Contribution Of Environmental Factors To The Risk Of Male Infertility

Human Reproduction Vol.16, No.8 pp. 1768–1776, 2001

Contribution of environmental factors to the risk of male
infertility

Alejandro Oliva1, Alfred Spira2 and Luc Multigner2,3
1Unidad de Andrologia, Hospital Italiano Garibaldi, 2000 Rosario, Argentina and 2INSERM U 292, Hopital de Bicetre, 94276
ˆ ˆ
ˆ
Kremlin-Bicetre, France
3To ´
whom correspondence should be addressed at: INSERM U 435, Campus de Baulieu, Universite de Rennes 1, 35042 Rennes
Cedex, France. E-mail: luc.multigner@rennes.inserm.fr

BACKGROUND: An increasing number of reports suggest that chemical and physical agents in the environment,
introduced and spread by human activity, may affect male fertility in humans. We investigated the relationships
between exposure to environmental agents and seminal characteristics, and the concentrations of reproductive
hormones in the serum of men seeking infertility treatment. METHODS: We studied 225 male partners from
consecutively recruited couples, who had their first infertility consultation between 1995 and 1998, in the Litoral
Sur region of Argentina, one of the most productive farming regions in the world. RESULTS: A multivariate
logistic regression model showed that exposure to pesticides and solvents is significantly associated with sperm
threshold values well below the limit for male fertility. We also found that men exposed to pesticides had higher
serum oestradiol concentrations, and that men exposed to solvents had lower LH concentrations than non-exposed
men. All of these effects were greater in men with primary infertility than in men with secondary infertility.
CONCLUSION: We have shown that environmental factors contribute to the severity of infertility, and that this
may worsen the effects of pre-existing genetic or medical risk factors.

Key words: environment/hormones/male infertility/spermatozoa

Introduction Male reproductive function in the general population has
Humans are exposed to many environmental agents that may attracted increasing attention due to reports suggesting that the
be hazardous to their reproductive capacity. Male reproductive occurrence of several biological problems affecting the male
function is known to be highly sensitive to many chemicals genital tract have increased during the last 50 years (Toppari
and physical agents generated by industrial or agricultural et al., 1996). These include an increased incidence of testicular
activities (Bonde, 1996; Spira and Multigner, 1998). Such cancer and some congenital anomalies, such as cryptorchidism
agents are commonly present in some occupational activities or hypospadias, and an apparent decline of sperm production in
and in the general environment. the overall male population. Such events have been tentatively
Environmental hazards to male reproductive function were linked to the widespread use of chemicals with hormonal
revealed 30 years ago, when pesticide manufacturers and properties, also called endocrine disruptors (Sharpe and
agricultural workers in contact with the nematocide, 1,2- Skakkebaek, 1993).
dibromo-3-chloropropane (DBCP), suffered from severely An obvious undesirable consequence of reproductive
impaired spermatogenesis, leading to infertility (Whorton et al., toxicants is infertility. Infertility, defined as the inability to
1977; Slutsky et al., 1999). Since then, different chemical conceive after 1 year of unprotected intercourse, concerns
classes of pesticides and solvents have been demonstrated to ~15% of couples in Western countries (Thonneau and Spira,
be male reproductive toxicants in animal models (Sundaram 1991; Irvine, 1998). A male contributory factor is involved in
and Witorsch, 1995). However, the number of substances that approximately half of these cases (Irvine, 1998). However, in
have been proven to have a deleterious effect on human a considerable proportion of men with semen anomalies, no
spermatogenesis is very small, and these observations are medical or surgical factors are implicated, and the aetiology
limited to a few cross-sectional studies done on occupational remains unclear. Nowadays, there is an increasing awareness
populations that were exposed to these substances at very high of the potential risk of genetic, chemical, and physical environ-
concentrations (Cohn et al., 1978; Wyrobek et al., 1981; mental agents on male infertility.
Ratcliffe et al., 1987; Schrader et al., 1988; Ratcliffe et al., The involvement of environmental factors in male infertility
1989). Due to the widespread use of such chemicals, and their and the suspected increased incidence of male-related infertility
potential for leakage into the environment, they constitute a induced by such agents are of great concern. It is not known
putative hazard to male fertility. if the reported decline of semen quality, or the increased
1768 © European Society of Human Reproduction and Embryology

