Acrylonitrile

Epidemiological data

Administrative data

Endpoint:

epidemiological data

Type of information:

other: published review

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Published, peer-reviewed review article

Data source

Reference

Reference Type:

review article or handbook

Title:

Unnamed

Year:

2009

Materials and methods

Study type:

other: review of literature data

Endpoint addressed:

toxicity to reproduction / fertility

developmental toxicity / teratogenicity

Test guideline

Qualifier:

no guideline available

Principles of method if other than guideline:

The authors review the studies reporting developmental and reproductive effects associated with acrylonitrile exposure in humans.

GLP compliance:

no

Remarks:

: not relevant

Test material

Test material information

Constituent 1

Reference substance name:

Acrylonitrile

EC Number:

203-466-5

EC Name:

Acrylonitrile

Cas Number:

107-13-1

Molecular formula:

C3H3N

IUPAC Name:

prop-2-enenitrile

Method

Type of population:

occupational

Ethical approval:

not specified

Details on study design:

Various studies are critically reviewed.

Exposure assessment:

measured

Details on exposure:

Exposure levels are generally absent, not well documented or are baseed on area rather than personal measurements.

Results and discussion

Results:

Some Chinese epidemiological studies reported increased rates of birth defects, stillbirths, premature delivery, infertility, and spontaneous abortions; these findings may be artifacts of differential ascertainment, reporting bias, confounding exposures, or other factors. The epidemiological
studies are not sufficiently robust, free of potential confounders, or adequately documented for exposure to be an appropriate basis for risk assessment..

Confounding factors:

The study of Ivanescu et al (1990) does not record potential exposures to other chemicals.

Strengths and weaknesses:

The review of four Chinese epidemiological studies by Collins et al (2003) notes a number of serious limitations, including very limited measurements of AN exposure that could be used to estimate individual exposures; reliance on area measurements rather than personal exposures; lack of information assuring that exposure occurred before health outcome; possible use of inappropriate comparison groups; unmeasured potential confounding exposures such as excessive heat, heavy labor, prolonged standing, or exposures to other chemicals; possibility of recall and selection biases; absence of medical record confirmation of adverse outcomes; lack of criteria for use of multiple reproductive outcomes from one couple; and, for some studies, lack of consideration of prior adverse reproductive outcomes or births; reliance on proxy reporting in some studies; lack of evaluation of other potential reproductive risks to parents in most of the studies; and a lack of personal exposure data over a period of time.


The study of Czeizel et al (2004) is considered to have major limitations. Some pregnancy outcomes preceded possible exposure; the comparison group from same factory may have had incidental exposure; incorrect exposure classification for analyses of congenital abnormalities (exposed and unexposed workers grouped together); exposure to other chemicals possibly related to birth defects; and lack of recent exposure data.

The study of Czeizel et al (1999) is also considered too limited to be reliable because of both the absence of personal exposure data over time, and the use of proximity to a factory as a surrogate for level or duration of exposure.

The study of IIvanescu et al (1990) does not record exposure levels or potential exposures to other chemicals.

Any other information on results incl. tables

The Chinese studies report findings (which could reflect either male- or female-mediated toxicity) reported in spouses of workers potentially exposed to AN, including relatively consistent increased risks of stillbirths, spontaneous abortions, and birth defects (not further described). These studies also describe a suggestive increased risk of sterility in wives of male workers.

Czeizel et al (2004) found no association of AN exposure with birth defects in the children of male or female workers. The same investigators concluded that there was no effect on birth outcomes of AN exposure for people living near to the AN factory (Czeizel et al, 1999).

Ivanescu et al (1990) report decreased testosterone concentrations in male workers in a chemical factory with exposure to AN at levels not measured and where there was also likely exposure to many other chemicals and agents

Applicant's summary and conclusion

Conclusions:

Some Chinese epidemiological studies reported increased rates of birth defects, stillbirths, premature delivery, infertility, and spontaneous abortions; these findings may be artifacts of differential ascertainment, reporting bias, confounding exposures, or other factors. The epidemiological
studies are not sufficiently robust, free of potential confounders, or adequately documented for exposure to be an appropriate basis for risk assessment.

Executive summary:

Summary of Chinese epidemiology studies

Collins et al (2003) reviewed findings in four Chinese epidemiological studies of AN workers (including male workers, female workers, and/or spouses of potentially exposed workers) that included reproductive and developmental endpoints. Findings reported in female workers potentially exposed to AN in Chinese factories include relatively consistent increased relative risks (i.e., ≥1.5 and lower bound confidence interval [C.I.] ≥1.0) in two or more studies of total birth defects (not further defined), stillbirths, and premature delivery. These studies also describe a suggestive increased relative risk (i.e., relative risk ≥1.5) in at least one study of pregnancy complications (not further defined), sterility, infant mortality, post-term delivery, and maternal mortality. Findings (which could reflect either male- or female-mediated toxicity) reported in spouses of workers potentially exposed to AN in Chinese factories include relatively consistent increased risks of stillbirths, spontaneous abortions, and birth defects (not further described). These studies also describe a suggestive increased risk of sterility in wives of male workers. These findings, while somewhat consistent, could result from differential ascertainment, reporting bias, and confounding exposures. Study limitations and inadequate description of these studies make it difficult to use them for either causal assessment or as the basis for risk assessment.  Based on the Collins et al (2003) review of the Chinese epidemiological studies, it appears that there are sufficient data to hypothesize that exposure to AN may be associated with several adverse reproductive outcomes, but that the evidence is not sufficient to establish causation, or to draw conclusions regarding no-effect or threshold exposure levels for the reported findings.

One study of Hungarian factory workers (Czeizel et al, 2004) found no association of AN exposure with birth defects in the children of male or female workers. However this study is considered to have major limitations. The same investigators had previously looked at birth outcomes in the proximity of an AN manufacturing plant and concluded that there was no effect of AN exposure for people living near to the AN factory (Czeizel et al, 1999). This study is also considered too limited to be reliable.

One group of investigators reported decreased testosterone concentrations in male workers in a chemical factory with exposure to AN and many other chemicals and agents (Ivanescu et al, 1990). Exposure levels to AN or other potential adverse exposures were not documented.  The statistical tests were not well documented and appeared to be selectively employed. Conclusions in this report do not appear to be supported by the data presented.

A recent Chinese occupational health study evaluated sperm quality in AN-exposed workers but did not assess reproductive outcomes (Xu et al, 2003). Decreased sperm counts were reported in semen samples from AN workers. Exposures to AN were not well documented in this study and other potential adverse exposures were not presented. The sperm counts among AN workers were reported to be slightly lower than the comparison group, but were within the normal range for a population of 25- to 30-year old men in another study (Martin et al., 2003). There were no differences in sperm viability, motility, or morphology between the exposed and comparison group. Single-cell gel electrophoresis (COMET) and fluorescence in situ hybridization (FISH) assays were reported to show significant increases in sperm with evidence of DNA damage and/or aneuploidy in the AN workers compared to the comparison group. However, the statistical analyses of the COMET and FISH assays were not appropriate, and the conclusion that there were adverse effects from AN exposure is not considered reliable.  Because of the limited evidence for effects on conventional measures of sperm quality and concerns regarding the COMET and FISH data analyses, it is considered that the Xu et al. (2003) study does not provide convincing evidence of adverse effects on sperm from worker exposure to AN. Unless decreased sperm counts are replicated in another study of AN workers, it is difficult to give much weight to this study.