are
more useful for interpretation than numerically
Fig. 9.9 Long liver core is floated off the microsieve into more but incomplete portal structures, which
formalin jar without excessive handling may be at the edge of the liver core.
The earlier studies on EUS-LB measured
Considerations for the Pathologist aggregate specimen length. However, the metric
that we feel to be most relevant is “length of the
The surgical pathology receiving laboratory is longest piece” (LLP). Highly fragmented cores
typically used to handling small specimens. It is are harder to interpret, particularly for evaluation
highly recommended to discuss EUS-LB with of fibrosis such as in the cases of early cirrhosis.
the pathologist, to ensure optimal handling. For Liver disease itself can contribute to biopsy frag-
liver cores, like other small specimens, excessive mentation, of course. But, routine specimen pro-
handling should be avoided to limit artifactual cessing contributes to fragmentation [25] which
fragmentation of the specimen. Advances and is more marked with smaller gauge biopsy
refinements in biopsy technique and needle tech- needles.
nology will continue to yield better (i.e., less Over the past 2 decades, pathologists have
fragmented) specimens for histological interpre- become more expert at rendering diagnoses with
tation, which will make the pathologist’s job smaller and smaller amounts of tissue. However,
easier (Figs. 9.10 and 9.11) [31–34]. longer cores make the job of liver biopsy inter-
The advantage of the sieve and washing step pretation significantly easier than trying to “read
described above is the delivery to the pathology the tea leaves” that a highly fragmented specimen
laboratory of an enriched liver specimen with presents. Based on the early reports and trends
little or no blood. This greatly simplifies handling utilizing newer needle technologies, it seems
of the specimen by the pathology technicians likely that the 19G core needles will be favored
because they do not have to manually separate for EUS-LB because of their ability to deliver
the blood and tissue. Blood and tissue processed longer cores with less fragmentation than non-
together (i.e., no tissue and clot manual separa- core needles. Development of a reliable Tru-Cut
tion step) makes it more difficult to interpret the needle that can be used for EUS-LB may also
liver biopsy specimen (Fig. 9.12). prove useful.
Standard adequacy metrics that have been
used in the published literature include measures
of specimen length and number of portal struc- Post-Procedure Recovery
tures. Initially, measurement of “complete portal After EUS-LB
triads” was reported (defined as portal structures
with an identifiable artery, vein, and bile duct). A Self-limited pain after percutaneous large gauge
recent study quantified “complete” and “incom- liver biopsy is common, has been described in at
plete” portal structures [35]. Other pathologists least 80% of patients [36, 37]. More severe pain
may describe portal triads in terms of how much is seen in a smaller number [38]. It is possible
liver parenchyma is visible around them (e.g., that there is less pain after EUS-LB due to the
less than or more than 180°) (Fig. 9.13a, b). ability of the real-time ultrasound to allow avoid-
90 D. L. Diehl

Fig. 9.10 Good liver cores demonstrated on glass slide after processing (a) low power and (b) medium power (tri-
chrome stain)

Fig. 9.11 A small fragment of gastric (a) or duodenal (b) mucosa indicates if the biopsy was transgastric or
transduodenal

ance of intervening vasculature as well as the procedure. Increased risk of bleeding after
ability to use smaller gauge needles. EUS-LB has not been demonstrated, even with
A common practice after percutaneous liver bilobar biopsy.
biopsy is to have the patient lie on their right side We reviewed recovery data on 124 patients
for 2–4 h after the biopsy, presumably because who underwent EUS-LB by 2 practitioners [39].
this offers “tamponade” of the peritoneal mem- One used a 1-h recovery period and the other a
brane to the liver capsule at the site of puncture. 2-h recovery time. About 30% of patients experi-
There is little or no available literature on the enced pain after the procedure; it was easily con-
advantage of this practice. With EUS-LB, there is trolled by a small dose of IV pain medication
no opportunity to obtain “tamponade,” since the given after they arrived in the recovery room. The
point of puncture is not adjacent to the abdominal vast majority (92%) were pain free by 1 h, and
wall. In our practice, we have the patients recover the other 3 (8%) had pain that resolved within 2
in a supine position, like every other endoscopic h. These findings would indicate that a 1-h recov-
9 Endoscopic Ultrasound-Guided Liver Biopsy 91

Fig. 9.12 Highly fragmented liver biopsy specimen with blood clots

Fig. 9.13 Complete portal triads in central (a) or peripheral (b) location in the core

ery period is sufficient in almost every case, with lower incidence of AEs would probably be
the need for longer observation for pain control expected compared to percutaneous approach.
necessary in only a few. There are limited reports of EUS-LB-specific
complications. A single case of bleeding was
reported in a retrospective multicenter study of
Adverse Effects 110 patients [15]. This patient had evolving
­diffuse intravascular coagulation (DIC) and in
Adverse effects (AEs) of traditional liver biopsy retrospect should not had an EUS-LB. In a study
are infrequent but can be severe, and include life-­ of 75 patients comparing diagnostic yields
threatening bleeding, organ perforation, and pain between the Quick Core Tru-Cut needle and a
[40–45]. EUS-LB features “real-time” monitor- regular 19G needle, 2 patients (both in the Tru-­
ing of the needle trajectory during biopsy, so a Cut group) were seen in the emergency room for
92 D. L. Diehl

abdominal pain, but perforation and bleeding further development of new devices, it is possible
were excluded. that EGD and EUS will be the preferred approach
We have not personally encountered, but are to provide a comprehensive evaluation of patients
aware of, several instances of inadvertent splenic with chronic liver disease, being able to screen
puncture during EUS-LB. This is presumably for varices, measure portal pressures, and obtain
from misidentification of the left lobe of the liver a liver biopsy. Newer research is looking into the
and confusing it with the spleen. Indeed, we have possibility of insertion of intrahepatic portosys-
encountered cases where the echotexture of the temic shunts by endoscopic ultrasound [49]. If
spleen and left lobe of the liver are remarkably this comes to pass, then a comprehensive EUS-­
similar. Care must be taken in identifying the based approach to diagnosis and treatment of
biopsy target; the liver has larger vessels, and portal hypertension could be realized.
they typically can be traced back to the larger There continues to be development of nonin-
venous origins of hepatic veins and main portal vasive methods of liver assessment and these
vein. In cases of fatty liver, the venous anatomy, have certainly supplanted the need for biopsy in
particularly portal, can be obscured. many cases. However, there remains a clear need
for liver biopsy in clinical practice as well as for
clinical research [34, 50], and a method which is
Future Directions safe, efficient, and effective will remain impor-
tant in the foreseeable future.
Standard 19G FNA needles can reproducibly
produce usable core samples from liver lobes.
Preliminary data suggests that 19G core needles References
may be able to deliver core samples with better
yields than regular FNA needles, which will aid 1. Rockey DC, Caldwell SH, Goodman ZD, Nelson
RC, Smith AD, American Association for the
the pathologist in interpretation. Future evolution Study of Liver Diseases. Liver biopsy. Hepatology.
and improvement of the EUS Tru-Cut needle 2009;49(3):1017–44.
platform may theoretically be able to more repro- 2. Caldwell S. Liver biopsy: the reports of its demise
ducibly deliver cores with good length, due to the are greatly exaggerated. Clin Transl Gastroenterol.
2016;7:e171.
“automatic” nature of the cutting element of the 3. Bravo AA, Sheth SG, Chopra S. Liver biopsy. N Engl
needle, which is not dependent on operator-­ J Med. 2001;344(7):495–500.
variable techniques. 4. Lebrec D, Goldfarb G, Degott C, Rueff B, Benhamou
As more endosonographers get comfortable JP. Transvenous liver biopsy: an experience based on
1000 hepatic tissue samplings with this procedure.
with EUS-guided liver biopsy, a niche for this tech- Gastroenterology. 1982;83(2):338–40.
nique will continue to emerge. The ability to com- 5. Bull HJ, Gilmore IT, Bradley RD, Marigold JH,
plete both liver biopsy and endoscopy or EUS Thompson RP. Experience with transjugular liver
during the same session is a step forward in conve- biopsy. Gut. 1983;24(11):1057–60.
6. Kalambokis G, Manousou P, Vibhakorn S, Marelli L,
nience and comfort for the patient as well as pro- Cholongitas E, Senzolo M, et al. Transjugular liver
viding more efficient use of resources. Economic biopsy—indications, adequacy, quality of specimens,
analyses of these combined procedures are con- and complications—a systematic review. J Hepatol.
templated, and would serve to drive further use. 2007;47(2):284–94.
7. Ring EJ, Lake JR, Roberts JP, Gordon RL, LaBerge
Development of devices and techniques for JM, Read AE, et al. Using transjugular intrahepatic
EUS-guided portal pressure measurement may portosystemic shunts to control variceal bleed-
also expand the role of EUS in the investigation ing before liver transplantation. Ann Intern Med.
of suspected portal hypertension [46–48]. 1992;116:304–9.
8. tenBerge J, Hoffman BJ, Hawes RH, Van Enckevort
Patients without overt cirrhosis who require both C, Giovannini M, Erickson RA, et al. EUS-guided
liver biopsy and portal pressure measurement fine needle aspiration of the liver: indications, yield,
currently require a transjugular approach for liver and safety based on an international survey of 167
biopsy and portal pressure measurement. With cases. Gastrointest Endosc. 2002;55(7):859–62.
9 Endoscopic Ultrasound-Guided Liver Biopsy 93

9. DeWitt J, LeBlanc J, McHenry L, Ciaccia D, Imperiale 23. Shah ND, Sasatomi E, Baron TH. Endoscopic
T, Chappo J, et al. Endoscopic ultrasound-guided fine ultrasound-­guided parenchymal liver biopsy: single
needle aspiration cytology of solid liver lesions: a center experience of a new dedicated core needle.
large single-center experience. Am J Gastroenterol. Clin Gastroenterol Hepatol. 2017;15(5):784–6.
2003;98(9):1976–81. 24. Ching-Campanioni R, Diehl DL, Confer B, et al.
10. Mathew A. EUS-guided routine liver biopsy in selected ClinicalTrials.gov Identifier: NCT03408171.
patients. Am J Gastroenterol. 2007;102(10):2354–5. 25. Mok SR, Diehl DL, Johal AS, Khara HS, Diehl M,
11. Gleeson FC, Clayton AC, Zhang L, Clain JE, Gores Mudireddy PR, Confer B, Chen ZE. Mo1245 19
GJ, Rajan E, et al. Adequacy of endoscopic ultrasound versus 22-gauge fine needle biopsy for endoscopic
core needle biopsy specimen of nonmalignant hepatic ultrasound guided liver biopsy (EUS-LB): a pro-
parenchymal disease. Clin Gastroenterol Hepatol. spective randomized trial. Gastrointest Endosc.
2008;6(12):1437–40. 2017;85(5):AB473–4.
12. Dewitt J, McGreevy K, Cummings O, Sherman S, 26. Katanuma A, Itoi T, Baron TH, Yasuda I, Kin T, Yane
Leblanc JK, McHenry L, et al. Initial experience with K, et al. Bench-top testing of suction forces generated
EUS-guided Tru-cut biopsy of benign liver disease. through endoscopic ultrasound-guided aspiration nee-
Gastrointest Endosc. 2009;69(3):535–42. dles. J Hepatobiliary Pancreat Sci. 2015;22(5):379–85.
13. Acosta R, Twadell W, Darwin P. Adequacy of EUS 27. Berzosa M, Villa N, Bartel M, Wallace M, Tau J, Trang
guided true cut liver biopsy. Gastrointest Endosc. T, et al. Mo1420 pilot study comparing hybrid vs.
2009;69(5):AB331. wet vs. dry suction techniques for EUS-FNA of solid
14. Stavropoulos SN, Im GY, Jlayer Z, Harris MD, Pitea lesions. Gastrointest Endosc. 2014;79((5):AB430.
TC, Turi GK, et al. High yield of same-session EUS-­ 28. Attam R, Arain MA, Bloechl SJ, Trikudanathan G,
guided liver biopsy by 19-gauge FNA needle in Munigala S, Bakman Y, et al. “Wet suction technique
patients undergoing EUS to exclude biliary obstruc- (WEST)”: a novel way to enhance the quality of EUS-­
tion. Gastrointest Endosc. 2012;75(2):310–8. FNA aspirate. Results of a prospective, single-blind,
15. Diehl DL, Johal AS, Khara HS, Stavropoulos randomized, controlled trial using a 22-gauge needle
SN, Al-Haddad M, Ramesh J, et al. Endoscopic for EUS-FNA of solid lesions. Gastrointest Endosc.
ultrasound-­guided liver biopsy: a multicenter experi- 2015;81(6):1401–7.
ence. Endosc Int Open. 2015;3(3):E210–5. 29. Diehl DL, Mok SR, Johal AJ, et al. Heparin priming
16. Pineda JJ, Diehl DL, Miao CL, Johal AS, Khara HS, of EUS-FNA needles does not adversely affect tissue
Bhanushali A, et al. EUS-guided liver biopsy provides cytology or immunohistochemical staining. Endosc
diagnostic samples comparable with those via the per- Int Open. 2018;6:E356.
cutaneous or transjugular route. Gastrointest Endosc. 30. Nieto J, Khaleel H, Challita Y, Jimenez M, Baron TH,
2016;83(2):360–5. Walters L, et al. EUS-guided fine-needle core liver
17. Stine JG, Liss G, Lewis JH. The safety of same-day biopsy sampling using a novel 19-gauge needle with
endoscopy and percutaneous liver biopsy. Dig Dis modified one pass, one actuation wet suction tech-
Sci. 2011;56(4):1201–6. nique. Gastrointest Endosc. 2017;67(2):469–75.
18. Johal AS, Khara HS, Maksimak MG, Diehl 31. Crawford AR, Lin XZ, Crawford JM. The normal
DL. Endoscopic ultrasound-guided liver biopsy adult human liver biopsy: a quantitative reference
in pediatric patients. Endosc Ultrasound. standard. Hepatology. 1998;28(2):323–31.
2014;3(3):191–4. 32. Rocken C, Meier H, Klauck S, Wolff S, Malfertheiner
19. Bhatia V, Hijioka S, Hara K, Mizuno N, Imaoka P, Roessner A. Large-needle biopsy versus thin-­
H, Yamao K. Endoscopic ultrasound description needle biopsy in diagnostic pathology of liver dis-
of liver segmentation and anatomy. Dig Endosc. eases. Liver. 2001;21(6):391–7.
2014;26(3):482–90. 33. Colloredo G, Guido M, Sonzogni A, Leandro
20. Larson SP, Bowers SP, Palekar NA, et al. G. Impact of liver biopsy size on histological evalua-
Histopathologic variability between the right and left tion of chronic viral hepatitis: the smaller the sample,
lobes of the liver in morbidly obese patients undergo- the milder the disease. J Hepatol. 2003;39(2):239–44.
ing Roux-en-Y bypass. Clin Gastroenterol Hepatol. 34. Riaz Z, Wright M. Advances in clinical hepatology
2007;5(11):1329–32. and what the hepatologist expects from a liver biopsy
21. Sey MS, Al-Haddad M, Imperiale TF, McGreevy result. Diagn Histopathol. 2014;20(3):95–101.
K, Lin J, DeWitt JM. EUS-guided liver biopsy for 35. Nakanishi Y, Mneimneh WS, Sey M, Al-Haddad M,
parenchymal disease: a comparison of diagnostic DeWitt JM, Saxena R. One hundred thirteen consecu-
yield between two core biopsy needles. Gastrointest tive transgastric liver biopsies for hepatic parenchy-
Endosc. 2016;83(2):347–52. mal diseases: a single-institution study. Am J Surg
22. Schulman AR, Thompson CC, Odze R, Chan WW, Pathol. 2015;39(7):968–76.
Ryou M. Optimizing EUS-guided liver biopsy sam- 36. Eisenberg E, Konopniki M, Veitsman E, Kramskay R,
pling: comprehensive assessment of needle types and Gaitini D, Baruch Y. Prevalence and characteristics
tissue acquisition techniques. Gastrointest Endosc. of pain induced by percutaneous liver biopsy. Anesth
2017;85(2):419–26. Analg. 2003;96(5):1392–6.
94 D. L. Diehl

37. Akhtar M, Ahmad M, Rehman A. Frequency of pain non-cirrhotic patients: an evaluation of 3806 biopsies.
associated with ultrasound guided percutaneous liver Gut. 2007;56(5):736–7.
biopsy. J Postgrad Med Inst. 2012;26(4):24. 45. Myers RP, Fong A, Shaheen AA. Utilization rates,
38. Firpi RJ, Soldevila-Pico C, Abdelmalek MF, Morelli complications and costs of percutaneous liver biopsy:
G, Judah J, Nelson DR. Short recovery time after a population-based study including 4275 biopsies.
percutaneous liver biopsy: should we change our Liver Int. 2008;28(5):705–12.
current practices? Clin Gastroenterol Hepatol. 46. Huang JY, Samarasena JB, Tsujino T, Lee J, Hu KQ,
2005;3(9):926–9. McLaren CE, et al. EUS-guided portal pressure gradient
39. Shah N, Huseini MS, Chen ZM, Persing K, Zeiber measurement with a simple novel device: a human pilot
H, Diehl DL, Johal AS, Khara HS. Effective study. Gastrointest Endosc. 2017;85(5):996–1001.
resource utilization with an early discharge protocol 47. Huang JY, Samarasena JB, Tsujino T, Chang
for EUS-guided liver biopsy. Am J Gastroenterol. KJ. EUS-guided portal pressure gradient measure-
2015;110:S978–80. ment with a novel 25-gauge needle device versus
40. Perrault J, McGill DB, Ott BJ, Taylor WF. Liver standard transjugular approach: a comparison animal
biopsy: complications in 1000 inpatients and outpa- study. Gastrointest Endosc. 2016;84(2):358–62.
tients. Gastroenterology. 1978;74(1):103–6. 48. Schulman AR, Thompson CC, Ryou M. EUS-guided
41. Piccinino F, Sagnelli E, Pasquale G, Giusti portal pressure measurement using a digital pressure
G. Complications following percutaneous liver wire with real-time remote display: a novel, minimally
biopsy. A multicentre retrospective study on 68,276 invasive technique for direct measurement in an ani-
biopsies. J Hepatol. 1986;2(2):165–73. mal model. Gastrointest Endosc. 2016;83(4):817–20.
42. McGill DB, Rakela J, Zinsmeister AR, Ott BJ. A 49. Schulman AR, Ryou M, Aihara H, Abidi W, Chiang
21-year experience with major hemorrhage after A, Jirapinyo P, Sakr A, Ajeje E, Ryan MB, Thompson
percutaneous liver biopsy. Gastroenterology. CC. EUS-guided intrahepatic portosystemic shunt
1990;99(5):1396–400. with direct portal pressure measurements: a novel
43. Janes CH, Lindor KD. Outcome of patients hospital- alternative to transjugular intrahepatic portosystemic
ized for complications after outpatient liver biopsy. shunting. Gastrointest Endosc. 2017;85(1):243–7.
Ann Intern Med. 1993;118(2):96–8. 50. Rowe IA, Parker R. Liver biopsy for the selection of
44. Huang JF, Hsieh MY, Dai CY, Hou NJ, Lee LP, Lin patients with nonalcoholic steatohepatitis for clinical
ZY, et al. The incidence and risks of liver biopsy in trials. Gastroenterology. 2015;148:262.
 EUS-Guided Fiducial Placement
10
Aamir N. Dam and Jason B. Klapman

These gold seeds measured approximately

Introduction 2.5–5 mm in length, ranged from 0.8 to 1.2 mm in
diameter, and required a 19-gauge needle to
Image-guided radiation therapy (IGRT) uses deploy them [12, 15, 16]. Visicoil (Radio Med
real-time imaging to precisely localize tumors Corporation, Tyngsboro, MA, Core Oncology,
and deliver focused high-dose beams of radio- Santa Barbara, CA) fiducials were subsequently
therapy [1]. Fiducials are radiopaque markers introduced into the market and unlike traditional
implanted at the site of a tumor or a lymph node fiducials, they are flexible and have a coiled
that enhance lesion localization, and serve as ref- design to theoretically reduce risk of migration
erence points for targeting radiation therapy [2]. (Fig. 10.1). Visicoil fiducials are longer in length
Historically, fiducial markers were placed surgi- (10 mm) and produced in two different diameters
cally or percutaneously using ultrasound or CT (0.35 mm, 0.75 mm). The smaller diameter coiled
guidance [3]. In the past decade, an endoscopic fiducials can be used with a 22-gauge needle, pro-
ultrasound (EUS)-guided approach has evolved viding more flexibility in anatomic areas requir-
and shown to be a safe method for fiducial marker ing increased angulation or torque [4, 5, 8, 9].
placement. This contrasts with the larger 0.75-mm fiducial
which requires a 19-gauge needle for deployment.
In addition, Visicoil fiducials utilize a specific
Fiducial Types needle-carrier delivery system to facilitate their
insertion into the tip of the EUS needle (Fig. 10.2).
Many types of fiducial markers have been devel- A retrospective study comparing traditional
oped and described in the literature. Table 10.1 fiducials (0.8 mm × 5 mm) to the flexible Visicoil
outlines various fiducial types that have been fiducials (0.35 mm × 10 mm) in patients with
placed using EUS. In early published studies, tra- advanced pancreatic cancer demonstrated com-
ditional cylindrical gold seeds were investigated. parable technical success with no difference in
migration or complication rates when fiducials
Electronic supplementary material: The online version were placed into tumors via EUS guidance.
of this chapter (https://doi.org/10.1007/978-3-319-97376- However, the visibility of traditional fiducials
0_10) contains supplementary material, which is available
was significantly better than the Visicoil fidu-
to authorized users.
cials on CT scans and during subsequent IGRT,
A. N. Dam · J. B. Klapman (*)
possibly related to their larger diameter [10]. In
Department of Gastrointestinal Oncology, Moffitt
Cancer Center, Tampa, FL, USA ­contrast, Machiels et al. reported higher rates of
e-mail: [email protected] visibility in esophageal cancer with the newer

© Springer Nature Switzerland AG 2019 95

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_10
96 A. N. Dam and J. B. Klapman

Table 10.1 Summary of studies describing different gold fiducial types used with EUS
Size
Fiducial type (Trademark) (Diameter × Length)a Needle gaugea
Visicoil flexible gold coiled fiducial (Radio 0.75 mm × 10 mm 19G (Cook Endoscopy, Winston Salem,
Med Inc., Tyngsboro, MA) [4–7] 0.35 mm × 10 mm NC)
22G (Cook Endoscopy, Winston Salem,
NC)
Visicoil flexible gold linear fiducial (Core 0.35 mm × 10 mm 22G (Cook Endoscopy, Winston Salem,
Oncology, Santa Barbara, CA) [8–11] 0.35 mm × 2–20 mm NC)
Gold cylindrical fiducial (Best Medical 0.8 mm × 3 or 5 mm 19G (MEDI-Globe, Achenmuhle,
International, Springfield, VA) [12–14] Germany, or Cook Endoscopy, Winston
Salem, NC)
Gold cylindrical fiducial (Alpha Omega 0.8 mm × 2.5 or 19G (Cook Endoscopy, Winston Salem,
Services Inc, Bellflower, CA) [15, 16] 5 mm NC)
Gold cylindrical fiducial (Northwest Medical 0.8 mm × 3 mm 19G (Cook Endoscopy, Winston Salem,
Physics Equipment Inc., Lynwood, WA) [17] NC)
Gold cylindrical fiducial (CIVCO Medical 0.8 mm × 3 mm 19G (Cook Endoscopy, Winston Salem,
Solutions, Orange County, IA) [18] NC)
Gold anchor ball shaped or line shaped fiducial 0.28 mm × 10 mm 22G (Cook Endoscopy, Winston Salem,
(Naslund Medical AB, Huddinge, Sweden) [8] NC)
X-MARK gold fiducial (IZI Medical Products, 0.85 mm × 1, 2, or 19G
Owings Mills, MA, ONC Solutions Inc., Acton, 3 cm
MA) [14]
Sizes of fiducials and needle gauge listed are limited to the ones used in the studies
a

Fig. 10.1 Visicoil gold

fiducial marker

flexible Visicoil markers that were ≥5 mm in were easier to see radiographically [6]. Given
length when compared to the solid gold and liq- the limitations due to study design and sample
uid hydrogel fiducials markers. The authors sug- size in many studies, firm conclusions cannot be
gested that length may play a critical role in made regarding the optimal type of fiducial to be
improved visibility [11]. Fernandez et al. found placed by EUS guidance. Based on retrospective
no significant difference in visibility between the and limited prospective data, fiducials with
0.35 mm × 10 mm and 0.75 mm × 10 mm diam- increased length and diameter appear to
eter Visicoil fiducial markers in patients with have im proved v­ isibility and may be prefer-
esophageal cancer, except in patients with larger able if the positioning of the echoendoscope
body habitus where the larger diameter fiducials allows their use.
10 EUS-Guided Fiducial Placement 97

Fig. 10.2 Visicoil

fiducial preloaded on a
needle-carrier device

in the patient before the target tissue has been

 iducial Set Up and Deployment
F reached. The needle is inserted into the operating
Techniques channel of the echoendoscope and advanced into
the target lesion using Doppler ultrasound to
Most studies have had successful outcomes with avoid intervening blood vessels. A small “track”
regard to delivery of fiducials into the target is made in the target tissue to facilitate insertion
lesion (Table 10.2). Different techniques have and the fiducial is deployed by advancing the sty-
been described with slight variations in technique let completely while simultaneously retracting
with regard to loading of the fiducial marker into the needle an equal distance. The fiducial can be
the needle and deployment into the target tissue. seen to deploy endosonographically and via fluo-
An antegrade method for loading fiducials roscopy, if utilized. The needle is removed and
was first described in two case series [5, 12]. reloaded with a new fiducial and the method is
Ammar et al. preferred this approach because it repeated until the desired number of fiducials
prevented handling of the sharp end of the needle have been placed. In both of these techniques, the
and minimized the risk of fiducial loss while stylet is used to deploy the fiducial. The back-­
accessing the lesion [4]. In this method, the nee- loading technique and intra-tumoral deployment
dle is inserted into the target lesion and the stylet is demonstrated in Video 10.1.
is withdrawn. Next, the fiducial is manually An alternative to the stylet-push method has
loaded into the needle lumen and the stylet is been developed. This technique uses a hydro-
reinserted to push the fiducial forward into the static technique to deploy the fiducial into the
target lesion. Another more commonly used target lesion. In this method, the stylet is removed,
method involves a back-loading technique using the needle is first flushed with sterile water or
a 19- or 22-gauge EUS-FNA needle that has been normal saline, and the fiducial is back-loaded
described in numerous studies [6–10, 13, 15–18]. into the needle. Then, the needle is inserted into
First, the stylet is withdrawn from the needle the tumor and 1–2 mL of sterile water or normal
approximately 7–8 mm, and a fiducial marker is saline is instilled into the needle to deploy the
back-loaded into the needle tip in a retrograde fiducial [13, 15]. The advantages reported include
manner using sterile forceps or using the needle-­ reducing air artifact and aiding delivery during
carrier delivery device (Fig. 10.3). Once the fidu- difficult scope positions.
cial itself is within the lumen of the needle, the To improve efficiency, other specialists have
needle tip is sealed with sterile bone wax to pre- described preloading two fiducials into the tip of
vent loss of the fiducial in the echoendoscope or the needle with the ability of placing multiple
98 A. N. Dam and J. B. Klapman

Table 10.2 Summary of efficacy and safety of EUS-guided fiducial placement

Location of Needle Technical
Study Design malignant lesion gauge (G) Fiducial type (mm) success Adverse events
Pishvaian et al. PS Mediastinum 19G Gold (0.8 × 3 or 5) 11/13 Cholangitis (1)
[12] (n = 13) Esophagus (85%)
Pancreas
Metastatic lesions
in abdomen
Sanders et al. PS Pancreas 19G Gold (0.8 × 5) 46/51 Mild
[16] (n = 51) (90%) pancreatitis
(1)
Park et al. [15] PS Pancreas 19G Gold (0.8 × 2.5) 50/53 Minor bleeding
(n = 57) (94%) (1)
Ammar et al. RS Pancreas 22G VC (0.35 × 10) 13/13 None
[4] (n = 13) Abdominal lymph (100%)
node
Liver lesion
Adrenal gland
Bile duct (CCA)
DiMaio et al. RS Esophagus 22G VC (0.35 × 10) 29/30 Fever (1)
[9] (n = 30) Pancreas (97%)
Gastric
Bile duct (CCA)
Metastatic liver
lesion
Varadarajulu RS Pancreas 19G Gold (0.8 × 3) 9/9 None
et al. [13] (n = 9) (100%)
Khashab et al. RS Pancreas 19G Gold (0.8 × 5) 39/39 None
[10] (n = 39) 22G VC (0.35 × 10) (100%)
Fernandez RS Esophagus 19G VC (0.75 × 10) 60/60 Abdominal
et al. [6] (n = 60) 22G VC (0.5 × 10) (100%) pain (1)
VC (0.35 × 10)
Majumder RS Pancreas 19G Gold (0.8 × 5) 35/39 Abdominal
et al. [19] (n = 77) (90%) pain (3)
Mild
pancreatitis (1)
Choi et al. [18] RS Pancreas 19G Gold (0.8 × 3) 32/32 Mild
(n = 32) Liver lesion (100%) pancreatitis (1)
Metastatic lymph
node
Chandran et al. PS Gastric 19G VC (0.35 × 10) 7/8 (88%) None
[20] (n = 8)
Davila Fajardo PS Pancreas 22G VC (0.35 × 5–20) 23/23 Minor bleeding
et al. [8] (n = 23) Gold anchor (100%) (1)
(0.28 × 10)
Moningi et al. RS Rectum 19G Gold (0.8 × 5) 11/11 None
[14] (n = 11) X-mark fiducial (100%)
(0.85 × 10–30)
Machiels et al. PS Esophagus 22G Gold 30/30 Pneumothorax
[11] (n = 32) (0.43–0.64 × 5) (100%) (1)
Visicoil Mediastinitis
(0.35 × 2–10) (2)
Hydrogel marker
(continued)
10 EUS-Guided Fiducial Placement 99

Table 10.2 (continued)

Location of Needle Technical
Study Design malignant lesion gauge (G) Fiducial type (mm) success Adverse events
Dhadham et al. RS Mediastinum 19G VC (0.75 × 10) 513/514 Minor bleeding
[7] (n = 514) Esophagus 22G VC (0.35 × 10) (99.8%) (9)
Pancreas
Rectum/anal canal
Metastatic lesions
in abdomen and
liver
PS prospective study, RS retrospective study, VC Visicoil, CCA cholangiocarcinoma

markers at the same time [15]. Currently, pre-

loaded needles are commercially available for
use. The Beacon FNF needle (Medtronic,
Minneapolis, MN) is available in two sizes and
preloaded with two solid gold fiducial markers—
22-gauge (0.43 mm × 5 mm) and 19-gauge
(0.75 mm × 5 mm). In addition, the 22-gauge
EchoTip Ultra preloaded needle (Cook Medical,
Bloomington, IN) has been developed and been
shown to be effective in a live porcine models
[21]. The Cook needle system comes preloaded
with four gold fiducials that are each 5 mm in
length and 0.43 mm in diameter. A current ran-
Fig. 10.3 Visicoil fiducial loaded on the distal tip of EUS
domized controlled trial is underway comparing needle
overall efficiency and technical success between
the 22-gauge EchoTip Ultra preloaded fiducial
needle versus the traditional back-loading tech- Fiducial Tumor Targets
nique in patients with pancreatic cancer.
The optimal number of fiducials to be placed Pancreatic Cancer
into a lesion has not been well established. In the
literature, most studies have placed between 2 Pancreatic cancer has recently become the third
and 5 fiducials for each tumor/lymph node/target most common cause of cancer-related deaths,
lesion. In our experience, we attempt to place at and only 20% of patients are surgically resectable
least three fiducials in different locations within at the time of diagnosis [22, 23]. For patients
pancreatic lesions and one fiducial marker at both with borderline resectable or locally advanced
the proximal and distal margins of luminal disease, neoadjuvant chemotherapy and radiation
tumors if feasible, although practice in this regard play an important role in controlling tumor
varies between centers. growth and influencing overall survival [24–26].
Technical difficulties that have been encoun- While EUS has traditionally aided in the diagno-
tered include resistance while pushing the fiducial sis and staging of pancreatic cancer, more thera-
with the stylet [8, 12, 15], and the presence of peutic options have emerged including celiac
intervening vasculature [7, 16] which makes safe plexus neurolysis, EUS-guided biliary access and
deployment challenging. As described above, to drainage, fine needle injection, and fiducial
overcome difficult anatomic positions, techniques placement (Fig. 10.4) [27]. In 2006, Pishvaian
that have been successfully reported include repo- et al. performed the first case series evaluating
sitioning the scope, using a smaller size fiducial/ EUS-guided gold fiducial placement in mediasti-
needle or trying a different deployment technique nal and abdominal malignancies which included
such as the hydrostatic technique. five patients with advanced pancreatic cancer and
100 A. N. Dam and J. B. Klapman

Fig. 10.4 Endosonographic image of a hyperechoic fiducial placed within the pancreatic body mass

one with recurrent cancer post-Whipple. The (3.1%) mainly involving intervening blood ves-
technique followed the same principle of EUS-­ sels, and minor bleeding that resolved spontane-
guided FNA and delivered an average of 3–4 ously in seven patients (1.3%) [7].
fiducials in each of the five patients using a In early studies, fluoroscopy was used in con-
19-gauge needle. One failure occurred secondary junction with EUS to help achieve appropriate
to gastric outlet obstruction in a patient with a angulation and distance between fiducial markers
tumor in the pancreatic head. The study showed (Fig. 10.5). More recent studies have shown suc-
an overall technical success rate of 85% and was cessful placement of EUS-guided fiducial mark-
the first to demonstrate the safety and feasibility ers without the use of fluoroscopy, suggesting
of EUS-guided fiducial placement for tumor that fluoroscopy can be used if available but is not
marking to guide radiotherapy [12]. Since that considered essential for safe and successful EUS-­
report, multiple prospective and retrospective guided fiducial placement [6, 7, 9, 18]. In addi-
case series have described the feasibility of fidu- tion, a recent retrospective study by Majumder
cial placement, specifically in pancreatic cancer, et al. found that achieving ideal fiducial geometry
with high success rates ranging from 88 to 100% may be unnecessary for successful tracking and
[10, 13, 15, 16, 18]. Four studies demonstrated delivery of radiation in patients with pancreatic
success with the use of a 22-gauge needle to cancer [19].
place smaller diameter Visicoil fiducial markers
in patients with pancreatic cancer [4, 8–10].
There are no prospective data comparing the 19- Esophageal Cancer
and 22-gauge needles for fiducial placement, but
experts report that the 22-gauge needle may help Radiotherapy plays an important role in esopha-
overcome issues of angulation in pancreatic geal cancer as many patients also present with
lesions in the head and uncinate process [8, 9]. advanced stage disease [28]. Several studies have
In the largest retrospective series involving specifically evaluated EUS-guided fiducial place-
188 patients with pancreatic cancer, a 22-gauge ment in patients with esophageal cancer and have
needle was used to place 414 Visicoil fiducials shown favorable results with high technical suc-
(0.35 mm × 10 mm) in 80% of patients, and a cess [6, 7, 9, 11].
19-gauge needle was used to place 93 Visicoil Fiducials can be placed proximal and distal to
fiducials (0.75 mm × 10 mm) in 20% of patients. the tumor and provide accurate delineation of the
Technical difficulty occurred in 16 patients extent of the lesion (Fig. 10.6) [6, 7, 11].
10 EUS-Guided Fiducial Placement 101

