Letter to the editor on a recent publication titled “Occupation and male infertility: glycol ethersSir, Recently a paper by Cherry et al. mentioned above was published in this journal (1). The paper reports the results of a CEFIC co-sponsored case-referent study on male infertility and occupational exposures (2). The study reported an association between male infertility and glycol ethers. We would like to describe several critical points in the analysis. These points not only pertain to the research question but also the a priori hypothesis and further information that was collected during this research project, but not reported in the paper. 1. A priori hypothesis. The paper mentions that the study was “designed primarily to test the one a priori hypothesis, that organic solvents and specifically solvent mixtures containing glycol ethers were associated with a reduced number of motile sperm”. However, the objectives of the project as set out in the final study proposal (2) were more extensive, as follows: “The specific objectives are to determine, for male patients newly attending fertility clinics in 10 UK cities: 1. Whether the distribution of occupations or inferred exposures to chemicals differ between infertile men (cases) and men attending the same clinics but with normal semen analysis (referents) 2. Whether concentrations of organic solvents and heavy metals in blood, urine and seminal plasma, differ between cases and referents. 3. The distribution of inferred, non-occupational exposures between cases and referents. 4. The distribution of cotinine in urine and seminal seminal plasma, declared alcohol intake, age, previous illness, medications and wearing of constrictive clothing between cases and referents.” From the description of these broader objectives it seems clear that the possible association with glycol ethers was just one of many hypotheses under investigation. 2. No overall association between any glycol ether exposure and male infertility. In the peer review paper it is concluded that there is an association between glycol ether exposure and male infertility. The data in table 4 of the article allow calculating the crude Odds Ratio (OR) between any glycol ether exposure and male infertility, by combining the 3 exposure categories. 653 cases and 949 referents were assessed not to be exposed and 221 cases and 949 controls were assed to be exposed to glycol ethers, giving an unadjusted OR of 1.09, which is not statistically significantly different from unity. 3. Extremely poor agreement between exposure assessors. The journal paper provides little information on the quality of the exposure assessments. However, the full project report contained a specific analysis of inter-assessor agreement (Table 40). Agreement between the two assessors was extremely poor, particularly for the high exposure category, on which the conclusion of an association is based. The high exposure group consisted of 33 subjects. 29 were rated high by rater 1 and 6 by rater 2. This implies that for only 2 subjects out of the 33 (6%) was there concordance. We believe this is a major weakness in the study and should have been reported in the journal paper. A more reliable approach would have been to only regard as exposed those subjects about whom the assessors were in agreement. 4. Validation of exposure assessment by means of biomonitoring. Despite being a part of the full project, the paper by Cherry et al only briefly mentions the biomonitoring work that was conducted. However, in the project report information is presented on this. Blood, urine and semen samples were collected from each man in the study. If a solvent-exposed individual, as judged by the industrial hygienists, was not expected having been exposed to solvents in the 24 hours prior to the collection, an additional clinic visit was arranged specifically choosing a clinic time that would ensure the subject had been exposed to solvents on the day of the second sample. For all subjects in the high, 56 out of the 129 men in the moderate and 55 of the 1507 men in the no exposure groups the urine samples were analyzed for glycol ether metabolites (ethoxy-, methoxy- and butoxy-acetic acid) focusing on those study subjects with the potentially highest exposure and compare these with referents. In none of the 200 samples (including all potentially highest exposed workers) methoxyacetic acid was detected, ethoxyacetic acid was detected in only 2 samples and butoxyacetic acid in 8 samples. Any glycol ether metabolite gave an OR of 0.19 with low fertility. The four Pearson correlations, for butoxy- and ethoxyacetic acid and two independent raters were all below 0.14 indicating very poor correlations between the metabolites in the biological samples and the ratings. We therefore seriously doubt the accuracy of the exposure rating assigned by the hygienists and hence the validity of the proposed relationship between glycol ether exposure and reduced sperm motility. 5. Incorrect time window of exposure. Although the questionnaire focused on the 24 months before semen sample collection, the associations between specific solvents and metals and male infertility was only based on the information on the 3-month period immediately prior to semen sample collection. The conclusion that a positive association between glycol ether exposure and male infertility exists, is therefore limited to this 3-month window. However all couples that had a period of less than 12 months attempting to conceive were excluded from the study. Infertility therefore already existed 9 months before the time period over which the exposure information was used. Information from the preceding 21 months was not used at all in the statistical analysis. Although several shortcomings were acknowledged in the sponsor’s report (2), these do not come across clearly in the paper as published. In particular, findings could have been interpreted quite differently depending on the weight given to either the biomonitoring or assessors’ data. Had the biomonitoring data (which include all highly exposed subjects) been taken as the gold standard, the study would rather have indicated that there was no association between the presence of glycol ethers metabolite or an inverse association with male infertility. The data warrant a more cautious interpretation. Gerard M.H. Swaen, PhD Senior Epidemiologist The Dow Chemical Company P.O. Box 444 4530 AK Terneuzen The Netherlands Peter J. Boogaard, PhD, PharmD, DABT, ERT Senior Toxicologist Shell Health Shell International bv P.O. Box 162 2501 AN The Hague The Netherlands Reference 1. Cherry N, Moore H, McNamee R, et al. Occupation and male infertility: Glycol ethers and other exposures. Occup Environ Med April 2008; doi:10.1136/oem.2007.035824. 2. The study details and final report can be found at the CEFIC-LRI website at: http://www.cefic-lri.org/index.php?mact=MCLibrary,cntnt01,details,0&cntnt01library_id=2&cntnt01template=&cntnt01group_by=pages%2C114&cntnt01origid=21&cntnt01item_id=1202820616&cntnt01returnid=21
This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of
View free sample issue >>
Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.