Customize your JAMA Network experience by selecting one or more topics from the list below.
Khattak S, K-Moghtader G, McMartin K, Barrera M, Kennedy D, Koren G. Pregnancy Outcome Following Gestational Exposure to Organic Solvents: A Prospective Controlled Study. JAMA. 1999;281(12):1106–1109. doi:10.1001/jama.281.12.1106
Author Affiliations: The Motherisk Program, Division of Clinical Pharmacology and Toxicology (Drs Khattak, Kennedy, and Koren, Mr Moghtader, and Ms McMartin), and the Department of Psychology (Dr Barrera), The Hospital for Sick Children, and the University of Toronto, Toronto, Ontario.
Context Numerous women of childbearing age are exposed
occupationally to organic solvents. Previous retrospective studies have
reported conflicting results regarding teratogenic risk.
Objective To evaluate pregnancy and fetal outcome following
maternal occupational exposure to organic solvents.
Design A prospective, observational, controlled study.
Setting An antenatal counseling service in Toronto, Ontario.
Patients One hundred twenty-five pregnant women who were exposed
occupationally to organic solvents and seen during the first trimester
between 1987 and 1996. Each pregnant woman who was exposed to organic
solvents was matched to a pregnant woman who was exposed to a
nonteratogenic agent on age (±4 years), gravidity
(±1), and smoking and drinking status.
Main Outcome Measure Occurrence of major congenital malformations.
Results Significantly more major malformations occurred among
fetuses of women exposed to organic solvents than controls (13 vs 1;
relative risk, 13.0; 95% confidence interval, 1.8-99.5). Twelve
malformations occurred among the 75 women who had symptoms temporally
associated with their exposure, while none occurred among 43
asymptomatic exposed women (P<.001). (One malformation
occurred in a woman for whom such information was missing.) More of
these exposed women had previous miscarriage while working with organic
solvents than controls (54/117 [46.2%] vs 24/125 [19.2%];
P<.001). However, exposed women who had a previous
miscarriage had rates of major malformation that were similar to
exposed women who had no previous miscarriage.
Conclusions Occupational exposure to organic solvents during
pregnancy is associated with an increased risk of major fetal
malformations. This risk appears to be increased among women who report
symptoms associated with organic solvent exposure. Women's exposure to
organic solvents should be minimized during pregnancy. Symptomatic
exposure appears to predict higher fetal risk for
Many women of
childbearing age are occupationally exposed to organic solvents. The
most important women-dominated occupations with potential chemical
exposures are health care professions and work tasks in the clothing
and textile industries, all of which involve exposure to organic
Many industrial solvents are teratogenic in laboratory animals.
There are reports of limb and central nervous system defects in mice,
marked developmental toxic effects and retardation of skeletal growth
in rats, and congenital malformations in rabbits.3-10
However, the animal studies typically use high doses of single solvents
and a variety of routes of administration. In the occupational setting,
exposure usually occurs to a multitude of solvents at much lower doses
by inhalation, making extrapolation from animals to humans problematic.
A recent meta-analysis of studies in humans detected an
apparent increased risk of major malformations and a trend toward
increase in rates of miscarriage11 in women who
self-reported occupational exposures to solvents. However, all
available published experience is based on retrospective
studies.12-14 The present study is the first to
prospectively evaluate pregnancy and fetal outcome following maternal
occupational exposure to organic solvents.
The study group consisted of all pregnant women occupationally exposed
organic solvents and counseled between 1987 and
1996 by the Motherisk Program at the Hospital for Sick Children in
Toronto, Ontario. Each mother who was occupationally exposed to organic
solvents was paired with a pregnant woman who was exposed to a
nonteratogenic agent, attended the Motherisk clinic, and matched the
index woman on age (±4 years), gravidity (±1), and
smoking and drinking status. A nonteratogenic agent was defined as a
medicinal or environmental substance that has been proved not to
increase the baseline risk for major malformations or miscarriages.
Organic solvents to which women were occupationally exposed included
aliphatic and aromatic hydrocarbons, phenols, trichloroethylene,
xylene, vinyl chloride, acetone, and related compounds.