Environment and male infertility

incidence in genital anomalies, is linked to the prevalence of demographic, medical, surgical, and reproductive history, recent
infertility. It has been stressed that the percentage of men illnesses and treatment, familial history of infertility, sexuality, occupa-
whose sperm count has fallen below the level associated with tional history, and lifestyle. During the second visit, which took place
optimal fertility has increased (Carlsen et al., 1992; Jouannet 2–4 weeks later, a complete physical and andrological examination
was carried out and men gave a semen specimen and a blood sample.
and Auger, 1996; Bonde et al., 1999).
Toxic damage to the testes can result in many effects, Exposure assessment
namely, reduced sperm production, the production of defective
The degree of exposure was assessed from the initial interview, by
spermatozoa, and impaired androgen production. In man the studying the detailed history of past and present jobs and lifestyle
non-invasive method for assessing male fertility involves habits. Men were asked about their contact with chemical substances
semen analysis, measurement of reproductive hormones in the or physical agents during the previous 10 years. An industrial hygienist
blood, and evaluation of previous success in impregnating a verified the correlation between jobs and the declared exposures.
partner. Semen analysis allows the male reproductive function Patients were divided into five groups: non-exposed (men who did
to be evaluated directly and the relationship between exposure not report any exposure and whose occupation did not expose them
to environmental agents and fertility to be investigated. How- to any of the agents); and three groups of men who were only exposed
ever, semen samples are difficult to obtain in general population to one kind of agent: pesticides (herbicides, fungicides, insecticides,
studies and the participation rate, which is usually 20%, may fumigants, and rodenticides), solvents (paints, varnish, lacquers,
impair conclusions (Bonde, 1996). Studies of populations in thinners, degreasers, and inks), and heat (prolonged sitting position
or radiant heat). Finally, due to the small number of individuals, we
which men are seeking infertility treatment avoid this problem,
regrouped men who were in contact with more than one of these
because sperm analysis is a key part of their fertility evaluation. exposures and/or exposed to miscellaneous agents, including heavy
If the different bias and confounding factors are taken into metals, oils, acids, explosives, and low temperatures (mixed-group).
account, this population provides the opportunity to study the We excluded any individual who could not provide clear information
associations between risk factors and outcomes. However, on their job circumstances or exposure risks.
environmental or occupational exposures may be low in the
population that the consulting men come from, and thus the Semen analysis
statistical power may be insufficient to obtain a significant Semen samples were obtained by masturbation after a recommended
conclusion. We studied a population of men consulting infertil- period of 3–5 days sexual abstinence. Seminal volume was measured
ity clinics in the Litoral Sur region of Argentina. This area is in a graduated pipette, accurate to within 0.1 ml. Sperm concentration
characterized by intensive agricultural and industrial activity, was determined by haemocytometer (improved Neubauer counting
and thus favours an elevated prevalence of environmental chamber), after an appropriate dilution. Sperm motility was assessed
by direct observation under a microscope ( 400). Sperm morphology
exposures. We analysed the effect of environmental exposures
was assessed by staining slides (May–Grunwald–Giemsa) and direct
on semen characteristics and reproductive hormone concentra- observation under a microscope ( 1000). All procedures followed
tions in men having their first infertility consultation in one WHO guidelines (World Health Organization, 1992). The three
of three health centres in the Litoral Sur. laboratories were examined every month by the same quality control
officer. Measurements were evaluated in each laboratory by the same
Materials and methods person during the study period. Between- and within-assay coefficients
of variation were 7 and 6% for sperm concentration, 17 and 9% for
Study area
sperm motility, and 14 and 8% for sperm morphology respectively.
This study took place in the Santa Fe and Entre Rios provinces in
Argentina. These provinces constitute the Litoral Sur region, which Reproductive hormones
is one of the most fertile farmland zones in the world. These Blood samples were collected in a 10 ml plastic syringe, and kept at
neighbouring provinces are separated by the Parana river, and occupy room temperature until serum was separated. After centrifugation,
78 781 km2 and 133 007 km2, respectively. The combined population the serum was transferred to a new tube, kept at –20°C, and assayed
of these provinces was estimated at 4 173 601 inhabitants, in 1991. within 4 weeks. FSH and LH were measured by an immunoradiometric
The Litoral Sur is a rich agricultural and cattle ranching area. There assay, using the Coat-a-Count kit (Diagnostic Products Corporation,
is also much industrial development, centred mainly around the three Los Angeles, CA, USA). The dose-response curve of FSH was
largest cities, Santa Fe capital (449 000 inhabitants), Rosario calibrated according to the WHO 2nd International Reference Prepara-
(1 200 000 inhabitants) and Parana (320 000 inhabitants).
tion (IRP) 78/549, and the LH dose-response curve according to the
Subjects WHO 2nd IRP 80/552 and WHO 1st IRP 68/40. Prolactin was
The study sample comprised 253 consecutively recruited male partners measured by a double-antibody radioimmunoassay (Diagnostic Prod-
from couples having their first infertility consultation at the Andrology ucts Corporation) and calibrated according to the WHO 1st IRP
Unit of one of three private institutions (Hospital Italiano Garibaldi, 75/504 and the 3rd International Standard 84/500. Total testosterone
Rosario; Centro de Urologia, Santa Fe; Sanatorio Adventista del and oestradiol were measured by radioimmunoassay, using the Coat-
Plata, Libertador General San Martin, Entre Rios), between January a-Count kit (Diagnostic Products Corporation). Between- and within-
1995 and December 1998. They represented all causes of male assay coefficients of variation were 9.2 and 8.5% for FSH, 10.8 and
infertility, except testicular cancer. Approval for this study was 12.1% for LH, 9.8 and 9.5% for prolactin, 12.5 and 8.5% for
obtained from the Institutional review board. testosterone, and 16.1 and 12.2% for oestradiol respectively.

Data sources Outcomes
Information was collected at several stages. A structured interview Continuous outcomes (as evaluated by the laboratory) were seminal
was conducted, during the first visit, to obtain information on basic volume (ml), sperm concentration ( 106/ml), total sperm output