Fig. 10.5 Fluoroscopic image of fiducials placed within the: (a) pancreatic head, (b) uncinate process of the pancreas,
and (c) pancreatic body

Fig. 10.6 Endosonographic imaging of a hyperechoic fiducial placed just proximal to an esophageal mass

In approximately one-third of cases, a single reduce migration rates especially after tumor
fiducial marker was placed given that the tumor regression from treatment (Fig. 10.7) [6, 7, 11].
was obstructing and prevented passage of the DiMaio et al. assessed EUS-guided fiducial
echoendoscope [6, 7]. Most studies have placement (Visicoil 0.35 mm × 10 mm) using a
described securing the fiducial into the submu- 22-gauge needle in 12 patients with esophageal
cosa or muscularis propria adjacent to the tumor, tumors; all were technically feasible except for
instead of into the tumor itself, to theoretically one in which the lesion could not be identified
102 A. N. Dam and J. B. Klapman

Fig. 10.7 Endosonographic image of a hyperechoic fiducial clearly placed within the muscularis propria proximal to a
distal esophageal mass

[9]. Fernandez and colleagues reported a retro- Rectal Cancer

spective series of 60 patients with esophageal
cancer who underwent EUS-guided fiducials. In Two studies have evaluated the role of EUS-­
the majority of patients, Visicoil fiducial markers guided fiducial placement in rectal cancer. The
(0.75 mm × 10 mm) were placed with a 19-gauge first report described EUS-guided fiducial place-
needle, and in a few patients, the smaller diame- ment used in the management of rectal cancer
ter (0.35 mm × 10 mm) fiducials were used. A with high-dose rate endorectal brachytherapy. In
total of 105 markers were placed, 33% had a this study, 11 patients underwent EUS-guided
single fiducial marker, 58% had two fiducial placement of two different types of gold fidu-
markers, and 8% had three fiducial markers cials. All fiducials were placed at the superior and
inserted proximal and distal to the lesion if pos- inferior extents as well as in the center of the
sible. The investigators concluded that implanta- tumor, and the mean number of fiducials placed
tion of fiducials for esophageal cancer was per patient was 3.6. All fiducials, regardless of
feasible, allowed for more confident target delin- type, were clearly visible, and all 11 patients
eation, and improved assessment of respiratory underwent IGRT with subsequent successful
tumor motion on CT simulation [6]. Another ret- resection [14]. In a subsequent study, 54 patients
rospective study involved 207 patients with with rectal cancer had 103 fiducials inserted,
esophageal cancer in which 348 fiducials were 70% fiducials were placed at both the proximal
inserted. The 0.75 mm × 10 mm Visicoil fiducial and distal margins, 16.6% at the proximal margin
marker was used in 91% of patients using a only, and 13.1% at the distal margin only.
19-gauge needle. In addition, there were 33 Minimal complications were reported with mild
patients with gastroesophageal junction tumors, bleeding occurring in one patient [7]. Figure 10.8
of which 64% had two fiducials placed and 36.4% demonstrates an endoscopic image of a rectal
of patients had one fiducial placed. These patients cancer and CT performed 1 month later with
successfully underwent radiation therapy with no fiducials remaining visible at site of rectal tumor.
significant complications related to fiducial
placement [7]. A recent retrospective analysis
showed the placement of fiducial markers cou- Other Sites
pled with 3D PET/CT aided in planning tumor
volume, specifically along the inferior border of Several studies have described the feasibility and
the tumor, and offered more accurate radiation technical success of EUS-guided fiducial place-
treatment delivery for locally advanced esopha- ment in a variety of other malignancies including
geal cancer [29]. prostate cancer [30], gastric cancer [20], anal
10 EUS-Guided Fiducial Placement 103

Fig. 10.8 (a) Endoscopic image of rectal cancer along nant left iliac lymph node near known rectal cancer. (d)
the posterior wall of the rectum, (b) CT scan confirming Fiducial needle inserted in a transrectal manner into the
the placement of multiple fiducial at the proximal margin malignant node. (e) Fiducials after deployment into the
of the rectal tumor. (c) EUS image of a peritumoral malig- malignant node

cancer [7], cholangiocarcinoma [4, 9], and meta- tion, most studies have reported that over 90% of
static lesions in the abdomen, liver, or mediasti- patients with successfully placed EUS-guided
num (Fig. 10.9) [4, 7, 12, 18]. fiducials completed radiation therapy [6–8, 13,
16, 18]. However, data on long-term outcomes in
fiducial placement are limited and have not been
Durability of Fiducial Placement clearly defined. In addition, studies assessing
improved overall survival with fiducials are lack-
In regard to fiducial placement and feasibility as ing. Various endpoints that have been evaluated
stated above, high rates of technical success rang- include the presence of markers at simulation CT
ing from 85 to 100% have been reported. In addi- scan, visibility during treatment period, and
104 A. N. Dam and J. B. Klapman

Fig. 10.9 Endosono­

graphic image of two
hyperechoic fiducials
placed within a
metastatic pancreatic
tail mass

migration rates. Figure 10.10 demonstrates visi- include fever, cholangitis, mild acute pancreatitis,
bility of fiducials on CT scan and PET-CT. minor bleeding, and post procedure abdominal
DiMaio et al. evaluated fiducial placement in pain. Rare cases of pneumothorax, mediastinitis,
30 patients with various GI malignancies and and intramural duodenal hematoma have also
fiducials were identified in 83% of patients at the been reported [11, 31].
time of CT simulation for radiation therapy [9]. Fiducial migration rates have been measured
Fernandez and colleagues investigated long-term on simulation exams and during therapy and have
stability of fiducial placement in the setting of ranged from 0.4 to 9.5%. There was one report of
esophageal cancer. In their study, 105 Visicoil migration of a fiducial into the lung in a patient
markers were placed; 94% of markers were still with esophageal cancer, although the patient
present at CT simulation, and 88% were still remained asymptomatic [11].
present in their initial position at a median time The use of prophylactic antibiotics for EUS-­
of 107 days. In patients who did not undergo sur- guided fiducial placement is debatable and mul-
gery, 90% of fiducials were visible at a median tiple studies have used them in their protocol [4,
time of 165 days following implantation [6]. 10, 13, 15, 16]. Infectious complications rates
Machiels et al. reported in a small prospective were not increased in other studies that did not
study that 63% of solid gold markers and 80% of routinely give antibiotics [7, 8]. There are no pro-
Visicoil markers placed in esophageal tumors spective data on this topic, and based on the cur-
remained visible during the treatment period. In a rent literature, there is no firm evidence to support
subgroup analysis, 91% of Visicoil markers the routine use of antibiotics during EUS-guided
≥5 mm in length were visible at the end of their fiducial placement.
treatment period. Most markers that lost visibility
were related to detachment and small size, and
rarely related to migration [11]. Dhadham et al. Conclusion
also reported a low fiducial migration rate of
0.4% evaluated during IGRT in 207 patients with EUS-guided fiducial placement is a safe, effec-
locally advanced esophageal cancer [7]. tive technique to enhance IGRT and provides
precise targeted radiation while limiting dosage
to normal surrounding tissue. EUS may be the
Adverse Events preferred approach as diagnosis, staging, and
therapeutic interventions can be performed in
EUS-guided fiducial placement is safe with a low the same session and expedite treatment. Many
reported adverse event rate between 1 and 5%. studies have investigated EUS-guided fiducial
Common adverse events were self-limited and ­placement in pancreatic tumors, but there is
10 EUS-Guided Fiducial Placement 105

Fig. 10.10 Fiducial markers seen on: (a) CT scan within the pancreatic head, (b) CT scan within the pancreatic body,
and (c) PET-CT within the pancreatic head

increasing evidence for its use in other GI in choosing the correct size, number, and
malignancies including esophageal, gastric, type of fiducial/needle to use in specific
rectal, anal, and hepatobiliary cancers. As malignancies.
described in this chapter, the technique and fea-
sibility for EUS-guided fiducial placement has
been well delineated in the current literature References
with high technical success. More prospective
1. Reese AS, Lu W, Regine WF. Utilization of intensity-­
studies are needed to assess the short- and long- modulated radiation therapy and image-guided radia-
term clinical impact of fiducial placement on tion therapy in pancreatic cancer: is it beneficial?
IGRT, and to help further guide the endoscopist Semin Radiat Oncol. 2014;24(2):132–9.
106 A. N. Dam and J. B. Klapman

2. Han J, Chang KJ. Endoscopic ultrasound-guided implanted during EUS for patients with localized rec-
direct intervention for solid pancreatic tumors. Clin tal cancer receiving high-dose rate endorectal brachy-
Endosc. 2017;50(2):126–37. therapy. Gastrointest Endosc. 2015;81(3):765–9.e1.
3. Kothary N, Heit JJ, Louie JD, Kuo WT, Loo BW Jr, 15. Park WG, Yan BM, Schellenberg D, Kim J, Chang
Koong A, et al. Safety and efficacy of percutaneous DT, Koong A, et al. EUS-guided gold fiducial inser-
fiducial marker implantation for image-guided radia- tion for image-guided radiation therapy of pancre-
tion therapy. J Vasc Interv Radiol. 2009;20(2):235–9. atic cancer: 50 successful cases without fluoroscopy.
4. Ammar T, Cote GA, Creach KM, Kohlmeier C, Gastrointest Endosc. 2010;71(3):513–8.
Parikh PJ, Azar RR. Fiducial placement for stereotac- 16. Sanders MK, Moser AJ, Khalid A, Fasanella KE, Zeh
tic radiation by using EUS: feasibility when using a HJ, Burton S, et al. EUS-guided fiducial placement
marker compatible with a standard 22-gauge needle. for stereotactic body radiotherapy in locally advanced
Gastrointest Endosc. 2010;71(3):630–3. and recurrent pancreatic cancer. Gastrointest Endosc.
5. Ghassemi SFD. EUS-guided placement of fidu- 2010;71(7):1178–84.
cial markers using a 22-guage needle. Gastrointest 17. Owens DJ, Savides TJ. EUS placement of metal fidu-
Endosc. 2009;69(5):AB337–AB8. cials by using a backloaded technique with bone wax
6. Fernandez DC, Hoffe SE, Barthel JS, Vignesh S, seal. Gastrointest Endosc. 2009;69(4):972–3.
Klapman JB, Harris C, et al. Stability of endo- 18. Choi JH, Seo DW, Park DH, Lee SK, Kim
scopic ultrasound-guided fiducial marker place- MH. Fiducial placement for stereotactic body radia-
ment for esophageal cancer target delineation and tion therapy under only endoscopic ultrasonography
image-guided radiation therapy. Pract Radiat Oncol. guidance in pancreatic and hepatic malignancy: practi-
2013;3(1):32–9. cal feasibility and safety. Gut Liver. 2014;8(1):88–93.
7. Dhadham GC, Hoffe S, Harris CL, Klapman 19. Majumder S, Berzin TM, Mahadevan A, Pawa R,
JB. Endoscopic ultrasound-guided fiducial marker Ellsmere J, Sepe PS, et al. Endoscopic ultrasound-­
placement for image-guided radiation therapy without guided pancreatic fiducial placement: how important is
fluoroscopy: safety and technical feasibility. Endosc ideal fiducial geometry? Pancreas. 2013;42(4):692–5.
Int Open. 2016;4(3):E378–82. 20. Chandran S, Vaughan R, Efthymiou M, Sia J,
8. Davila Fajardo R, Lekkerkerker SJ, van der Horst A, Hamilton C. A pilot study of EUS-guided fiducial
Lens E, Bergman JJ, Fockens P, et al. EUS-guided insertion for the multidisciplinary management of
fiducial markers placement with a 22-gauge needle gastric cancer. Endosc Int Open. 2014;2(3):E153–9.
for image-guided radiation therapy in pancreatic can- 21. Draganov PV, Chavalitdhamrong D, Wagh
cer. Gastrointest Endosc. 2014;79(5):851–5. MS. Evaluation of a new endoscopic ultrasound-­
9. DiMaio CJ, Nagula S, Goodman KA, Ho AY, guided multi-fiducial delivery system: a ­prospective
Markowitz AJ, Schattner MA, et al. EUS-guided fidu- non-survival study in a live porcine model. Dig
cial placement for image-guided radiation therapy in Endosc. 2013;25(6):615–21.
GI malignancies by using a 22-gauge needle (with 22. Kamisawa T, Wood LD, Itoi T, Takaori K. Pancreatic
videos). Gastrointest Endosc. 2010;71(7):1204–10. cancer. Lancet. 2016;388(10039):73–85.
10. Khashab MA, Kim KJ, Tryggestad EJ, Wild AT, 23. Siegel RL, Miller KD, Jemal A. Cancer statistics,
Roland T, Singh VK, et al. Comparative analysis 2017. CA Cancer J Clin. 2017;67(1):7–30.
of traditional and coiled fiducials implanted during 24. Gillen S, Schuster T, Meyer Zum Buschenfelde C,
EUS for pancreatic cancer patients receiving stereo- Friess H, Kleeff J. Preoperative/neoadjuvant therapy
tactic body radiation therapy. Gastrointest Endosc. in pancreatic cancer: a systematic review and meta-­
2012;76(5):962–71. analysis of response and resection percentages. PLoS
11. Machiels M, van Hooft J, Jin P, van Berge Med. 2010;7(4):e1000267.
Henegouwen MI, van Laarhoven HM, Alderliesten 25. Sultana A, Tudur Smith C, Cunningham D, Starling
T, et al. Endoscopy/EUS-guided fiducial marker N, Tait D, Neoptolemos JP, et al. Systematic review,
placement in patients with esophageal cancer: a com- including meta-analyses, on the management of locally
parative analysis of 3 types of markers. Gastrointest advanced pancreatic cancer using radiation/combined
Endosc. 2015;82(4):641–9. modality therapy. Br J Cancer. 2007;96(8):1183–90.
12. Pishvaian AC, Collins B, Gagnon G, Ahlawat 26. Gutt R, Liauw SL, Weichselbaum RR. The role
S, Haddad NG. EUS-guided fiducial placement of radiotherapy in locally advanced pancre-
for CyberKnife radiotherapy of mediastinal and atic carcinoma. Nat Rev Gastroenterol Hepatol.
abdominal malignancies. Gastrointest Endosc. 2010;7(8):437–47.
2006;64(3):412–7. 27. Oh SY, Irani S, Kozarek RA. What are the current and
13. Varadarajulu S, Trevino JM, Shen S, Jacob R. The potential future roles for endoscopic ultrasound in the
use of endoscopic ultrasound-guided gold markers in treatment of pancreatic cancer? World J Gastrointest
image-guided radiation therapy of pancreatic cancers: Endosc. 2016;8(7):319–29.
a case series. Endoscopy. 2010;42(5):423–5. 28. Lagergren J, Smyth E, Cunningham D,
14. Moningi S, Walker AJ, Malayeri AA, Rosati LM, Lagergren P. Oesophageal cancer. Lancet.
Gearhart SL, Efron JE, et al. Analysis of fiducials 2017;390(10110):2383–96.
10 EUS-Guided Fiducial Placement 107

29. Oliver JA, Venkat P, Frakes JM, Klapman J, Harris before targeted radiation therapy for prostate
C, Montilla-Soler J, et al. Fiducial markers coupled cancer recurrence. Gastrointest Endosc. 2011;73(6):
with 3D PET/CT offer more accurate radiation treat- 1302–5.
ment delivery for locally advanced esophageal cancer. 31. El H II, Easler JJ, Sherman S, Al-Haddad
Endosc Int Open. 2017;5(6):E496–504. M. Intramural duodenal hematoma post EUS-guided
30. Yang J, Abdel-Wahab M, Ribeiro A. EUS-guided placement of fiducial radiopaque markers. Dig Liver
fiducial placement after radical prostatectomy Dis. 2018;50(2):201.
 EUS-Guided Therapies for Solid
Pancreatic Tumors Including Drug 11
Delivery and Brachytherapy

Gursimran Singh Kochhar and Michael Wallace

also aids in the management of certain advanced

Abbreviations lesions by allowing local tissue ablation or place-
ment of fiducial markers to help our radiology
CT Computed tomography colleagues.
DC Dendritic cells In this chapter, we will cover the role of EUS
EUS Endoscopic ultrasound in managing solid pancreatic lesions and local
FNA Fine-needle aspiration EUS-guided therapies like ethanol ablation,
FNI Fine-needle injection radiofrequency ablation (RFA), brachytherapy,
HCC Hepatocellular carcinoma and fiducial placements.
PDT Photodynamic therapy
PNET Pancreatic neuroendocrine tumor
RFA Radio frequency ablation EUS-Guided Ethanol Ablation
US Ultrasound
Ethanol has long been used in the management of
locally advanced cancers by radiologists, under
computed tomography (CT) guidance and/or
Introduction ultrasound guidance [1]. In one case, EUS was
used to inject ethanol in a patient with metastatic
Since its advent, endoscopic ultrasound (EUS) hepatocellular carcinoma (HCC) successfully
has quickly progressed from a diagnostic tool to [2]. Its use in the pancreas was then widely stud-
a therapeutic tool adding many indications for its ied in animal models [3]. Early animal studies
use over time. One such use is the management showed that EUS-guided ethanol injection is fea-
of solid pancreatic lesions and cancers in a grow- sible, safe, and resulted in a wide area of abala-
ing variety of ways. EUS is used for diagnosis tion [4]. EUS-guided ethanol injection showed a
and obtaining tissue samples from lesions, but it linear dose–response relationship to the concen-
tration of ethanol used and the area of tissue
Electronic supplementary material: The online version
ablated [4].
of this chapter (https://doi.org/10.1007/978-3-319-97376- EUS-guided ethanol injection therapy has not
0_11) contains supplementary material, which is available been widely studied in pancreatic adenocarci-
to authorized users. noma, although it has been reported in multiple
G. S. Kochhar · M. Wallace (*) case reports and in small series of patients with
Department of Gastroenterology and Hepatology, pancreatic neuroendocrine tumors (PNET). In one
The Mayo Clinic, Jacksonville, FL, USA
of the earlier case reports, EUS-guided ethanol
e-mail: [email protected]

© Springer Nature Switzerland AG 2019 109

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_11
110 G. S. Kochhar and M. Wallace

Fig. 11.1 EUS-guided

ethanol ablation of a
functioning insulinoma

injection was used to treat a patient with an insuli- tions better before it becomes more accepted as a
noma (Fig. 11.1) [5]. The patient had a successful mainstream therapy. These cases highlight the
response to ablation of the insulinoma, although ability of EUS-guided therapy to identify and
the patient had to be hospitalized due to severe access lesions for injection-guided ablation
abdominal pain, likely representing some degree therapy.
of pancreatitis. Subsequently, Deprez et al.
reported treating an elderly patient with an insuli-
noma successfully via EUS-guided ethanol abla-  US-Guided Radio Frequency
E
tion [6]. Levy and colleagues reported a case series Ablation
where eight patients with symptomatic insulino-
mas were treated with ethanol ablation [7]. Using RFA is a technique for transmitting electromag-
99% ethanol, five patients were injected with etha- netic energy to induce heating in the targeted
nol under EUS guidance, and three patients were tissue [11]. Based on the type of electromag-
injected intraoperatively. All patients achieved netic source, RFA can be divided into two
successful ablation after a median follow-up of types: monopolar and bipolar RFA. In monopo-
13 months. Three patients treated intraoperatively lar RFA, the patient forms the part of a circuit
had minor complications including pancreatitis, that also includes an electrode needle, RF gen-
bleeding at the tumor site, and fluid collection or erator, and an electrode grounding pad [12].
pseudocyst [7]. Recently, Park et al. showed that The electrode delivers the RF energy to the tis-
EUS-guided ethanol ablation was successful in a sue, and, depending on the time of current
larger case series of 11 patients with 14 lesions application and the temperature achieved in the
with PNET [8]. The lesions were successfully tissue, it results in tissue necrosis. In bipolar
ablated in all patients, with seven lesions requiring RFA, the current oscillates between two inter-
only a single session. Three patients developed stitial nodes, thereby avoiding the need for a
self-limiting pancreatitis. grounding pad [13].
EUS-guided ethanol injections have been suc- Traditionally, RFA was done under CT or
cessfully used to treat other malignancies includ- ultrasound (US) guidance, externally. With the
ing a gastrointestinal stromal tumor in one patient advent of modern tools and techniques, RFA is
[9] and adrenal metastatic disease in a patient now done under EUS guidance (Video 11.1)
with lung cancer [10]. While this early evidence (Fig. 11.2).
is promising, we still need large-scale clinical tri- Currently, we have four different EUS-guided
als to understand its indications and complica- RF probes for the pancreas. They can be broadly
11 EUS-Guided Therapies for Solid Pancreatic Tumors Including Drug Delivery and Brachytherapy 111

Fig. 11.2 (a) Needle electrode (EUS-guided radiofre- Handle of the needle electrode attached to the accessory
quency ablation). (b) Close-up of the tip of the needle channel of the echoendoscope. (e) Viva Combo RFA gen-
electrode showing the uncovered 1-cm tip. (c) Needle erator, front view. (f) Viva Combo RFA generator, side
electrode projecting from the echoendoscope tip. (d) view. (g) Viva pump

classified as a “through-the-needle” device and as Inc., Burlington, MA, USA), hybrid cryotherm
“EUS-FNA [fine-needle aspiration] needle-type” probe (HybridTherm, ERBE Elektromedizin
device. The needle-type RFA devices are rigid and GmbH, Tübingen, Germany), and EUSRA RF
resemble an EUS-FNA needle. They have a vari- electrode (STARmed, Koyang, Korea). The hybrid
able gauge (14–19G). Through-the-needle devices cryotherm probe is the only bipolar probe; the oth-
include Habib™ EUS-RFA catheter (EMcision ers are monopolar probes. All the RF probes are
Ltd., London, UK). The remaining three probes connected to their respective generators to deliver
are 19G EUS-FNA needle electrode (Radionics, accurate energy to the target lesion.
112 G. S. Kochhar and M. Wallace

The procedure is very similar to standard EUS cant post-procedure adverse events. Most recently,
procedure. The echoendoscope is inserted through Song et al. studied six patients with advanced
the esophagus into the stomach and duodenum. pancreatic cancer [21]. Song et al. used an 18-G
After the lesion is located, a 19- or 22-G FNA nee- needle electrode (STARmed), giving 20–50 W of
dle or RFA probe is inserted through the working energy for 10 s. The average number of EUS-RFA
channel of the echoendoscope into the target lesion. sessions in the group was 1.3, and necrosis was
The echogenic needle tip or probe is positioned at observed in all patients at the ablation site, with a
the far end inside the lesion. After confirming the mean ablation size of 38 mm. No major adverse
exact location, energy is delivered to the target events were reported in this study.
lesion. After a slight lag, one can start seeing echo- EUS-RFA has also been used to treat pancre-
genic bubbles at the target site. The wattage and atic cystic neoplasms. Pai et al. performed one
exposure time for the lesions has not yet been stan- such study in eight patients with pancreatic
dardized. However, in pilot studies, RF energy was cysts [6]. Four patients had a mucinous cyst,
applied for 90–120 s at the 5- to 25-W setting [14, two patients had pancreatic neuroendocrine
15]. The ablation was repeated two to six times in tumors (one had intraductal papillary mucinous
each session in prior clinical studies. neoplasm (IPMN), and one had a microcystic
Goldberg et al. described the first experience adenoma). They used Habib EUS-RFA needles
with EUS-RFA in 1999, in porcine models [16]. at 5–25 W, with exposure time ranging from 90
In 2008, Carrara et al. used a cryotherm probe to to 120 s. The mean number of RFA sessions was
do EUS-guided RFA of solid organs like the liver, 4.5 (range, 2–7), and at the 10-week follow-up,
spleen, and pancreas in pigs [17]. In 2009, two cysts were completely resolved, while four
Varadarajulu et al. performed EUS-RFA of the were reduced in size, and there was a 50%
liver using an umbrella-shaped monopolar retract- reduction in size in patients with PNET. Only
able electrode array in five pigs [18]. This device two patients reported mild abdominal pain in
is similar to RFA devices used by interventional the study. Recently, Lakhtakia et al. reported
radiology. This technique was used to provide a treating symptomatic insulinoma with EUS-
large area of coagulative necrosis. No complica- RFA [22]. They used 19G needles (STARmed),
tions arose from the procedure. The mean zone of at 50 W for 10–15 s. The average ablation size
ablation was 2.6 cm. These early animal studies was 19 mm. Treatments were successful in all
paved the way for human use. three patients; they had no more hypoglycemia
In one such study, Arcidiacono et al. performed symptoms during the 12-month follow-up
EUS-RFA in 22 patients with advanced metastatic period.
pancreatic cancer [19]. They used a cryotherm Overall, EUS-guided RFA seems to be a very
probe with 18 W of energy and 650 psi. The aver- promising therapy in the management of pancre-
age RFA time was 107 s. They found that 16 atic neoplasms. Its role in PNET is even more
patients had significant volume reduction in the encouraging (Figs. 11.3 and 11.4). The above
lesions. No major complications were observed in data suggest that EUS-guided RFA is safe and
the study. The procedure failed in six patients due can potentially become a mainstream therapy in
to the excessive thickness of the stomach wall and the management of pancreatic cancers. Although
tumor. The median survival time was 6 months in the initial results are very encouraging, there are
the study. In another study, Pai et al. included still a few limitations to its widespread use.
seven patients with advanced pancreatic adeno- Further technological advancements in needles
carcinoma. The target lesions were predominantly are necessary for easy tumor penetration.
located in the head of the pancreas (in five Sometimes the flexible cryotherm probe poses a
patients) [20]. RF was applied at 5–15 W, with a challenge in piercing the tumor. We also need
mean duration of 90 s. In follow-­up examinations, more data on wattage setting and the number of
the size of the lesion was reduced in two of seven RFA sessions required for different types of
patients. Again, researchers reported no signifi- pancreatic neoplasm. Future studies will also
11 EUS-Guided Therapies for Solid Pancreatic Tumors Including Drug Delivery and Brachytherapy 113

Fig. 11.3 (a) Abdominal contrast-enhanced CT in the arterial phase shows an enhancing lesion (insulinoma) in the
pancreatic genu (arrow). (b) Well-defined hypoechoic oval-shaped lesion (insulinoma) in the pancreatic genu (arrow)

[23]. It successfully achieved localized tissue

necrosis and tumor destruction without major
complications. Permanent seeds of iodine (I125) or
palladium (Pd103) can be easily planted in the
tumor using EUS (Figs. 11.5 and 11.6). There are
several advantages to the procedure. First, the
procedure can be done in an outpatient setting.
Locally placed iodine or palladium beads emit
low-dose radiation; hence, they do not require
extensive precautions or preparations at home.
The success of the EUS-guided brachytherapy
has been reported in various studies involving
cancers of the head and neck and in pancreatic
cancers [24, 25]. In a pilot study, patients with
Stage III and IV pancreatic adenocarcinoma were
Fig. 11.4 Post-EUS-guided radiofrequency ablation at selected. They underwent EUS-guided brachy-
6 weeks
therapy with iodine beads (I125) [26]. A mean of
hopefully better define the role of EUS in the 22 beads were placed per patient. Thirty-three
management of pancreatic cancer. From its current percent of the tumors stabilized after therapy, and
use of palliation, EUS-guided therapy might 30% of the patients experienced pain relief from
become a first-line treatment strategy to downgrade the procedure. No major adverse outcomes were
and/or debulk tumors. reported. In another trial, EUS-guided
­brachytherapy was performed in patients with
unresectable pancreatic cancer [27]. A total of 85
EUS-Guided Brachytherapy patients were enrolled in the trial, and they
received gemcitabine chemotherapy in conjunc-
Brachytherapy is a well-known treatment strategy tion with RFA. Patients showed longer median
for the management of various solid organ tumors survival up to 7.8 months after implantation com-
like prostate cancer. Its role in the gastrointestinal pared with patients who were not treated (median
tract, with the help of EUS, was first described in survival, 4 months). Again, no major side effects
animal models with pancreatic cancer in 2007 were reported from the trial.
114 G. S. Kochhar and M. Wallace

Fig. 11.5 New 22-gauge fiducial marker needle device that preloads four markers into the needle for sequential deploy-
ment (Cook Medical, Winston-Salem, NC, USA)

EUS-Guided Fiducial Placements

Fiducials are radiopaque markers used to direct

radiation therapy. Fiducial markers have distinct
advantages over traditional radiotherapy in that
they allow delivery of large doses of radiation pre-
cisely at the target tissue without damaging adja-
cent tissues. These markers are also traceable
during patient respirations, which allow health care
providers to avoid damaging the surrounding tis-
sues during radiation therapy. While CT guidance
can place fiducials around pancreatic malignancy,
EUS-guided placement is far more precise [28].
Currently, two types of fiducial markers are
commercially available [29]. The standard fidu-
cials are gold seeds measuring 3 or 5 mm in
length and 0.8–1.2 mm in diameter, which require
Fig. 11.6 Fluoroscopic image of four markers placed
into the target tissue in the pancreas using new 22-gauge a 19-G EUS needle for deployment. The newer
needle device (Cook Medical) fiducials are Visicoil™ fiducials measuring
11 EUS-Guided Therapies for Solid Pancreatic Tumors Including Drug Delivery and Brachytherapy 115

Fig. 11.7 Endosonography-­guided fiducial placement. Three fiducials are seen (red arrows) within a hypoechoic mass
previously determined to be pancreatic adenocarcinoma

10 mm in length and 0.35 mm in diameter (Core

Oncology, Santa Barbara, CA), which can be Other EUS-Guided Therapies
deployed via 22-G EUS needles.
Fiducials can be deployed in two ways. In the Cytoimplant
front-loading technique, the needle tip is
advanced to the tumor, then the stylet is removed, Cytoimplants are an allogenic mix of lymphocytic
and the fiducials are loaded manually into the tissue obtained from the tumor patient and a healthy
needle lumen. The stylet is reinserted in the nee- donor after coincubation. Chang et al. initially
dle channel, then the stylet is advanced, pushing described this technique in a Phase I clinical trial
the fiducial through the needle lumen, with ulti- [32]. Eight patients with unresectable pancreatic
mate deployment of fiducials in the target tissue. adenocarcinoma underwent EUS-­guided fine-nee-
In the back-loading technique, the fiducials dle injection (FNI) of cytoimplants. The median sur-
are inserted into the needle tip with the stylet vival was 13.2 months. There were no immediate
slightly withdrawn followed by insertion of bone post-­procedural complications. This technique has
wax at the tip to prevent premature dislodgement. not gained much popularity due to reports of severe
The needle tip is then advanced into the tumor, drug-related sepsis but has at least shown technical
and the stylet is advanced with the deployment of feasibility. A Phase II/III trail comparing gem-
the markers. Recently, Park et al. described a citabine with cytoimplant was prematurely stopped
technique using saline flushed in the needle due to a better response rate and survival benefit in
lumen to deploy the fiducials instead of using a patients undergoing gemcitabine therapy [33].
stylet [30]. Based on the size of the tumor, 2–6
markers are placed at the margins of the tumor.
EUS-guided placement of fiducials has been Gene Therapy
reported to be successful in 85–100% of cases
(Fig. 11.7) [31]. Failures are encountered in cases ONYX-015 is an E1B-55kD gene-deleted
where it is difficult to advance the echoendoscope, replication-­selective adenovirus that preferen-
or there is a failure to maneuver the 19-G needle in tially replicates inside malignant tumor cells and
especially difficult locations like the uncinate pro- causes cell death [34]. The use of this agent was
cess where the scope is in a rotated position. shown in one study including 21 patients with
Fiducial placement is an overall safe procedure advanced pancreatic cancer with no liver metasta-
with few side effects. Common complications ses [35]. The study participants underwent eight
include minor bleeding and fiducial migration [31]. EUS-guided injections; the last four of which
116 G. S. Kochhar and M. Wallace

Fig. 11.8 Endosonography (EUS)-guided pancreatic No residual tumor was found at the time of surgical resec-
tumor injection with TNFerade. (a) Tumor size is 3.9 cm tion. (Reprinted with the permission from Springer: From
at baseline (week 1) before treatment with EUS-guided Chang KJ, Lee JG, Holcombe RF, et al. Endoscopic ultra-
gene therapy. (b) Tumor size has decreased to 2.8 cm after sound delivery of an antitumor agent to treat a case of
1 week. (c) One month after completing treatment, the pancreatic cancer. Nat Clin Pract Gastroenterol Hepatol
tumor size had decreased to 1.8 cm and a fine-needle aspi- 2008;5(2):107–11)
ration performed at that time was negative for malignancy.