During the initial assessment (at or up to several weeks after
the point at which pregnancy was determined), we collected all
available data on exposure during pregnancy to medicinal and
recreational drugs, smoking, alcohol, lifestyle, medical and
nutritional status, and sexually transmitted diseases. Other
reproductive hazards were elucidated by taking a detailed medical,
genetic, and obstetric history. Although we recorded the medical
history of the father of the child and his use of drugs, most fathers
did not work with organic solvents and were not exposed to medications.
Details concerning the time of exposure to organic solvents were
recorded for determination of temporal relationship between exposure
and conception. The details on chemical exposure were recorded,
including occupation, chemicals involved, duration of exposure, type of
protective equipment used, and other safety features, including
ventilation fans. Adverse effects were defined as those known to be
caused by organic solvents (eg, irritation of the eyes or respiratory
system, breathing difficulty, headache). Temporal relationship to
exposure was investigated to separate these symptoms from those
associated with pregnancy.
The postnatal assessment occurred between 6 and 9 months after
the expected date of confinement. During this interview, the mother was
questioned about the course of her pregnancy subsequent to the first
meeting. This included verification of length of exposure to organic
solvents during pregnancy. Possible medical or obstetric complications
and details about the birth and the prenatal period were collected. All
major malformations were recorded and corroborated by a written report
from the physician caring for the child. The attainment of
developmental milestones was recorded with the use of the Denver Scale
from the maternal reports.
The following cases were excluded from the cohort of organic
solvent exposure: paternal exposure only; short-term maternal exposure
that did not occur in an occupational setting (eg, household painting);
women whose main task at work included heavy lifting, which might
increase the rate of miscarriages; cases in which during the first
interview it became apparent that the exposure to organic solvents had
occurred only before conception; any case in which during the first
interview (ie, during the first trimester) or the second interview
(postnatally) it became evident that the woman was exposed to known
teratogen(s) or neurotoxin(s) during the index pregnancy; and any
case in which the mother refused to give consent for participation in
our follow-up program. The time of conception was verified by
identification of the last menstrual period. When the time of the last
menstrual period was questionable, an ultrasonographic examination
was performed following the counseling session.
The primary outcome of interest was the rate of major
malformations. A major malformation was defined as any anomaly that has
an adverse effect on either the function or the social acceptability of
the child. The expected rate is 1% to 3%. Secondary outcomes of
interest were the rates of minor malformations, miscarriages or
therapeutic abortions, and premature births (<37 weeks' gestation);
birth weight and gestational age at delivery; and presence of fetal
distress or other neonatal complications. A minor malformation was
defined as a structural anomaly that does not pose any significant
health or social burden. Fetal distress was defined as the presence of
meconium and/or abnormal fetal heart rate monitoring during delivery or
the requirement of resuscitation or a neonatal intensive care unit.
Neonatal complications were defined as health complications that were
not structural in nature. This analysis was approved by the hospital's
research ethics board. The physician who counseled the woman, as well
the consultation letter sent to her physician, introduced the planned
follow-up program. At the time of follow-up, subjects were asked for
their consent for the follow-up interview. Rates of major malformations
in the study group were compared with those in the matched control
group by χ2 analysis. Relative risk was calculated with
95% confidence intervals. Secondary end points (ie, rates of
miscarriage, prematurity, and birth weight) were compared by
χ2 or Mann-Whitney rank sum test whenever appropriate.
All data are expressed as mean (SD).
Between 1986 and 1996, 256 women were seen in the Motherisk
Program because of occupational exposure to organic solvents. Of these,
42 women (16.4%) were not pregnant at the time of the study, 26
(10.2%) were lost to follow-up, 18 (7%) refused to consent to
participation in the study, and 45 (17.6%) were excluded from the
study, based on our exclusion criteria. The remaining 125 women were
matched to 125 control women. All exposed women worked with organic
solvents for at least the entire first trimester of pregnancy. The most
common occupations were factory worker (n=37),
laboratory technician (n=21), professional
artist/graphic designer (n=16), and printing industry
worker (n=14) (Table 1). The organic solvents most commonly
involved were aliphatic and aromatic hydrocarbons, phenols,
trichloroethylene, xylene, vinyl chloride, acetone, and related
The characteristics of subjects in the 2 groups are shown in
Table 2. There was a statistically
significant difference in the rates of previous miscarriage between
women in the exposed vs control groups. Subanalysis revealed that in
women, the previous miscarriages occurred during
occupational exposure to organic solvents. Previous miscarriage
occurred prior to such work in only 8 cases, yielding a significantly
higher rate of previous miscarriage among those working with organic
solvents (Table 2). As a result, women exposed to organic solvents had
a lower parity prior to the index pregnancy despite similar gravidity (Table 2). However, for the rest of the characteristics, there were no
differences between the 2 groups (Table 2). During the index pregnancy,
there were significant differences in the birth weight and rates of
fetal distress and neonatal complications (mainly eczema) (Table 3).