1769

A.Oliva, A.Spira and L.Multigner

( 106; the product of seminal volume sperm concentration), percent-
age of total motile spermatozoa, percentage of spermatozoa with Table I. Occupational circumstances of exposure
normal morphology, and concentration of reproductive hormones in
the serum. To study the effect of exposure on these outcomes, seminal Groups Activity n
characteristics were categorized as dichotomous variables. Seminal
Non-exposed Professionals 23
volume, sperm concentration and sperm output were categorized
(n 80) Administrative 21
according to the 25th, 50th or 75th percentiles, determined from the Technicians 20
distribution in the study sample. To determine threshold values which Sales workers 10
gave the maximum likelihood ratio, we successively compared the Others 6
Pesticides Farmers 23
exposure association below and above the 25th, 50th or 75th percentile
(n 40) Animal husbandry 5
limits of distribution. We used thresholds of 50 and 30%, for sperm Fumigators 5
motility and sperm morphology respectively, based on the WHO Pesticide factory workers 4
(1992) guidelines for defining reference limits. We compared concen- Others 3
Solvents Mechanics 12
trations of hormones in the serum as continuous variables between
(n 22) Painters 5
groups. As serum hormone values are highly skewed and do not Printers 3
follow a normal distribution, we normalized the distributions by Woodworkers 2
applying a log10 transformation (confirmed by Kolmogorov–Smirnov Heat Drivers 13
(n 21) Bakers 3
test for normality).
Cooks 2
Others 3
Data classification Mixed Farmers 6
(n 14) Chemical factory workers 2
We recorded and classified the following information as continuous Welders 2
variables: age, weight, height, time trying to conceive with the present Others 4
partner with unprotected intercourse before consultation (waiting
time), average monthly frequency of intercourse, duration of the
length of abstinence before the semen analysis, and testicular volume.
Body mass index (BMI) was calculated as weight/height2 (kg/m2).
We also recorded the following data: season of sperm analysis (spring,
summer, autumn, winter); annual income ( $12 000, between $12 000 Results
and $36 000, $36 000); smoking (current smoker versus non- Characteristics of the population
smoker), alcohol consumption ( 20 g/alcohol per day versus
20 g/alcohol per day); fertility status of men whether they had Of the initial 253 consulting men, 189 attended the second
fathered a child or impregnated the present partner, regardless of the consultation and gave a sperm sample. Of these 189 men, 80
outcome, before consultation (secondary infertility), or not (primary were classified as non-exposed, 40 as exposed to pesticides,
infertility). All retrospective health information and data obtained 22 as exposed to solvents, 21 as exposed to heat, 14 as exposed
from the physical examination and the laboratory were used to define to a mixture of the above. Twelve men were unclassified, and
the potential risk factors for male infertility, in accordance with WHO excluded, as described in Materials and methods. The age,
recommendations (1993). Medical or surgical male infertility risk infertility status and exposure characteristics of the 64 men
factors were categorized as follows (yes or no): reported family who did not supply sperm samples did not differ from the 189
history of infertility, cryptorchidism (treated or not), varicocele
men who did supply a sperm sample (data not shown). The
(treated or not), testicular torsion, testicular trauma, genital infections,
orchitic mumps, orchido-epididymitis, sexual dysfunction, systemic
occupational exposures of the 177 men included in this analysis
diseases (tuberculosis, diabetes mellitus), allergies, inguinal hernia are shown in Table I. The median exposure times were 7, 4,
(treated), urinary infections and treatments which may interfere with 3.5, and 4 years for the pesticide, solvent, heat, and mixed
testicular function. exposure groups respectively.
Comparisons between some general characteristics in each
Statistical analysis of the exposed groups and the non-exposed group are shown
Comparisons between groups were made using the χ2-test for categor- in Table II. The mean age of the study population was 33.6
ized variables, and the t-test and analysis of variance/co-variance for years, this value did not differ between groups. Nearly 40%
continuous variables. In the analysis involving a categorized outcome of the men were overweight (BMI between 25 and 30), obesity
variable, we used multiple logistic regression analysis to produce (BMI 30) was more frequent in the solvent-exposed group
odds ratio (OR) and the 95% confidence intervals (95% CI) for the (40% of cases compared with 20% in other groups). Alcohol
association between seminal characteristics and exposure variables, consumption was similar in each group, but there was a higher
adjusted for confounding factors. Univariate logistic regression ana- proportion of smokers in the solvent- and heat-exposed groups.
lysis was used to evaluate potential confounding factors in each
The main risk factors identified for infertility were: varicocele
exposed group, separately. Factors were considered confounding if
(37% of the overall population), orchido-epididymitis (28%),
their inclusion in the model modified the estimate of the OR by
10% (Greenland and Rothman, 1998). Age, length of abstinence, cryptorchidism (8.5%) and orchitic mumps (6.9%). The propor-
income, health center, BMI, and smoking were always included as tion of medical and surgical infertility risk factors was equal
confounding factors. All analyses were carried out using the Statview in groups (data not shown). Men with primary infertility were
software package (SAS Institute Inc., Cary, USA). All P-values were younger, and had lower seminal characteristics to men with
two-sided, and considered to be significant if P 0.05. secondary sterility (Table III).
1770


Table II. General characteristics of exposure groups

Non-exposed Exposed to (n 83)
(n 80)
Pesticides Solvents Heat Mixed
(n 40) P (n 22) P (n 21) P (n 14) P

Age (years)a 33.7 (5.3) 33.2 (4.2) ns 33.7 (7.0) ns 33.6 (6.5) ns 33.8 (6.4) ns
Body mass indexa 25.8 (3.3) 26.7 (3.8) ns 28.9 (4.0) 0.003 26.0 (3.6) ns 28.1 (5.6) ns
Testicular volume (ml)a 43.5 (9.2) 44.4 (10.0) ns 47.3 (8.2) ns 42.4 (9.8) ns 41.9 (11.9) ns
Waiting time before consultation 24 24 ns 36 ns 24 ns 24 ns
(months)b
Monthly frequency of intercoursea 11.6 (5.9) 11.9 (7.5) ns 10.5 (6.0) ns 11.3 (6.5) ns 10.6 (4.4) ns
Sexual abstinence (days) a 4.1 (0.6) 4.3 (0.6) ns 4.2 (0.6) ns 4.2 (0.6) ns 4.1 (0.5) ns
Infertility statusc
Secondary 32 (40.0) 10 (25.0) ns 11 (50.0) ns 9 (42.9) ns 5 (38.5) ns
Primary 48 (60.0) 30 (75.0) 11 (50.0) 12 (57.1) 8 (61.5)
Seasons of sperm analysisc
Spring 22 (27.5) 8 (20.0) ns 6 (27.3) ns 5 (23.8) ns 3 (21.4) ns
Summer 14 (17.5) 9 (22.5) 1 (4.5) 7 (33.3) 5 (35.7)
Autumn 4 (5.0) 3 (7.5) 1 (4.5) 0 (0.0) 2 (14.3)
Winter 40 (50.0) 20 (50.0) 14 (63.6) 9 (42.9) 4 (28.6)
Tobaccoc
Non-smokers 63 (78.8) 33 (82.5) ns 7 (31.8) 0.0001 49 (42.9) 0.0012 10 (71.4) ns
Current smokers 17 (21.2) 7 (17.5) 15 (68.2) 12 (57.1) 4 (28.6)
Alcoholc
20 g/day 62 (77.5) 27 (67.5) ns 16 (72.7) ns 17 (81.0) ns 12 (85.7) ns
20 g/day 18 (22.5) 13 (32.5) 6 (27.3) 4 (19.0) 2 (14.3)

aMean (SD).
bMedian.
cn (%).

Table III. Age and seminal characteristics according to infertility status

All Primary infertility Secondary infertility
(n 177) (n 110) (n 67)

Mean (SD) Median Mean (SD) Median Mean (SD) Median

Age (years) 33.6 (5.5) 33.0 31.9 (4.3) 31.6 36.4 (6.1) 35.3
Seminal volume (ml) 3.1 (1.6) 2.8 3.0 (1.7) 2.5 3.1 (1.4) 3.0
Sperm concentration ( 106/ml) 28.6 (46.0) 14.2 21.9 (29.6) 10.9 39.6 (63.3) 21.6
Sperm output ( 106) 82.8 (110.7) 41.6 70.3 (106.1) 23.8 103.4 (115.8) 65.0
Sperm motilitya (%) 36.6 (24.4) 41.0 33.5 (24.5) 38.5 41.7 (23.6) 50.0
Sperm morphologya (%) 29.8 (15.7) 28.5 27.3 (15.7) 26.0 33.6 (14.9) 36.0

aSeven azoospermic men were not included (five primary infertility and two secondary infertility).

Seminal characteristics motile sperm cells. Exposure to pesticides was also associated,
Dichotomized dependent variables were used to assess the but not significantly, with 30% of sperm cells being morpho-
association between exposure and seminal characteristics by logically normal. After stratification, men with primary infertil-
logistic regression. Table IV shows the adjusted OR; the ity were found to have an increased risk for all these anomalies,
reference group (OR 1) is always the non-exposed group. and men with secondary infertility were found to have an
Exposure to pesticides significantly increased the risk of a increased risk of low sperm motility. We observed a dose-
seminal volume of 3.8 ml (corresponding to the 75th related response in men with primary infertility exposed to
percentile). However, after stratification, this increase was only pesticides. These associations were higher for sperm concentra-
significant in men with secondary infertility. Moreover, we tion [4.4 (1.2–15.7)], sperm output [3.8 (1.1–14.3)], sperm
observed a higher risk in men who were frequently exposed motility [7.8 (1.5–41.0)], and sperm morphology [3.6 (1.1–
to pesticides [OR 8.5 (95% CI 1.2–58.8)], than that in those 11.5)], in men who were frequently exposed than in men
who were only exposed occasionally 4.4 (0.5–38.3). Exposure who reported occasional exposure [3.1 (0.2–43.3) for sperm
to pesticides was associated with 1 106 spermatozoa per concentration, 3.4 (0.2–48.4) for sperm output, 0.7 (0.1–9.6)
ml (corresponding to the 25th percentile), a sperm output of for sperm motility, and 1.2 (06–11.5) for sperm morphology].
3 106 (corresponding to the 25th percentile), and 50% Men exposed to solvents showed elevated OR for inferior
1771

1772

Table IV. Odds ratios (OR) and 95% conﬁdence intervals (95% CI) for the relationship between seminal characteristics and exposure

Seminal volume Sperm concentration Sperm output Sperm motilitya Sperm morphologya
3.8 ml 1 106/ml 3 106 50% 30%

OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P

All
Non-exposed 1.0 1.0 1.0 1.0 1.0
Pesticides 2.8 (1.2–6.6) 0.019 3.0 (1.2 –7.4) 0.015 2.7 (1.1–6.7) 0.031 4.5 (1.8–11.5) 0.002 2.1 (1.0 –4.8) 0.064
Solvents 2.4 (0.8–7.6) 0.130 2.7 (0.9–8.3) 0.088 2.5 (0.8–7.9) 0.105 3.1 (1–9.5) 0.050 3.0 (1.0–9.0) 0.055
Heat 1.6 (0.5 –5.2) 0.459 1.9 (0.6 –6.3) 0.265 2.3 (0.7 –7.3) 0.141 1.3 (0.5–3.6) 0.612 0.9 (0.3–2.4) 0.811
Mixed 0.8 (0.1–6.8) 0.794 1.8 (0.3–11.1) 0.529 3.2 (0.9–11.1) 0.073 2.6 (0.6–12.0) 0.221 0.8 (0.2–3.7) 0.822
Primary infertility
Non-exposed 1.0 1.0 1.0 1.0 1.0
Pesticides 2.2 (0.7 –6.5) 0.164 3.4 (1.2–9.7) 0.024 3.0 (1.0–8.5) 0.048 3.6 (1.1 –11.4) 0.031 4.1 (1.4–12.0) 0.011
Solvents 2.3 (0.4–12.6) 0.318 3.9 (0.9–17.3) 0.078 3.6 (0.8–16.4) 0.092 10.6 (1.1–105.6) 0.044 4.4 (0.7–26.2) 0.103
Heat 1.1 (0.2–6.7) 0.875 1.7 (0.4–7.8) 0.470 2.4 (0.6–10.4) 0.236 2.2 (0.5–9.6) 0.300 1.0 (0.3–3.9) 0.982
Mixed ncb 1.4 (0.1–16.6) 0.773 3.9 (0.9–13.7) 0.080 2.0 (0.3–15.3) 0.512 1.5 (0.2–10.8) 0.712
Secondary infertility
Non-exposed 1.0 1.0 1.0 1.0 1.0
Pesticides 6.6 (1.4–31.4) 0.019 1.8 (0.3–12.0) 0.534 1.8 (0.3–12.0) 0.534 5.8 (1.0 –32.7) 0.046 0.4 (0.1–2.1) 0.265
Solvents 2.3 (0.5–11.7) 0.310 2.3 (0.3–18.3) 0.426 2.3 (0.3–18.3) 0.426 2.1 (0.5–9.3) 0.321 2.7 (0.6–13.2) 0.208
Heat 2.3 (0.4–12.6) 0.352 2.7 (0.3 –21.6) 0.338 2.7 (0.3–21.6) 0.338 0.9 (0.2–4.4) 0.874 0.6 (0.1–3.2) 0.534
Mixed 1.8 (0.1–23.7) 0.649 1.7 (0.1–49.2) 0.752 1.4 (0.1–17.1) 0.783 9.1 (0.6–131.9) 0.103 0.3 (0.1–4.2) 0.377