were combined with gemcitabine (1000 mg/m2). the tumor along with standard chemoradiotherapy
The results of the trial were mixed. Two patients (Fig. 11.8) [36]. The major advantage of this
had partial regression, two had a minor response, approach is the potential to use anti-TNF-α locally
six patients’ disease stabilized while in 11 patients without systemic side effects. In a recent study by
the disease progressed. Two patients developed Hecht et al., 50 patients with advanced pancreatic
sepsis, and two patients experienced duodenal adenocarcinoma underwent TNFerade therapy via
perforations; this led to a change in the trial proto- EUS-guided (n = 27) and percutaneous injection
col, and injections were then administered using a (n = 23) [36]. The study aimed to determine the
transgastric approach rather than transduodenal. maximally tolerated dose, safety, and feasibility of
No patient developed pancreatitis, although eleva- TNFerade with chemoradiotherapy. Over a 5-week
tions in lipase levels were observed in patients treatment period, weekly intratumoral injections
after the procedure. of TNFerade (4 × 109, 4 × 1010, and 4 × 1011 parti-
TNFerade is the newest EUS-guided antitumor cle units in 2 mL) were given in combination with
therapy. A local injection of TNFerade allows intravenous 5-­ fluorouracil (200 mg/m2/day,
delivery of tumor necrosis factor-α (TNF-­α) into 5 days/week) and radiation (50.4 Gy). The long-
11 EUS-Guided Therapies for Solid Pancreatic Tumors Including Drug Delivery and Brachytherapy 117

term results showed that toxicities potentially Conclusion

related to TNFerade were mild, and the procedure
was well tolerated with only two reported cases of EUS has come a long way from its first use as a
acute pancreatitis. The higher dose group (n = 11) tool for diagnosing pancreatic malignancies to a
was associated with higher locoregional control of means to treat them now. The EUS-FNI tech-
the tumor with a longer disease-free survival. Four nique is well-described and safe. With direct
patients became surgically resectable and achieved tumor-guided therapy with EUS FNI, many of
pathologically negative margins, and three patients the systemic side effects of other therapies can be
survived more than 24 months. avoided. Although, to date, the data are very
encouraging for the management of pancreatic
cancer with EUS-guided therapies, its use is still
Immunotherapy limited due to a lack of large randomized clinical
trials. More studies with larger numbers of
Dendritic cells (DCs) are potent antigen-­presenting patients are needed to study exact indications and
cells that can stimulate a T-cell-­dependent immune safety profiles of these interventions to define a
response. In a pilot study, seven patients with stage more clear role for EUS-guided therapies in the
IV pancreatic cancer refractory to gemcitabine ther- management of pancreatic cancer.
apy received intratumor injections of immature
DCs by EUS-FNI [37]. DCs were administrated Conflict of Interest The authors declared no
every 7 days and the number of EUS-FNIs ranged conflict of interest.
from 2 to 21. All injections were well tolerated
without significant complications. The median sur-
vival was 9.9 months with two patients having a
mixed response, two patients had stable disease, References
and three patients had disease progression. While
1. Shiina S, Teratani T, Obi S, Hamamura K, Koike Y,
immunotherapy is an exciting prospect for manag-
Omata M. Percutaneous ethanol injection therapy for
ing cancer, additional studies with more effective liver tumors. Eur J Ultrasound. 2001;13:95–106.
antitumor agents are needed for pancreatic cancer. 2. Barclay RL, Perez-Miranda M, Giovannini M. EUS-­
guided treatment of a solid hepatic metastasis.
Gastrointest Endosc. 2002;55:266–70.
3. Aslanian H, Salem RR, Marginean C, et al. EUS-­
Photodynamic Therapy guided ethanol injection of normal porcine pancreas:
a pilot study. Gastrointest Endosc. 2005;62:723–7.
Photodynamic therapy (PDT) was initially suc- 4. Giday SA, Magno P, Gabrielson KL, et al. The utility of
contrast-enhanced endoscopic ultrasound in monitoring
cessful in managing advanced cholangiocarci-
ethanol-induced pancreatic tissue ablation: a pilot study
noma, but its use in pancreatic cancer in humans in a porcine model. Endoscopy. 2007;39:525–9.
is still in its infancy. In a pilot study, Choi et al. 5. Jurgensen C, Schuppan D, Neser F, et al. EUS-guided
included four patients with advanced pancreati- alcohol ablation of an insulinoma. Gastrointest Endosc.
2006;63:1059–62.
cobiliary cancer [38]. Patients underwent EUS-­
6. Deprez PH, Claessens A, Borbath I, Gigot JF, Maiter
guided PDT with a chlorin e6 derivative D. Successful endoscopic ultrasound-guided ethanol
(Photolon, Belmedpreparaty, Minsk, Republic of ablation of a sporadic insulinoma. Acta Gastroenterol
Belarus) and a flexible laser-light catheter (Photo Belg. 2008;71:333–7.
7. Levy M, Topazian M. Sp694 ultrasound guided etha-
Glow, South Yarmouth, MA, USA). The proce-
nol ablation of insulinomas: a new treatment option.
dure was safe, feasible, and all patients showed Gastrointest Endosc. 2011;73:AB102.
stable disease at the end of the 5-month follow- 8. Park DH, Choi J-H, Oh D, et al. Endoscopic
­up period. It was shown to be a feasible proce- ultrasonography-­ guided ethanol ablation for small
pancreatic neuroendocrine tumors: results of a pilot
dure in all patients. Additional studies and data
study. Clin Endosc. 2015;48:158–64.
are needed for the use of PDT in pancreatic can- 9. Gunter E, Lingenfelser T, Eitelbach F, Muller H, Ell
cer management. C. EUS-guided ethanol injection for treatment of a
118 G. S. Kochhar and M. Wallace

GI stromal tumor. Gastrointest Endosc. 2003;57(1): 25. Sun S, Qingjie L, Qiyong G, Mengchun W, Bo Q,
113–5. Hong X. EUS-guided interstitial brachytherapy of
10. Artifon ELA, Lucon AM, Sakai P, et al. EUS-guided the pancreas: a feasibility study. Gastrointest Endosc.
alcohol ablation of left adrenal metastasis from non-­ 2005;62:775–9.
small-­cell lung carcinoma. Gastrointest Endosc. 26. Sun S, Xu H, Xin J, et al. Endoscopic ultrasound-­
2007;66:1201–5. guided interstitial brachytherapy of unresectable pan-
11. Lakhtakia S, Seo D. Endoscopic ultrasonography-­ creatic cancer: results of a pilot trial. Endoscopy. 2006;
guided tumor ablation. Dig Endosc. 2017;29:486–94. 38:399–403.
12. Ahmed M, Brace CL, Lee FT, Goldberg SN. Principles 27. Jin Z, Du Y, Li Z. Su1575 long-term effect of gem-
of and advances in percutaneous ablation. Radiology. citabine-combined endoscopic ultrasonography-­guided
2011;258:351–69. brachytherapy in pancreatic cancer. Gastrointest
13. McGahan JP, Gu W-Z, Brock JM, Tesluk H, Darryl Endosc. 2013;77:AB373.
Jones C. Hepatic ablation using bipolar radiofre- 28. Pishvaian AC, Collins B, Gagnon G, Ahlawat S,
quency electrocautery. Acad Radiol. 1996;3:418–22. Haddad NG. EUS-guided fiducial placement for
14. Pai M, Yang J, Zhang X, et al. PWE-055 endoscopic CyberKnife radiotherapy of mediastinal and abdominal
ultrasound guided radiofrequency ablation (EUS-­ malignancies. Gastrointest Endosc. 2006;64(3):412–7.
RFA) for pancreatic ductal adenocarcinoma. Gut. 29. Mukewar S, Muthuswamy V. Recent advances in
2013;62:A153. therapeutic endosonography for cancer treatment.
15. Pai M, Senturk H, Lakhtakia S, et al. 351 endo- Gastrointest Endosc Clin N Am. 2017;27(4):657–80.
scopic ultrasound guided radiofrequency ablation 30. Park WG, Yan BM, Schellenberg D, et al. EUS-­guided
(EUS-RFA) for cystic neoplasms and neuroendocrine gold fiducial insertion for image-guided radiation ther-
tumors of the pancreas. Gastrointest Endosc. 2013; apy of pancreatic cancer: 50 successful cases without
77:AB143–4. fluoroscopy. Gastrointest Endosc. 2010;71(3):513–8.
16. Goldberg SN, Mallery S, Gazelle GS, Brugge 31. Dhadham GC, Hoffe S, Harris CL, et al. Endoscopic
WR. EUS-guided radiofrequency ablation in the pan- ultrasound-guided fiducial marker placement for
creas: results in a porcine model. Gastrointest Endosc. image-guided radiation therapy without fluoroscopy:
1999;50:392–401. safety and technical feasibility. Endosc Int Open.
17. Carrara S, Arcidiacono PG, Albarello L, et al. 2016;4(3):E378–82.
Endoscopic ultrasound-guided application of a new 32. Chang KJ, Nguyen PT, Thompson JA, et al. Phase I clin-
hybrid cryotherm probe in porcine pancreas: a pre- ical trial of allogeneic mixed lymphocyte culture (cyto-
liminary study. Endoscopy. 2008;40:321–6. implant) delivered by endoscopic u­ltrasound-­ guided
18. Varadarajulu S, Jhala NC, Drelichman ER. EUS-­guided needle injection in patients with advanced pancreatic
radiofrequency ablation with a prototype elec- carcinoma. Cancer. 2000;88:1325–35.
trode array system in an animal model (with video). 33. Levy MJ, Alberts SR, Bamlet WR, et al. EUS-­
Gastrointest Endosc. 2009;70:372–6. guided fine-needle injection of gemcitabine for
19. Arcidiacono PG, Carrara S, Reni M, et al. Feasibility locally advanced and metastatic pancreatic cancer.
and safety of EUS-guided cryothermal ablation in Gastrointest Endosc. 2017;86(1):161–9.
patients with locally advanced pancreatic cancer. 34. Han J, Chang K. Endoscopic ultrasound guided direct
Gastrointest Endosc. 2012;76:1142–51. intervention for solid pancreatic tumor. Clin Endosc.
20. Pai M, Habib N, Senturk H, et al. Endoscopic ultra- 2017;50:126–37.
sound guided radiofrequency ablation, for pancreatic 35. Hecht JR, Bedford R, Abbruzzese JL, et al. A phase
cystic neoplasms and neuroendocrine tumors. World I/II trial of intratumoral endoscopic ultrasound injec-
J Gastrointest Surg. 2015;7:52–9. tion of ONYX-015 with intravenous gemcitabine in
21. Song TJ, Seo DW, Lakhtakia S, et al. Initial expe- unresectable pancreatic carcinoma. Clin Cancer Res.
rience of EUS-guided radiofrequency ablation of 2003;9:555–61.
unresectable pancreatic cancer. Gastrointest Endosc. 36. Hecht JR, Farrell JJ, Senzer N, et al. EUS or percuta-
2016;83:440–3. neously guided intratumoral TNFerade biologic with
22. Lakhtakia S, Ramchandani M, Galasso D, et al. 5-fluorouracil and radiotherapy for first-line treatment
EUS-guided radiofrequency ablation for management of locally advanced pancreatic cancer: a phase I/II
of pancreatic insuli-noma by using a novel needle study. Gastrointest Endosc. 2012;75(2):332–8.
electrode (with videos). Gastrointest Endosc. 2016; 37. Irisawa A, Takagi T, Kanazawa M, et al. Endoscopic
83:234–9. ultrasound-guided needle injection of immature den-
23. Sun S, Wang S, Ge N, et al. Endoscopic ultrasound-­ dritic cells into advanced pancreatic cancer refrac-
guided interstitial chemotherapy in the pancreas: tory to gemcitabine: a pilot study. Pancreas. 2007;
results in a canine model. Endoscopy. 2007;39:530–53. 35:189–90.
24. Maier W, Henne K, Krebs A, Schipper J. Endoscopic 38. Choi JH, Oh D, Lee JH, et al. Initial human experi-
ultrasound-guided brachytherapy of head and neck ence of endoscopic ultrasound-guided photodynamic
tumours. A new procedure for controlled application. therapy with a novel photosensitizer and a flexible
J Laryngol Otol. 1999;113:41–8. laser-light catheter. Endoscopy. 2015;47:1035–8.
 EUS-Guided Enhanced Imaging
and Sampling of Neoplastic 12
Pancreatic Cysts

Shivangi Kothari, Enqiang Linghu,

Truptesh H. Kothari, and Vivek Kaul

challenging and poses a clinical dilemma in

Introduction being able to distinguish neoplastic from
­nonneoplastic cysts. Distinguishing mucinous
Advances in recent imaging and the high cystic neoplasm (MCN) from nonmucinous
­number of cross-sectional imaging studies (CT cysts is extremely important given the malig-
and MRI) being performed for a variety of gas- nant potential of the mucinous lesions, includ-
trointestinal complaints has led to an increase ing intraductal papillary mucinous neoplasm
in detection of pancreatic cystic lesions. (IPMN), and to avoid unnecessary surgical
Approximately, 2.5% of the cross-sectional interventions on benign cysts. Current guide-
imaging studies report pancreatic cysts and lines recommend surgical resection for all sur-
this can be as high as 10% in patients older gically fit patients with MCNs, all patients
than 70 [1]. Pancreatic cysts can be seen in 3% with main duct-IPMNs and branch duct-­IPMNs
of CT scans and up to 20% of MRI [2, 3]. An with worrisome features (cyst ≥3 cm, thick
autopsy study on 300 elderly patients reported enhancing cyst wall, main duct 5–9 mm, mural
a 24.3% prevalence of pancreatic cysts [4]. The nodules, and positive cytology) [8]. Non­
rate of malignant transformation of IPMN mucinous cysts such as pseudocysts and serous
cysts can range from 10% to >70% depending cystadenomas are considered benign and do
on the presence of high risk features (mural not require continued surveillance or interven-
nodules, main duct involvement, multifocal tion in asymptomatic patients.
cysts, etc.) [5–8]. The management of these The limitations in achieving a definitive diag-
incidentally discovered pancreatic cysts is nosis of these cysts and the uncertainty regarding
the potential for malignant transformation of
these cystic lesions leads to patient anxiety,
Electronic supplementary material: The online version
of this chapter (https://doi.org/10.1007/978-3-319-97376- unnecessary surveillance, and even surgical inter-
0_12) contains supplementary material, which is available ventions (some of which will ultimately be found
to authorized users. to have been unnecessary), posing significant
S. Kothari (*) · T. H. Kothari · V. Kaul management challenges to the patient and the
Division of Gastroenterology and Hepatology, physician. Therefore, diagnostic tools that can
Department of Medicine, University of Rochester help differentiate these cysts are essential to the
Medical Center, Rochester, NY, USA
e-mail: [email protected] endosonographer to be able to accurately assess
these cystic lesions and only refer the truly malig-
E. Linghu
Department of Gastroenterology, Chinese People’s nant and high risk cystic lesions for surgery, while
Liberation Army General Hospital, Beijing, China avoiding surgery for the more benign lesions.

© Springer Nature Switzerland AG 2019 119

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_12
120 S. Kothari et al.

Endoscopic ultrasound (EUS) is widely used teomic analysis) are being studied and developed
in the evaluation of pancreatic cystic lesions but their widespread clinical use is yet to be
(PCL) but the sensitivity, specificity, and accu- established. In addition, the cost associated with
racy of EUS imaging alone in PCL evaluation has these additional tests, their availability, and addi-
been reported to be low [9]. EUS is operator tive value to currently available cytology needs to
dependent and has a low accuracy in differentiat- be evaluated.
ing mucinous from nonmucinous cysts based on All these factors have led to the development
imaging alone. Even among experienced endo- of new techniques to help overcome the limita-
sonographers, there is a poor rate of interobserver tions of EUS-FNA and also better characterize
agreement between neoplastic and nonneoplastic pancreatic cystic lesions. In this chapter, we will
pancreatic cystic lesions. Other than the serous discuss novel EUS- and FNA-based imaging and
cystadenoma with a classic “honey-comb” micro- tissue acquisition tools that can help clinicians
cystic appearance, differentiating premalignant better distinguish benign from neoplastic pancre-
cysts solely based on EUS imaging can be very atic cysts.
difficult.
According to the 2012 guidelines, the pres-
ence of “high-risk” features such as obstructive Contrast Harmonic EUS
jaundice in a patient with a cystic lesion of the
head of the pancreas, enhancing solid component Recently, contrast harmonic EUS (CH-EUS) has
within cyst, main pancreatic duct >10 mm in size, been reported as a useful adjunct in the evalua-
or “worrisome” features such as cyst >3 cm in tion and differential diagnosis of pancreatic solid
size, thickened/enhancing cyst walls, nonenhanc- tumors which has led to its application in the
ing mural nodule, main duct 5–9 mm in size, assessment of pancreatic cysts. CH-EUS uses a
abrupt change in caliber of pancreatic duct with microbubble-based contrast to enable evaluation
distal pancreatic atrophy, and lymphadenopathy of the microcirculation of lesions. A 2–5-micron
indicate higher risk of malignancy in a patient gas bubble core, which is stabilized by a shell, is
with a pancreatic cystic lesion [8]. Due to the used as the contrast. The injection of this IV con-
poor interobserver variability and low sensitivity trast is used to visualize the blood flow even in
of EUS imaging alone in differentiating benign very small vessels and that in turn allows for
from malignant PCLs, many patients require cyst evaluation of the vascularity of the cyst wall,
fine needle aspiration (FNA) for obtaining fluid mural nodules, and septa. This also helps differ-
for evaluation and analysis. Fluid CEA, amylase, entiate small neoplastic solid components in the
and cytology are the most commonly performed PCLs which would show signs of vascularization
tests on PCL fluid. However, cyst fluid cytology in comparison to the debris and mucus in a cyst
has limited diagnostic yield, with a recent meta-­ which would appear avascular on the CH-EUS
analysis showing a pooled sensitivity of 54%, but (Fig. 12.1).
a high specificity of 93%, in differentiating muci-
nous from nonmucinous cysts. High CEA levels
(> 192 ng/mL) are associated with mucinous Technique
cysts, with a meta-analysis reporting a 63% sen-
sitivity and 88% specificity for a high cyst fluid A 16- or 18-gauge IV is placed in the patient with
CEA level in the diagnosis of a mucinous pancre- a 3-way stopcock to avoid breaking down the
atic cyst [10]. However, a high cyst fluid CEA microbubbles in the contrast. The contrast is
level alone cannot help distinguish malignant injected followed by a saline flush. The area of
from benign cysts and thus has limited overall interest is imaged with the fundamental B mode
accuracy. imaging and then simultaneous imaging is per-
New intracystic markers (mutated KRSA formed on a split screen with the B mode imag-
DNA, mutated GNAS DNA, glucose, and pro- ing on one half of the screen and CH-EUS image
12 EUS-Guided Enhanced Imaging and Sampling of Neoplastic Pancreatic Cysts 121

Fig. 12.1 IPMN with nodule: nonenhancement seen on CH-EUS (Courtesy of Dr. Pietro Fusaroli, Italy)

on the other side. The arterial phase starts about associated with malignancy, particularly invasive
10–20 s after the injection of the contrast and cancer, 88.9% and 91.7% of times, respectively
lasts for about 30–45 s. After the arterial phase, [14]. In 2013, Yamashita et al. used CH-EUS
the venous phase persists for about 30–120 s dur- (with Sonazoid) in 17 patients with IPMN with
ing which there is progressive washout of the mural lesions. CE-EUS demonstrated vascularity
contrast [11, 12]. in all 12 cases with pathologically confirmed
During CH-EUS, the vascular portions of the mural nodules, whereas all four cases without
cyst are echogenic and the intracystic debris, mucin, vascularity had mucous “clots.” [15] One case of
and blood clots remain nonechogenic or invisible. If a cystic septum which was interpreted as a hyper-
the cyst or IPMN has a mural nodule, it is usually enhanced solid nodule on CH-EUS accounted for
difficult to distinguish the nodule (especially small a false-positive. The sensitivity, specificity, posi-
nodules) based on CT and MRI; however, CH-EUS tive predictive value, negative predictive value,
can help distinguish these nodules due to the echo- and accuracy of CH-EUS for mural nodule detec-
genicity and microvascular perfusion of the nodule tion were 100%, 80%, 92%, 100%, and 94%,
(Figs. 12.2 and 12.3 and Video 12.1). respectively. Five of the 12 mural nodules (three
The use of CH-EUS in evaluation of pancre- of them larger than 10 mm) in this study were not
atic cysts was first reported in 2009 [13]. In a detected by multidetector CT.
study of 87 patients with IPMN with mural nod- Studies of indeterminate pancreatic cysts have
ules, CH-EUS findings were compared with the reported that CH-EUS could not differentiate
pathologic findings. Mural nodules were classi- between serous and mucinous cysts due to the simi-
fied into four types based on the morphology: larity in enhancement of the cystic walls and septae.
type I: low papillary nodule, type II: polypoid However, CH-EUS was helpful in targeting FNA of
nodule, type III: papillary nodule, and type IV: cysts that revealed malignant nodule enhancement
invasive nodule. The study reported that in cysts and thereby helping avoid FNA of cysts
hyperenhanced nodules type III and IV were
­ containing mucus plug and debris [16, 17].
122 S. Kothari et al.

Fig. 12.2 Hyperenhancement seen on CH-EUS in serous cyst (Courtesy of Dr. Pietro Fusaroli, Italy)

Fig. 12.3 IPMN nodule seen enhancing on CH-EUS (Courtesy of Dr. Pietro Fusaroli, Italy)
12 EUS-Guided Enhanced Imaging and Sampling of Neoplastic Pancreatic Cysts 123

CH-EUS has been shown to have a higher copy, cystoscopy, the cytobrush, and the use of
accuracy than B mode EUS imaging in d­ iagnosing intracystic biopsy forceps.
a malignant cyst with a mural nodule >4 mm in
height. Standard B mode imaging had a low spec-
ificity (40%) compared to CH-EUS (75%) in the  eedle-Based Confocal Laser
N
evaluation of mural nodules [18]. Endomicroscopy (nCLE)
In a study of time-intensity curve parameters
and evaluation of microvessel density of mural Probe-based confocal laser endomicroscopy
nodules, the diagnostic accuracy of CH-EUS in (Cellvizio, Mauna Kea Technologies, Paris,
differentiating the grade of dysplasia inside nod- France) has been used for real-time imaging at
ules (low-grade, intermediate grade vs. high-­ the microscopic level in Barrett’s esophagus and
grade dysplasia/carcinoma) was reported [19]. In in the biliary tree to evaluate for dysplasia and
30 patients with resected IPMNs (14 LGD/IGD, carcinoma [20, 21]. Recently, its application has
16 HGD/invasive carcinoma), the authors been extended in the evaluation of pancreatic
observed that the nodule/pancreatic parenchyma cystic lesions, using a submillimeter probe that
contrast ratio was significantly higher in the fits through the 19-G FNA needle [22] (Fig. 12.4).
HGD/invasive carcinoma group than in the LGD/
IGD group (p < 0.05). Technique
In summary, the preliminary experience shows A 19-gauge EUS-FNA needle is used in this pro-
that CH-EUS may not be able to universally dis- cedure. Ex vivo, the stylet of the needle is
tinguish between enhancing patterns of cystic
wall and septa but can help in differentiating
intracystic solid components to help detect malig-
nant cysts and may help identify prime targets for
FNA within PCLs.

New EUS-FNA-Based Technology

CT/MRCP characteristics, EUS-FNA, fluid anal-

ysis (CEA, amylase, and other markers), cytol-
ogy, fluid characteristics (viscosity), serum tumor
markers (CA 19-9), and changes in cyst size and/
or morphology over time are currently used to
help evaluate pancreatic cysts. This approach is
not always diagnostic and is limited at times in
allowing an accurate differentiation between the
various PCLs, especially in distinguishing
IPMNs from MCNs. It is for these reasons that
additional diagnostic modalities and newer EUS-­
FNA-­based platforms have been investigated for
facilitating characterization between mucinous
and nonmucinous cysts. These new approaches
may also help differentiate branch duct IPMN
and MCN. Some of the recently developed tech-
nologies that have been used in combination
EUS-FNA that will be discussed in this chapter Fig. 12.4 nCLE probe at FNA needle tip (Courtesy of
include needle-based confocal laser endomicros- Mauna Kea Technologies)
124 S. Kothari et al.

Fig. 12.5 EUS-nCLE evaluation of large pancreatic cyst with a mural nodule

removed and a proprietary locking device is Table 12.1 nCLE diagnostic criteria as proposed by
Krishna and Lee [26]
attached to the needle Luer Lock. The AQ-Flex-19
Cyst type nCLE features
nCLE probe is inserted into the needle and locked
IPMN • Finger-like projections
into a predetermined position, extending approx- • Dark rings
imately 2 mm from the beveled edge of the FNA • Parallel thick bands
needle. The probe is then retracted 1 cm into the • Absence of “superficial vascular
needle. Then, under real-time EUS guidance the network”
• Absence of “bright floating
cyst is punctured using the 19-gauge FNA heterogeneous particles”
needle. Mucinous • Solitary epithelial bands
Upon entering the cyst, the probe is slowly cystadenoma • Large caliber blood vessels
advanced into the needle, locked in place and • Clusters of bright particles
then real-time imaging of the cyst wall is per- Serous • “Superficial vascular network”
cystadenoma • Multiple blood vessels
formed in vivo. Intravenous injection of fluores- • Absence of finger-like
cein (2.5–5 mL of 10% fluorescein sodium) projections
immediately prior to the actual imaging is Pseudocyst • Clusters of bright, floating,
required to facilitate and enhance the image. heterogeneous particles
Fluorescein stains the vessels and helps delineate • Absence of finger-like
projections
the tissue structures. The nuclei are not stained
and appear as dark spots on the exam. The probe
is placed gently against the cyst wall without 2. Serous cystadenoma—Blood vessels are
pressure and various parts of the cyst wall and superficial and closer to the cystic lumen
mural nodules if present are evaluated using a (superficial vascular network) (Fig. 12.6 and
fanning approach by moving the FNA needle Video 12.1).
(Fig 12.5). The endomicroscopy images and vid- 3. Intraductal papillary mucinous neoplasm—
eos are then recorded. Finger-like “papillary” projections, dark ring
Diagnostic criteria for various pancreatic cysts with white core (cross-section), which corre-
as represented by nCLE examination: [23–25] spond to the villous changes of the intestinal-­
(Table 12.1 as described by Krishna and Lee [26]). type IPMN lesion and presence of fine caliber
vessels characterize benign IPMN compared
1. Mucinous cystadenoma—Large white or gray to dark clumps with neovascularization and
bands with rare vessels. Vessels are deeper in large vessels (>20 μ diameter) which repre-
the ovarian like stroma. sent malignant IPMN. (Fig. 12.7).
12 EUS-Guided Enhanced Imaging and Sampling of Neoplastic Pancreatic Cysts 125

4. Pseudocysts: Three types of structures are ately prior to the procedure. Technical feasibility
noted with nCLE (Fig. 12.8 and Video 12.1): to perform nCLE with good imaging was noted
(a) Heterogeneous floating bright particles in 17 out of 18 cases. Two patients (11.1%) devel-
(b) Small black floating particles oped post-procedure pancreatitis—the first
(c) Large, dark, round homogenous floating patient developed mild pancreatitis requiring a
structures short hospital stay and the second patient devel-
5. Cystic neuroendocrine tumor: Black neoplas- oped moderate pancreatitis requiring a 5-day
tic cell clusters with white fibrous areas. hospitalization. Out of the 17 patients, ten
patients had very good images, five had “moder-
The use of nCLE to evaluate pancreatic cysts ate” quality images, and two had “poor” images.
was first assessed in a porcine model in 2010. Overall, there were few technical difficulties with
The first human experience was reported by loading of the nCLE probe and performing nCLE
Konda et al. in 2011 to evaluate the feasibility of via the transduodenal approach.
nCLE in evaluation of PCLs [22]. Eighteen In 2013, Konda et al. conducted an interna-
patients were enrolled in the study (16 cysts and tional multicenter pilot study (INSPECT trial) to
two solid masses). Patients received intravenous develop descriptive criteria for the image inter-
injection of 2.5 mL of 10% fluorescein immedi- pretation of nCLE findings in various PCLs and
also to assess both safety and diagnostic potential
of nCLE in differentiating PCLs [23]. Sixty-six
patients at eight referral centers underwent nCLE
imaging of which 14 (21.2%) had confirmation
by surgical histopathology. Images from eight
patients were excluded due to insufficient infor-
mation for consensus reference diagnosis. Villous
structures could be identified in IPMNs as dem-
onstrated by INSPECT trial, which confirmed the
preliminary findings of the feasibility trial [22].
The presence of epithelial villous-like structures
on nCLE was strongly associated with neoplastic
cystic lesions (IPMNs, MCN, or adenocarci-
Fig. 12.6 Superficial vascular network (seen in serous noma). Patients who were identified with villous-
cystadenoma) (Courtesy of Mauna Kea Technologies) or finger-like structures via nCLE were felt to be

Fig. 12.7 Finger-like projections and dark ring with white core seen in IPMN (Courtesy of Mauna Kea Technologies)
126 S. Kothari et al.

lesions—DETECT trial (Diagnosis of Pancreatic

Cysts: EUS, Through-the-Needle Confocal Laser
Endomicroscopy, and Cystoscopy Trial) [24].
This was a single center study with 30 patients
with PCLs (only two had surgical histopathol-
ogy) who underwent dual modality evaluation;
cystoscopy performed with the spyglass probe
followed by nCLE. The main outcome evaluated
was achieving a firm clinical diagnosis of the
PCL, using a combination of cystoscopy and
nCLE. Clinical diagnoses were established with
high probability in 18 patients. Mucin seen dur-
ing cystoscopy and papillary projections seen via
nCLE were characteristic findings for mucinous
Fig. 12.8 Bright floating particles and lack of blood ves-
cystic lesions. The sensitivity of cystoscopy was
sels seen in a pseudocyst 90% (9/10) and that of nCLE was 80% (8/10),
and the combination yielded 100% sensitivity for
diagnosis of pancreatic cystic neoplasm. The pro-
likely to have IPMN despite equivocal fluid anal- cedures were technically successful with the
ysis and nondiagnostic cytology. This trial dem- exception of one probe exchange failure. nCLE
onstrated a sensitivity of 59%, specificity of had a specificity of 100%, PPV of 100%, NPV of
100%, positive predictive value of 100%, and a 80%, and accuracy of 89% for diagnosis of muci-
negative predictive value of 50% in differentiat- nous cysts. Two patients developed post-­
ing the different PCLs. Overall complication rate procedure pancreatitis (6.6%) requiring 4–5 days
was 9%, which included pancreatitis (3%, n = 2) of hospitalization without intensive care unit
(one patient developed mild and other patient admission or intervention (7%).
developed moderate pancreatitis), intracystic A recent multicenter study on the Clinical
bleeding (n = 3), and transient abdominal pain (n evaluation of nCLE in cystic lesions of the pan-
= 1). The lower rate of pancreatitis in this study creas (CONTACT) was published in 2015. This
was attributed to the investigators limiting the study was performed in two phases. Phase 1
cyst imaging time to <10 min. involved retrospective validation of specific
Apart from the potential for complications nCLE diagnostic criteria of pancreatic cysts and
(although typically mild and self-limited), one Phase 2 was the prospective validation of the spe-
major limitation of this technology is the ability to cific criteria. A superficial vascular network was
only evaluate a very small portion of the cyst wall, found to be a unique feature of serous cystadeno-
i.e., a small area on the contralateral side of the cyst mas in this study, of note. In the multicenter eval-
relative to the access point through the cyst wall. In uation of 31 patients with PCLs, six nonblinded
addition, there is a complete inability to image the investigators reviewed the nCLE sequences and
cyst wall adjacent to the entry point of the FNA identified the superficial vascular network as a
needle in the cyst. Also, the ultrathin straight gray single feature that was only present in serous
bands seen in serous cystadenoma are also seen in cystadenoma (SCN). For nCLE-based diagnosis
adenocarcinoma, representing the desmoplastic of SCN, the sensitivity, specificity, PPV, and
fibrous reaction. Other limitations include limited NPV were 69%, 100%, 100%, 82%, and 87%,
data, lack of datasets on large numbers of patients, respectively. The interobserver agreement for the
and an unknown inter-reader reliability. nCLE SCN findings was high (k = 0.77) [27].
In 2015, Nakai et al. assessed the feasibility, Thus, the specificity of nCLE in identifying
safety, and diagnostic yield of the combination of branch duct IPMN (finger-like papillae) and SCN
cystoscopy (using the spyglass probe) and nCLE (superficial vascular network) is high (nearing
in the clinical diagnosis of pancreatic cystic 100%); however, in the absence of these “classic”
12 EUS-Guided Enhanced Imaging and Sampling of Neoplastic Pancreatic Cysts 127

findings, the accuracy of nCLE evaluation This platform has been used to visualize the
remains low. Also, for MCN and pseudocysts, the contents of pancreatic cysts and also direct biop-
validation of the nCLE findings in large clinical sies. This platform allows for direct visualization
trials is lacking. of the cyst wall and contents to help distinguish
In conclusion, nCLE is a novel FNA-based between various PCLs. Its successful use in eval-
tool that can help classify certain PCLs with high uation of pancreatic cysts has been reported in
accuracy and represents a recent advance in this single case reports, case series, and also more
realm. There are some limitations and there is a recently in larger prospective studies [24,
learning curve for image interpretation, as well as 31–33].
cost associated with the technology. Pancreatitis,
although mild to moderate, remains a potential  echnique (as Described by Chai et al.)
T
risk. Future studies with higher volume of [33]
patients and long-term outcomes will help further The cyst of interest is evaluated with EUS and
clarify and validate the role of nCLE in pancre- punctured using a 19-gauge FNA needle and the
atic cyst evaluation. cyst fluid is aspirated. The color and turbidity of
the fluid is assessed. The authors graded the cyst
fluid from A to C based on clarity (A—clear
EUS-Guided Cystoscopy background, B—blurred background, C—back-
ground not visible). If the cyst fluid is turbid, then
The Spyglass Direct Visualization System saline injection is performed to replace the turbid
(Boston Scientific, Natick MA) has been used cyst fluid and facilitate the intracystic visualiza-
widely for various applications in the bile and tion with the fiber optic probe. Following this, the
pancreatic ducts for visualization, stone manage- fiber optic probe is advanced through the needle
ment, stricture evaluation, etc. [28–30]. It allows into the cyst to directly image and visualize the
for direct visualization and targeted biopsies and/ cystic contents and the cyst wall.
or therapy and ability to assess epithelial abnor- Intracystic imaging characteristics that have
malities. The first generation of this device uti- been evaluated include:
lized a 0.035″ wide fiber optic probe that can be
advanced through the lumen of a 19-gauge FNA 1. Blood vessels:
needle into a target structure for endoscopic eval- Blood vessels have been characterized as
uation. Of note, the second generation of the the thick main blood vessel and the branch
­spyglass device utilizes a digital imaging system vessels (Fig. 12.9 and Video 12.1) which are
and no longer uses this fiber optic probe. then subcategorized as:

Fig. 12.9 Blood vessels in wall of cyst on cystoscopy (Courtesy of Dr. Enqiang Linghu, China)
128 S. Kothari et al.