There were significantly more major malformations among the exposed
women compared with the control group (Table 3
). The relative risk of
major malformation among the exposed women was 13.0 (95% confidence
interval, 1.8-99.5). The major malformations in the study group are
detailed in Table 4. Rates of major
malformations did not differ between women who had a previous
miscarriage while working with organic solvents vs those who did not
have a previous miscarriage.
Among the 125 women occupationally exposed to organic solvents,
75 reported symptoms temporally associated with their exposure, 43 were
asymptomatic, and in 7 cases, such information was missing. Twelve of
the 13 major malformations occurred among the symptomatic women vs 0
among the 43 asymptomatic women (P<.001). In a further
subanalysis, women exposed to organic solvents were stratified
according to whether they were exposed to organic solvents for more
than 7 months or for 3 to 7 months. Sixteen women who were exposed for
more than 7 months had labor with fetal distress requiring
resuscitative measures vs only 1 among those with shorter exposures
(P = .002). Also, birth weights were lower among those with
longer exposure (mean [SD], 2975.2 [976.2] g vs 3431.4 [579.3] g;
P = .03). Gestational age was also lower, although differences
were not significant (mean [SD], 38.0 [7.41] weeks vs 40.0 [1.86]
weeks; P = .60).
There are controversial reports regarding fetal outcome following
prenatal exposure to organic solvents. Among them are increased rates
of miscarriage, central nervous system and cardiovascular
malformations, fetal solvent/gasoline syndrome, and perinatal
mortality; in addition, maternal fertility is
Fat-soluble organic solvents can
pass through biological membranes, including the
placenta.10 There is a paucity of information regarding the
impact of in utero exposure to organic solvents on the developing
brain. Animal studies have clearly shown that a variety of solvents
readily cross the placenta and that maternal inhalation of organic
solvents results in neurodevelopmental deficits in neonatal
Our recent meta-analysis has found that occupational exposure to
organic solvents is associated with increased risk of major
malformation.11 We have also shown a trend toward more
miscarriages, although it failed to reach statistical
significance.11 Yet, none of the eligible studies was
prospective. Recall bias may affect the accuracy of assessment of fetal
outcome in such studies. Moreover, the retrospective design of these
studies does not allow validation of crucial details
regarding the nature or extent of the exposure; or of
symptoms associated with the exposure. Also, most available
retrospective studies did not match patients for smoking, alcohol use,
and other potentially confounding reproductive risks.
The Motherisk protocol has allowed us to record in a systematic manner
all exposure data and other maternal and paternal medical details at
the time of exposure during the first trimester of pregnancy and to
follow up pregnancy outcomes prospectively in this cohort. The control
group was assessed in an identical manner.
This prospective study confirms the results of our recent
meta-analysis.11 Women exposed occupationally to organic
solvents had a 13-fold risk of major malformations as well increased
risk for miscarriages in previous pregnancies while working with
organic solvents. Moreover, women reporting symptoms associated with
organic solvents during early pregnancy had a significantly higher risk
of major malformations than those who were asymptomatic, suggesting a
dose-response relationship. Other factors (eg, type of solvent) might
have accounted for the presence of symptoms in some women.
Although some human teratogens have been shown to cause a
homogeneous pattern of malformation(s), in other cases no specific
syndrome has been described.19 No homogenous pattern of
malformations is obvious from the present study. However, organic
solvents, although traditionally clustered together, are a diverse
group of compounds that should not be expected to cause similar
patterns of reproductive toxic effects.
Although more prospective studies will be needed to confirm the present
results, it is prudent to minimize women's exposure to organic
solvents during pregnancy. Moreover, symptomatic exposure appears to
confer an unacceptable level of fetal exposure and should be avoided by
appropriate protection and ventilation. Health care professionals who
counsel families of reproductive age should inform their patients that
some types of employment may influence reproductive outcomes.
Create a personal account or sign in to: