Are endocrine-disrupting chemicals linked to endometriosis?

July 18, 2019
Mary Beth Nierengarten
Mary Beth Nierengarten

A recently published Chinese study is the first of its kind to summarize all evidence on endometrial risk associated with exposure to EDCs.

A pooled analysis of 30 studies by Chinese investigators shows a link between exposure to endocrine-disturbing chemicals (EDCs) and incidence of endometriosis. The findings are from a meta-analysis published in Gynecologic Endocrinology, which the authors said is the first to summarize evidence on endometrial risk associated with exposure to EDCs while they acknowledged the limitations of the underlying data.  

For the analysis, researchers from Zhongnan Hospital of Wuhan University reviewed and synthesized data from 20 English-language papers published up to January 2018 that documented results of 30 independent studies. Of the studies, 21 were case-control, eight cohort, and one cross-sectional. Bisphenol A (BPA) exposure was tested in four studies, polychlorinated biphenyls (PCB) in 12 studies, organochlorine pesticides (OCP) in eight studies, and phthalate esters (PAE) in six studies. 

Overall risk of endometriosis across all EDC exposures in the meta-analysis was 1.41 (95% CI, 1.23-1.6). Consistent increases in risk of endometriosis were associated with exposure to PCBs (odds ratio [OR] 1.58; 95% CI, 1.18-2.12), OCPs (OR, 1.40; 95% CI, 1.02-1.92), and PAEs (OR of 1.27; 95% CI, 1.00-1.60). No significant risk of endometriosis was found with exposure to BPAs (OR of 1.40; 95% CI, 0.94-2.08). 

According to the authors, lack of risk associated with BPAs may be due to the small size of the four studies that looked at BPA exposure and endometrial risk. As such, they urged caution in interpreting this result and emphasized the need for “well-designed epidemiology studies” to confirm the association of BPA exposure with endometrial risk.

Some other limitations of the meta-analysis as a whole included the small number of studies; lack of control for co-exposure to an EDC not focused on in the study, which may have obscured a true correlation between one specific EDC and endometrial risk; collection of a single spot serum or urine specimen to assess EDC exposure in most studies; and inclusion in control groups of most of the studies of subfertile women rather than women from the general population.

Given these limitations, the authors acknowledged that further research is needed, including multicenter clinical studies and longitudinal investigations composed of adequate sample sizes and occupational exposure; research on mechanisms of exposure dose, duration, and phase of EDCs; studies that control co-exposure of EDCs to improve credibility of the epidemiological evidence; and use of well-defined biomarkers of and detection of EDCs in research. 

That said, the authors emphasized that their meta-analysis is the first to include all published epidemiological evidence fulfilling inclusion criteria examining the link between exposure to the four types of EDCs and endometriosis risk.