(a) Type I: Sparse tree-like branching pattern, DETECT trial [24]. In this study, cystoscopy had
seen in SCN (61.5%) a sensitivity of 90% with an accuracy of 83% in
(b) Type II: Dense grid-like pattern, seen in diagnosing mucinous cysts. If surgical pathology
MCN (66.7%) is used as a gold standard for mucinous cyst diag-
(c) Type III: Vine-like pattern, seen surround- nosis on cystoscopy, the criteria of finger-like
ing papilla like protrusions or partitions projections has a low sensitivity of 22%, accu-
(19.4%) racy of 42% but a 100% specificity [34]. Biopsy
2. Papillary protrusions: proven mucinous cystadenomas have also been
Papillary structures can be seen on cystos- reported to have smooth cyst walls on cystoscopy
copy and have been classified into two types: evaluations [32].
(a) Yellow-white: Seen in MCN or IPMN In a preliminary study of 43 patients with
(Fig. 12.10). PCLs using the single operator cholangioscopy
(b) Red: Seen mostly in IPMN due to a richer fiber optic probe for visualizing the cyst contents,
blood supply. no complications were seen with the platform
3. Imaging characteristics of various PCL: [33]. The study was performed in cysts >1 cm in
(a) SCN: Smooth cyst walls, mainly type I size and it provided the image interpretations of
like vessel distribution. Also, partitions the cystoscopy findings of various PCL that were
frequently seen within the cyst with type definitively diagnosed with histopathology. Cyst
II blood vessel distribution next to the fluid clarity is very important for visualizing the
partitions. cyst wall and contents and thus this platform
(b) MCN: Type II blood vessels distribution, requires removal of the turbid cyst fluid and
opaque cyst fluid. White-yellow deposits replacing it with saline to be able to visualize the
on cyst wall. cyst wall. In this study, the tree-like branching
(c) IPMN: White roe-shaped or red papilla-­ pattern of blood vessel distribution was found to
like structures can be seen. Fluid can be a common characteristic of a serous cystic
sometimes be white mucus or jelly like. neoplasm. Intracystic papillary structures were
(d) Pseudocyst: Yellow or back necrosis/ an important characteristic for diagnosing muci-
necrotic deposit can be seen in the cyst nous cystic neoplasms in this study.
wall with scant blood vessels. Flocculent Widespread application of cystoscopy is still
particles can be seen in the cyst. limited due to the cost of the cholangioscopy
probe, limited availability of the probe, need for
Cystoscopy has been combined with nCLE for saline/clear fluid in the cyst to facilitate visualiza-
evaluation of PCLs and the results reported in the tion, and lack of large prospective studies validat-

Fig. 12.10 White yellow deposits seen in mucinous cyst (Courtesy of Dr. Enqiang Linghu, China)
12 EUS-Guided Enhanced Imaging and Sampling of Neoplastic Pancreatic Cysts 129

ing the cystoscopy findings with the gold standard specimen. The brush is then removed and final
of surgical pathology. In addition, given the aspirate of the cyst with the needle is performed
advancement of this technology to a digital plat- to collapse the cyst and the cyst is completely
form, it is unknown for how much longer the ven- aspirated.
dor will continue to manufacture the fiber optic Studies have reported a higher yield of epithe-
probe. lial cells using the cytobrush compared with stan-
dard EUS-FNA for cystic lesions of the pancreas
(mean size >2 cm). Complication including GI
EUS-Guided Cytobrushing bleeding and pancreatitis has been reported with
the cytobrush [35, 36].
Another EUS-based platform to improve the In another study of 30 patients with cysts >15
yield of FNA in the pancreatic cyst evaluation mm, the technique failed in eight patients (27%). A
described in the literature is the cytology brush. cellular diagnosis was obtained using the brush in
A “through-the-needle” cytology brush system 20/22 cases (91%) and the EUS cytology brushing
(EchoBrush; Cook Endoscopy, Winston-Salem, was superior to the aspirated fluid for detecting
NC) that was FDA approved for cytology sam- diagnostic cells (73% vs. 36%, p = 0.08) and muci-
pling during EUS evaluation of cystic lesions of nous cells (50% vs. 18%, p = 0.016). However, the
the pancreas was developed (Fig. 12.11). procedure had a 10% complication rate with 13.6%
morbidity and 4.5% mortality [37].
Technique In 2011, Lozano et al. published their cyto-
After aspirating 50% of the cyst volume using a brush data with a total of 127 cystic lesions of the
standard 19-gauge needle FNA, the EchoBrush is pancreas from 120 patients. Mean size of the cys-
introduced into the needle and advanced into the tic lesions was 23.43 ± 21.67 mm. Diagnostic
cyst under EUS guidance. After ensuring that the material was obtained in 85.1% (40 of 47) cases
needle is in the cyst, the brush is moved back and using the cytobrush compared to the 66.3% (53
forth repeatedly for 30 s ensuring adequate tan- of 80) with conventional EUS-FNA (p < 0.05).
gential contact with the cyst wall. The brush can Three patients had self-limited intracystic bleed-
also be rotated on its axis to gain maximal c­ ontact ing and were observed in recovery room post-­
with the cystic wall and obtaining the cytology procedure, and then discharged home. One

Fig. 12.11 EUS cytology brush at tip of FNA needle and at the FNA needle handle (Courtesy of Cook Medical)
130 S. Kothari et al.

patient developed perigastric abscess, which Technique

required hospitalization [38]. The cyst of interest is imaged with EUS. The cyst
Despite encouraging results, the studies have is then punctured using a 19-gauge FNA needle
had mixed results due to the complications and and the stylet is withdrawn. Then, keeping the
technical failure rates. Development of a new and grasping jaws of the forceps in the closed posi-
improved commercially available cytobrushing tion, the microforceps is advanced slowly through
platform is required for clinical use in the future. the FNA needle to avoid kinking of the sheath.
Then, maintaining EUS visualization at all times
the forceps is slowly advanced through the nee-
EUS-Guided Cyst Biopsy dle and the jaws are then opened to capture the
desired tissue (Fig. 12.13 and Video 12.1). The
EUS-FNA-guided intracystic biopsy using a tissue grabbed with the forceps can dislodge the
220 cm × 0.8 mm biopsy forceps has been needle tip and thus it is very important for the
reported in 2010 in a pilot study of two patients endosonographer to maintain control of the nee-
with PCLs. Mucinous cystic neoplasm was diag- dle handle at all times and to keep the needle tip
nosed in both the patients; however, one patient and forceps under constant EUS vision. An
developed severe acute pancreatitis 1 month after extraction pick may be used to assist in specimen
the procedure which was possibly thought to be a removal from the forceps if necessary.
late complication of the procedure [32]. The lat- The microforceps has been used in many case
est FDA approved EUS-FNA-based pancreas reports and series of indeterminate pancreatic
cyst tissue sampling device is the disposable cysts and confirmatory diagnosis was established
Moray microforceps (US Endoscopy, Ohio, in these cases revealing IPMN, mucinous cyst,
USA). The device is 230 cm in length, with an and lymphoepithelial cyst [39–41]. In a case of a
open jaw width of 4.3 mm. The jaws are serrated 68-year-old male with an 8-cm pancreas cyst
and attached to a flexible 0.8-mm stainless steel with mixed solid/liquid components with
spring sheath for easy passage through a 19-gauge ­inconclusive CEA, amylase, and cytology, the
FNA needle and allowing its use in tortuous posi- microforceps was successful in obtaining ade-
tions (Fig. 12.12). quate tissue samples from the cyst wall and a

Fig. 12.12 EUS microforceps open and at tip of FNA needle (Courtesy of US Endoscopy)
12 EUS-Guided Enhanced Imaging and Sampling of Neoplastic Pancreatic Cysts 131

Fig. 12.13 Biopsy of pancreatic cyst wall using EUS microforceps (Courtesy of Dr. Harshit Khara, USA)

diagnosis of lymphoepithelial cyst was estab- References

lished and surgery was avoided [42].
1. de Jong K, et al. High prevalence of pancreatic
The preliminary limited data on the EUS cysts detected by screening magnetic resonance
microforceps with the case reports and case imaging examinations. Clin Gastroenterol Hepatol.
series is promising; however, large prospective 2010;8(9):806–11.
trials are needed to evaluate the safety and effi- 2. Megibow AJ, et al. The incidental pancreatic cyst.
Radiol Clin N Am. 2011;49(2):349–59.
cacy of the forceps and also to assess its diag- 3. Lee KS, et al. Prevalence of incidental pancreatic
nostic accuracy when compared to the gold cysts in the adult population on MR imaging. Am
standard with surgical histopathology of the J Gastroenterol. 2010;105(9):2079–84.
lesions. 4. Kimura W, et al. Analysis of small cystic lesions of
the pancreas. Int J Pancreatol. 1995;18(3):197–206.
5. Pergolini I, et al. Long-term risk of pancreatic
malignancy in patients with branch duct intraductal
Summary papillary mucinous neoplasm in a referral center.
Gastroenterology. 2017;153(5):1284–1294.e1.
6. Kamata K, et al. Value of EUS in early detection
With the rising frequency of incidental pancre- of pancreatic ductal adenocarcinomas in patients
atic cyst detection, accurate diagnosis and dif- with intraductal papillary mucinous neoplasms.
ferentiation of malignant from nonmalignant Endoscopy. 2014;46(1):22–9.
cysts continues to remain a challenge. There are 7. Fritz S, et al. Clinicopathologic characteristics of
patients with resected multifocal intraductal papil-
many new promising technologies and tech- lary mucinous neoplasm of the pancreas. Surgery.
niques to enhance image quality and improve 2012;152(3 Suppl 1):S74–80.
the diagnostic yield of EUS-FNA, thereby 8. Tanaka M, et al. International consensus guidelines
facilitating a more definitive and accurate eval- 2012 for the management of IPMN and MCN of the
pancreas. Pancreatology. 2012;12(3):183–97.
uation of pancreatic cystic lesions. Although 9. Brugge WR. The role of EUS in the diagnosis of
promising, all these platforms need large pro- cystic lesions of the pancreas. Gastrointest Endosc.
spective trials to determine their safety, effi- 2000;52(6 Suppl):S18–22.
cacy, reproducibility, and accuracy. At this 10. Thornton GD, et al. Endoscopic ultrasound guided
fine needle aspiration for the diagnosis of pancreatic
time, all these techniques are only adjuncts to cystic neoplasms: a meta-analysis. Pancreatology.
EUS-FNA and the role of the abovementioned 2013;13(1):48–57.
technologies from a cost–benefit standpoint and 11. Barr RG. Off-label use of ultrasound contrast
applicability to everyday practice is yet to be agents for abdominal imaging in the United States.
J Ultrasound Med. 2013;32(1):7–12.
established.
132 S. Kothari et al.

12. Alvarez-Sanchez MV, Napoleon B. Contrast-­ laser endomicroscopy (nCLE): proposition of a

enhanced harmonic endoscopic ultrasound imaging: comprehensive nCLE classification confirmed by
basic principles, present situation and future perspec- an external retrospective evaluation. Surg Endosc.
tives. World J Gastroenterol. 2014;20(42):15549–63. 2016;30(6):2603–12.
13. Kitano M, Takagi T, Sakamoto H, Das K, Komaki T, 26. Krishna SG, Lee JH. Appraisal of needle-based confo-
Noda K, Yamao K, Kudo M. Dynamic imaging by cal laser endomicroscopy in the diagnosis of pancreatic
contrast enhanced harmonic EUS with long-­lasting cysts. World J Gastroenterol. 2016;22(4):1701–10.
contrast: role in diagnosis of pancreatic tumors. 27. Napoleon B, et al. A novel approach to the diag-
Gastrointest Endosc. 2009;69:AB235. nosis of pancreatic serous cystadenoma: needle-­
14. Ohno E, et al. Intraductal papillary mucinous neo- based confocal laser endomicroscopy. Endoscopy.
plasms of the pancreas: differentiation of malignant 2015;47(1):26–32.
and benign tumors by endoscopic ultrasound findings 28. Kalaitzakis E, et al. Diagnostic and therapeutic util-
of mural nodules. Ann Surg. 2009;249(4):628–34. ity of single-operator peroral cholangioscopy for
15. Yamashita Y, et al. Usefulness of contrast-enhanced indeterminate biliary lesions and bile duct stones. Eur
endoscopic sonography for discriminating mural nod- J Gastroenterol Hepatol. 2012;24(6):656–64.
ules from mucous clots in intraductal papillary muci- 29. Nagayoshi Y, et al. Peroral pancreatoscopy
nous neoplasms: a single-center prospective study. using the SpyGlass system for the assessment
J Ultrasound Med. 2013;32(1):61–8. of intraductal papillary mucinous neoplasm
16. Hocke M, et al. Pancreatic cystic lesions: the value of the pancreas. J Hepatobiliary Pancreat Sci.
of contrast-enhanced endoscopic ultrasound to 2014;21(6):410–7.
influence the clinical pathway. Endosc Ultrasound. 30. Kurihara T, et al. Diagnostic and therapeutic single-­
2014;3(2):123–30. operator cholangiopancreatoscopy in biliopancreatic
17. Fusaroli P, et al. Contrast harmonic-endoscopic diseases: prospective multicenter study in Japan.
ultrasound is useful to identify neoplastic fea- World J Gastroenterol. 2016;22(5):1891–901.
tures of pancreatic cysts (with videos). Pancreas. 31. Antillon MR, et al. Taking SpyGlass outside the GI
2016;45(2):265–8. tract lumen in conjunction with EUS to assist in the
18. Kamata K, et al. Contrast-enhanced harmonic endo- diagnosis of a pancreatic cystic lesion (with video).
scopic ultrasonography for differential diagnosis of Gastrointest Endosc. 2009;69(3 Pt 1):591–3.
pancreatic cysts. Endoscopy. 2016;48(1):35–41. 32. Aparicio JR, et al. Direct intracystic biopsy and pancre-
19. Yamamoto N, et al. Contrast-enhanced harmonic atic cystoscopy through a 19-gauge needle EUS (with
endoscopic ultrasonography with time-intensity curve videos). Gastrointest Endosc. 2010;72(6):1285–8.
analysis for intraductal papillary mucinous neoplasms 33. Chai N, et al. Preliminary study of single-operator
of the pancreas. Endoscopy. 2016;48(1):26–34. cholangioscopy for diagnosing pancreatic cystic
20. Gaddam S, et al. Novel probe-based confocal laser lesions. Gastrointest Endosc. 2017;86(1):208–18.
endomicroscopy criteria and interobserver agreement 34. Tsujino T, Samarasena JB, Huang JY, Nakai Y, Lee
for the detection of dysplasia in Barrett’s esophagus. JG, Chang KJ. EUS-guided through-the-needle dual
Am J Gastroenterol. 2011;106(11):1961–9. imaging with confocal laser-induced endomicros-
21. Meining A, et al. Direct visualization of indetermi- copy (NCLE) and cystoscopy in pancreatic cysts:
nate pancreaticobiliary strictures with probe-based correlation with gold-standard surgical pathology.
­confocal laser endomicroscopy: a multicenter experi- Gastrointest Endosc. 2016;83(5S):AB 336.
ence. Gastrointest Endosc. 2011;74(5):961–8. 35. Al-Haddad M, et al. Safety and efficacy of cytol-
22. Konda VJ, et al. First assessment of needle-based ogy brushings versus standard fine-needle aspiration
confocal laser endomicroscopy during EUS-FNA in evaluating cystic pancreatic lesions: a controlled
procedures of the pancreas (with videos). Gastrointest study. Endoscopy. 2010;42(2):127–32.
Endosc. 2011;74(5):1049–60. 36. Al-Haddad M, Raimondo M, Woodward T,
23. Konda VJ, et al. A pilot study of in vivo iden- Krishna M, Pungpapong S, Noh K, Wallace
tification of pancreatic cystic neoplasms with MB. Safety and efficacy of cytology brushings
needle-based confocal laser endomicroscopy versus standard FNA in evaluating cystic lesions
under endosonographic guidance. Endoscopy. of the pancreas: a pilot study. Gastrointest Endosc.
2013;45(12):1006–13. 2007;65(6):894–8.
24. Nakai Y, et al. Diagnosis of pancreatic cysts: EUS-­ 37. Sendino O, et al. Endoscopic ultrasonography-­
guided, through-the-needle confocal laser-induced guided brushing increases cellular diagnosis of pan-
endomicroscopy and cystoscopy trial: DETECT creatic cysts: a prospective study. Dig Liver Dis.
study. Gastrointest Endosc. 2015;81(5):1204–14. 2010;42(12):877–81.
25. Napoleon B, et al. In vivo characterization of pan- 38. Lozano MD, et al. EchoBrush may be superior to
creatic cystic lesions by needle-based confocal standard EUS-guided FNA in the evaluation of cys-
12 EUS-Guided Enhanced Imaging and Sampling of Neoplastic Pancreatic Cysts 133

tic lesions of the pancreas: preliminary experience. 41. Shakhatreh MH, et al. Use of a novel through-the-­
Cancer Cytopathol. 2011;119(3):209–14. needle biopsy forceps in endoscopic ultrasound.
39. Coman RM, et al. EUS-guided, through-the-needle Endosc Int Open. 2016;4(4):E439–42.
forceps: clenching down the diagnosis. Gastrointest 42. Huelsen A, et al. Endoscopic ultrasound-guided,
Endosc. 2016;84(2):372–3. through-the-needle forceps biopsy in the assessment
40. Pham KD, et al. Diagnosis of a mucinous pancreatic of an incidental large pancreatic cystic lesion with
cyst and resection of an intracystic nodule using a prior inconclusive fine-needle aspiration. Endoscopy.
novel through-the-needle micro forceps. Endoscopy. 2017;49(S 01):E109–10.
2016;48 Suppl 1:E125–6.
 EUS-Guided Pancreatic Cyst
Ablation 13
Kristopher Philogene and William R. Brugge

carcinoembryonic antigen (CEA) and CA 19-9

Introduction have been used to differentiate a malignant ver-
sus benign mucinous lesion with variable results.
With the increasing use of cross-sectional imag- Cyst fluid analysis remains in widespread use
ing over recent years, there have also been an despite its clear limitations. A significant propor-
increasing number of incidental pancreatic cysts tion of cystic lesions are found to be indetermi-
discovered. The prevalence of incidental pancre- nate with regard to perceived risk of malignancy
atic cysts based on screening computed tomogra- or potential malignant transformation even after
phy (CT) is 2.6% and magnetic resonance rigorous investigation, including cystic fluid
imaging (MRI) is 13.5%, respectively [1, 2]. aspiration and analysis [3].
Although pancreatic cysts are usually found inci- Chiaro et al. released a study of patients who
dentally, there is a wide variety of histopathol- were diagnosed with pancreatic cysts and who
ogy, with some pancreatic cysts that are inherently had underwent surgical resection found that 8.5%
neoplastic. Some cystic lesions that have been of the patients who underwent surgery had expe-
historically classified by pathologists as neoplas- rienced an error in their preoperative diagnosis,
tic include mucinous cystic neoplasms (MCN) showing that these patients may not have required
and intraductal papillary mucinous neoplasms surgical intervention at all based off of the histol-
(IPMN). What is essential when it comes to ogy of their lesions [4]. Although surgical resec-
knowing what histological cystic lesions have tion has a mortality rate of <1% (with the most
malignancy potential is being able to differenti- commonly resected lesion being IPMN) compli-
ate between a benign and malignant cyst [1]. cations developed in up to 18% of patients, with
Cyst fluid analysis for certain tumor markers like pancreatic fistula being the most common com-
plication, underscoring the relatively high-risk
Electronic supplementary material: The online version nature of pancreatic surgery (even in the modern
of this chapter (https://doi.org/10.1007/978-3-319-97376- era). The 90-day mortality rate in patients who
0_13) contains supplementary material, which is available required a total pancreatectomy was also as high
to authorized users.
as 7% [5, 6]. Therefore, a clinical decision often
K. Philogene needs to be made on whether to observe the cyst
Department of Medicine, Mt. Auburn Hospital,
or go through with surgical resection of an inde-
Cambridge, MA, USA
terminate incidental pancreatic cyst, which can
W. R. Brugge (*)
be a difficult decision to make.
Department of Gastroenterology, Mt. Auburn
Hospital, Cambridge, MA, USA Because of the unclear management strategy
e-mail: [email protected] for these patient populations, there was an

© Springer Nature Switzerland AG 2019 135

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_13
136 K. Philogene and W. R. Brugge

important need to find a safe and effective treat- (Fig. 13.1). Cyst fluid should be evacuated from the
ment modality for pancreatic cysts. Endoscopic cyst prior to injection and sent for cyst fluid analy-
ultrasonography-­ guided fine needle aspiration sis. Subtotal evacuation of the cyst is followed by
(EUS-FNA) of cystic fluid has been a widely injection with the volume of fluid removed equal-
used diagnostic tool for pancreatic cyst analysis ing the volume of fluid being replaced. The cyst is
and identification. With EUS-FNA, pancreatic then allowed to undergo lavage with ethanol for
cyst evaluation then turned towards manage- 3–5 min. The ethanol is typically aspirated as much
ment and treatment through pancreatic cyst as possible from the cyst after lavage (Fig. 13.2 and
ablation by EUS-guided ethanol injections Video 13.1). If a chemotherapeutic agent is being
along with injections of other potentially abla- injected following an ethanol injection into the
tive agents. There are several accounts and stud- cyst, the chemotherapeutic agent should be injected
ies that have shown that ethanol ablation through and left within the cyst, with the total volume of
EUS-guided injections can be performed safely injection not exceeding the amount of fluid that
with few complications, particularly evidenced was aspirated. To avoid leakage within the cyst
by the successful ablation attempts on insulino- wall or parenchymal injury, the needle tip must be
mas, thyroid nodules, splenic, liver, and renal maintained in the cyst cavity. After lavage and aspi-
cysts [7–10]. The use of antitumor agents during ration is complete, the needle is then removed from
these injections has also been found to be safe the cyst cavity.
and feasible in treating pancreatic cancer [11–
13]. EUS-guided pancreatic cyst ablation has
been studied as a possible alternative to surgical Ablative Agents
resection by several clinical trials with special
attention on ethanol injections and the use of Ethanol has been the most frequently used abla-
Paclitaxel injections in conjunction with ethanol tive agent for cyst ablations because of its low
injections. The purpose of this review is to cost compared to other agents. Ethanol has an
examine the procedural basics, safety, and effi- extremely low cost and it has a low viscosity that
cacy of cyst ablation along with the treatment makes it easy to inject when using a small gauge
response and future application that derives needle. In hepatic cyst injections, ethanol can
from this treatment modality. lead to cell membrane lysis, protein denaturation,
and vascular occlusion [14, 15]. It is felt that sim-
ilar benefits can be seen in pancreatic cyst
 ow to Perform an EUS-Guided Cyst
H injections.
Ablation Paclitaxel is a widely used chemotherapeutic
agent whose mechanism of action is to bind to
The equipment that is required to assess the inter- the β-subunit of the tubulin protein leading to sta-
nal structure of the pancreatic cyst must be able to bilization of microtubules ultimately inhibiting
determine the number of septations within the cyst, normal mitotic spindle formation [16]. Paclitaxel
the presence of a mass or nodule, and the overall is hydrophobic and viscous, which reduces the
wall thickness of the cyst. To determine the struc- risk of leakage within the cystic cavity when used
ture of the cyst, traditional EUS imaging utilized a as an ablative agent. However, given the high vis-
radial scanning echoendoscope. The curvilinear- cosity of paclitaxel and its cosolvent, castor oil,
array echoendoscope with 7.5-MHz transducer is paclitaxel has to be diluted 1:1 with 0.9% normal
also an instrument that can be used to image pan- saline solution for injection. There is another for-
creatic cysts, and is favored by many operators. mula of the paclitaxel solution that uses a poly-
Both have high-resolution imaging capabilities. meric micelle that is a less viscous delivery
Curvilinear-array echoendoscopes are used to mechanism and can be administered without
puncture the cyst through a transgastric or trans- dilution [17]. Paclitaxel injection is less common
duodenal route using an EUS aspiration needle than ethanol injection, of note.
13 EUS-Guided Pancreatic Cyst Ablation 137

Fig. 13.1 Technique summary of EUS-guided injection and lavage of a pancreatic cyst

Fig. 13.2 Linear EUS image of ethanol lavage before (a) and after injection (b): note the presence of injected bubbles
of ethanol in the cyst fluid

Special Considerations was predictive of complete resolution of

for EUS-­Guided Cyst Ablation pancreatic cysts that underwent ablation. Cyst
­
injection therapy and its overall efficacy are also
The morphologic characteristics of a pancreatic affected by the number of loculations that the cyst
cyst will guide the approach of how to best man- may have as well as the number of septations. A
age these cysts through injection and ablation unilocular or oligolocular cyst with 2–3 locules
while also giving a better sense of efficacy and has the greatest chance of having a successful
resolution of a cyst. An initial cyst size of <35 mm ablation with first needle pass as the entire cyst
138 K. Philogene and W. R. Brugge

can be accessed via a single injection (especially have shown that pancreatitis is a relatively rare
if septations are perforate in nature). A second adverse effect of cyst ablation ranging from 2%
needle pass may be required in cysts that have to 10% of patients [19].
more than 2–3 locules [18]. When all locules can- Portal vein thrombosis has also been seen as a
not be visualized through endoscopic imaging, complication, evidenced by a case report of a
needle passage through a septation may be indi- 68-year-old woman who underwent her second
cated. To determine good distribution of the abla- pancreatic cyst ablation and was found to have
tive agents within the cyst after needle passage portal vein thrombosis on CT imaging. Portal
through a septation, there will be formation of vein thrombosis can be precipitated by local
echogenic bubbles across the septation along with inflammation seen in pancreatitis and diverticuli-
collapse of the locules. Sometimes, if a locule is tis. EUS-guided pancreatic cyst ablation induces
missed, there may be regrowth of the cyst, indi- inflammation locally within the cyst leading to
cating inadequate treatment or treatment failure. atrophy of the epithelial lining of the cyst
It is important to determine the optimal needle (Fig. 13.3). With that, however, cyst ablation can
angle in order to maximize the number of targeted lead to extensive inflammation around the cyst and
locules with the fewest passes as there are associ- within the cyst. Splenic vein thrombosis/oblitera-
ated risks with injection therapy. Pancreatitis tion is also another rare complication of cyst abla-
related to the ablative agent, particularly if there is tion, seen in a prospective double-blind randomized
a communication between the main pancreatic control trial where one patient developed splenic
duct and the cyst (although relatively uncom- vein obliteration [19]. Any leakage of the ablative
mon), is one of those complications [17]. Repeated agent from the cyst can also induce inflammation
lavage and injection can lead to an outflow tract to that could spread to nearby vessels which can lead
form and thus diminish the ablative effect as well to portal vein thrombosis [20]. Similar outcomes
due to reduced time of contact with the cyst. were observed in another study of 52 patients who
Because of this inherent risk, multiple injections underwent EUS-guided ablation where one patient
and lavages should be avoided when possible. developed splenic vein thrombosis/obliteration
Near-complete evacuation of the cyst prior to with collateral formation [17].
injection therapy also leads to increasing the sur- When it comes to the concern of using chemo-
face area that is directly exposed to the ablative therapeutic agents as a means for cyst ablation, it
agent, which increases the effectiveness of the raises the question of possible systemic effects
ablation. Ethanol lavage before using other abla- after chemotherapy injection. In a case series of
tive/chemotherapeutic agents may reduce the vis- ten patients who underwent cyst ablation with
cosity of the thick mucin and improve the delivery alcohol followed by paclitaxel, the plasma pacli-
of the ablation agent within the cyst locules. taxel concentrations were nearly undetectable
and rarely caused any adverse effect [21].

 afety and Controversy of Cyst

S
Ablation

With any budding treatment, modality, safety,

and efficacy must be considered when it comes to
investigating and, eventually, implementing man-
agement. Therefore, the complications related to
cyst ablation can be described in several clinical
trials. Most complications were self-limited and
mild. Abdominal pain is the most common com-
plication after cyst ablation. Pancreatitis is also a Fig. 13.3 Histology of ethanol ablated cyst epithelium:
possible complication. However, several studies note the thin attenuated epithelium
13 EUS-Guided Pancreatic Cyst Ablation 139

In recent years, there has been the question of  linical Trials for EUS-Guided Cyst
C
how to minimize adverse effects of EUS-guided Ablation
cyst ablation while maintaining the efficacy of
the procedure. It has been thought that the use of Several studies have been published on EUS-­
alcohol as an ablative agent is what leads to the guided cyst ablation since the initial pilot study in
serious complications of ablation (pancreatitis, 2005, focusing both on ethanol injection alone
splenic vein obliteration, etc.) due to alcohol and ethanol followed by paclitaxel injection. In
extravasation or due to the known inflammatory the initial pilot study, ethanol lavage was admin-
effects of alcohol on the pancreatic parenchyma istered alone during EUS-guided cyst ablation
and its surrounding tissue [22]. Ablation of and the patients were followed up in a 6–12-­
benign cysts should also be considered. These month period. Thirty-five percent of 23 patients
procedures and interventions do have their own had complete resolution. All septated cysts per-
risk of complications. Using cystic fluid analysis sisted. Five patients from this same study under-
can help guide the management strategy that can went surgical resection, all of which were MCN
be pursued but despite cystic fluid analysis, there with a variable degree of epithelial ablation [23].
are still a cohort of patients who will be have an A retrospective study done at two tertiary care
indeterminate pancreatic cyst. centers had 13 patients undergo ethanol lavage
It is widely accepted to continue to monitor through EUS-guided ablation and 11 of the 13
pancreatic cysts in asymptomatic patients who (85%) patients had complete resolution with a
are not good surgical candidates. However, life- mean follow-up of 26 months [24] (Fig. 13.4).
long surveillance is time consuming, economi- One study that included the longest follow-up
cally challenging, and a burden on the patients, and largest number of patients within a clinical
particularly the elderly patients who are most trial for EUS-guided cyst ablation had 91 patients
commonly diagnosed with pancreatic cystic who were categorized as having indeterminate
lesions. Cyst ablation could be an alternative for pancreatic cystic lesions undergo the procedure.
patients who are not surgical candidates and used The resolution rate for MCN was 50% as com-
to promote early management of possible prema- pared to IPMN where the resolution rate was a
lignant lesions. With the proposed eradication of disappointing 11%, suggesting that communica-
premalignant lesions through EUS-guided cyst tion with the pancreatic duct may reduce the effi-
ablation, it may be a reasonable treatment modal- cacy of ethanol [4].
ity especially because of its low risk and rela- To increase the ablative effect, Paclitaxel, a
tively high efficacy. chemotherapeutic agent used for treatment of

Fig. 13.4 Histology of cyst epithelium after saline lavage (a) and after ethanol (b): note the intact epithelium after
saline lavage and the attenuated epithelium after ethanol lavage
140 K. Philogene and W. R. Brugge

Fig. 13.5 Histology of

injected Paclitaxel gel
into the pancreas: note
the lack of inflammatory
response to the gel

Fig. 13.6 CT scan of pancreatic body cyst before (a) and after (b) ablation therapy

several malignancies, has been combined with lavage, 62% of patients had complete resolution,
ethanol injection (Fig. 13.5). It was proposed that with smaller cystic lesions having a higher likeli-
ethanol can be used to distort the epithelial lining hood of resolution [17].
of the cyst, which would allow for paclitaxel or There has also been increased interest in
any other ablative agent to diffuse through the ­eliminating ethanol injections from cyst ablation
damaged epithelium leading to additional inhibi- altogether given that it is thought that the compli-
tory effects through apoptosis. In a pilot study of cations that arise with cyst ablation come from
14 patients who underwent ablation therapy the ethanol injections (Fig. 13.7). The CHARM
using ethanol injection followed by paclitaxel trial, a prospective, randomized double-­blinded
injection, 11 of the 14 patients saw complete res- pilot study of ten patients with mucinous cysts,
olution at 6-month follow-up (Fig. 13.6). This had patients divided into two groups: those
could represent a synergistic effect given that undergoing ablation using ethanol injections fol-
ethanol alone had a resolution rate of 33% in the lowed by a combination of paclitaxel and gem-
previous investigations [25]. Another study of 52 citabine and those getting normal saline injections
patients who underwent ethanol and paclitaxel followed by the chemotherapeutic agents
13 EUS-Guided Pancreatic Cyst Ablation 141

Fig. 13.7 EUS-guided cyst injection into a 2-cm unilocular cyst (a) complicated by acute pancreatitis as seen on a CT
scan with an air-fluid level in the cyst (b). The patient made a quick recovery

described. At 6 months and 12 months, the 12 months had regrowth of the cyst thought to be
alcohol-­free group had a resolution rate of 67% due to the presence of, and subsequent prolifera-
while the alcohol group had a resolution rate of tion of, remnant mucinous epithelium in missed
50% and 75%, respectively. This study suggests locules, confirmed by histopathology [28].
that alcohol use may not be required for effective Therefore, careful selection and review of the
cyst ablation [26]. In a single-center, prospective, patients and the morphology of the cystic lesion
double-blind clinical trial, 39 patients with muci- is important to consider for improving the effi-
nous pancreatic cysts also were divided into two cacy of using cystic ablation as an effective treat-
groups both receiving paclitaxel and gemcitabine ment method.
with one group receiving prior normal saline The short-term outcomes of EUS-guided cyst
injection and the other group receiving ethanol ablation appear to be promising. However, there
injection to determine the efficacy of an alcohol-­ had been concern for the overall efficacy in the
free ablation as well as assess its effect on the long term when it comes to complete resolution
complication rates. Sixty-seven percent of the with no recurrence. In a single-center, prospec-
alcohol-free group had complete resolution of tive study of 164 patients with pancreatic cysts
cysts compared to 61% for the alcohol group. Six undergoing EUS ablation using ethanol and
percent of patients within the alcohol group had paclitaxel, complete resolution of the cyst
serious adverse effects (e.g. pancreatitis) and occurred in 114 (72.2%) patients with only two
22% developed minor side effects (e.g. mild patients (1.7%) showing cyst recurrence at a
abdominal pain). The alcohol-free group had no median follow-up of 72 months. This study may
reports of any complications, minor or serious indicate that EUS-guided cyst ablation is an
[27]. Therefore, it is worth noting that removing effective and durable alternative therapy to sur-
alcohol from the treatment modality for cyst gery [29].
ablations may be a safer technique, but with very To improve the ablative effect and resolution
good efficacy. of pancreatic cysts using this treatment modality,
Septations in a pancreatic cyst is an important a second needle passed at different angles along
morphological factor when it comes to the effec- with booster ablation have been proposed and
tiveness and efficacy of cyst ablation. In a case ­trialed. Cysts that have six or fewer locules are
series of ten patients who had septated pancreatic usually preferred when evaluating which cysts
cysts, complete resolution occurred in six of the are amenable to ablation because the presence of
ten patients. Two patients had an initial response septa may prevent the delivery of the ablative
to ethanol and paclitaxel ablation but by agents to all locules. To minimize the risk of
142 K. Philogene and W. R. Brugge

Fig. 13.8 MRCP of a side branch IPMN located deep in the uncinate with multiple septations (a) making the EUS-­
guided injection technically difficult (b). A 19-gauge needle was required because of the viscosity of the fluid

missing locules, multiple needle passes at differ- unilocular or oligolocular cysts without any obvi-
ent angles may be warranted. In a case series of ous communication between the cyst and the
13 patients with suspected IPMN, complete reso- main pancreatic duct on imaging, (2) cysts that
lution seen on computed tomography (CT) or have increased in size at follow-up, (3) when
magnetic resonance imaging (MRI) occurred in FNA is required for characterization, and (4)
five (38%) patients after two EUS-guided ethanol patients who have a high surgical risk or decline
lavages compared to no patients showing resolu- surgery [29]. Physicians should also have a mul-
tion with a single needle pass, measured by the tidisciplinary approach to discussing the patient’s
decrease in size of the cyst and its surface area. case prior to making the decision to manage the
The results of this study demonstrate that there cyst using cyst ablation.
may be a need for multiple ethanol lavage ses- The ideal cyst for treatment with cyst ablation
sions in the presence of viscous fluid and/or septa appears to be mucinous cystic neoplasms. These
to allow for more epithelial surfaces to encounter lesions, however, can often be removed easily
the ablative agent, resulting in higher cyst abla- with laparoscopic distal pancreatectomy with
tion rates [18]. (Fig. 13.8) However, multiple minimal mortality and no risk of recurrence after
needle passes may increase the complications resection [31]. It is still too early to suggest that
associated with cyst ablation and, therefore, a surgery or cyst ablation alone would be the only
second needle pass should be performed during treatment modality that should be offered for
the same session with caution [30]. pancreatic cysts and would require further study.

I ndications for EUS-Guided Cyst  uture Use and Modifications

F
Ablation to Therapy of EUS-Guided Cyst
Ablation
EUS-guided pancreatic cyst ablation is still being
investigated and should be used only for a select EUS-guided cyst ablation has been studied and
group of patients based on the overall efficacy of has the potential to be an acceptable alternative to
this treatment modality while also keeping the surgical resection of cysts without the associated
procedure-related risks/complications low. The risks of surgery. However, the acceptance of abla-
ideal candidates for ablation are (1) 2–3.5-cm tion by surgeons as an alternative to surgery and
13 EUS-Guided Pancreatic Cyst Ablation 143

for oncologists, institutions, and institutional Photodynamic therapy (PDT) has also been
review boards to permit gastroenterologists to shown to be an effective means of inducing coag-
use chemotherapy to treat pancreatic cysts would ulation with the photosensitizing agent Porfimer
have to be addressed before EUS-guided ablation sodium and has been shown in the previous stud-
can become more commonplace [31]. ies to be effective in tissue ablation within por-
Some procedural modifications that can be cine liver, pancreas, kidney, and spleen.
made to improve the efficacy of ablation include Brachytherapy is a type of radiation therapy
a second needle pass technique for septated cysts, where the radiation source is inserted within or
booster ablation for larger cyst that are unrespon- adjacent to known cancer tissue. Brachytherapy
sive to treatment, maintenance of ethanol con- has been utilized in the management of several
centrations during lavage, and developing slow localized cancers. However, it is still under inves-
release ablative agents [32]. Long-term follow- tigation for pancreatic cancer. One clinical trial
­up in patients who undergo cyst ablation will also of 15 patients with pancreatic adenocarcinoma
need to be established, particularly because there who underwent brachytherapy showed some
are no conventional imaging techniques available improvement in pain for a limited period.
that could accurately ensure effective identifica- However, there was no survival benefit and three
tion of resolution. Oh et al. determined that some patients had complications of pancreatitis and
patients who underwent ethanol/paclitaxel injec- pseudocyst formation. High-intensity focused
tion and, eventually, surgical resection had at ultrasound (HIFU) is another rapidly developing
least 50% of the epithelial lining intact. These treatment modality that is being used more often
patients continue to be at risk for tumorigenesis for noninvasive and minimally invasive ablation
and this risk should be kept in mind in selecting of benign and malignant tumors. HIFU works by
patients for EUS-guided ablation [33]. delivering ultrasound energy to the tumor which
There are several other treatment modalities ultimately leads to heating of the tumor tissue
that are currently being studied in place of EUS-­ and tissue denaturation. HIFU may be both cura-
guided ablation that have been emerging in recent tive and palliative for patients with pancreatic
years. Radiofrequency ablation (RFA) is a well-­ cancer. Studies have shown pain reduction in
studied antitumor treatment for dysplastic patients with unresectable pancreatic cancer [38].
Barret’s esophagus and hepatocellular carcinoma
that utilizes local thermal-induced coagulative
necrosis. This technique has been applied to pan- Conclusions
creatic cancers. However, the postprocedural
morbidities were high and most likely due to the A large number of trials have now demonstrated
local effects of heat damage. Unlike hepatic the technical ease and safety of EUS-guided
tumors, which have a surrounding parenchyma injection and ablation therapy of pancreatic cys-
that is protective, pancreatic tumors are usually tic neoplasms. Morphologically, the ideal candi-
wrapped around blood vessels or the distal bile date for ablation is a 2–4-cm unilocular cyst in
duct, which can sustain thermal injury during the the body or tail of the pancreas. Mucinous cystic
procedure. However, multiple studies have dem- neoplasms appear to have a better response rate
onstrated that radiofrequency ablation is a feasi- as compared to side branch IPMN lesions.
ble, safe treatment modality. RFA does require Macrocystic serous cystadenomas have not been
further investigation in larger study populations well studied, but will probably respond well to
[34–36]. A new EUS needle prototype is being injection therapy. Malignant cysts and neuroen-
developed that would be able to connect to stan- docrine cystic neoplasms should not be treated
dard electrosurgical units found in many endos- with ablation therapy. In order to achieve com-
copy laboratories. This would help reduce costs plete elimination of the cyst, it is ideal to provide
in purchasing new equipment and decrease the injection therapy every 3 months until eradica-
need for additional training [37]. tion has been achieved. MRCP should be used as
144 K. Philogene and W. R. Brugge

the imaging guide that can provide highly accu- endoscopic ultrasound injection of ONYX-015 with
intravenous gemcitabine in unresectable pancreatic
rate measurements of the cysts. This type of ther- carcinoma. Clin Cancer Res. 2003;9(2):555–61.
apy is still considered “investigational” and 13. Chang KJ, Lee JG, Holcombe RF, Kuo J, Muthusamy
should be performed under the guidance of a for- R, Wu ML. Endoscopic ultrasound delivery of an anti-
mal protocol approved by an institutional review tumor agent to treat a case of pancreatic cancer. Nat
Clin Pract Gastroenterol Hepatol. 2008;5(2):107–11.
board. In the future, EUS-guided radiofrequency 14. Bean WJ, Rodan BA. Hepatic cysts: treatment with
ablation therapy might be used in conjunction alcohol. AJR Am J Roentgenol. 1985;144:237–41.
with injection therapy to provide long-term reso- 15. Gelczer RK, Charboneau JW, Hussain S, Brown
lution of cystic neoplasms. DL. Complications of percutaneous ethanol ablation.
J Ultrasound Med. 1998;17(8):531–3.
16. Rowinsky EK, Donehower RC. Paclitaxel (taxol). N
Engl J Med. 1995;332(15):1004–14.
References 17. Oh HC, Seo DW, Song TJ, Moon SH, Park DH,
Soo LS, et al. Endoscopic ultrasonography-guided
1. Levy A. Prevalence of unsuspected pancreatic cysts ethanol lavage with paclitaxel injection treats
on MDCT. Yearb Diagn Radiol. 2009;2009:273–4. patients with pancreatic cysts. Gastroenterology.
2. Sekhar A, Lee KS, Brown A, Pedrosa I. Prevalence of 2011;140(1):172–9.
incidental pancreatic cysts in the adult population on 18. Dimaio CJ, Dewitt JM, Brugge WR. Ablation of pan-
MR imaging. Pancreas. 2008;37(4):495. creatic cystic lesions: the use of multiple endoscopic
3. Bhutani M, Jana T, Shroff J. Pancreatic cystic neo- ultrasound-guided ethanol lavage sessions. Pancreas.
plasms: review of current knowledge, diagnostic 2011;40(5):664–8.
challenges, and management options. J Carcinog. 19. Seo D-W, Cho M-K, Choi J-H. Endoscopic
2015;14(1):3. ultrasound-­guided ablation therapy for pancreatic
4. Park JK, Song BJ, Ryu JK, Paik WH, Park JM, cysts. Endosc Ultrasound. 2015;4(4):293.
Kim J, et al. Clinical outcomes of endoscopic 20. Oh H-C, Seo DW, Kim SC. Portal vein thrombosis
ultrasonography-­ guided pancreatic cyst ablation. after EUS-guided pancreatic cyst ablation. Dig Dis
Pancreas. 2016;45(6):889–94. Sci. 2012;57(7):1965–7.
5. Ferrone CR. Current trends in pancreatic cystic neo- 21. Oh H-C, Seo DW, Kim S-H, Min B, Kim J. Systemic
plasms. Arch Surg. 2009;144(5):448–54. effect of endoscopic ultrasonography-guided pancre-
6. Gerry JM, Poultsides GA. Surgical management of atic cyst ablation with ethanol and paclitaxel. Dig Dis
pancreatic cysts: a shifting paradigm toward selective Sci. 2014;59(7):1573–7.
resection. Dig Dis Sci. 2017;62(7):1816–26. 22. Dewitt J, Mcgreevy K, Schmidt CM, Brugge
7. Qin SY, Lu XP, Jiang HX. EUS-guided ethanol abla- WR. EUS-guided ethanol versus saline solution
tion of insulinomas: case series and literature review. lavage for pancreatic cysts: a randomized, double-­
Medicine (Baltimore). 2014;93(14):e85. blind study. Gastrointest Endosc. 2009;70(4):710–23.
8. Larssen TB, Viste A, Horn A, Haldorsen IS, Espeland 23. Gan SI, Thompson CC, Lauwers GY, Bounds BC,
A. Single-session alcohol sclerotherapy of symptom- Brugge WR. Ethanol lavage of pancreatic cys-
atic liver cysts using 10–20 min of ethanol exposure: tic lesions: initial pilot study. Gastrointest Endosc.
no recurrence at 2–16 years of follow-up. Abdom 2005;61(6):746–52.
Radiol. 2016;41(9):1776–81. 24. Caillol F, Poincloux L, Bories E, Cruzille E, Pesenti
9. Basu N, Dutta D, Maisnam I, Basu S, Ghosh S, C, Darcha C, et al. Ethanol lavage of 14 mucinous
Chowdhury S, et al. Percutaneous ethanol ablation in cysts of the pancreas: a retrospective study in two ter-
managing predominantly cystic thyroid nodules: an tiary centers. Endosc Ultrasound. 2012;1(1):48–52.
eastern India perspective. Indian J Endocrinol Metab. 25. Oh H-C, Seo DW, Lee TY, Kim JY, Lee SS, Lee SK,
2014;18(5):662–8. et al. New treatment for cystic tumors of the pancreas:
10. Yang X, Yu J, Liang P, Yu X, Cheng Z, Han Z, et al. EUS-guided ethanol lavage with paclitaxel injection.
Ultrasound-guided percutaneous ethanol ablation for Gastrointest Endosc. 2008;67(4):636–42.
primary non-parasitic splenic cysts in 15 patients. 26. Moyer MT, Dye CE, Ancrile B, Sharzehi S, Mathew
Abdom Radiol. 2016;41(3):538–44. A, Mcgarrity TJ, et al. 104 is alcohol required for
11. Chang KJ, Nguyen PT, Thompson JA, Kurosaki TT, effective pancreatic cyst ablation? The prospective
Casey LR, Leung EC, et al. Phase I clinical trial of randomized CHARM preliminary trial pilot study.
allogeneic mixed lymphocyte culture (cytoimplant) Gastrointest Endosc. 2015;81(5):AB114.
delivered by endoscopic ultrasound guided fine-­ 27. Moyer MT, Sharzehi S, Mathew A, Levenick JM,
needle injection in patients with advanced pancreatic Headlee BD, Blandford JT, et al. The safety and effi-
carcinoma. Cancer. 2000;88(6):1325–35. cacy of an alcohol-free pancreatic cyst ablation proto-
12. Hecht JR, Bedford R, Abruzzese JL, Lahoti S, Reid TR, col. Gastroenterology. 2017;153(5):1295–303.
Soetikno RM, et al. A phase I/II trial of ­intratumoral 28. Oh HC, Seo DW, Kim SC, Yu E, Kim K, Moon SH,
et al. Septated cystic tumors of the pancreas: is it
13 EUS-Guided Pancreatic Cyst Ablation 145

p­ ossible to treat them by endoscopic ultrasonography-­ cystic tumors: is it justified? Gastrointest Endosc.
guided intervention? Scand J Gastroenterol. 2016;83(5):921–3.
2009;44(2):242–7. 34. Lakhtakia S, Seo D-W. Endoscopic ultrasonography-­
29. Choi J-H, Seo D, Song T, Park D, Lee S, Lee S, et al. guided tumor ablation. Dig Endosc.
Long-term outcomes after endoscopic ultrasound-­ 2017;29(4):486–94.
guided ablation of pancreatic cysts. Endoscopy. 35. Kim J. Endoscopic ultrasound-guided treatment
2017;49(09):866–73. of pancreatic cystic and solid masses. Clin Endosc.
30. Oh H-C, Seo DW, Song TJ. Resolution of a sep- 2015;48(4):308.
tated pancreatic cyst by booster endoscopic 36. Kirtane T, Bhutani MS. EUS for pancreatic cystic
ultrasonography-­ guided ablation. J Dig Dis. neoplasms: the roadmap to the future is much more
2011;12(6):497–9. than just a few shades of gray. Asian Pac J Trop Med.
31. Dewitt J. Pancreatic cyst ablation: why are we 2016;9(12):1218–21.
not doing more of these procedures? Endoscopy. 37. Wallace M, Moris M, Atar M, Kadayifci A, Krishna
2017;49(09):839–41. M, Librero A, et al. Thermal ablation of pancreatic
32. Xu XX, Du Y, Yang HF, Zhang Q, Li Y, Zee cyst with a prototype endoscopic ultrasound capable
CS. CT-guided sclerotherapy with ethanol concen- radiofrequency needle device: a pilot feasibility study.
tration monitoring for treatment of renal cysts. Am Endosc Ultrasound. 2017;6(2):123.
J Roentgenol. 2011;196(1):W78–82. 38. Ende AR, Hwang JH. Endoscopic ultrasound-­
33. Vazquez-Sequeiros E, Maluf-Filho guided tumor ablation. Gastrointestinal Intervention.
F. Endosonography-guided ablation of pancreatic 2014;3(1):27–9.
 Endoscopic Ultrasound-Guided
Access to the Stomach in Patients 14
with Prior Gastric Bypass
to Facilitate Endoscopic
Retrograde
Cholangiopancreatography

Christine Boumitri, Bhupinder Romana,

and Michel Kahaleh

pancreatography (ERCP) in these patients has its

Background own obstacles since the normal anatomy to reach
the papilla has been altered. Multiple endoscopes
Over the last four decades the prevalence of obe- and techniques have been described to gain access
sity among adults in the United States (US) has into the excluded pancreatobiliary system with
increased significantly. According to the 2013– varying success rates and limitations (Table 14.1).
2014 national health and nutrition examination Enteroscopy-­assisted ERCP using double-balloon
survey the estimated percentage of US adults with enteroscopy, single-balloon enteroscopy, or spiral
obesity is 37.7 that means more than 1 in 3 adults enteroscopy has been used in patients with Roux-­
were considered to be obese [1, 2]. Surgical treat- en-­Y reconstruction. However, these scopes do
ment of obesity is an effective method of achiev- not have elevators and thus the manipulation of
ing weight loss and the most commonly performed the accessories may be challenging, if not impos-
bariatric surgery worldwide is Roux-­en-­Y gastric sible. The overall success rate using enteroscopy-­
bypass (RYGB) [3]. RYGB patients are predis- assisted ERCP has been evaluated in retrospective
posed to develop cholelithiasis and other pancrea- analysis and multicenter clinical trial and was
tobiliary pathologies that pose a different found to be around 63% [4–6]. This is signifi-
challenge to the gastroenterologist. Performing cantly lower compared to when ERCP is per-
conventional endoscopic retrograde cholangio- formed through a gastrostomy tract into the
excluded stomach [4]. Gastrostomy-assisted
Electronic supplementary material: The online version
ERCP is another approach in this patient popula-
of this chapter (https://doi.org/10.1007/978-3-319-97376- tion. It consists of accessing the gastric remnant
0_14) contains supplementary material, which is available by placing a gastrostomy tube and then proceed-
to authorized users. ing with conventional ERCP. This often requires
C. Boumitri · B. Romana aggressive dilation of the gastrostomy tract, which
Division of Gastroenterology and Hepatology, can be painful for patients. This procedure has a
University of Missouri, Columbia, MO, USA
higher success rate; however, the rates of adverse
e-mail: [email protected]
events are more common when compared to dou-
M. Kahaleh (*)
ble balloon-assisted ERCP and mostly related to
Division of Gastroenterology and Hepatology,
Rutgers Robert Wood Johnson, the gastrostomy creation (14.5% vs 3% p = 0.022)
New Brunswick, NJ, USA [4]. Access into the excluded stomach can be

© Springer Nature Switzerland AG 2019 147

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_14
148 C. Boumitri et al.

Table 14.1 Success rate and adverse events of device-assisted enteroscopy facilitated ERCP versus transgastric ERCP
Overall success
Route of access rate (%) Adverse events Limitations
Device-assisted enteroscopy 63 [4, 5] 3–12.4% [4, 5]  – Length of Roux-en-Y limb
facilitated ERCP  – Limited manipulations of accessories
due to lack of elevator
 – Forward vision enteroscope
Transgastric ERCP 98.5 [7] 14% complication  – Need for staged procedure with
techniques rate [7] antecedent G tube placement
 1. Laparoscopic-assisted 98.9   • 83%  – Increased cost of the procedure
ERCP gastrostomy  – Need for multidisciplinary
 2. Open surgery 100 related collaboration
 3. Antecedent gastrostomy 96.4   • 17% ERCP
tube placement related
 4. EUS directed 93.8
transgastric ERCP

obtained laparoscopically, by open surgery, via Overview of the EDGE Procedure

previous gastrostomy tube placement by interven-
tional radiology and staged ERCP or with endo- Equipment Needed for Internal EDGE
scopic ultrasound (EUS) guidance.
The overall success rate of transgastric ERCP a. Linear echoendoscope (GF-UCT180;
has been evaluated in a recent systematic review Olympus, Central Valley, PA
by Banerjee et al. and was found to be 98.5% b. Nineteen-gauge EUS needle (ECHO-19;
regardless of the technique used to gain access Cook Medical, Winston-Salem, NC), or
into the excluded stomach with a 14% complica- similar
tion rate (Table 14.1) [7]. In this chapter, we will c. Diluted contrast with 120 mL water
focus on an innovative approach of endoscopic d. 0.025- or 0.035-inch guidewire
ultrasound-guided access into the gastric rem- e. Four to six millimeter dilation balloon
nant in patients with RYGB for assisted (Hurricane RX; Boston Scientific, Natick
ERCP. The use of endoscopic ultrasound to MA)
facilitate creation of a gastrostomy in patients f. Lumen apposing metal stent (LAMS) 15 mm
with RYGB has been described as early as 2011 (Axios; Boston Scientific, Natick MA)
[8]. Multiple approaches have been used and are g. Dilation balloon: 15- to 18-mm balloon (con-
noteworthy such as EUS-guided sutured gastro- trolled radial expansion balloon dilator [CRE];
pexy for transgastric ERCP (ESTER) [9], exter- Boston Scientific, Natick MA)
nal EUS-­ directed transgastric ERCP (EDGE) h. Duodenoscope and other equipment needed
[10], and internal EDGE [11]. Percutaneous for conventional ERCP
assisted transprosthetic endoscopic therapy i. Optional plastic double-pigtail stents
(PATENT) is another minimally invasive endo- (10 F × 10-cm) for anchoring LAMS if con-
scopic approach to access the gastric remnant cerns for migration
and create a gastrostomy; however, it does not j. Endoscopic snare or grasping forceps for
involve endoscopic ultrasound and access into LAMS removal
the excluded stomach is obtained via deep bal- k. Endoscopic suturing for gastrogastric fistula
loon enteroscopy [12]. The same group has created (Overstitch; Apollo Endosurgery,
reported recently a modified PATENT using Austin, Tex) or over-the-scope clip (diameter
endoscopic ultrasound [13]. 12 mm; Ovesco, Los Gatos, California, USA)
14 Endoscopic Ultrasound-Guided Access to the Stomach in Patients with Prior Gastric Bypass… 149

Procedure Techniques stomach followed by dilation of the tract using

8- Fr to 16-Fr dilating catheters. A 16-Fr gastros-
The external EDGE and ESTER procedures are tomy tube is then sutured in place (Fig. 14.2).
similar in principals with some minor changes in The second stage of the procedure consists of
techniques and instruments used. The first step dilating the tract and placing a metallic stent
consists of accessing the gastric remnant. This is across the tract itself. This is achieved by passing
achieved by advancing a linear echoendoscope to a guidewire through the PEG tube and coiling it
the gastric pouch or Roux limb, identifying the into the stomach then the PEG tube is removed
gastric remnant and puncturing it with a 19-Gauge and the extra thin scope (GIF-XP 180; Olympus)
fine needle aspiration needle under sonographic is advanced over the wire into the excluded stom-
guidance (Fig. 14.1). A small amount of contrast ach. Three T fasteners are then placed under fluo-
(5–10 mL) is then injected to confirm placement roscopic and endoscopic guidance to maintain
in the gastric remnant. The gastric remnant is apposition of the gastric and ventral walls. The
then inflated to obtain apposition with the gastric fistula tract is then progressively dilated using
wall for a gastrostomy placement. Inflation has Maloney dilators up to 54-Fr followed by over
been described with air (400–500 mL) (ESTER the wire deployment of a fully covered metal
technique) or with 120 mL sterile water and esophageal stent (Fig. 14.3). The metal stent is
120 mL of air (EDGE technique): both are effec- sutured to the surrounding skin and ERCP is then
tive. A guidewire is subsequently advanced performed using the antegrade approach.
through the needle and coiled into the stomach. In the ESTER approach, once the inflation of
The second step consists of creating a gastros- the excluded stomach is achieved, an 18-gauge
tomy that will be used as an access route for needle is used to percutaneously access the rem-
ERCP. Again, different instruments and nant. Once position confirmed fluoroscopically
approaches can be used. two 0.018 inch guidewires are passed though the
The external EDGE describes using an needle and coiled into the stomach. One guide-
18-gauge needle for creating the gastrostomy. In wire is used to pass a 20 mm stone extraction bal-
this procedure, once adequate position in the gas- loon that will be used to provide counter traction
tric remnant is confirmed, a stiff guidewire is while a sequential dilation up to 24-Fr is per-
advanced through the needle and coiled into the formed through the second wire. This is followed

Fig. 14.1 EUS-guided

puncture of the excluded
stomach before
insufflation with air
150 C. Boumitri et al.

which the duodenoscope is inserted to proceed

with the conventional antegrade ERCP. These
two procedures were described in 2014 and both
studies included a small number of patients.
These novel and minimally invasive
approaches were just the introduction to another
innovative approach, the internal EDGE proce-
dure, that bypasses the need to create a percuta-
neous gastrostomy with sutured gastropexy. The
procedure was first described by Kedia et al. in
2014 (Video 14.1). The initial steps of the inter-
nal EDGE procedure are similar to the external
EDGE where a linear echoendoscope is advanced
into the gastric pouch and the excluded stomach
is identified and then punctured using a 19-gauge
Fig. 14.2 Placement of a percutaneous gastrostomy needle (Fig. 14.4). The access into the remnant
can be obtained from the pouch or via the afferent
limb. Contrast mixed with water is then injected
through the needle to distend the gastric remnant
followed by advancing a 0.035 inch guidewire
through the needle and coiling it within the lumen
of the stomach (Figs. 14.5 and 14.6). This creates
a gastrogastric fistula between the pouch and
excluded stomach which is subsequently dilated
over the wire using 4 mm balloon (Hurricane RX,
Boston Scientific). The next step is to deploy a
transluminal stent to provide apposition of the
gastric pouch and remnant through which ERCP
will be performed. The delivery system of the
lumen apposing metal stent (LAMS) is then
advanced into the fistula (Fig. 14.7) and the distal
flange deployed into the excluded stomach
(Fig. 14.8) and proximal flange into the gastric
Fig. 14.3 Placement of a transcutaneous fully covered pouch using both fluoroscopic and sonographic
esophageal stent guidance (Figs. 14.9 and 14.10). The lumen of
the stent is then dilated using a 15–18 mm dilat-
ing balloon (CRE; Boston Scientific) (Figs. 14.11
by the insertion of 20-Fr “peel away” sheath and 14.12). The duodenoscope is then advanced
(Cook endoscopy) instead of the 16-Fr gastros- through the stent and conventional antegrade
tomy tube used in the external EDGE technique. ERCP is performed (Figs. 14.13 and 14.14).
The ultrathin scope (GIF-XP 180; Olympus) is Once further access into the pancreatobiliary tree
then advanced through the sheath into the is no longer needed the LAMS is removed using
excluded stomach and endoscopic sutured gas- a 25 mm snare and the fistula tract is closed using
tropexy is performed using a 2 mm laparoscopic over-the-scope clip (OTSC; OVESCO, Los
suture passing needle under endoscopic visual- Gatos, CA, USA) or endoscopic suturing device
ization. Once the apposition of the gastric and (Overstitch; Apollo Endosurgery, Austin, TX,
ventral wall is obtained the “peel away” sheath is USA). In some cases, the stent can be left in place
replaced by a laparoscopic trocar system through if repeated ERCPs are needed. The tract can also
14 Endoscopic Ultrasound-Guided Access to the Stomach in Patients with Prior Gastric Bypass… 151

Fig. 14.4 EUS-guided

puncture of the excluded
stomach

Fig. 14.5 Injection of contrast into the excluded stomach

under fluoroscopy
Fig. 14.6 Placement of a guidewire in the excluded
stomach under fluoroscopy
152 C. Boumitri et al.

be left to close secondarily after transluminal Outcomes

stent removal. Also, the use of argon plasma
coagulation (APC) to de-epithelialize the fistula Success Rate
tract and enhance transgastric fistula closure has The use of endoscopic ultrasound to assist trans-
been described. gastric ERCP in patients with gastric bypass is of
no doubt a new technique and innovation with
limited overall experience. The outcomes dis-
cussed here are mostly based on results from
­different case series reported in the literature.
Most case series report a technical and clinical
success rates (Table 14.2).
The reported success rate of the ESTER tech-
nique is 90% [9]. In a case series of six patients
Kedia et al. reported an 83% success rate of EUS-­
guided gastrostomy tube placement with 100%
success rate of ERCP [10]. The percutaneous
access using the standard percutaneous endo-
scopic gastrostomy (PEG) tube needle kit could
not be obtained in one patient and subsequently
patient required PEG placement by interven-
tional radiology.
In a midterm analysis of internal EDGE,
Tyberg et al. reported a 100% technical success
rate and 91% clinical success rate in sixteen
patients undergoing EDGE at two academic cen-
Fig. 14.7 Advancement of a cautery enhanced lumen
apposing metal stent in the excluded stomach under ters [14]. Technical success rate was defined by
fluoroscopy successful creation of gastrogastric fistula

Fig. 14.8 Deployment

of inner flange of the
lumen apposing metal
stent under
ultrasonography
14 Endoscopic Ultrasound-Guided Access to the Stomach in Patients with Prior Gastric Bypass… 153

Fig. 14.9 Deployment

of the outer flange in the
pouch under endoscopic
visualization

success rate of 100% [15]. In summary, the over-

all success rate of EUS-guided transgastric ERCP
from the reported case series ranges between
83% and 100%. This depends on the local exper-
tise and approach used.

Complications and Limitations

The complications that can occur with EUS-­
guided transgastric ERCP can be related to the
EUS-guided gastrogastric or jejunogastric fistula
creation in the case of EDGE procedure, to the
percutaneous gastrostomy creation in case of
external EDGE and ESTER, or to the ERCP pro-
cedure itself. Attam et al. reported no immediate
complication related to the ESTER procedure,
Fig. 14.10 Fluoroscopic vision of complete deployment
however there was no mention of how long
of the lumen apposing metal stent
patients were followed and whether there were
any delayed complications [9]. With the external
(37.5%) or jejunogastric fistulas (62.5%). EDGE, the reported ERCP complication rate was
Clinical success rate was defined by successful 0% and 33% of cases had complications related
ERCP or EUS through LAMS which was per- to gastrostomy tube placement (localized PEG
formed in only 11 of the 16 patients since 5 site infection). Internal EDGE reported compli-
patients were awaiting fistula tract maturation. cations included lumen apposing metal stent dis-
Four patients had ERCP at the same index lodgement requiring repositioning or bridging
procedure. with a fully cover metal stent at a rate 19% [14].
Another multicenter experience with EDGE Ngamruengphong et al. reported 16% rate of
has been published by Ngamruengphong et al. LAMS dislodgment during ERCP requiring stent
[15]. The study included a total of 13 patients repositioning [15]. In their report, patients who
with RYGB with different indications for underwent ERCP using the therapeutic duodeno-
ERCP. The group reported a technical and c­ linical scope had a higher stent migration rate when
154 C. Boumitri et al.

Fig. 14.11 Endoscopic

placement of a 15 mm
wire guided balloon into
the lumen apposing stent

Fig. 14.13 Advancement of the duodenoscope until the

Fig. 14.12 Balloon dilation of the lumen apposing metal second portion of the duodenum though the lumen appos-
stent up to 15 mm under fluoroscopy ing metal stent (indicated by arrow)

compared to no cases of migration when the slim despite attempted closure which is reported to
duodenoscope was used (33% vs 0%) [15]. The range between 8 and 12.5% [14, 15]. The persis-
risk of stent migration can therefore be mitigated tence of a fistula (failed closure or staged proce-
by using a diagnostic duodenoscope (if one is dures) between the pouch or afferent limb and the
available), avoiding excessive stent dilation and gastric remnant is a concern for weight gain. This
allowing the fistula tract to mature by performing did not appear to be a major concern in the
ERCP at a later stage if possible in non-urgent reported series, however the number of cases and
cases. the duration of follow-up in these reports are not
Another concern with the creation of a gastro- enough to make a final conclusion about the risks
gastric or jejunogastric fistula is persistent fistula of weight gain. Another observed complication is
14 Endoscopic Ultrasound-Guided Access to the Stomach in Patients with Prior Gastric Bypass… 155

performed surgery for obesity in the USA),

RYGB remains the most common surgery world-
wide. These patients have an altered anatomy
which limits access to their pancreatobiliary
tree. While enteroscopy-assisted ERCP has a
success rate of 63%, the rate increases to 98%
when a transgastric approach used. When a sur-
gical approach is used to create the gastrostomy
access this entails a multidisciplinary approach
and an increase in the cost, time, and complica-
tions. There is no doubt that the indication for
ERCP, the urgency, and the existence of other
surgical indications to explore the abdomen play
an important decisive role for the endoscopist
with regard to which route to use (surgical,
enteroscopy assisted, or ultrasound guided).
With the advances of therapeutic endoscopic
ultrasound, we learned that access into the gas-
Fig. 14.14 Cholangiogram performed with the duodeno- tric remnant is possible using endoscopic ultra-
scope advanced though the lumen apposing metal stent sound without the need for a surgical/
(indicated by arrow) interventional radiology approach. The introduc-
tion of lumen apposing metal stent has revolu-
bowel perforation which can typically be closed tionized the word of therapeutic interventions
endoscopically when diagnosed in a timely and proves its utility in creating a gastrogastric
fashion. or jejunogastric fistula in patients with RYGB
Although the introduction of these novel tech- for ERCP. This procedure offers the benefit of
niques of EUS-guided transgastric ERCP in avoiding a percutaneous gastrostomy and its
patients with RYGB offers many advantages with complications, doing ERCP during the same
a significant success rate, each above described index procedure or doing a staged procedure
technique has its own limitations. In addition, all which allows the fistula tract to mature and per-
case series reported in the literature are retrospec- form non-urgent ERCP as well as repetitive pro-
tive, included a small number of patients and cedure as needed (i.e., benign biliary strictures,
some series included a multicenter experience etc.). The experience is still limited and larger
which entails different endoscopist technique randomized trials are needed to prove the long-
with possible variation in instruments used. term outcomes of this procedure and the risk of
weight gain in this population. Cost-­benefit anal-
ysis of these innovative approaches compared to
Conclusion and Future Directions laparoscopy-assisted ERCP or enteroscopy-
assisted ERCP will certainly shape the future of
Despite the trend over the last 10 years to per- endoscopic ultrasound-guided access to the
form fewer RYGB and the rise of sleeve gastrec- stomach in patients with RYGB to facilitate
tomy (which is currently the most commonly ERCP.
 156

Table 14.2 Characteristics of studies that used endoscopic ultrasound-guided access to the stomach in patients with prior gastric bypass to facilitate endoscopic retrograde
cholangiopancreatography
Number
of
Study patients Success rate Complication rate Procedure time Median follow-up Weight gain in kg
Attam et al. [9] 10 90% (9/10) Immediate procedure 88 min (median) Not reported Not reported
(ESTER approach) related adverse events:
0%
Kedia et al. [10] 6 EUS access into remnant success Gastrostomy tube related EUS-guided gastrostomy tube Not reported Not reported
(external EDGE, rate: 100% adverse events 33% placement: 81 ± 26 (median ±
two stages) Gastrostomy tube placement ERCP related adverse SD, minutes)
success rate: 83% events 0% ERCP: 98 ± 24 (mean ± SD,
ERCP success rate: 100% minutes)
Tyberg et al. [14] 16 Technical success rate: 100% LAMS dislodgement: Not reported Not reported Mean −2.85
(internal EDGE) Clinical success rate in 10 19%
patients: 91% Jejunal perforation:
6.25%
Persistent fistula after
attempted closure 12.5%
(1/8)
Ngamruengphong 13 Technical success rate: 100% Procedure related 0% 30 ± 17 (mean ± SD, minutes) 68 days Mean −3.6 ± 4.8
et al. [15] Clinical success rate: 100% Post-procedure (SD)
(internal EDGE)  1. LAMS migration
16% (2/12)
 2. Persistent fistula
after attempted closure
8% (1/12)
C. Boumitri et al.
14 Endoscopic Ultrasound-Guided Access to the Stomach in Patients with Prior Gastric Bypass… 157