aSeven azoospermic men were not included (two non-exposed, two pesticide-exposed, one solvent-exposed and one mixed-exposed.
bNon-calculable.


Table V. Serum reproductive hormones in exposed groups

Non-exposed Exposed to:

Pesticides Solvents Heat Mixed

Mean (SD) Mean (SD) P Mean (SD) P Mean (SD) P Mean (SD) P

All
FSH (IU/l) 10.6 (15.3) 8.2 (6.9) 0.965 5.8 (4.3) 0.225 10.5 (12.6) 0.945 9.9 (12.8) 0.200
LH (IU/l) 5.1 (4.3) 6.0 (5.0) 0.441 3.3 (2.1) 0.073 3.8 (2.5) 0.333 5.2 (4) 0.884
Testosterone (ng/ml) 5.3 (2.3) 5.5 (2.8) 0.927 5.2 (5.5) 0.419 4.6 (1.8) 0.292 4.5 (3.6) 0.107
Oestradiol (pg/ml) 23.0 (15.8) 36.8 (25.7) 0.002 23.1 (17.2) 0.647 27.3 (22.9) 0.428 25.8 (18.4) 0.416
Prolactin (ng/ml) 12.3 (7.8) 11.5 (7.3) 0.910 11.2 (6.4) 0.705 14.4 (9.1) 0.221 14.8 (11.4) 0.290
Oestradiol/testosterone ratio 5.2 (4.8) 8.6 (10.2) 0.005 5.0 (4.6) 0.763 6.5 (7.0) 0.267 6.3 (7.0) 0.415
Primary infertility
FSH (IU/l) 10.2 (12.2) 8.7 (7.0) 0.907 7.0 (5.5) 0.608 9.4 (9.1) 0.863 10.9 (15.6) 0.393
LH (IU/l) 5.4 (3.8) 6.7 (5.4) 0.509 2.6 (0.9) 0.004 4.2 (3.2) 0.185 5.0 (4.7) 0.210
Oestradiol (pg/ml) 24.0 (16.7) 34.1 (20.7) 0.019 20.7 (8.9) 0.882 27.3 (21.7) 0.606 31.0 (16.2) 0.161
Prolactin (ng/ml) 13.3 (8.9) 11.7 (7.9) 0.598 10.5 (5.5) 0.553 13.4 (7.2) 0.621 14.7 (13.3) 0.899
Oestradiol/testosterone ratio 5.1 (4.4) 6.9 (4.7) 0.041 5.1 (2.3) 0.527 6.0 (5.0) 0.432 6.5. (7.7) 0.347
Secondary infertility
FSH (IU/l) 11.4 (19.4) 6.8 (6.7) 0.786 4.6 (2.6) 0.300 11.9 (16.8) 0.773 8.0 (6.6) 0.659
LH (IU/l) 4.6 (5.1) 3.8 (2.9) 0.889 3.9 (2.6) 0.956 3.2 (0.9) 0.921 5.4 (2.3) 0.436
Oestradiol (pg/ml) 21.5 (14.4) 45.9 (38.8) 0.061 24.9 (22.0) 0.587 27.2 (25.4) 0.522 19.9 (20.9) 0.692
Prolactin (ng/ml) 10.8 (5.4) 10.6 (5.0) 0.755 12.0 (7.5) 0.865 15.9 (11.5) 0.211 15.0 (8.3) 0.326
Oestradiol/testosterone ratio 5.5 (5.4) 14.6 (19.2) 0.043 4.9 (5.8) 0.705 7.0 (8.9) 0.276 5.8 (6.0) 0.652

sperm characteristics, but this was only significant for sperm concentrations in the serum of male partners of infertile
motility. After stratification, the OR were higher in men with couples. We studied a population of men seeking infertility
primary infertility. We did not find any significant association treatment, which allowed us to investigate the association
between categorized seminal outcomes, and heat exposure, between high prevalent risk factors and the outcomes related
before or after stratification in either infertility status. Finally, to infertility. Nevertheless, these men constitute a selected
for the mixed exposure group, an increased risk for low sperm population, thus conclusions derived from such studies should
output was observed, although the 95% CI includes one, and be interpreted with caution.
this was still present in men with primary infertility. Over half the men included in this study were exposed to
either chemical or physical factors that may be associated with
Hormonal characteristics
infertility. This proportion suggests a relationship between
Table V shows the mean concentration of reproductive hor- these factors and male infertility, but may also express a
mones in the serum in each group. It also shows the statistical selection bias. The exposure prevalence may be artificially
comparison, as continuous variables, between each exposure increased by the chosen geographic site which was an industrial
group and the non-exposed group, after adjustment for potential and agricultural area. Moreover, not all couples with an
confounding factors (age, BMI, and smoking). The pesticide- infertility problem seek medical help. If the decision to seek
exposed group showed significantly higher oestradiol values medical care is related to occupational exposure, this may
than the non-exposed group, and after stratification this result in increased prevalence of such factors. Patients with
remained significant in men with primary infertility, and was primary infertility are known to seek medical help more
nearly significant in men with secondary infertility. Moreover, promptly than those who have already conceived. The propor-
there was a significantly higher oestradiol/testosterone ratio in tion of primary infertility was higher in the pesticide-exposed
the pesticide-exposed group than in the non-exposed group, group, but we do not know how one factor may influence the
in men with both primary and secondary infertility. The other. Therefore, we cannot exclude a selection bias, with
solvent-exposed group had a lower LH concentration than the people exposed to certain agents consulting more frequently.
non-exposed group, and after stratification this difference To counteract this potential bias we stratified our analysis,
was highly significant in men with primary infertility. No according to the infertility status.
differences were observed for any reproductive hormones, Another source of bias is the misclassification of the type
before or after stratification, in the heat-exposed group or in the of exposure. We evaluated the type of exposure based on
mixed-exposed group, compared with the non-exposed group. detailed questionnaires, mostly consisting of occupational
questions, which provided a qualitative assessment. We are
Discussion aware that many different active compounds are combined in
This study aimed to evaluate the effect of environmental each of the exposure groups and that the exposure conditions
agents on seminal characteristics and reproductive hormone differ between individuals (i.e. intensity and frequency).
1773


However, further sub-grouping and quantitative evaluation pesticides, in both primary and secondary infertility, is
were impaired by the limited size of our sample. Biological surprising. The oestrogen/androgen imbalance was evaluated
assessment of exposure would have been more precise indic- by calculating the oestradiol/testosterone ratio. This ratio
ators, but this was limited by the cost and the large number was significantly increased in the pesticide-exposed group
of suspected chemicals that individuals were exposed to. and may reflect the quality of the endocrinological milieu
Despite these limitations, questionnaires have provided good of the testis (Itoh et al., 1994). An increased oestradiol/
estimates of exposures (Tielemans et al., 1999a). testosterone ratio may be a marker of infertility in obese
This study showed that, in an infertility-consulting popula- men (Jarow et al., 1993). Nevertheless, we found that the
tion, environmental exposure, particularly to pesticides and oestradiol/testosterone ratio was significantly increased in
solvents, is associated with dramatic changes in seminal pesticide-exposed group, even after adjustment for the BMI.
characteristics. Our results suggest that toxicants act on the Many pesticides act as direct testicular toxicants (Schrader
testes and post-testicular sites, including the accessory sex and Kesner, 1992; Bonde, 1996), but some of them are
glands. Our observations are consistent with those found in now believed to exert toxicity due to their similarity to
previous cross-sectional studies on workers exposed to specific reproductive steroid hormones (Cheek and McLachlan, 1998).
pesticides, such as DBCP (Lipshultz et al., 1980), chlordecone Pesticides can therefore bind to endocrine receptors, and
(Cohn et al., 1978), carbaryl (Wyrobek et al., 1981), ethylene may act as hormonal antagonists or agonists, disrupting
dibromide (Ratcliffe et al., 1987; Schrader et al., 1988) or biological responses (Johnson et al., 2000). Further studies
2-4D (Lerda and Rizzi, 1991), and to solvents, such as glycol are required to determine whether these mechanisms cause
ethers (Welch et al., 1988; Ratcliffe et al., 1989), carbon the observed seminal modifications in the pesticide-
disulphide (Lancranjan, 1972), perchloroethylene (Eskenazi exposed group.
et al., 1991) or 2-bromopropane (Kim et al., 1996). A study Little is known about the effects of solvents on reproductive
of men attending a sperm bank (Bigelow et al., 1998) found hormones. It is noteworthy that exposure to carbon disulphide,
higher seminal volumes and lower sperm concentrations and an industrial organic solvent, is accompanied by reduced
motility, in farm workers than in non-farm workers. Previous circulating concentrations of LH (Lancranjan, 1972). Most
studies on patients from infertility clinics showed that solvents seem to be direct testicular toxicants (Schrader and
glycol ethers (Veulemans et al., 1993) and aromatic solvents Kesner, 1992; Bonde, 1996). The low concentrations of LH
(Tielemans et al., 1999b) are associated with reduced semen that accompany exposure to solvents suggests a feedback
quality. Although scrotal thermal irradiation and sitting for mechanism induced by a gonadal action; however, an effect
prolonged periods of time are strongly suspected to be infertility on the pituitary gland or the hypothalamus cannot be
risk factors (Thonneau et al., 1998), we did not observe an excluded.
increased risk of poor sperm characteristics in the heat-exposed Only values under a certain limit are considered to
group. This discrepancy may be due to low exposure times or contribute to infertility and reduce the likelihood to pregnancy.
intensities. Most of the men in this group spend prolonged However, these limits are variable and are not strictly
periods in the sitting position (e.g. taxi drivers), which may defined. Seminal volume, except for very low volumes, does
be not sufficient to cause dramatic seminal modifications. not seem to be significantly related to fertility. However,
Despite the small number of men in the mixed-exposure group, very high volumes may reduce the sperm concentration
an increased risk of low sperm production was observed. This considerably and prove critical when sperm output is already
group comprised individuals exposed to heavy metals, which low. Sperm concentrations below 5, 10 or 20 106/ml have
are known testicular toxicants, and individuals exposed to been associated with increased infertility, and 5 106/ml is
multiple agents which may increase the risk of seminal considered to be the clinically significant threshold of male
anomalies. infertility (Jouannet et al., 1988). The proportion of sperm
We observed higher values of oestradiol in the pesticide- with normal morphology was strongly correlated to the
exposed group, and lower LH values in the solvent-exposed likelihood of pregnancy, and was independent of sperm
group, than in the non-exposed group. Few studies have concentration. Thus, the probability of pregnancy was
investigated the impact of pesticide exposure on blood reduced when the proportion of spermatozoa with normal
reproductive hormones. These studies only investigated the morphology was 40% (Bonde et al., 1998). The likelihood
effect of DBCP (Whorton et al., 1977; Egnatz et al., 1980; of pregnancy was also decreased if 60% of sperm cells
Lipshultz et al., 1980; Lantz et al., 1981; Eaton et al., were motile, but after adjustment for sperm concentration
1986) and para-tertiary butyl benzoic acid (Whorton et al., and morphology this decrease was not significant (Bonde
1981) on FSH, LH and testosterone production. These et al., 1998). In this study, we showed that exposure to
studies gave contradictory results, and interpretation was pesticides and solvents is significantly associated with
rendered difficult because hormonal values vary according threshold sperm values, much lower than the considered
to the time between the last toxicant exposure and the time limits for male fertility.
of hormonal evaluation (Lipshultz et al., 1980). Moreover, We noted that the significant associations between exposure
serum concentrations of reproductive hormones fluctuate to agents and all seminal characteristics, except seminal
considerably over time, and conclusions obtained from one volume, were more frequent in men with primary infertility
sample should be interpreted with caution. The high serum than in men with secondary infertility. This may be because
concentrations of oestradiol in the group exposed to environmental factors potentiate the adverse effects of
1774


predisposing genetic, medical or surgical factors for infertility Eaton, M., Schenker, M., Whorton, M.D. et al. (1986) Seven-year follow-up
of workers exposed to 1,2-dibromo-3-chloropropane. J. Occup. Med., 28,
frequently found in men with primary infertility. 1145–1150.
In conclusion, we believe that environmental factors, particu- Egnatz, D.G., Ott, M.G., Townsend, J.C. et al. (1980) DBCP and testicular
larly exposure to pesticides and solvents, may contribute to effects in chemical workers: an occupational survey in Midland, Michigan.
J. Occup. Med., 22, 727–732
the severity of sperm parameters, and that infertile patients
Eskenazi, B., Wyrobek, A.J., Fenster, L. et al. (1991) A study of the effect
constitute a highly susceptible group. The testicles are one of of perchloroethylene exposure on semen quality in dry cleaning workers.
the most vulnerable organs to environmental physical and Am. J. Ind. Med., 20, 575–591.
chemical agents. The use of these agents has increased substan- Food and Agriculture Organization of the United Nations (2001) FAO
tially since the 1940s, due to industrial and agricultural Statistical Databases. http://www.fao.org/
Grenland, S. and Rothman, K.J. (1998) Introduction to stratified analysis. In
activities, and Argentina has become a major user of pesticides. Rothman, K.J. and Greenland, S. (ed.), Modern Epidemiology. Lippincot-
The main agricultural activities are cereal culture (mainly Raven, Philapdelphia, pp. 253–279.
wheat and soya beans) and fruit trees, which require large Irvine, D.S. (1998) Epidemiology and aetiology of male infertility. Hum.
quantities of insecticides, fungicides and herbicides. The recent Reprod., 13 (Suppl. 1), 33–44.
Itoh, N., Tsukamoto, T., Nanbu, A. et al. (1994) Changes in the endocrinological
development of transgenic plants that are resistant to herbicides milieu after clomiphene citrate treatment for oligozoospermia: the clinical
has also led to extended use of these products. In the absence significance of the estradiol/testosterone ratio as a prognostic value.
of reliable data on the quantitative distribution of the various J. Androl., 15, 449–455.
active compounds used, only the economic data can be Jarow, J.P., Kirkland, J, Koritnik, D.R. and Cefalu, W.T. (1993) Effect of
obesity and fertility status on sex steroid levels in men. Urology, 42, 171–174.
compared with other countries. Thus, between 1990 and 1998,
Johnson, R.A., Harris, R.E. and Wilke, R.A. (2000) Are pesticides really
the turnover represented by the import and export of pesticides endocrine disruptors? Wisconsin Med. J., 99, 34–38
by Argentina increased by 270%. In comparison, during the Jouannet, P. and Auger, J. (1996) Declining sperm counts? More research is
same period, this increase was 49 and 62% for France and the needed. Andrologia, 28, 302–303.
USA respectively (Food and Agriculture Organization of the Jouannet, P., Ducot, B., Feneux, D. and Spira, A. (1988) Male factors and the
likelihood of pregnancy in infertiles couples. I. Study of sperm
United Nations, 2001). characteristics. Int. J. Androl., 11, 379–394.
There has been increased concern in many Western countries Kim, Y., Jung, K., Hwang, T. et al. (1996) Hematopoietic and reproductive
regarding the deleterious effects of environmental chemical hazards of Korean electronic workers exposed to solvents containing
2-bromopropane. Scand. J. Work Environ. Health, 22, 387–391.
agents on male reproduction. Attention to this issue should
Lancranjan, I. (1972) Alterations of spermatic liquid in patients chronically
now be given by developing countries. Finally, our conclusions poisoned by carbon disulphide. Med. Lavoro, 63, 29–33.
should promote further evaluation of male reproductive toxicity Lantz, G.D., Cunningham, G.R., Huckins, C. et al. (1981) Recovery from
of commonly used substances or those that are likely to be in severe oligospermia after exposure to dibromochloropropane. Fertil. Steril.,
contact with human populations, on male fertility. 35, 46–53.
Lerda, D. and Rizzi, R. (1991). Study of reproductive function in persons
occupationally exposed to 2,4-dichlorophenoxyacetic acid. Mutat. Res., 262,
47–50.
Acknowledgements Lipshultz, L.I., Ross, C.E., Whorton, D. et al. (1980) Dibromochloropropane
and its effect on testicular function in man. J. Urol., 124, 464–468.
This study was supported by funds from INSERM. We acknowledge
Ratcliffe, J.M., Schrader, S.M., Steenland, K. et al. (1987) Semen quality in
the support provided by the Fundacion Universitaria Italiana de
papaya workers with long term exposure to ethylene dibromide. Br. J. Ind.
Rosario and the Hospital Italiano Garibaldi. We would like to thank Med., 44, 317–326.
the technicians from the laboratories of each study centre for carrying
Ratcliffe, J.M., Schrader, S.M., Clapp, D.E. et al. (1989) Semen quality in
out the semen analyses and hormonal measurements. We thank workers exposed to 2-ethoxyethanol. Br. J. Ind. Med., 46, 399–406.
J.M.Guigena for helping us with the exposure assessment, and
Schrader, S.M., Turner, T.W. and Ratcliffe, J.M. (1988) The effects of ethylene
Dr S.Cordier for her thoughtful review and comments. dibromide on semen quality: a comparison of short-term and chronic
exposure. Reprod. Toxicol., 24, 191–198.
Schrader, S.M. and Kesner, J.S. (1992) Male reproductive toxicology. In
References Maurenu, P. (ed.), Occupational and Environmental Reproductive Hazards.
Williams and Wilkins, Baltimore, pp. 3–17.
Bigelow, P.L., Jarrell, J., Young, M.R. et al. (1998) Association of semen
quality and occupational factors: comparison of case-control analysis and Sharpe, R.N. and Skakkebaek, N.E. (1993) Are oestrogens involved in falling
analysis of continuous variables. Fertil. Steril., 69, 11–18. sperm counts and disorders of the male reproductive tract? Lancet, 343,
1392–1395.
Bonde, J.P. (1996) Environmental Factors. In Comhaire, F.H. (ed.), Male
Infertility, Clinical Investigation, Cause Evaluation and Treatment. Chapman Slutsky, M., Levin, J.L. and Levy, B.S. (1999) Azoospermia and oligospermia
and Hall, London, pp. 267–284. among a large cohort of DBCP applicators in 12 countries. Int. J. Occup.
Bonde, J.P., Ernst, E., Jensen, T.K. et al. (1998) Relation between semen Environ. Health, 5, 116–122.
quality and fertility: a population-based study of 430 first-pregnancy Spira, A. and Multigner, L. (1998) The effect of industrial and agricultural
planners. Lancet, 352, 1172–1177. pollution on human spermatogenesis. Hum. Reprod., 13, 2041–2042
Bonde, J.P., Hjollund, N.H., Kolstad K.A. et al. (1999) Environmental semen Sundaram, K. and Witorsch, R. (1995) Toxic effects on the testes. In Witorsch,
studies: is infertility increased by a decline in sperm count? Scand. J. Work. R. (ed.), Reproductive Toxicology. Raven Press, New York, pp. 99–122,
Environ. Health, 25 (Suppl. 1), 76–78. Thonneau, P. and Spira, A. (1991) Prevalence of infertility: international data
Carlsen, E., Giwercman, A., Keiding, N. and Skakkebaek, N. (1992) Evidence and problems of measurement. Eur. J. Obstet. Gynecol. Reprod. Biol., 38,
for decreasing quality of semen during the past 50 years. Br. Med. J., 305, 43–52.
609–613. Thonneau, P., Bujan, L., Multigner, L. and Mieusset, R. (1998) Occupational
Cheek, A.O. and McLachlan, J.A. (1998) Environmental hormones and the heat exposure and male fertility: a review. Hum. Reprod., 13,
male reproductive system. J. Androl., 19, 5–10. 2122–2125.
Cohn, W., Boylan, J.J., Blanke, R.V. et al. (1978) Treatment of chlordecone Tielemans, E., Heederik. D., Burdorf, A. et al. (1999a) Assessment of
(Kepone) toxicity with cholestyramine. Results of a controlled clinical trial. occupational exposures in a general population: comparison of different
N. Engl. J. Med., 298, 243–248. methods. Occup. Environ. Med., 56, 145–151.