References tube placement in patients with Roux-en-Y gastric

bypass: a novel technique for access to the gastric
remnant. Gastrointest Endosc. 2011;74(3):677–82.
1. Flegal KM, Kruszon-Moran D, Carroll MD,
9. Attam R, Leslie D, Arain MA, Freeman ML,
Fryar CD, Ogden CL. Trends in Obesity Among
Ikramuddin S. EUS-guided sutured gastropexy for
Adults in the United States, 2005 to 2014. JAMA.
transgastric ERCP (ESTER) in patients with Roux-­
2016;315(21):2284–91.
en-­Y gastric bypass: a novel, single-session, minimally
2. Fryar CDCM, Ogden CL. Prevalence of overweight,
invasive approach. Endoscopy. 2015;47(7):646–9.
obesity, and extreme obesity among adults aged 20
10. Kedia P, Kumta NA, Widmer J, Sundararajan
and over: United States, 1962–1962 through 2011–
S, Cerefice M, Gaidhane M, et al. Endoscopic
2014. National Center for Health Statistics Data,
ultrasound-­directed transgastric ERCP (EDGE) for
Health E-Stats, July 2016. Available from https://
Roux-en-Y anatomy: a novel technique. Endoscopy.
www.cdc.gov/nchs/data/hestat/obesity_adult_13_14/
2015;47(2):159–63.
obesity_adult_13_14.htm.
11. Kedia P, Tyberg A, Kumta NA, Gaidhane M, Karia
3. Angrisani L, Santonicola A, Iovino P, Formisano G,
K, Sharaiha RZ, et al. EUS-directed transgastric
Buchwald H, Scopinaro N. Bariatric surgery world-
ERCP for Roux-en-Y gastric bypass anatomy: a
wide 2013. Obes Surg. 2015;25(10):1822–32.
minimally invasive approach. Gastrointest Endosc.
4. Choi EK, Chiorean MV, Cote GA, El H II, Ballard
2015;82(3):560–5.
D, Fogel EL, et al. ERCP via gastrostomy vs. double
12. Law R, Wong Kee Song LM, Petersen BT, Baron
balloon enteroscopy in patients with prior bariat-
TH. Single-session ERCP in patients with previ-
ric Roux-en-Y gastric bypass surgery. Surg Endosc.
ous Roux-en-Y gastric bypass using percutaneous-­
2013;27(8):2894–9.
assisted transprosthetic endoscopic therapy: a case
5. Shah RJ, Smolkin M, Yen R, Ross A, Kozarek RA,
series. Endoscopy. 2013;45(8):671–5.
Howell DA, et al. A multicenter, U.S. experience
13. Law R, Grimm IS, Baron TH. Modified percutaneous
of single-balloon, double-balloon, and rotational
assisted transprosthetic endoscopic therapy for trans-
overtube-assisted enteroscopy ERCP in patients with
gastric ERCP in a gastric bypass patient. Endoscopy.
surgically altered pancreaticobiliary anatomy (with
2016;48(Suppl 1):E16–7.
video). Gastrointest Endosc. 2013;77(4):593–600.
14. Tyberg A, Nieto J, Salgado S, Weaver K, Kedia
6. Maaser C, Lenze F, Bokemeyer M, Ullerich H,
P, Sharaiha RZ, et al. Endoscopic ultrasound
Domagk D, Bruewer M, et al. Double balloon enter-
(EUS)-directed transgastric endoscopic retro-
­
oscopy: a useful tool for diagnostic and therapeutic
grade cholangiopancreatography or EUS: mid-term
procedures in the pancreaticobiliary system. Am
analysis of an emerging procedure. Clin Endosc.
J Gastroenterol. 2008;103(4):894–900.
2017;50(2):185–90.
7. Banerjee N, Parepally M, Byrne TK, Pullatt RC, Cote
15. Ngamruengphong S, Nieto J, Kunda R, Kumbhari V,
GA, Elmunzer BJ. Systematic review of transgastric
Chen YI, Bukhari M, et al. Endoscopic ultrasound-­
ERCP in Roux-en-Y gastric bypass patients. Surg
guided creation of a transgastric fistula for the
Obes Relat Dis. 2017;13(7):1236–42.
management of hepatobiliary disease in patients
8. Attam R, Leslie D, Freeman M, Ikramuddin S, Andrade
with Roux-en-Y gastric bypass. Endoscopy. 2017;
R. EUS-assisted, fluoroscopically guided gastrostomy
49(6):549–52.
 Endoscopic Ultrasound-Guided
Gastroenterostomy (EUS-GE) 15
Steven P. Shamah and Uzma D. Siddiqui

jejunostomy (open or laparoscopic) or, more

Background commonly, endoscopic enteral stent placement.
Enteral stents that are currently available have
Gastric outlet obstruction (GOO) is a term used been used in malignant obstruction for over
to define any mechanical obstruction, typically in 15 years with high success rates (>90%) when
the distal stomach or proximal duodenum, compared to surgical gastrojejunostomy [2].
impeding gastric emptying. Presenting signs and Although surgical gastrojejunostomy offers bet-
symptoms include nausea, vomiting, early sati- ter long-term clinical outcomes, it is associated
ety, and abdominal distention. Workup can with higher rates of morbidity and mortality
include upper endoscopy and cross-sectional when compared to endoscopic stenting in some
imaging, such as computerized tomography studies, although many surgeons prefer not to
(CT). These allow for better delineation of the operate on patients with advanced unresectable
location, severity, and etiology of the obstruction. malignancy. Retrospective data comparing
Currently, the most common etiology of GOO enteral stenting to surgery showed stenting had
has become pancreatic cancer, but in some cases significantly less complications and shorter hos-
may be from benign disease such as chronic pan- pital stays as well as decreased costs, but there
creatitis or peptic ulcer disease, among other was a higher re-intervention rate [3]. Enteral
causes. Other malignancies that can cause GOO stents are designed for patients with malignancy
include cholangiocarcinoma, ampullary cancer, who have short life expectancies and may not be
or gallbladder cancer [1]. Unfortunately, many ideal for use in benign conditions [4–6].
patients with GOO may not be candidates for pri- Endoscopic ultrasound (EUS)-guided creation
mary tumor resection and are therefore managed of a gastroenterostomy has recently been
with palliative measures such as surgical gastro- described and studied as a viable alternative
treatment for gastric outlet obstruction (GOO)
arising from benign and malignant conditions.
Electronic supplementary material: The online version EUS-guided gastroenterostomy (EUS-GE), also
of this chapter (https://doi.org/10.1007/978-3-319-97376- referred to as EUS-gastrojejunostomy (EUS-GJ),
0_15) contains supplementary material, which is available was made possible with the advent of lumen
to authorized users. apposing metal stents (LAMS) used to create
S. P. Shamah · U. D. Siddiqui (*) luminal anastomoses. In 2012, Binmoeller and
Center for Endoscopic Research and Therapeutics Shah first described this technique using a porcine
(CERT), University of Chicago, Chicago, IL, USA
e-mail: [email protected]; model [7]. The procedure was performed using
[email protected] an anchor wire to appose the lumen of the small

© Springer Nature Switzerland AG 2019 159

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_15
160 S. P. Shamah and U. D. Siddiqui

bowel to the stomach and a LAMS was deployed precluding safe creation of a gastrojejunostomy,
under EUS guidance to create the gastroenteric a distance of >1 cm between stomach and small
anastomosis. The procedure was technically suc- bowel walls, and large volume ascites.
cessful in all animals without complications.
Another animal study by Itoi et al. showed similar
results with a successful gastroenterostomy cre- Pre-procedure Care
ation and no adverse events [8]. In this study, the
authors utilized a novel double balloon enteric Regardless of the technique, all patients undergo-
tube to access and stabilize the small bowel. ing EUS-GE should receive IV antibiotics. Some
Since then, there have been multiple studies in suggest the patient be positioned in the supine
patients and these have utilized various EUS-GE position, and be intubated prior to initiating the
techniques and types of LAMS. procedure as well, but these recommendations are
In the USA, a cautery-enhanced (CE) LAMS not standardized. IV glucagon can be adminis-
system (Hot Axios, Boston Scientific, Natick, tered to decrease small bowel contractions during
MA) allows for direct puncture through the stom- the procedure, however minimal data currently
ach and into the small bowel and obviates the exists to support its routine use.
need for tract dilation or fluoroscopy with stent The main procedural steps in EUS-GE include
deployment. Furthermore, the single-step access filling the target small bowel with water or con-
to the small bowel may minimize the chance for trast to distend it for better identification and
separation between the stomach and small bowel. apposition with the gastric wall, creation of a gas-
However, the use of LAMS and CE-LAMS for troenterostomy, and finally stent deployment
gastroenterostomy is an off-label indication. across the gastroenterostomy with the distal
The following chapter will review the differ- flange in the small bowel and the proximal flange
ent EUS-GE techniques that have been described in the stomach. In the USA, LAMS come in
in the literature, as well as the data on safety and 10 mm, 15 mm, and 20 mm diameters (our pref-
efficacy. erence is 15 mm). We will henceforth describe
the four EUS-GE techniques that have been
described in the literature.
Endoscopic Technique
EUS GJ Techniques
EUS-GE using LAMS was developed as a way to
bypass the obstructed portion of the GI tract in Water Immersion Method [9]
GOO with direct placement of a stent between A large volume of isotonic saline with or without
the stomach and more distal duodenum or proxi- a readily identifiable dye such as methylene blue
mal jejunum. This new endoscopic technique is infused through the working channel of the lin-
continues to evolve as endosonographers gain ear echoendoscope, via a puncture of the distal
more clinical experience and as more devices are small bowel with a 22-gauge (G) fine needle
developed for the creation of endoscopic entero-­ aspiration (FNA) needle, or via a nasojejunal
enteric anastomosis. There is no “ideal method” tube. This infusion of fluid distends the target
for how to perform this procedure and the tech- small bowel, allowing for better visualization
nique itself has multiple steps that require an under EUS. A 19 G FNA needle is then used to
expert operator and a multidisciplinary plan of puncture the gastric wall and through the small
care. EUS-GE indications include both malig- bowel wall into the distended loop of small
nant and benign obstruction; while contraindica- bowel. Aspiration of fluid (with or without dye)
tions include multi-focal obstruction or blockage to confirm proper placement in the small bowel is
distal to small bowel puncture site, coagulopathy helpful. A 0.035-inch or 0.025-inch guidewire is
15 Endoscopic Ultrasound-Guided Gastroenterostomy (EUS-GE) 161

passed through the FNA needle into the small place across the obstruction. Over the guidewire
bowel and then the 19 G needle is exchanged a specialized double balloon enteric tube (Tokyo
over the wire, leaving the wire connecting the Medical University, Japan) is passed in combi-
gastric cavity and the small bowel. A 10-French nation with a guidewire that allows for better
cautery-enhanced LAMS catheter system is maneuverability. Once the enteric tube is in
passed over the guidewire and deployed under across the obstruction the distal and proximal
endosonographic guidance with the distal pha- balloons are filled with saline. The lumen
lange in the small bowel and the proximal end in between the balloons is filled with saline and
the gastric cavity. The lumen of the LAMS is then contrast, thereby distending a focal area of small
dilated to the diameter of the stent diameter using bowel and limiting the amount of fluids infused.
a through the scope (TTS) dilation balloon, A linear echoendoscope is advanced into the
although dilation of the LAMS is not considered gastric cavity and the lumen between the bal-
a mandatory step as the stent may be left to fully loons is identified endosonographically. A 19 G
efface on its own. FNA needle is used to puncture the small bowel
in the space between the balloons and fluid is
Balloon Assisted Method [10] aspirated confirming location. If a non-cautery-
A standard endoscope is advanced to the level of enhanced LAMS is used, then a dilation balloon
the obstruction. A stiff guidewire is passed across and fluoroscopy is needed prior to stent deploy-
the obstruction into the small bowel, under fluo- ment. If the CE-LAMS is used, then the delivery
roscopic guidance. The gastroscope is withdrawn system is placed over the guidewire and deployed
leaving the wire in the small bowel distal to the using cautery under EUS guidance. The distal
obstruction. Under fluoroscopic guidance, a large phalange is deployed into the distal small bowel
caliber (18–20 mm) TTS dilation balloon is and the proximal end into the gastric cavity. The
passed over the wire into the small bowel and stent lumen is balloon dilated to LAMS maximal
inflated with a contrast agent. The echoendo- diameter if so desired.
scope is then passed into the gastric cavity and is
used to locate the dilation balloon using EUS Direct (“Freehand”) Method
imaging. Once located, a 19 G FNA needle is The small bowel lumen is first distended with
used to puncture across the gastric wall, the small copious amounts of isotonic saline, a contrast
bowel wall, and the balloon itself across the small agent, and dye (usually methylene blue) [12]. An
bowel to ensure proper access. A second guide- echoendoscope is advanced to the distal stomach
wire is passed downstream into the small bowel and the closest loop of small bowel is identified.
through the 19 G needle. A CE-LAMS is then Once the small bowel site is confirmed, the
deployed over the guidewire creating the LAMS CE-LAMS catheter system is used to puncture
assisted gastroenterostomy. The LAMS is then through the stomach and into the small bowel and
balloon dilated to the maximal stent diameter the LAMS is deployed creating the gastroenter-
(either 10 mm or 15 mm) if so desired (Figs. 15.1, ostomy without the use of a wire, dilation bal-
15.2, 15.3, 15.4, 15.5, 15.6, 15.7, 15.8, and 15.9 loon, or fluoroscopy [12]. This technique mirrors
and Video 15.1 demonstrate balloon assisted the common “wire free” techniques used to drain
technique). pancreatic fluid collections via CE-LAMS.
The above techniques described do not repre-
EUS-guided Double-Balloon-Occluded sent an exhaustive description of the procedures
Gastrojejunostomy Bypass: EPASS available to perform EUS-GE; however, they are
Method [11] the most commonly used techniques for the
A standard endoscope is advanced to the level of establishment of an EUS-guided gastroenteros-
the obstruction. A guidewire is passed under tomy. As expertise with currently available tools
fluoroscopic guidance across the obstruction. grows and/or new technology is created, these
The endoscope is exchanged leaving the wire in techniques will continue to evolve.
162 S. P. Shamah and U. D. Siddiqui

Fig. 15.1 (a) Passage of a guidewire across the gastric outlet obstruction. (b) Contrast injection confirms guidewire
passage into the small bowel distal to the level of the obstruction

Fig. 15.2 (a) Dilation balloon passed over the guidewire into the small bowel under fluoroscopy. (b) The dilation bal-
loon is inflated via filling it with contrast dye

Post-procedure and Follow-up Care demonstrate adequate tolerance to oral diet, at

least to the level of liquids in most cases [14]. In
There is no consensus on the post-procedural terms of long-term management, there are no
management of patients undergoing available data on (a) optimal duration of LAMS
EUS-GE. Patients will generally be admitted for placement or (b) whether or not the LAMS
close observation; will be kept NPO for the first should be removed at some time in the future
24 h and then advanced to clear liquids for when the anastomosis becomes chronic and sta-
2–3 days. Antibiotic regimens vary from any- ble. Closure of the anastomosis after LAMS
where from 3 days to 7 days post-procedure [9, removal is a possibility, but the frequency of this
13]. Patients are discharged home when they remains unknown.
15 Endoscopic Ultrasound-Guided Gastroenterostomy (EUS-GE) 163

Fig. 15.3 Linear echoendoscope passed into the stomach and used to locate the dilation balloon in the adjacent small
bowel

Fig. 15.4 19 G needle used to puncture the balloon in the small bowel under EUS guidance
164 S. P. Shamah and U. D. Siddiqui

Fig. 15.5 (a) Second guidewire passed through the EUS needle from the stomach and into the small bowel. (b)
Electrocautery-enhanced LAMS is passed over the wire and into the small bowel beyond the obstruction

Fig. 15.6 (a) Distal flange of LAMS deployed into the small bowel. (b) Endoscopic view of the proximal flange of the
lumen apposing metal stent (LAMS) after deployment. (c) Fluoroscopic view of fully deployed LAMS before dilation
15 Endoscopic Ultrasound-Guided Gastroenterostomy (EUS-GE) 165

Fig. 15.7 Balloon dilation of the LAMS to ensure maximal opening diameter

Fig. 15.9 Final endoscopic view of LAMS when viewed

from the gastric side

Fig. 15.8 Fluoroscopic view of the fully deployed

LAMS between the stomach and small bowel creating the
gastroenterostomy
166 S. P. Shamah and U. D. Siddiqui

Review of Data nostomy. There were five reported adverse events

in the EUS-GE group, which included three
Technical and Clinical Success patients with stent misdeployments into the peri-
toneum and two patients with severe abdominal
Most of the data for EUS-GE can be found in case pain requiring hospitalization [15]. All five
reports and case series, it is only recently that com- patients were managed conservatively and there
parative studies have emerged from the literature. were no fatalities. In this study, surgical gastroje-
The largest retrospective case series, published by junostomy had a higher technical success rate but
Tyberg et al., was a multicenter international col- a similar clinical success rate when compared to
laboration encompassing 26 patients with GOO EUS-GE.
(17 malignant, 9 benign) [13]. Technical success Misdeployment rates of LAMS range from 4
was 92%, defined as a successful creation of an to 6% across the multiple case series emphasiz-
EUS-GE. Clinical success, defined as the ability to ing the difficulty in maintaining lumen apposi-
tolerate PO diet was slightly lower at 85%. In a tion between the small bowel and the gastric
case series of 20 patients; Itoi and his colleagues in cavity, even in expert hands [11–16]. Adverse
Japan published their experience with a double events can be potentially severe or life threaten-
balloon enteric tube that they used to facilitate gas- ing and therefore close collaboration with surgi-
troenterostomy formation [14]. In their experi- cal colleagues, review of the technique, and
ence, similar technical success rate of 90% was careful patient selection are keys to ensure suc-
obtained. cess and safety during this intervention.

 afety and Adverse Events Related

S EUS-GE vs Enteral Stenting
to EUS-GE
Enteral stenting (ES) has historically offered a
Since EUS-GE is still in its investigational stages less invasive palliative approach to relieve malig-
with comparative studies just beginning to be nant gastric outlet obstruction in comparison to its
published, most of the data regarding safety and counterpart surgical gastrojejunostomy. However,
adverse events is derived from a limited number with recent technical advances in therapeutic
of patients. The studies include patients with endoscopic ultrasound and stent technology
benign and malignant gastric outlet obstruction EUS-GE has emerged as another palliative option.
and reported an overall adverse event rate of less Chen and colleagues conducted the largest retro-
than 12% [11–16]. spective analysis to date comparing EUS-GE vs
Tyberg and colleagues reported three adverse enteral stenting (ES). Primary outcomes included
events in their patient series (11.5%) described as symptom recurrence rates and rates of re-inter-
bleeding, worsening abdominal pain and, notably, vention, while secondary outcomes included tech-
one death caused by peritonitis [13]. The patient nical and clinical success rates between these two
with peritonitis had ascites and carcinomatosis endoscopic techniques [17].
and experienced a LAMS misdeployment. Fifty-two patients were included into the ES
Khashab in 2015 published a case series of ten group and thirty in the EUS-GE group. Other
patients and reported no procedure related adverse than age, with the ES group being significantly
events [12]. The case series by Itoi et al. reported younger, both groups had similar baseline char-
one adverse event (5%) out of twenty patients, acteristics. Similar rates of technical (86.7%
which resulted in pneumoperitoneum [14]. EUS-GE vs 94.2% ES; P = 0.20) and clinical
In a recently published comparative study success (83.3% EUS-GE vs 67.3% ES; P = 0.12)
with surgical gastrojejunostomy by Khashab et. were identified between the two groups. Enteral
al., 93 patients were studied: 30 undergoing stenting had a higher rate of re-intervention and
EUS-GE and 63 undergoing surgical gastrojeju- symptom recurrence (28.6%) in comparison to
15 Endoscopic Ultrasound-Guided Gastroenterostomy (EUS-GE) 167

EUS-GE (4%). Enteral stenting is still in wide- $16,541) compared to the cost of an EUS-GE at
spread practice and is often technically simple, $4515 (95% confidence interval, $4079–$4905.5)
whereas EUS-GE is still a relatively uncommon (P < 0.00001) [16].
procedure

Conclusion and Future Direction

 US-GE vs Surgical
E
Gastrojejunostomy (SGJ) Several case series and comparative studies have
confirmed that EUS-GE can be performed with
Two recent studies have been published compar- high technical and clinical success rates. Recent
ing surgical gastrojejunostomy (SGJ) and EUS-GE data has suggested that this approach for the
by evaluating the difference in clinical success, treatment of GOO may also be a more cost-­
technical success, adverse events, length of hospi- effective option compared to surgical gastrojeju-
tal stay (LOHS), and symptom recurrence. nostomy and more durable than enteral stenting.
Khashab and colleagues conducted a non-­ Several challenges and unanswered questions
randomized, retrospective multicenter study that still exist. Further comparative studies need to be
compared open surgical gastrojejunostomy vs conducted to establish which EUS-GE technique
EUS-GE [15]. A total of 93 patients (63 SGJ and can offer the highest rates of success and limit
30 EUS-GE) were included in the analysis. There adverse events. There also needs to be develop-
was no standardization in the technique of ment of better tools and devices to reproducibly
EUS-GE. All EUS-GE used LAMS to create the maintain close apposition of the small bowel with
gastroenteric anastomosis. The EUS-GE group the stomach to prevent misdeployment of
had a higher rate of peritoneal carcinomatosis. LAMS. With advancements in scope technology,
While technical success rates significantly stent design, and device development we look for-
favored open SGJ vs EUS-GE (100% vs 87% ward to EUS-GE, a procedure currently in its
p = 0.009), the clinical success rates (defined as infancy, developing into a weathered alternative to
the ability to tolerate oral intake without vomit- SGJ and enteral stenting for the treatment of GOO.
ing) were similar (90% SGJ vs 87% EUS-GE
p = 0.18). Rate of adverse events was lower in the
EUS-GE group but was not statistically signifi- References
cant (16% vs 25% p = 0.3). The mean length of
hospital stay, time to re-intervention, and rate of 1. Conrad C, Lillemoe KD. Surgical palliation of pan-
creatic cancer. Cancer Journal. 2012;18(6):577–83.
GOO recurrence were similar between the two 2. Fukami N, Anderson MA, Khan K, et al. ASGE
groups, concluding that EUS-GE is a less inva- Standards of Practice Guidelines: the role of endos-
sive and comparable palliative intervention [14]. copy in gastroduodenal obstruction and gastroparesis.
In a similar study Perez-Miranda et al. com- Gastrointest Endosc. 2011;74(1):13–21.
3. Khashab M, Alawad AS, Shin EJ, et al. Enteral
pared EUS-GE to laparoscopic surgical gastroje- stenting versus gastrojejunostomy for palliation of
junostomy and demonstrated similar clinical and malignant gastric outlet obstruction. Surg Endosc.
technical success rates [16]. However, this study 2013;27(6):2068–75.
demonstrated lower adverse events in the EUS-GE 4. Tringali A, Didden P, Repici A, et al. Endoscopic treat-
ment of malignant gastric and duodenal strictures: a
group (EUS-GE 12% vs laparoscopic GJ 41%, prospective, multicenter study. Gastrointest Endosc.
p = 0.03) and a significant healthcare cost benefit 2014;79(1):66–75.
of approximately $10,000 per procedure in favor 5. Jeurnink SM, Steyerberg EW, Hof G, et al.
of EUS-GE.18 Gastrojejunostomy versus stent placement in patients
with malignant gastric outlet obstruction: a compari-
In the era of cost-effective medicine, the lower son in 95 patients. J Surg Oncol. 2007;96:389–96.
cost of EUS-GE compared to SGJ should be con- 6. Phillips MS, Gossain S, Bonatti H, et al. Enteral stents
sidered. The expected cost of a laparoscopic SGJ for malignancy: a report of 46 consecutive cases
is $14,778.80 (95% confidence interval, $14,807– over 10 years, with critical review of complications.
168 S. P. Shamah and U. D. Siddiqui

J Gastrointest Surg. 2008;12(11):2045–50. 13. Tyberg A, Perez-Miranda M, Sanchez-Ocana R, et al.

7. Binmoeller KF, Shah JN. Endoscopic ultrasound-­ Endoscopic ultrasound-guided gastrojejunostomy with a
guided gastroenterostomy using novel tools designed lumen-apposing metal stent: a multicenter, international
for transluminal therapy: a porcine study. Endoscopy. experience. Endosc Int Open. 2016;4(3):E276–81.
2012;44(5):499–503. 14. Itoi T, Ishii K, Tanaka R, et al. Current status and
8. Itoi T, Itokawa F, Uraoka T, et al. Novel EUS-guided perspective of endoscopic ultrasonography-guided
gastrojejunostomy technique using a new double-­ gastrojejunostomy: endoscopic ultrasonography
balloon enteric tube and lumen-apposing metal stent guided double balloon-occluded gastrojejunos-
(with videos). Gastrointest Endosc. 2013;78(6):934–9. tomy (with videos). J Hepatobiliary Pancreat Sci.
9. Amin S, Sethi A. Endoscopic ultrasound guided gas- 2015;22:3–11.
trojejunostomy. Gastrointest Endoscopy Clin N Am. 15. Khashab MA, Bukhari M, Baron TH, et al.
2017;27:707–13. International multicenter comparative trial of endo-
10. Irani S, Baron T, Itoi T, et al. Endoscopic gastro- scopic ultrasonography-guided gastroenterostomy
enterostomy: techniques and review. Curr opin versus surgical gastrojejunostomy for the treatment of
Gastroenterol. 2017;33(5):320–9. malignant gastric outlet obstruction. Endosc Int Open.
11. Itoi T, Ishii K, Ikeuchi N, et al. Prospective evalua- 2017;5(4):E275.
tion of endoscopic ultrasonography-guided double-­ 16. Perez-Miranda M, Tyberg A, Poletto D, et al. EUS-­
balloon-­occluded gastrojejunostomy bypass (EPASS) guided gastrojejunostomy versus laparoscopic gastro-
for malignant gastric outlet obstruction. Gut. jejunostomy: An International Collaborative Study.
2016;65(2):193–5. J Clin Gastroenterol. 2017;51(10):896–9.
12. Khashab MA, Kumbhari V, Grimm IS, et al. EUS-­ 17. Chen Y-I, Itoi T, Baron TH, et al. EUS-guided gas-
guided gastroenterostomy: the first U.S. clini- troenterostomy is comparable to enteral stenting with
cal experience (with video). Gastrointest Endosc. fewer re-interventions in malignant gastric outlet
2015;82(5):932–8. obstruction. Surg Endosc. 2017;31(7):2946–52.
 Endoscopic Ultrasound-Guided
Portal Pressure Measurement 16
Jason B. Samarasena, Allen R. Yu,
and Kenneth J. Chang

Introduction EUS-Guided Portal Venous

Angiography
A growing number of studies have explored
endoscopic ultrasound (EUS)-guided vascular The portal vein is well seen from both the stom-
catheterization due to the relative proximity of ach and the duodenum during EUS. The vessel
the gastrointestinal tract to the major blood ves- itself is usually in very close proximity to the tip
sels of the mediastinum and abdomen and the use of the echoendoscope, making this an ideal target
of Doppler during EUS to ensure the absence of for vascular access. Portal venous angiography is
hemorrhage with needle puncture and withdrawal a modality to assess the anatomy of the hepatic
without additional administration of ionizing vasculature. Initial cases of successful in vivo
radiation. In particular, EUS-guided access to the EUS-guided PV catheterization were performed
portal vein (PV) may be favorable given the rela- in porcine models. In 2004, Lai and colleagues
tive difficulty of PV access via standard percuta- reported an EUS-guided transduodenal approach
neous routes. Two major diagnostic applications to access the extrahepatic portal vein in 21 swine
of EUS-guided vascular access include angiogra- with a 22 G fine needle aspiration (FNA) needle.
phy and assessment of intravascular pressure. A small amount of contrast was injected through
This review will outline the different devices and the needle for fluoroscopic confirmation of
techniques employed to obtain angiographic proper placement [1]. This study proved the fea-
visualization and/or direct pressure measure- sibilty on a technical level of EUS-guided portal
ments of the portal circulation. Ease of access, vein access.
safety, and important lessons learned from each The first study solely assessing PV angiogra-
approach will be highlighted. phy was a porcine study reported in 2007 by
Magno and colleagues [2]. 19 G, 22 G, and 25 G
Electronic supplementary material: The online version needles were inserted under EUS guidance into
of this chapter (https://doi.org/10.1007/978-3-319-97376-
0_16) contains supplementary material, which is available
the celiac, splenic, superior mesenteric artery, the
to authorized users. thoracic and abdominal aorta, and the splenic,
J. B. Samarasena · A. R. Yu · K. J. Chang (*)
portal, and hepatic veins. All vessels were suc-
Department of Gastroenterology, cessfully identified and punctured in 5 of 5 pigs.
H. H. Chao Comprehensive Digestive Disease Center, No signs of intraprocedural hemodynamic insta-
University of California, Irvine Medical Center, bility were observed. Immediate post-procedure
Orange, CA, USA
e-mail: [email protected]; [email protected];
necropsy showed no signs of injury with the 25 G
[email protected] needle. The 22 G needle left puncture marks

© Springer Nature Switzerland AG 2019 169

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_16
170 J. B. Samarasena et al.

without bleeding, and the 19 G needle caused a  US-Guided Portal Pressure

E
vascular hematoma in large-caliber vessels with Gradient Measurement
intraabdominal bleeding in 1 of the 5 pigs.
Injection of contrast provided good opacification Portal hypertension (PH), resulting from increased
of smaller vessels—the celiac trunk, splenic resistance of hepatic sinusoids to blood flow, is
artery, and hepatic veins—with only transient most commonly a complication of liver cirrhosis.
opacification in larger caliber vessels. As would The pathogenesis involves alteration of the liver
be expected, the amount of resistance associated vasculature due to fibrosis as well as increased
with instilling the iodinated contrast was inversely production of vasoconstrictive mediators relative
correlated with needle caliber. to endogenous vasodilators. Complications of PH
Giday and colleagues attempted EUS-guided include esophageal varices, portal hypertensive
PV access in 2007 using a transgastric, transhe- gastropathy, ascites, and hepatorenal syndrome.
patic approach with a 25 G needle and a modified Measurement of portal hypertension has been use-
endoscopic retrograde cholangiopancreatogra- ful in determining the stage, progression, and
phy (ERCP) catheter [3]. This protocol was again prognosis of cirrhosis in individual patients. Portal
performed in 2008 as part of another PV catheter- pressure gradient measurement (PPG) of
ization study [4]. Angiography was achieved ≥10 mmHg is associated with development of
using both standard iodinated contrast and medi- esophageal varices [8] and PPG of ≥12 mmHg
cal grade carbon dioxide (CO2). PV catheteriza- with variceal hemorrhage [9]. Reduction of PPG
tion was achieved in 6 of 6 swine in 2007 and 6 by 20% or to below 12 mmHg with pharmacother-
of 6 swine in 2008, and no complications were apy has been found to decrease risk of future
noted in either study. Necropsy showed no evi- bleeding or rebleeding episodes [10, 11].
dence of bleeding, hematoma formation, or liver Previously, PPG values were obtained directly
injury. The transgastric, transhepatic approach is via either a percutaneous approach or using a
postulated to be safer than the transduodenal transjugular intrahepatic portosystemic shunt
approach by allowing for natural tamponade of (TIPS). The current standard for evaluation of PH
the needle track by liver parenchyma during is indirect measurement of the hepatic venous
withdrawal [3, 5]. The use of CO2 as a contrast pressure gradient (HVPG). In this technique, a
medium allowed for better visualization of the catheter is inserted into the hepatic vein percuta-
PV as well as easier intravascular administration neously via either the jugular or femoral vein. The
through the small-caliber FNA needle when com- free hepatic venous pressure is recorded and sub-
pared to the viscous iodine-based contrast. These tracted from the wedged hepatic venous pressure
studies as a whole suggested that needle puncture to determine the HVPG. Both percutaneous PV
of these vessels would not necessarily lead to catheterization and HVPG measurement are inva-
intraabdominal hemorrhage or vascular injury. sive procedures and require a high level of techni-
The safety of CO2 use has been evaluated in cal expertise. Direct PV catheterization has been
both animals and humans. It is highly soluble associated with a high complication rate [12, 13]
and easily cleared by the lungs [6] and, unlike and is not commonly performed. Despite the
iodinated contrast, is not associated with neph- overall safety profile of HVPG measurement, it is
rotoxicity or increased risk for hepatorenal syn- only routinely performed at tertiary medical cen-
drome [7]. The current data suggest that ters [14, 15]. Furthermore, HVPG has been shown
combining CO2 with a 25 G needle may allow to correlate poorly with directly measured portal
for easier injection of contrast, adequate visual- pressure in cases of presinusoidal PH, which may
ization of the portal circulation, and possibly be seen in cases of non-cirrhotic portal fibrosis
decreased risk of needle- and contrast-related and presinusoidal PH, including portal vein
complications. thrombosis and schistosomiasis [4, 16, 17].
16 Endoscopic Ultrasound-Guided Portal Pressure Measurement 171

Animal Studies during ERCP. Values of IVC pressure, as well as

of PVP for EGD and colonoscopy, were similar
Lai and colleagues were the first to report EUS-­ between baseline and procedure time.
guided PVP measurement in a porcine model [1]. Schulman and colleagues demonstrated a novel
In a cohort of 21 pigs, a PH model was generated method of measuring PVP in 2016 using an
in 14 animals using polyvinyl alcohol injection EUS-guided 22 G needle through which a wire
and a coagulopathy model generated in 7 animals with a digital pressure sensor was passed [19].
with heparin administration. A transduodenal Conventional transjugular catheterization was per-
EUS approach was used to access the portal vein formed as a control. Successful device placement
in 21 pigs with a 22 G FNA needle and a transab- and PVP measurement were achieved in 5 of 5
dominal ultrasound (TAUS)-guided transhepatic pigs with no hemorrhage or thrombosis noted on
approach in 14 of 21 pigs via a 22-gauge needle. both EUS and post-procedural necropsy.
PVP measurements were obtained in 18 of 21 Comparison of EUS-measured PVP with transjug-
swine. Minor complications found at necropsy ular HVPG measurements showed a difference of
included small subserosal hematomas at the EUS within 1 mmHg for all pigs. The study endosco-
puncture site in all 21 pigs and a 25 mL blood col- pists rated the procedure as having overall low
lection between the liver and duodenum in 1 of 7 subjective workload. The authors used the same
anticoagulated pigs. Failure to measure pressures device to perform PVP measurement in 5 other
in 3 subjects may have occurred due to thrombo- pigs that were then survived for 14 days before
sis within the FNA needle. There was a strong necropsy [20]. PVP was again measured on day
correlation between EUS- and transhepatic-­ 14. No signs of complications were observed dur-
measured PVP (r = 0.91). The development of ing the 2-week survival period, and necropsy again
hematomas in this study suggests that a transduo- showed no abnormalities. PVP values on day 0
denal approach that does not traverse the liver and day 14 were similar for all 5 pigs.
may increase risk of bleeding and therefore an Our group developed a method of EUS-guided
approach traversing through liver parenchyma portal pressure measurement using a 25 G needle
may be favorable. and simple transducer setup. The apparatus for
In 2007, Giday and colleagues used the trans- PPG measurement included a linear echoendo-
gastric approach with a 19 G needle and modified scope, a 25 G FNA needle, and a compact manom-
ERCP catheter to obtain continuous PVP mea- eter (Fig. 16.1) with non-compressible tubing
surement without an echoendoscope in place [3]. [21]. Prior to echoendoscope insertion, the
Five of 5 pigs were successfully catheterized, and manometer was zeroed at the mid-axillary line.
no hemorrhage or liver injury was noted on nec- Measurements were conducted in the portal vein
ropsy in all subjects despite the use of a signifi- (PV) and hepatic vein (HV) and the inferior vena
cantly larger caliber needle. Two of 5 pigs were cava (IVC). When the PV was targeted, manom-
survived for two weeks and exhibited no signs of etry was performed via a transgastric, and less
adverse events prior to and after necropsy. In a often a transduodenal, transhepatic approach and
later study, the same group used the same meth- only the intrahepatic portion near the PV bifurca-
ods to measure fluctuations in PVP and inferior tion was accessed (Fig. 16.1). When evaluating
vena cava (IVC) pressures in pigs that underwent the HV, the needle tip was placed 2 cm distal to
common endoscopic procedures: esophagogas- the ostia where possible. Needle placement was
troduodenoscopy (EGD), colonoscopy, and meticulous to ensure consistency. One milliliter
ERCP [18]. PV and IVC were accessed using a of heparinized saline was flushed through the nee-
19 G needle and modified ERCP catheter. Access dle before pressure measurement to clear the nee-
and pressure measurements of both vessels were dle lumen and confirm intravascular placement.
achieved in 5 of 5 pigs. Necropsy showed no evi- We also measured pressures in a swine model of
dence of injury in all subjects. A threefold portal hypertension induced by Dextran-40
increase in PVP was noted between baseline and administration. Percutaneous measurements in
172 J. B. Samarasena et al.

with arteriovenous malformations secondary to

Noonan syndrome. The measured portal pressure
gradient was 1 mmHg and correlated with the gra-
dient obtained by interventional radiology at a
prior procedure. There was no evidence of bleed-
ing or hemodynamic instability after this proce-
dure [22].
Our group performed the first prospective
pilot study of PPG measurement in human
patients with suspected or confirmed cirrhosis
[23]. The setup employed the simple transducer
setup discussed above with our animal study. The
compact manometer was zeroed at the mid-­
axillary line of each patient, and care was taken
to consistently place the needle 2 cm distal to the
hepatic vein ostia. Pressure readings were taken
of the PV and either the HV or the IVC if anat-
omy was unfavorable for HV access. Needle
placement was achieved and PPG measurement
obtained in 28 of 28 patients, and no adverse
events including bleeding, perforation, or infec-
tion were noted. The time required to obtain pres-
sure measurements was short, under 30 min per
Fig. 16.1 Compact manometer used for EUS-guided patient. PPG measurements correlated well with
portal pressure measurement (Cook Medical, clinical and endoscopic parameters with signifi-
Bloomington, IN). Reprinted from Gastrointestinal
Endoscopy, 85(5), EUS-guided portal pressure gradient
cant differences in PPG noted in patients that
measurement with a simple novel device: a human pilot were high-risk versus low-risk for cirrhosis and
study, May 1, 2017, with permission from Elsevier in patients with esophageal varices, portal hyper-
tensive gastropathy, and thrombocytopenia rela-
the same vessels were obtained for comparison. tive to patients without these conditions. There
All vessels were successfully accessed and pres- were no complications in any of the 28 patients.
sures measured via EUS in all 3 pigs. Necropsy In addition, the majority of the patients in this
was not performed, but intraprocedural monitor- study had EUS-guided liver biopsies performed
ing showed no signs of cardiorespiratory instabil- at the same procedure suggesting that combining
ity. Correlations between EUS-guided and a PPG measurement and liver biopsy in the same
percutaneous pressure measurements were very session should be safe.
strong, with R values in all vessels greater than or
equal to 0.985.
 US-guided PPG Measurement
E
Technique
Human Studies
The EUS manometry apparatus used in our
The first human single case of EUS-guided PVP human study is a simple setup that includes a
measurement was reported by Fujii-Lau and col- 25 G FNA needle, non-compressible tubing, a
leagues in 2014, in which a 22 G FNA needle con- compact digital manometer, and heparinized
nected to an arterial pressure catheter was used to saline (see Video 16.1). The tubing is connected
rule out portal hypertension in a 27-year-­old man by a luer lock to the distal port of the manometer,
16 Endoscopic Ultrasound-Guided Portal Pressure Measurement 173

while the heparinized saline is connected the or fluctuation and to give a range of pressures
proximal port. The end of the tubing is connected from which to derive a mean pressure. The mean
via a luer lock to the inlet of the 25 G needle. The of the three pressures is then considered the
patient is positioned supine and during EUS-­ hepatic vein pressure. The FNA needle is slowly
guided pressure measurement reading the withdrawn from the vein into the liver paren-
manometer is placed at the patient’s mid-axillary chyma and then back into the needle sheath with
line (Fig. 16.2). We prefer monitored anesthesia Doppler flow on to ensure there is no flow within
care or general anesthesia for this procedure. the needle tract.
The hepatic vein measurement is conducted The portal vein measurement is conducted
first. Of the hepatic veins, the middle hepatic vein next and the umbilical portion of the left portal
is targeted most commonly due to its larger cali- vein is targeted (Fig. 16.5). Doppler flow is used
ber and better alignment with the needle trajec- to confirm the typical venous hum of portal
tory on linear EUS (Fig. 16.3). Doppler flow is venous flow (Fig. 16.6). Using the 25 G FNA
used to confirm the typical multiphasic waveform needle, a transgastric transhepatic approach is
of hepatic venous flow (Fig. 16.4). Using the used to puncture the portal vein. The procedure
25 G FNA needle, a transgastric transhepatic that follows is the same as what was performed
approach is used to puncture the hepatic vein. for the hepatic vein. Approximately 1 cm3 of hep-
Approximately 1 cm3 of heparinized saline is arinized saline is used to flush the needle which is
used to flush the needle which is visible on EUS visible on EUS confirming good position within
confirming good position within the vessel. the vessel. Following the flush, the pressure read-
Following the flush, the pressure reading on the ing on the manometer will immediately rise and
manometer will immediately rise and then fall then fall and equilibrate at a steady pressure
and equilibrate at a steady pressure which is which is recorded. This measurement should be
recorded. This measurement should be repeated repeated and second and third time. The mean of
and second and third time to minimize any error the three pressures is then considered the portal

Fig. 16.2 Endoscopic ultrasound-guided portal pressure pact manometer being placed at the mid-axillary line of
measurement apparatus showing non-compressible tubing the patient (left panel)
attached to the FNA needle inlet (right panel) and com-
174 J. B. Samarasena et al.

Fig. 16.3 A: EUS image of needle puncture of middle pressure gradient measurement with a simple novel
hepatic vein with 25 G FNA needle. Reprinted from device: a human pilot study, May 1, 2017, with permis-
Gastrointestinal Endoscopy, 85(5), B: EUS-guided portal sion from Elsevier

Fig. 16.4 EUS Doppler flow image of middle hepatic vein demonstrating multiphasic waveform

vein pressure. The FNA needle is slowly procedural antibiotics are usually given for
­withdrawn from the vein into the liver paren- 5 days post-procedure.
chyma and then back into the needle sheath with
Doppler flow on to ensure there is no flow within
the needle tract. Conclusion
The portal pressure gradient is calculated
by subtracting the mean portal vein pressure Recent advances in the field of hepatology have
from the mean hepatic vein pressure. The included new and effective treatment for viral
patient is recovered in a similar manner to a hepatitis, with an increased need for assessment
routine diagnostic EUS with FNA. Post- of liver function and histology. At the same time
16 Endoscopic Ultrasound-Guided Portal Pressure Measurement 175

Fig. 16.5 A:EUS image of needle puncture of left portal p­ressure gradient measurement with a simple novel
vein with 25 G FNA needle. Reprinted from Gastro­ device: a human pilot study, May 1, 2017, with permis-
intestinal Endoscopy, 85(5), B: EUS-guided portal sion from Elsevier

Fig. 16.6 EUS Doppler flow image of left portal vein demonstrating typical waveform

there have been a growing number of endoscopic the portal pressure gradient would be a great
procedures that are pertinent to liver patients. It advance in the field of Endo-Hepatology. As we
would be ideal if the assessment and treatment of have just covered, the current literature suggests
liver disease and portal hypertension could be EUS-guided measurement of the portal pressure
performed and assimilated by the primary l­iver/ gradient is becoming safe and feasible. We look
GI specialist. We have termed this area of inte- forward to the results of an international multi-
gration or overlap of endoscopic procedures center human trial using our recently designed
within the practice of hepatology as Endo-­ manometry apparatus to further evaluate the
Hepatology. Given the wide availability of EUS, safety and clinical utility of this approach for
an EUS-guided approach for the measurement of patients with liver disease.
176 J. B. Samarasena et al.

References pressure adequately reflects portal pressure in hepati-

tis C virus-related cirrhosis. Hepatology. 1999;30(6):
1393–7.
1. Lai L, Poneros J, Santilli J, Brugge W. EUS-guided
13. Tsushima Y, Koizumi J, Yokoyama H, Takeda A,
portal vein catheterization and pressure measure-
Kusano S. Evaluation of portal pressure by splenic
ment in an animal model: a pilot study of feasibility.
perfusion measurement using dynamic CT. AJR Am
Gastrointest Endosc. 2004;59(2):280–3.
J Roentgenol. 1998;170(1):153–5.
2. Magno P, Ko CW, Buscaglia JM, Giday SA, Jagannath
14. Suk KT. Hepatic venous pressure gradient: clini-
SB, Clarke JO, et al. EUS-guided angiography: a
cal use in chronic liver disease. Clin Mol Hepatol.
novel approach to diagnostic and therapeutic inter-
2014;20(1):6–14.
ventions in the vascular system. Gastrointest Endosc.
15. Thalheimer U, Bellis L, Puoti C, Burroughs
2007;66(3):587–91.
AK. Should we routinely measure portal pressure in
3. Giday SA, Ko CW, Clarke JO, Shin EJ, Magno P,
patients with cirrhosis, using hepatic venous pres-
Jagannath SB, et al. EUS-guided portal vein carbon
sure gradient (HVPG) as a guide for prophylaxis and
dioxide angiography: a pilot study in a porcine model.
therapy of bleeding and rebleeding? No. Eur J Intern
Gastrointest Endosc. 2007;66(4):814–9.
Med. 2011;22(1):5–7.
4. Giday SA, Clarke JO, Buscaglia JM, Shin EJ, Ko
16. Sarin SK, Khanna R. Non-cirrhotic portal hyperten-
CW, Magno P, et al. EUS-guided portal vein cath-
sion. Clin Liver Dis. 2014;18(2):451–76.
eterization: a promising novel approach for portal
17. Pomier-Layrargues G, Kusielewicz D, Willems B,
angiography and portal vein pressure measurements.
Villeneuve JP, Marleau D, Cote J, et al. Presinusoidal
Gastrointest Endosc. 2008;67(2):338–42.
portal hypertension in non-alcoholic cirrhosis.
5. Brugge WR. EUS is an important new tool for
Hepatology. 1985;5(3):415–8.
accessing the portal vein. Gastrointest Endosc. 2008;
18. Buscaglia JM, Shin EJ, Clarke JO, Giday SA, Ko CW,
67(2):343–4.
Thuluvath PJ, et al. Endoscopic retrograde cholangio-
6. Hawkins IF, Caridi JG. Carbon dioxide (CO2)
pancreatography, but not esophagogastroduodenos-
digital subtraction angiography: 26-year experi-
copy or colonoscopy, significantly increases portal
ence at the University of Florida. Eur Radiol.
venous pressure: direct portal pressure measurements
1998;8(3):391–402.
through endoscopic ultrasound-guided cannulation.
7. Liss P, Eklof H, Hellberg O, Hagg A, Bostrom-Ardin
Endoscopy. 2008;40(8):670–4.
A, Lofberg AM, et al. Renal effects of CO2 and iodin-
19. Schulman AR, Thompson CC, Ryou M. EUS-guided
ated contrast media in patients undergoing renovas-
portal pressure measurement using a digital pressure
cular intervention: a prospective, randomized study.
wire with real-time remote display: a novel, minimally
J Vasc Interv Radiol. 2005;16(1):57–65.
invasive technique for direct measurement in an ani-
8. Bosch J, Garcia-Pagan JC, Berzigotti A, Abraldes
mal model. Gastrointest Endosc. 2016;83(4):817–20.
JG. Measurement of portal pressure and its role in
20. Schulman AR, Thompson CC, Ryou M. Endoscopic
the management of chronic liver disease. Semin Liver
ultrasound-guided direct portal pressure measurement
Dis. 2006;26(4):348–62.
using a digital pressure wire with real-time remote
9. Groszmann RJ, Bosch J, Grace ND, Conn HO,
display: a survival study. J Laparoendosc Adv Surg
Garcia-Tsao G, Navasa M, et al. Hemodynamic
Tech A. 2017;27(10):1051–4.
events in a prospective randomized trial of proprano-
21. Huang JY, Samarasena JB, Tsujino T, Chang
lol versus placebo in the prevention of a first variceal
KJ. EUS-guided portal pressure gradient measure-
hemorrhage. Gastroenterology. 1990;99(5):1401–7.
ment with a novel 25-gauge needle device versus
10. Albillos A, Banares R, Gonzalez M, Ripoll C,
standard transjugular approach: a comparison animal
Gonzalez R, Catalina MV, et al. Value of the hepatic
study. Gastrointest Endosc. 2016;84(2):358–62.
venous pressure gradient to monitor drug ther-
22. Fujii-Lau LL, Leise MD, Kamath PS, Gleeson FC,
apy for portal hypertension: a meta-analysis. Am
Levy MJ. Endoscopic ultrasound-guided portal-­
J Gastroenterol. 2007;102(5):1116–26.
systemic pressure gradient measurement. Endoscopy.
11. D’Amico G, Garcia-Pagan JC, Luca A, Bosch
2014;46(Suppl 1):E654–6.
J. Hepatic vein pressure gradient reduction and pre-
23. Huang JY, Samarasena JB, Tsujino T, Lee J, Hu KQ,
vention of variceal bleeding in cirrhosis: a systematic
McLaren CE, et al. EUS-guided portal pressure gra-
review. Gastroenterology. 2006;131(5):1611–24.
dient measurement with a simple novel device: a
12. Perello A, Escorsell A, Bru C, Gilabert R, Moitinho
human pilot study. Gastrointest Endosc. 2017;85(5):
E, Garcia-Pagan JC, et al. Wedged hepatic venous
996–1001.
 Endoscopic Ultrasound-Guided
Drainage of Pelvic, Intra-­ 17
abdominal, and Mediastinal
Abscesses

Enad Dawod and Jose M. Nieto

ous drainage under ultrasound guidance through

Introduction the transrectal or transvaginal route or under CT
guidance through the transgluteal route. This
Abscess formation results from pus accumulat- modality however has some limitations, most
ing within a tissue that has formed a cavity caused importantly due to the complexity of the struc-
by contamination either by bacteria, injury, or tures surrounding the abscess, making percutane-
foreign substances. An abscess can persist from ous drainage hard to achieve [3]. Moreover,
weeks to months [1]. Abscesses can either drain transrectal and transvaginal drainage is only pos-
naturally or through a variety of medical, radio- sible when the abscess is proximal to the ultra-
logic, or surgical interventions. Due to the high sound probe. Percutaneous drainage is also
risk of septicemia and shock due to the proximity associated with complications such as leakage,
of deeper abscesses to vital organs, intervention pneumoperitoneum, pneumomediastinum, bleed-
for these deeper abscesses is usually necessary. ing, infection, pain at the procedural site, and
Pelvic abscesses have various etiologies. limitations to ambulation make these interven-
Pelvic abscesses can occur as a complication to tions inconvenient and risky [4].
surgery (i.e., low anterior resection), Crohn’s dis- Mediastinal abscesses are mostly secondary
ease, ulcerative colitis, ischemic colitis, divertic- to an infection, commonly odontogenic and peri-
ulitis, and sexually transmitted diseases causing tonsillar abscess (descending necrotizing medi-
pelvic inflammatory disease [2]. Pelvic abscesses astinitis). These lesions could also arise as a
are commonly located proximal to the rectum result of an esophageal perforation, postoperative
and sigmoid [3]. The standard therapy for pelvic leakage, or following cardiovascular and thoracic
abscesses has traditionally involved percutane- surgeries. Other causes include trauma, tubercu-
losis, skin infections, and hematogenous spread
Electronic supplementary material: The online version
[5–7]. Mediastinal abscesses require prompt
of this chapter (https://doi.org/10.1007/978-3-319-97376- action as they are potentially life threatening and
0_17) contains supplementary material, which is available are usually associated with prolonged hospital
to authorized users. stays with the vast majority of cases requiring
E. Dawod surgical intervention [8]. Mediastinal abscesses
Department of Gastroenterology and Hepatology, due to esophageal perforation or postoperative
Weill Cornell Medicine, New York, NY, USA
leakage nearly always require surgical interven-
J. M. Nieto (*) tion [9]. Alternatively, mediastinal collections
Department of Gastroenterology and Hepatology,
Borland Groover Clinic, Advanced Therapeutic could be treated with interventional radiological
Endoscopy Center, Jacksonville, FL, USA techniques and transcutaneous US and CT-guided

© Springer Nature Switzerland AG 2019 177

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0_17
178 E. Dawod and J. M. Nieto

thoracentesis with catheter drainage [10]. from the abscess can be sampled as needed if clin-
Endoscopic treatment for mediastinal abscess has ically indicated-often no such sampling is per-
been described in the literature the past two formed as the abscess is assumed to be
decades either through direct endoscopic vision polymicrobial and the patient is already on broad-
or, most recently, under endoscopic ultrasound spectrum antibiotics. A guidewire is passed
(EUS) guidance [9–11]. through the needle into the cavity and is coiled in
Abdominal abscesses have a variety of etiolo- the cavity. Once guidewire access to the abscess is
gies including Crohn’s disease, diverticular dis- obtained, drainage by the placement of one or
ease, and postoperative causes [4]. Liver more transluminal stents is performed, with or
abscesses are usually caused by biliary obstruc- without balloon or passage catheter dilation of the
tion, hepatic trauma, bacteremia, amebiasis, or a transluminal tract. If using an electrocautery-­
history of abdominal surgery [12]. Subphrenic enhanced lumen apposing metal stent (LAMS),
abscesses are a complication of gastric, hepatic, the steps of needle access, guidewire passage, and
and colonic disease, in addition to abdominal sur- tract dilation may be obviated.
gery trauma [13]. Bilomas can result from bile Any other intervention involved such as dila-
duct disruption or hepatic trauma [14]. Splenic tion, clipping, or cutting were considered in this
abscess could occur as a complication of surgery review as “other devices” used to facilitate
or in patients with concurrent infections, more abscess drainage. The type and number of stents
commonly in immunocompromised patients [15, used, other devices used for facilitating abscess
16]. Abdominal abscess are conventionally drainage, and site of drainage were correlated
treated with interventional radiology-guided per- with abscess resolution and adverse events/rele-
cutaneous drainage with concomitant use of anti- vant clinical complications from the procedure.
biotics [17, 18]. Abdominal abscesses that are not
amenable to percutaneous drainage are usually
managed via surgery. Although EUS-guided Management
drainage of pancreatic fluid collections has
become the standard of care, to date there have Pelvic Abscesses
only been limited reports of using EUS in treat-
ing intra-abdominal abscess [19, 20]. To date, EUS-guided drainage of pelvic abscess
As a result, over the past 15 years, EUS-­ has been reported in 105 patients (Figs. 17.1, 17.2,
guided drainage of abscesses has been studied in 17.3, 17.4, 17.5, 17.6, 17.7, 17.8, and 17.9). The
relation to stenting, dilation, drainage, clipping, total average size of all of the pelvic abscesses in
and cutting. This chapter will review the efficacy these studies was 59.47 × 46.31 mm (n = 49) with
and safety of EUS-guided drainage of pelvic, a range of 7.50 mm to 96.00 × 7.40 mm to
mediastinal, and intra-abdominal abscesses. This 83.00 mm. Seven cases had incomplete resolution
chapter will also analyze what specific conditions of their abscess (6.67%), 98 cases had complete
and procedural modifications can lead to better resolution of their abscess (93.33%), and 9 cases
results. had adverse events (8.57%) [3, 21, 22–24, 25–28].
Transrectally drained abscesses were thought
to have more complications in comparison to
Background trans-sigmoid drained abscesses. This is due to
the fact that transrectal stents can migrate or clog
The general technique for EUS-guided abscess easily by fecal matter or pus causing complica-
drainage is as follows: EUS is used to locate the tions [2]. Eighty-one out of 105 (77.1%) cases
abscess via either a transrectal/transcolonic, trans- had their abscesses drained transrectally, high-
gastric, transintestinal, or transesophageal win- lighting both the frequency of pelvic abscesses
dow. Once the abscess is identified, a 19-gauge and their amenability to EUS-guided drainage.
needle is used to puncture the abscess site. Fluid 14 (13.3%) cases had their abscesses drained
17 Endoscopic Ultrasound-Guided Drainage of Pelvic, Intra-abdominal, and Mediastinal Abscesses 179

Fig. 17.1 EUS image of a pelvic abscess prior to drainage

Fig. 17.2 EUS image of same abscess showing internal debris/contents

through the trans-sigmoid route. In 7 (6.67%) went drainage of their abscesses through the
patients, the abscess was drained through the trans-sigmoid route experienced adverse events
transcolonic route and through trans-abdominal versus approximately 5% of those who under-
route in 2 (1.9%) cases [3, 21, 22–24, 25–28]. went transrectal drainage. Due to the lower case
Contrary to our expectations, drainage through count for patients with trans-sigmoid abscess
the trans-sigmoid route led to a higher rate of drainage, the results are not significant. In
incomplete abscess resolution and adverse ­addition, while it has been quoted that transrectal
events. Fourteen percent of patients who under- drainage can lead to a higher chance of clogging
180 E. Dawod and J. M. Nieto

Fig. 17.3 Global view of abscess cavity on EUS

Fig. 17.4 EUS image of the abscess in proximity to the bladder

the lumen, because of the tortuosity of the sig- pain, nausea, vomiting, and left lower quadrant
moid colon, a similar issue could arise and pos- pain.
sibly be worse. Also, the drainage catheter used Historically, for biliary and pancreatic stent
would have to be significantly longer due to the placement, plastic stents have been known to
distance from the anus and this could also lead to have a higher rate of stent migration and stent
more complications such as accidental closure or occlusion. In general, metal biliary stents develop
leakage of the catheter inside the lumen. Adverse stent occlusion at a later date and with less fre-
events in both groups included fever, abdominal quency than plastic stents. In addition, there is a
17 Endoscopic Ultrasound-Guided Drainage of Pelvic, Intra-abdominal, and Mediastinal Abscesses 181

Fig. 17.5 Figure EUS view of same abscess after distal flange of the first LAMS deployed. Ultimately, two LAMS
were used to drain the abscess in its entirety

Fig. 17.6 Internal view from within the abscess showing Fig. 17.7 Endoscopic view through one LAMS showing
one of the flanges of a LAMS the 2nd LAMS in the abscess cavity

reduced risk of perforation due to a reduced need 7F double pigtail stent deployed, 25 cases had
for prior stricture dilatation [29, 30] As a result, two 7F double pigtail stents deployed, 1 case had
for pelvic abscess cases, we predicted a similar three 7F double pigtail stents deployed, 1 case
outcome, in which metal stents would have a had two 10F double pigtail stents deployed, 5
lower rate of incomplete resolution and adverse cases had one 8.5F double pigtail stent deployed,
events. 1 case had 1 lumen apposing metal stent (LAMS)
Out of 105 cases, 85 (80.95%) cases utilized deployed, and 3 cases had 1 fully covered
transluminal stents and 20 (19.05%) cases were ­self-­expandable metal stent (FCSEMS). Within
treated with aspiration alone. Among the 85 cases these studies, cases that had 1 full covered metal
that had a stent deployed, 29 cases had one 10F stent, 1 10F plastic stent, or 1 10F double pigtail
double pigtail stent deployed, 16 cases had one stent and 1 8.5F double pigtail stent had a higher
182 E. Dawod and J. M. Nieto

Fig. 17.8 Endoscopic view of pus draining into the colon

through a LAMS after deployment in the pelvic abscess Fig. 17.9 An additional view of pus draining through the
LAMS into the bowel lumen
rate of adverse events/relevant clinical complica-
tions compared to cases that had other stents associated with a worse outcome [3, 21, 22–24,
placed, suggesting the multiple stents may be 25–28]. These results were consistent with our
better than a single stent when treating abscesses. prediction that plastic stents had a higher rate of
Twenty patients had their abscess aspirated and incomplete resolution, but were not consistent
did not undergo stent placement and of those 3 with our hypothesis that plastic stents would have
had incomplete resolution of their abscess. a higher rate of adverse events compared to metal
Patients that did not have a stent placed had a stents. Due to the small sample size of patients
higher rate of adverse events compared to cases found for cases with metal stents placed, the
that had one or more stents placed [3, 21, 22–24, results for the rate of adverse events might not be
25–28]. representative of what was expected. As a result,
The average time from stent placement to more research has to be done regarding metal
stent removal was calculated in weeks. 72 cases stents deployed for pelvic abscess drainage
had usable data for average stent removal. The Taking a look at the other instruments used to
average number of weeks for stent removal for facilitate drainage in these cases, we believed
every case with usable data was 5.34 weeks dilation and catheter drainage would be superior
(0.29–30.10). The average stent dwell time in to the other modes of drainage. This is due to the
weeks for cases that had incomplete abscess res- fact that after using a dilator or cystotome, the
olution was 16.05 weeks (2.00–30.10) as com- placement of a stent or catheter should be easier.
pared to 5.03 weeks (0.29–30.10) in the patients Because no electrocautery is used during the dila-
who had complete resolution of their abscesses. tion procedure, bleeding or perforation is gener-
The average number of weeks from stent place- ally only seen in 1% of patients. Due to the use of
ment to stent removal for cases that had adverse electrocautery with cystotomes, perforation is a
events was 16.55 weeks (n = 6) with a range of frequent complication of the procedure [31]. In
4–30.10 weeks as opposed to 3.83 (n = 66) with a addition, catheter drainage poses the risk of acci-
range of 0.29–30.10 weeks in the patients who dental dislodgement of both the catheter and the
had no adverse events. Cases that had stents stent due to the catheter protruding from the anus,
placed for a longer period of time had an overall but allows for access to the abscess cavity and for
higher incidence of having incomplete abscess frequent irrigation to allow the drainage to com-
resolution and adverse events. This suggests that plete as quickly as possible [2].We hypothesized
it is possible that these were sicker patients to that in this retrospective case study, there would
begin with or that longer stent dwell time may be be the fewest incomplete abscess resolutions and
17 Endoscopic Ultrasound-Guided Drainage of Pelvic, Intra-abdominal, and Mediastinal Abscesses 183

adverse events in patients with both a catheter and

dilator used to facilitate drainage
The total number of cases that had other
devices and techniques used to facilitate abscess
drainage was 86 cases. Nineteen cases had dila-
tion and catheter drainage done along with the
main intervention. Ten cases had only catheter
drainage to facilitate abscess drainage along with
the main intervention, but no usable data was
found in the studies.
Fourteen cases had dilation done and a guiding
catheter placed. Thirty-seven cases had only dila-
tion done to facilitate abscess drainage along with
the main intervention and 6 cases used cystotomes
to drain the abscess. Thirty-seven had no other
intervention done along with the main interven- Fig. 17.10 EUS views of a hepatic abscess
tion [3, 21, 22–24, 25–28]. While these other
interventions do make a difference in the efficacy
and safety of the outcome of the results, stent
placement is more crucial for a beneficial out-
come according to the data collected in this study.

Intra-abdominal Abscess

Currently there are multiple published reports

which include 37 patients that have undergone
EUS-guided drainage of intra-abdominal abscesses,
including hepatic abscesses (Figs. 17.10, 17.11,
and 17.12) and (Video 17.1). The average size of
these abscesses was 59.83 × 52.72 mm (n = 18)
with a range of 25 mm to 150 × 21 mm to 170 mm.
Notably, 100% of the cases had complete resolu-
tion of the abscesses and 4 cases had adverse events Fig. 17.11 EUS view of same abscess showing internal
debris/contents
or relevant clinical complications (10.81%) [4, 12,
14, 16, 18, 32–37].
Procedures and cases that used the transgastric of various sizes with no adverse events being
route for drainage had more adverse events in reported. Two cases had no stents deployed and
comparison to cases that had abscesses drained one out of 2 cases had adverse events (50%). The
from the transduodenal route. Thirty-two abscesses type and number of stent had no effect on resolu-
were drained through the transgastric route, 4 via tion of the abscess, however, cases with FCSEMS
the transduodenal route, and 1 via the transjejunal stents had more adverse event outcomes com-
route [4, 12, 14, 16, 18, 32–37]. pared to cases who had other types of stents
Thirteen cases involved the use of a translumi- deployed [4, 12, 14, 16, 18, 32–37].
nal, fully covered self-expanding metal stent The average time from stent placement to
(FCSEMS) and 3 out of 13 cases had adverse stent removal was calculated in weeks. The aver-
events/relevant clinical complications (23.08%). age number of weeks throughout all the cases
Twenty-two cases had at least one double pigtail that had EUS-guided intra-abdominal abscess
184 E. Dawod and J. M. Nieto

Fig. 17.12 (a–c) Endoscopic views of the interior of the hepatic abscess as seen through the LAMS with an upper
endoscope

drainage stent placement was 5.55 weeks (n = 14) developed adverse events (21.43%). Ten cases
with a range of 1.57–12 weeks. In the 14 cases utilized only drainage catheters, while one case
that provided stent removal data, 0 cases had had an adverse event (10.00%). Twelve cases
incomplete abscess resolution or any adverse ­utilized only dilators to facilitate abscess drain-
events [4, 12, 14, 16, 18, 32–37] (Figs. 17.10, age in addition to the main interventions and 1
17.11, 17.12, and 17.13). additional case involved a hemostatic clip placed
When looking at hepatic abscesses in particu- to close the fistula in addition to the main inter-
lar, large-diameter metal stents provided effective ventions. These cases did not have any adverse
drainage of liver abscess. Covered metal stents events [4, 12, 14, 16, 18, 32–37].
have been used to facilitate hepatic abscess The mortality rate with the surgical method
debridement [37–39]. of treating hepatic abscesses has been reported
Fourteen cases involved the use of both dila- to be between 17 and 32%, and the percutane-
tors and drainage catheters, of which 3 cases ous method is associated with serious complica-
17 Endoscopic Ultrasound-Guided Drainage of Pelvic, Intra-abdominal, and Mediastinal Abscesses 185

Fig. 17.13 EUS view of a mediastinal abscess Fig. 17.14 EUS view of LAMS deployed into the medi-
astinal abscess
tions such as bleeding, biliary peritonitis, and
fistula formation. EUS-guided drainage ing metal stents (LAMS), FCSEMS, and double
decreases the risk of injury to intervening vas- pigtail stents (Figs. 17.14 and 17.15).
culature, resulting in decreasing the rate of com- The average number of weeks stents were
plications [12, 18, 33]. Additionally, EUS placed was 5.43 weeks (n = 4) with a range of
decreases the incidence of infections associated 0.71–16.00 weeks and reported adverse events in
with the transcutaneous route and also allows these patients included esophageal stenosis,
for potential replacement of the external stent esophageal ulceration, perforation, sepsis, fever,
with an internal stent that could prevent recur- pain, and bleeding [7, 9–11, 41–43].
rence [34]. EUS-guided drainage presents itself Due to the very small sample size, no signifi-
as a safe and superior modality in treating cant differences could be established in terms of
hepatic abcess [18, 36, 40]. One of the limita- which specific conditions and procedural modifi-
tions of EUS is its inability to visualize and cations can lead to better results and less adverse
access the right lobe of the liver. However, this outcomes. EUS visualizes and accurately local-
is not the case with left lobe, thus permitting full izes blood vessels and other vital structures
access and visualization [18]. within close proximity of the abscess and identi-
fies a clear and safe path for drainage.
Furthermore, EUS has an advantage where there
Mediastinal Abscesses is no mucosal indentation of the abscess in which
case blind per oral drainage might pose a high
The literature describes 6 patients who under- risk [7].
went EUS-guided drainage of mediastinal
abscesses via transesophageal or transgastric
approaches. The total average size of the cases’ Summary
abscesses was 45.18 × 33.85 mm (n = 5) with a
range of 17.70–63.00 mm in which there was The results from these studies demonstrate that
100% complete resolution of their abscess. One EUS-guided abscess drainage is an effective and
case had an adverse event/relevant clinical com- safe method. Out of the 148 total cases in these
plication (16.67%) [7, 9–11, 41–43]. Mediastinal studies, 95.27% of cases had complete abscess
abscesses have been drained using lumen appos- drainage, and 90.54% of cases had no adverse
186 E. Dawod and J. M. Nieto

Fig. 17.15 (a–d) Endoscopic views of the interior of the mediastinal abscess as seen through the LAMS with an upper
endoscope

events. While limited, the data suggests that dou- improvement and overall resolution of the
ble pigtail stents may produce better outcomes abscess and adverse events/relevant clinical
than metal stents. complications.
The studies cited above were heterogeneous Overall, EUS-guided abscess drainage
and reported their data in different ways, limiting offers a safe and effective alternative option for
their generalizability to some extent. As such, drainage of abscesses in patients who are poor
this leads to some degree of ambiguity when cor- surgical candidates or in patients who prefer
relating data between types and number of stents, not to undergo surgery or percutaneous cathe-
other devices used, and drainage route to clinical ter drainage [2].
17 Endoscopic Ultrasound-Guided Drainage of Pelvic, Intra-abdominal, and Mediastinal Abscesses 187