1775


Tielemans, E., Burdorf, A., de Velde, E.R. et al (1999b) Occupationally Whorton, M.D., Stubbs, H.A., Obrinsky, A. et al. (1981) Testicular function
related exposures and reduced semen quality: a case-control study. Fertil. of men occupationally exposed to para-tertiary butyl benzoic acid. Scand.
Steril., 71, 690–696. J. Work Environ. Health, 7, 204–213.
Toppari, J., Larsen, J.C., Cristiansen, P. et al. (1996) Male reproductive health. World Health Organization (1992) WHO Laboratory Manual for the
Environmental chemicals with estrogenic effects. Env. Health Perspect., Examination of Human Semen and Semen-Cervical Mucus Interaction.
104, 741–803. Cambridge University Press, Cambridge.
Veulemans, H., Steeno, O., Masschelein, R. and Groeseneken, D. (1993) World Health Organization (1993) WHO Manual for the Standardised
Exposure to ethylene glycol ethers and spermatogenic disorders in man: a Investigation and Diagnosis of the Infertile Couple. Cambridge University
case-control study. Br. J. Ind. Med., 50, 71–78. Press, Cambridge.
Welch, L.S., Schrader, S.M., Turner, T.W. and Cullen, M.R. (1988) Effects of Wyrobek, A.J., Watchmaker, G., Gordon, L. et al. (1981) Sperm shape
exposure to ethylene glycol ethers on shipyard painters: II. Male abnormalities in carbaryl-exposed employees. Environ. Health Perspect.,
reproduction. Am. J. Ind. Med., 14, 509–526. 40, 255–265.
Whorton, D., Krauss, R.M., Marshall, S. and Milby, T.H. (1977) Infertility in
male pesticide workers. Lancet, ii, 1259–1261. Received on December 1, 2000; accepted on March 14, 2001

1776

Contribution Of Environmental Factors To The Risk Of Male Infertility

More Related Content

Viewers also liked (20)

Similar to Contribution Of Environmental Factors To The Risk Of Male Infertility (20)

More from Biblioteca Virtual (20)

Contribution Of Environmental Factors To The Risk Of Male Infertility