References 17. Singhal S, Lane D, Anand S, Duddempudi

S. Endoscopic ultrasound-guided hepatic and perihe-
patic abscess drainage: an evolving technique. Therap
1. Bieluch VM, Tally FP. Pathophysiology of abscess
Adv Gastroenterol. 2014;7:93–8.
formation. Clin Obstet Gynaecol. 1983;10:93–103.
18. Noh SH, et al. EUS-guided drainage of hepatic
2. Trevino JM, Drelichman ER, Varadarajulu S. Modified
abscesses not accessible to percutaneous drainage
technique for EUS-guided drainage of pelvic abscess
(with videos). Gastrointest Endosc. 2010;71:1314–9.
(with video). Gastrointest Endosc. 2008;68:1215–9.
https://doi.org/10.1016/j.gie.2009.12.045.
https://doi.org/10.1016/j.gie.2008.07.016.
19. Koichiro Mandai KU, Yasuda K. Endoscopic
3. Choi EK, et al. Endoscopic ultrasound-guided peri-
ultrasound-­ guided drainage of postoperative intra-­
rectal abscess drainage without drainage catheter: a
abdominal abscesses. World J Gastroenterol. 2015;
case series. Clin Endosc. 2017;50:297–300. https://
21:3402–8.
doi.org/10.5946/ce.2016.123.
20. Chapman CG, Matthews JB, Siddiqui UD. EUS-­
4. Piraka C, Shah RJ, Fukami N, Chathadi KV, Chen
guided internal drainage of a deep abdominal postop-
YK. EUS-guided transesophageal, transgastric, and
erative abscess after Whipple procedure. Gastrointest
transcolonic drainage of intra-­abdominal fluid col-
Endosc. 2015;82:1132–3. https://doi.org/10.1016/j.
lections and abscesses. Gastrointest Endosc. 2009;70:
gie.2015.05.024.
786–92. https://doi.org/10.1016/j.gie.2009.04.049.
21. Giovannini M, et al. Drainage of deep pel-
5. Ji ZD. Diagnosis and treatment of mediastinal abscess.
vic abscesses using therapeutic echo endoscopy.
Zhonghua Wai Ke Za Zhi. 1990;28:610–1.
Endoscopy. 2003;35:511–4. https://doi.org/10.1055/s-
6. Carmichael LP, Bernard AC. A novel treatment for
2003-39673.
inferior mediastinal abscess via abdominal laparos-
22. Hadithi M, Bruno MJ. Endoscopic ultrasound-guided
copy. J Surg Case Rep. 2012;4:7.
drainage of pelvic abscess: a case series of 8 patients.
7. Balekuduru A, Dutta AK, Subbaraj SB. Endoscopic
World J Gastrointest Endosc. 2014;6:373–8. https://
ultrasound-guided transoral drainage of parapharyn-
doi.org/10.4253/wjge.v6.i8.373.
geal abscess. Dig Endosc. 2016;28:756. https://doi.
23. Kawakami H, et al. Endoscopic ultrasound-guided
org/10.1111/den.12696.
pelvic abscess drainage using a dedicated, wide,
8. Urschel J, Razzuk MA, Wood RE, Galbraith N,
flared-end, fully covered self-expandable metal stent.
Pockey M, Paulson DL. Improved management of
Endoscopy. 2015;47(Suppl 1):E265–6. https://doi.org
esophageal perforation: exclusion and diversion in
/10.1055/s-0034-1391953.
continuity. Ann Surg. 1974;179:587.
24. Lee M, Izzy M, Ho S. Novel use of fully covered
9. Wehrmann T, et al. Endoscopic debridement of para-
self-expandable metal stent for drainage of peri-
esophageal, mediastinal abscesses: a prospective case
rectal abscess. A case series. Arab J Gastroenterol.
series. Gastrointest Endosc. 2005;62:344–9. https://
2017;18:122–5. https://doi.org/10.1016/j.ajg.2017.05.
doi.org/10.1016/j.gie.2005.03.001.
017.
10. Kahaleh M, Yoshida C, Kane L, Yeaton P. EUS drain-
25. Mukai S, Itoi T, Tsuchiya T, Tonozuka R, Sofuni
age of a mediastinal abscess. Gastrointest Endosc.
A. EUS-guided pelvic abscess drainage with use of a
2004;60:158–60. https://doi.org/10.1016/S0016-5107
biflanged metal stent through the perianal transgluteal
(04)01310-0.
route. Gastrointest Endosc. 2016;84:1069–70. https://
11. Saxena P, Kumbhari V, Khashab MA. EUS-guided
doi.org/10.1016/j.gie.2016.05.049.
drainage of a mediastinal abscess. Gastrointest
26. Puri R, et al. Endoscopic ultrasound-guided pelvic and
Endosc. 2014;79:998–9. https://doi.org/10.1016/j.gie.
prostatic abscess drainage: experience in 30 patients.
2013.12.026.
Indian J Gastroenterol. 2014;33:410–3. https://doi.
12. Itoi T, et al. Endoscopic ultrasonography-guided
org/10.1007/s12664-014-0485-8.
drainage for tuberculous liver abscess drainage. Dig
27. Puri R, Eloubeidi MA, Sud R, Kumar M, Jain
Endosc. 2011;23(Suppl 1):158–61. https://doi.org/10.
P. Endoscopic ultrasound-guided drainage of pelvic
1111/j.1443-1661.2011.01115.x.
abscess without fluoroscopy guidance. J Gastroenterol
13. Seewald S, et al. EUS-guided drainage of subphrenic
Hepatol. 2010;25:1416–9. https://doi.org/10.1111/j.
abscess. Gastrointest Endosc. 2004;59:578–80. https://
1440-1746.2010.06328.x.
doi.org/10.1016/S0016-5107(03)02878-5.
28. Varadarajulu S, Drelichman ER. Effectiveness of
14. Shami VM, et al. EUS-guided drainage of bilomas: a
EUS in drainage of pelvic abscesses in 25 con-
new alternative? Gastrointest Endosc. 2008;67:136–
secutive patients (with video). Gastrointest Endosc.
40. https://doi.org/10.1016/j.gie.2007.07.040.
2009;70:1121–7. https://doi.org/10.1016/j.gie.2009.
15. Giovanna Ferraioli EB, Gulizia R, Mariani G, Marone
08.034.
P, Filice C. Management of splenic abscess: report on
29. Pfau PR, et al. Pancreatic and biliary stents.
16 cases from a single center. Int J Infect Dis. 2009;
Gastrointest Endosc. 2013;77:319–27. https://doi.org/
13:524–30.
10.1016/j.gie.2012.09.026.
16. Lee DH, Womeldorp CM, Horwhat JD. Endoscopic
30. Pavlides M, Gorard MPADA. Stents in gastrointes-
therapy of a splenic abscess: definitive treatment
tinal endoscopy. Therapeutic gastrointestinal endos-
via EUS-guided transgastric drainage. Clin Endosc.
copy. 2011. www.intechopen.com.
2006;64:631–4.
188 E. Dawod and J. M. Nieto

31. Prasad GA, Varadarajulu S. Endoscopic ultrasound-­ 38. Medrado BF, et al. Endoscopic ultrasound-guided
guided abscess drainage. Gastrointest Endosc Clin N drainage of giant liver abscess associated with trans-
Am. 2012;22:281–90. https://doi.org/10.1016/j.giec. gastric migration of a self-expandable metallic stent.
2012.04.002. Endoscopy. 2013;45(Suppl 2):E331–2. https://doi.org
32. Yukiko Ito HI, Nakai Y, Umefune G, Sato T, /10.1055/s-0033-1344128.
Nakahara S, Suwa J, Kato K, Nakata R. Successful 39. Ogura T, et al. Placement of a 6 mm, fully covered
endosonography-­ guided drainage of an intra-­ metal stent for main pancreatic head duct stricture
abdominal abscess in a 1-year-old infant. Gut Liver. due to chronic pancreatitis: a pilot study (with video).
2016;10:483–5. Ther Adv Gastroenterol. 2016;9:722–8. https://doi.
33. John Keohane CJD, Schattner MA, Gerdes H. EUS-­ org/10.1177/1756283X16651855.
guided transgastric drainage of caudate lobe liver 40. Ang TL, Seewald S, Teo EK, Fock KM, Soehendra
abscesses. J Interv Gastroenterol. 2011;1:139–41. N. EUS-guided drainage of ruptured liver abscess.
34. Kodama R, et al. Endoscopic ultrasonography-guided Endoscopy. 2009;41(Suppl 2):E21–2. https://doi.org
drainage of infected intracystic papillary adenocarci- /10.1055/s-0028-1103468.
noma of the liver. Clin J Gastroenterol. 2015;8:335–9. 41. Consiglieri CF, Escobar I. EUS-guided transesopha-
https://doi.org/10.1007/s12328-015-0607-6. geal drainage of a mediastinal abscess using a diabolo-­
35. Kumta NA, Torres-Ruiz F, Reinoso PJ, Kahaleh shaped lumen-apposing metal stent. Gastrointest
M. Endoscopic management of hepatic abscess Endosc. 2015;81:221–2.
after EUS-guided hepaticogastrostomy. Gastrointest 42. Tonouchi A, Kuwabara S, Furukawa K, Matsuzawa
Endosc. 2016;84:1054–5. https://doi.org/10.1016/j. N, Kobayashi K. Phlegmonous esophagitis treated by
gie.2016.07.023. endoscopic drainage. Esophagus. 2016;14:183–7.
36. Seewald S, et al. EUS-guided drainage of hepatic 43. Rodrigues-Pinto E, Baron TH. Endoscopic
abscess. Gastrointest Endosc. 2005;61:495–8. https:// ultrasound-­
guided transesophageal drainage of an
doi.org/10.1016/S0016-5107(04)02848-2. infected aneurysm after aortic stent graft placement.
37. Tonozuka R, Tsuchiya T. EUS-guided drainage of Gastroenterology. 2016;151:30–1. https://doi.org/
hepatic abscess and infected biloma using short and 10.1053/j.gastro.2016.04.039.
long metal stents (with videos). Gastrointest Endosc.
2015;81:1463.
 Index

A Antegrade approach, 149

Abdominal abscess, 178 Antegrade method, 97
Abdominal pain, 65, 66, 71, 72, 138 Antitumor therapy, 116
Absolute alcohol, 68 Argon plasma coagulation (APC), 152
Acquire™ needle, 76
Acute cholecystitis
diagnosis, requirements, 35 B
EUS-GBD, 35 Back-loading technique, 97, 115
laparoscopic cholecystectomy, 35 Balloon assisted method, 161–165
peri-procedural management, 38 Barrett’s esophagus, 123
recurrence, 39 Barrett’s surveillance, 84
systematic review and pooled analysis, 38 Bile duct stones (BDS), 17, 18, 21
Acute GI bleeding, 55 Biliary drainage (EUS-BD)
Acute necrotizing pancreatitis (ANP), 2 efficacy and safety, 17
Alcohol, 139 EUS-CDS, 17
Alcohol-free group, 141 FNA, 17
Angiotherapy vs. PTBD, 17
advantages, 55 success rates, 17
coils/glue, use of, 56 Biliary obstruction, 25, 31
complexity, 56 Bilomas, 178
FNA needle, 56 Bipolar RFA, 110
limitations, 55, 56 Blood noodles, 88
for nonvariceal bleeding, 62, 63 Brachytherapy, 113, 114, 143
rectal varices, 62 Bupivacaine, 65, 68, 71
technique, 56
varices
coil injection, EUS-guided, 56, 57 C
combination therapy, 59–61 Carbon dioxide (CO2), 170
cyanoacrylate injection, EUS-guided, 58, 59 Carcinoembryonic antigen (CEA), 135
vascular access, EUS-guided, 56 Cautery-enhanced (CE) LAMS system, 160, 161
Antegrade (EUS-AG) stenting Celiac plexus, 65
advantages, 17 Celiac plexus block (CPB)
bile duct drainage, 20 central vs. bilateral technique, 68, 70
bile duct stones removal, 18, 20 complications, 71
clinical benefits, 18 contraindications, 66
complications, 22, 23 vs. CPN, 70
contraindications, 18 CT-guided vs. EUS-guided CPB, 71
description, 17 description, 65
devices, 18, 20 EUS-guided, 70
fine needles, 18 indications, 66
guide wires, 18 unopposed parasympathetic activity, 66
indications, 18 Celiac plexus neurolysis (CPN)
success rate, 20–22 celiac artery and plexus, origin, 66, 68
trials, 18, 19 central vs. bilateral technique, 68, 70

© Springer Nature Switzerland AG 2019 189

D. G. Adler (ed.), Interventional Endoscopic Ultrasound,
https://doi.org/10.1007/978-3-319-97376-0
190 Index

Celiac plexus neurolysis (CPN) (cont.) ESTER, 148

complications, 71 gastric bypass, 152
contraindications, 66 gastrostomy, 147
vs. CPB, 70 indication, 155
description, 65 laparoscopy, 150, 155
EUS-guided, 70 mature and perform non-urgent, 155
indications, 66 RYGB, 148, 155
as intraoperative procedure, 65 success rate, 152
for persistent and intractable abdominal pain, 66 therapeutic duodenoscope, 153
techniques, 66, 68 transgastric, 148
CHARM trial, 140 Core biopsy, 75–77, 79, 80
Chemoradiotherapy, 116 Crohn’s disease, 178
Cholangiocarcinoma, 117 Curvilinear-array echoendoscope, 136
Choledochoduodenostomy Cyanoacrylate, 56, 60, 62
echoendoscope, 27 Cyanoacrylate injection, EUS-guided
19-gauge FNA needle, 27 adverse events, 59
and malignant gastric outlet obstruction, 28 clinical applications, 58, 59
retrograde access, 27 glue embolization, 58
retrograde drainage, 27 hemostasis and vascular occlusion, 58
SEMS, 27 oil-based contrast agent, 58
success rates, for biliary drainage procedures, 27 Cyst biopsy, 130, 131
transmural stenting, 27 Cystic duct stenting, 35
Coagulopathy, 84 Cystic neuroendocrine tumor, 125
Coil injection, EUS-guided Cystoscopy
clinical applications, 57 application, 128
FNA needle, 56 biopsy, 128
microcoils, 56–58 blood vessels, 127
risk, coil migration, 56 cyst fluid, 127
Colonoscopy, 171 DETECT trial, 128
Common bile duct (CBD), 25 EUS, 127
Contrast harmonic EUS (CH-EUS) fiber optic probe, 127
arterial phase, 121 imaging characteristics, 128
cystic wall and septa, 123 intracystic papillary structures, 128
echogenicity and microvascular perfusion, 121 pancreatic cysts, 127
evaluation and differential diagnosis, 120 papillary protrusions, 128
FNA (see Fine needle aspiration (EUS-FNA)) PCLs, 128
hyperenhanced solid nodule, 121 surgical pathology, 128
hyperenhancement, 121, 122 Cytobrushing, 129, 130
imaging and sampling, 120 Cytoimplants, 115
IPMNs, 120–123
microbubbles, 120
mural nodules, 121, 123 D
neoplastic solid components, 120 Deep balloon enteroscopy, 148
pancreatic cysts, 121 Dendritic cells (DCs), 117
PCLs, 123 DETECT trial, 126, 128
time-intensity curve parameters, 123 Diffuse intravascular coagulation (DIC), 91
vascular portions, 121 Direct (freehand) method, 161
Conventional endoscopic retrograde Doppler flow, 173–175
cholangiopancreatography (ERCP) Doppler imaging, 56, 59, 61, 63
access route, 149 Double-balloon enteroscopy, 147
anatomy, 147 Double pigtail plastic stents (DPPSs)
antegrade, 149, 150 and FCSEMS, 6
complications, 153 guidewire, in prospective study, 3
different indications, 153 infection after placement, 11
directed transgastric (see Endoscopic ultrasound vs. LAMSs, 9–11
(EUS) directed transgastric ERCP (EDGE)) placement, 3
double balloon, 147 success rates, 3
duodenoscope, 148 symptomatic, 3
enteroscopy, 147, 148, 155 transmural drainage, PFCs, 3, 4
and EUS, 153 Drainage
Index 191

gallbladder (see Gallbladder drainage (EUS-GBD)) inflation, 149

PFCs (see Endoscopic ultrasound (EUS)) internal, 150
Duodenoscope, 150 LAMS, 150, 152, 154, 155
limitations, 153–155
multicenter clinical trial, 147
E obesity, 147
E1B-55kD gene-deleted replication-selective pancreatobiliary system, 147
adenovirus, 115 PATENT, 148
Echoendoscope, 112 peel away sheath, 150
Electrocautery-enhanced LAMS (EC-LAMS), 10–11 PEG tube, 149
Electromagnetic energy, 110 percutaneous gastrostomy, 149, 150
Endoscopic retrograde cholangiopancreatography performance, 150
(ERCP), 17, 18, 22 procedure, 150
access positioning, 26 puncture, 149
in biliary obstruction, 25 retrospective analysis, 147
enteroscopy-assisted ERCP (e-ERCP), 52 RYGB (see Roux-en-Y gastric bypass (RYGB))
vs. EUS-guided biliary drainage, 31 sonographic guidance, 149
and EUS-PD, 52 success rate, 147, 148, 152, 153, 156
papilla/anastomosis, 46 transcutaneous fully covered esophageal stent, 149, 150
push position, 26 ultrathin scope, 150
Endoscopic transpapillary cystic duct stenting, 35 Endoscopic-ultrasound-guided gallbladder drainage
Endoscopic ultrasound (EUS) (EUS-GBD), see Gallbladder drainage
adverse events (EUS-GBD)
EC-LAMS, 11 Endoscopic-ultrasound-guided pancreatic duct drainage
infection, 11, 12 (EUS-PD), see Pancreatic duct drainage
metal vs. DPPS, 11 (EUS-PD)
migration and occlusion, stent, 12 Endoscopy unit/GI/hepatology clinic, 83
CPB (see Celiac plexus block (CPB)) Endosonography, 57
CPN (see Celiac plexus neurolysis (CPN)) and fluoroscopy, 97
DPPSs vs. metal stent, 9, 10 hyperechoic fiducials, 100–102, 104
drainage (see EUS-guided drainage) Enteral stenting (ES)
EC-LAMS, 10–11 vs. gastroenterostomy (EUS-GE), 166, 167
without endoluminal bulging, 2 and SGJ, 167
fiducial placement (see Fiducial markers) surgery, 159
GE (see Gastroenterostomy (EUS-GE)) Entero-enteric anastomosis, 160
LAMS (see Lumen-apposing metal stents (LAMS)) Enteroscopy-assisted ERCP, 147, 148, 155
LB (see Liver biopsy (LB)) EPASS method, 161
metal biliary stents, 3–6 Esophageal cancer, 95, 100–102
neoplastic (see Neoplastic pancreatic cysts) Esophagogastroduodenoscopy (EGD), 171
pancreatic cyst ablation (see Pancreatic cyst ablation) Ethanol, 65, 67, 68, 71, 136, 138
PPG (see Portal pressure gradient (PPG) Ethanol ablation, 109, 110
measurement) EUS-gastrojejunostomy (EUS-GJ), 159
repeat evaluation, 2 balloon assisted method, 161–165
solid pancreatic tumors (see Solid pancreatic tumors) direct (freehand) method, 161
standard approaches, 2, 3 double-balloon-occluded gastrojejunostomy bypass, 161
transmural drainage, 2 water immersion method, 160, 161
Endoscopic ultrasound (EUS) directed transgastric ERCP EUS-guided angiotherapy, see Angiotherapy
(EDGE) EUS-guided biliary drainage (EUS-BD), 17
adverse events, 147, 148 adverse events, 29
APC, 152 biliary vs. plastic stents, 29
complications, 153–155 choledochoduodenostomy (see
equipment, 148 Choledochoduodenostomy)
and ESTER, 149 complications, 29
external, 149 ERCP, 31
fluoroscopic and sonographic guidance, 150, 153, extrahepatic vs. intrahepatic rendezvous, 26
154 percutaneous approach, 30, 31
gastric pouch and excluded stomach, 150, 151 “rendezvous” procedure, 25, 26
gastric pouch/Roux limb, 149 EUS-guided core biopsy, 75–77, 79, 80
gastrostomy, 147, 149 EUS-guided double-balloon-occluded gastrojejunostomy
guidewire, 149 bypass, 161
192 Index

EUS-guided drainage diagnostic modalities, 123

abdominal abscess, 178 nCLE, 123–127
clipping, 178 PCLs, 123
cutting, 178 PPG
dilation, 178 and compact manometer, 171
intra-abdominal abscess, 183, 184 non-compressible tubing, 172
liver abscesses, 178 small-caliber, 170
mediastinal abscess, 177, 185 TAUS, 171
pelvic abscess, 177–183 ProCore™, in pancreatic patients, 74
percutaneous, 177 and RFA probe, 112
polymicrobial, 178 SharkCore needle, 76, 77
splenic abscess, 178 standard 19 gauge needle, 74
subphrenic abscesses, 178 and through-the-needle device, 111
transrectal and transvaginal, 177 tissue acquisition from lymph nodes, 77
EUS-guided rendezvous, 25–27, 29 Fine needle biopsy (EUS-FNB), 83, 86
EUS-guided sutured gastropexy for transgastric ERCP EchoTip Procore FNB, 73
(ESTER), 148, 149, 152, 153 of gastrointestinal subepithelial tumors, 76, 77
EUS retrograde biliary drainage, 25, 27, 31 of liver, 78
Extrahepatic method, 25, 26, 29 with 19-gauge needle, 79, 80
Ex vivo LB, 86 Tru-Cut biopsy, 78, 79
of lymphadenopathy, 77, 78
for malignant liver lesions, 80
F of solid pancreatic lesions, 74
Fiducial markers Acquire™ needle, 76
advantages, 114 ProCore™ needle, 74, 76
adverse events, 104 SharkCore™ needles, 76
deployment techniques, 97, 99 standard 19-gauge needle, 74
durability, 103–105 Quick-Core, 73
efficacy and safety, 97–99 TCB, 73
IGRT, 95 Fine-needle injection (FNI)
radiotherapy, 95 cytoimplants, 115
set up, 97, 99 immature DCs, 117
tumor targets and safe, 117
esophageal cancer, 100–102 Front-loading technique, 115
feasibility and technical success, 102, 104 Fully covered self-expandable metal stent
metastatic pancreatic tail mass, 103 (FCSEMS), 35, 39, 181
pancreatic cancer, 99–101
rectal cancer, 102, 103
types, 95–97, 114 G
Fiducial placements, 114, 115 Gallbladder, 85
Fine needle aspiration (EUS-FNA), 17, 18, 56 Gallbladder drainage (EUS-GBD)
advantages, histological core specimen, 73 for acute cholecystitis, 42
electrode, 111 complications, 36, 38
EUS-LB use, 79 large bore LAMS, usage, 39
vs. EUS-TCB, 73 management after PT-GBD, 39
and FNB, 78 patient with metastatic breast cancer, 37
focal liver lesions, 83 vs. PC-GBDs, outcomes, 38
19G, 86 peri-procedural care, 38
LB, 84 stent removal, optimal time, 39
liver lobes, 92 success rates, 38
malignancy, 116 surgical revision, 39
masses, 88 technique/procedure, 36
malignant liver lesions, 80 transgastric and transduodenal routes, 39, 40
on-site cytopathology, 73 transmural gallbladder drainage, 35
pancreatic cyst ablation, 136, 142 Gastric outlet obstruction (GOO), 159, 160, 166, 167
pancreatic cysts Gastric varices (GV), 57–61
biopsy, 130, 131 Gastroenterostomy (EUS-GE)
CT/MRCP characteristics, 123 adverse events, 166
cystoscopy, 127–129 clinical outcomes, 159
cytobrushing, 129, 130 cross-sectional imaging, 159
Index 193

enteral stents, 159 advances and refinements, 89

entero-enteric anastomosis, 160 advantages, 86, 89
vs. ES, 166, 167 adverse effects (AEs), 91, 92
follow-up care, 162 bilobar, 85
GOO, 159 blood clots, 89, 91
LAMS, 159, 160 complete portal triads, 89, 91
post-procedure care, 162 contraindications, 84
pre-procedure care, 160 FNA, 83, 84
retrospective data, 159 hepatic venous system, 83
safety, 166 histological interpretation, 89, 90
vs. SGJ, 167 identification, 84–86
signs and symptoms, 159 indications, 84
technical and clinical success, 166 LLP, 89
Gastrointestinal (GI) bleeding, 55, 63 management, 83
Gastrointestinal (GI) stromal tumor, 77 multicenter study, 84
Gastrostomy-assisted ERCP, 147 needle preparation, 86, 87
Gastrostomy tube, 147–150, 152, 153, 156 needle selection, 86, 87
Gene therapy, 115–117 needle technique, 87, 88
Glue, 56, 58–62 noninvasive hepatic assessment, 83
See also Cyanoacrylate parenchymal, 83
Gold seeds, 95 percutaneous, 85
post-procedure recovery, 89–91
quantitative, 89
H read the tea leaves, 89
Hemostasis, 55, 56, 58, 60 real-time FNB, 83
Heparinization, 88 specimen handling, 88, 89
Hepaticogastrostomy (EUS-HGS), 17, 18, 22 specimen length, 89
Hepatic vein (HV), 171–173 standard adequacy metrics, 89
Hepatic venous pressure gradient (HVPG), 170, 171 suction, 86, 87
Hepatic venous system, 83 technique, 84–92
Hepatocellular carcinoma (HCC), 109 TIPS, 83
High-intensity focused ultrasound (HIFU), 143 tissue yields, 84
Hybrid cryotherm probe, 111 training in percutaneous, 83
transjugular, 85
Tru-Cut needle, 83, 84
I Lumen-apposing metal stents (LAMS), 27, 30, 150, 152,
Image-guided radiation therapy (IGRT), 95, 102, 104, 105 154, 155
Immunotherapy, 117 adverse rate, 36
Inferior vena cava (IVC), 171, 172 AXIOS stent, 6
Intra-abdominal abscess, 183, 184 bowel lumen, 182
Intra-cystic markers, 120 vs. DPPSs, adverse events, 9, 11
Intraductal papillary mucinous neoplasm (IPMN), 112, EC-LAMS, 10–11
135, 139, 142, 143 electrocautery-enhanced, 178
branch duct, 119, 123, 126 in 181 EUS-GBD procedures, 38
intestinal type, 124 vs. FCSEMSs, 10
main duct, 119 10.8F LAMS delivery catheter, 36
malignant transformation, 119 gallbladder drainage, 35, 36
and MCNs, 123, 128 GE
mucinous lesions, 119 advent, 159
mural lesions, 121 anastomoses, 159
mural nodules, 121 anastomosis, 162
neoplastic cystic lesions, 125 balloon dilation, 165
with nodule, 121 catheter system, 161
villous structures, 125 and CE-LAMS, 160, 161
distal flange, 164
electrocautery-enhanced, 164
L gastroenteric anastomosis, 167
Laparoscopic-assisted ERCP, 150, 155 GOO, 160
Length of the longest piece (LLP), 89 maximal stent diameter, 161
Liver biopsy (EUS-LB), 78–80 misdeployment, 166, 167
194 Index

Lumen-apposing metal stents (LAMS) (cont.) SCN, 126

non-cautery-enhanced, 161 superficial vascular network, 126
optimal duration, 162 transduodenal approach, 125
and stomach, 160 villous/finger like structures, 125
TTS dilation balloon, 161 Needle maneuverability, 26
types, 160 Neoadjuvant chemotherapy, 99
graded dilation technique, 9 Neoplastic pancreatic cysts
large bore LAMS, for EUS-GBD, 39 CEA levels, 120
mediastinal abscess, 185 CH-EUS, 120–123
15mm diameter, patient with infected WON, 6, 7 CT scans, 119
NAGI stent, 6 cytology, 120
Niti-S Spaxus stent, 6 definitive diagnosis, 119
pelvic abscess, 181 EUS, 120
vs. plastic stents, 10 FNA (see Fine needle aspiration (EUS-FNA))
post-deployment, 36 gastrointestinal complaints, 119
safe and efficacious, 8 high-risk features, 120
success rate, 39 honey-comb microcystic appearance, 120
Lymphadenopathy, 77, 78 imaging studies, 119
intra-cystic markers, 120
IPMN (see Intraductal papillary mucinous neoplasm
M (IPMN))
Macrocystic serous cystadenomas, 143 management, 119
Malignant tumor cells, 115 MCNs, 119
Maloney dilators, 149 MRI, 119
Mediastinal abscess, 185, 186 non-mucinous cysts, 119, 120
Metal biliary stents, 3–6 PCLs (see Pancreatic cystic lesions (PCLs))
Monopolar RFA, 110 worrisome features, 120
MRCP, 142, 143 Noninvasive hepatic assessment, 83
Mucinous cyst, 128 Non-mucinous cysts, 119, 120
Mucinous cystic neoplasms (MCN), 119, 123, 125, 127,
128, 135, 139, 140, 142, 143
Mural nodules, 121 O
Obesity, 147, 155
Obstructive jaundice, 18
N Oligolocular cyst, 137
Necropsy, 170 ONYX-015, 115
Necrosectomy, 2, 7, 12
Needle based confocal laser endomicroscopy (nCLE)
AQ-Flex-19, 124 P
clinical diagnosis, 126 Paclitaxel, 136, 138–141, 143
clinical evaluation, 126 Pancreatic adenocarcinoma, 143
complications, 126 Pancreatic cancer, 74, 76, 99–101
cystic neuroendocrine tumor, 125 Pancreatic cyst ablation
and cystoscopy, 126 agents, 136
diagnostic criteria, 124 antitumor agents, 136
endomicroscopy images and videos, 124 chemotherapeutic agent, 136
evaluation, pancreatic cyst, 123, 124 classification, 135
ex-vivo, 123 clinical decision, 135
feasibility, 125 clinical trials, 136, 139–142
finger-like projections and dark ring, 124, 125 controversy, 138, 139
fluorescein stains, 124 CT, 135
FNA needle tip, 123, 127 curvilinear-array echoendoscope, 136
in vivo, 124 cyst fluid analysis, 135
INSPECT trial, 125 ethanol injection, 136, 137
MCN, 127 EUS-guided, 137, 138
pancreatitis, 127 FNA, 136
PCLs, 126 future use and modifications, 142, 143
pseudo-cysts, 125 indications, 142
real time imaging, 123 injection and lavage, 136, 137
safety and diagnostics, 125 management strategy, 135
Index 195

MRI, 135 “mature PFCs”, 2

safety, 138, 139 PPs (see Pancreatic pseudocysts (PP))
structure, 136 risk factors, for infection, 12
Pancreatic cystic lesions (PCLs) WON (see Walled-off pancreatic necrosis (WON))
accuracy, 127 Pancreatic neuroendocrine tumors (PNET), 109, 110, 112
center study, 126 Pancreatic pseudocysts (PP)
characterization, 120 definition, 1
cholangioscopy fiber optic probe, 128 with DPPSs, 11, 12
clinical diagnosis, 126 drainage, 6
cystoscopy, 128 etiology, 1
DETECT trial, 126, 128 with FCSEMSs, 6
detection, 119 LAMSs vs. DPPSs, 9
EUS-FNA, 131 and WON, 1, 2
evaluation, 120, 123 Pancreatitis
FNA, 123 ANP, 2
imaging characteristics, 128 infection, 11, 12
IPMNs, 123 Pancreatobiliary system, 147
malignant, 120 Papillary protrusions, 128
nCLE, 125 Parenchymal LB, 83
neoplastic and non-neoplastic, 120 Pelvic abscess
nonblinded investigators, 126 and amenability, 178
patient with, 120 biliary and pancreatic stent placement, 180
solid components, 120 cavity, 180
tests, 120 and dilation, 183
visualization, 127 dilator/cystotome, 182
Pancreatic cystic neoplasms, 112 distal flange, 181
See also Neoplastic pancreatic cysts electrocautery, 182
Pancreatic duct drainage (EUS-PD) etiologies, 177
adverse events, 52, 53 EUS-guided drainage, 178
antegrade orientation, 49–51 FCSEMS, 181
common contraindications, 46 7F double pigtail stents deployed, 181
common indications, 45 internal debris/contents, 179
opacification, PD, 45 LAMS, 181
percutaneous/surgical drainage, 45 plastic stents, 182
plastic pancreaticogastrostomy stent, 49–51 proximity to bladder, 180
procedural technique stent placement, 182
cautery, 46 therapy, 177
equipment, 46 transrectally drained abscesses, 178
vs. ERCP, 52 trans-sigmoid drained abscesses, 178, 179
outcomes, 49, 51 Percutaneous approach, 170
patient selection, 46 Percutaneous assisted transprosthetic endoscopic therapy
performing, EUS-PD, 46 (PATENT), 148
post-procedure management, 48 Percutaneous endoscopic gastrostomy (PEG), 152
procedural prerequisites, 46 Percutaneous gallbladder drainage (PC-GBD), 38
procedure times, 47 Percutaneous LB, 90
rendezvous technique, 46 Percutaneous transhepatic biliary drainage (PTBD), 17, 22
success rate, 49, 51, 52 Percutaneous transhepatic gallbladder drainage
technical success, 52 (PT-GBD), 35, 36, 38, 39
transgastric approach, 46 Photodynamic therapy (PDT), 117, 143
transluminal stent placement, 46 Porcine models, 171
transpapillary/anastomotic transluminal stent Portal hypertension (PH), 170
deployment, 47 Portal pressure gradient (PPG) measurement
and standardization, 53 anesthesia care/general anesthesia, 173
Pancreatic duct (PD) intervention, 46, 52, 53 animal studies, 171–172
Pancreatic fluid collections (PFCs) development of esophageal varices, 170
CT scans, 2 Doppler flow, 173–175
endoscopic approaches, 2 EUS-guided, 172
EUS (see Endoscopic ultrasound (EUS)) human studies, 172
fully covered metal biliary stent, 5 HV, 173
LAMSs vs. DPPSs, 11 HVPG, 170
196 Index

Portal pressure gradient (PPG) measurement (cont.) S

middle HV, 173, 174 Saline flushed, 115
non-compressible tubing, 173 Self-expandable metal stent (SEMS), 20, 22, 27
percutaneous approach, 170 Serous cystadenoma (SCN), 119, 126
PH, 170 SharkCore™ needles, 76
portal venous angiography, 169, 170 Single-balloon enteroscopy, 147
PV (see Portal vein (PV)) Solid pancreatic tumors
reduction, 170 brachytherapy, 113, 114
TIPS, 170 cytoimplant, 115
transgastric transhepatic ethanol ablation, 109, 110
approach, 173 fiducial placements, 114, 115
Portal vein (PV), 171, 172 gene therapy, 115–117
access, 169 immunotherapy, 117
catheterization, 170 management, 109
manometry, 171, 172, 175 PDT, 117
thrombosis, 138 RFA, 110–113
visualization, 170 Spiral enteroscopy, 147
Portal venous angiography, 169, 170 Splenic abscess, 178
ProCore™ needle, 74, 76 Splenic vein thrombosis/obliteration, 138
Pseudoaneurysm, 63 Spyglass Direct Visualization System, 127
Pseudocysts, 119 Standard endoscopy, 55, 56, 61, 62
See also Pancreatic pseudocysts (PP) Subphrenic abscesses, 178
PVP, 171, 172 Surgical gastrojejunostomy (SGJ)
clinical outcomes, 159
comparative study, 166
R and ES, 167
Radio frequency ablation (RFA), 143 vs. gastroenterostomy (EUS-GE), 167
abdominal contrast-enhanced CT, 112, 113 palliative measures, 159
bipolar, 110 success rate, 166
classification, 111 Surgically altered anatomy, 17, 18, 22
cryotherm probe, 112
description, 112
echoendoscope, 112 T
echogenic needle tip/probe, 112 T-cell-dependent immune response, 117
electrode delivers, 110 Through-the-needle devices, 111
electromagnetic energy, 110 Through the scope (TTS) dilation balloon, 161
EUS guidance, 110 Tissue tray, 84
mean number, 112 TNFerade, 116, 117
median survival time, 112 Transabdominal ultrasound (TAUS)-guided transhepatic
monopolar, 110 approach, 171
needle electrode, 110, 111 Transgastric transhepatic approach, 173
needle-type device, 111 Transjugular intrahepatic portosystemic
pancreatic cystic neoplasms, 112 shunt (TIPS), 83, 170
PNET, 110, 112 Transluminal biliary drainage, 26, 27
post-EUS-guided, 112, 113 Triamcinolone, 68, 71
post-procedure adverse events, 112 Tru-Cut biopsy (EUS-TCB), 73, 78, 79
symptomatic insulinoma, 112 Tumor necrosis factor-α (TNF-α), 116
target lesions, 112
through-the-needle devices, 111
treatments, 112 U
wattage and exposure time, 112 Umbrella-shaped monopolar retractable
wattage setting, 112 electrode array, 112
Radiotherapy, 95 Unilocular cyst, 137
Rectal cancer, 102, 103
Rectal varices, EUS-guided
angiotherapy, 62 V
Recurrent cholecystitis, 36, 38 Varices, 56, 57, 62
Rendezvous procedure, 25, 27 See also Angiotherapy
Roux-en-Y gastric bypass Visicoil fiducials, 95, 96
(RYGB), 147, 148, 153, 155 Visicoil gold fiducial marker, 96
Index 197

W LAMSs vs. DPPSs, 9

Walled-off pancreatic necrosis (WON) patient with infected WON, LAMS use, 6, 7
CECT criteria, 1 vs. PPs, 2
definition, 1 resolution failure, 10
DEN, 2 Water immersion method, 160, 161
drainage, 10