Plasma adipokines and endometriosis risk: a prospective nested case-control investigation from the Nurses' Health Study II

Abstract

study question: Do higher leptin levels and lower adiponectin levels predict subsequent development of endometriosis? summary answer: Plasma leptin and adiponectin levels were not associated with laparoscopically confirmed endometriosis when collected prior to disease diagnosis.what is known already: Case-control studies have identified altered levels of the inflammatory adipokines leptin and adiponectin in women with endometriosis, but it remains unclear whether inflammation results in endometriosis or whether the presence of endometriosis creates an inflammatory state. study design, size, duration: Nested, matched, case-control study within the prospective Nurses' Health Study II (NHS II) cohort. Blood samples were collected between 1996 and 1999 from 29 611 female nurses within the cohort. Women who reported endometriosis before blood collection were excluded.participants/materials, setting, methods: Plasma leptin and adiponectin levels were assayed by ELISA. Three hundred and fifty cases of laparoscopically confirmed endometriosis were matched 1:2 with 694 controls of comparable race, age, infertility history, menopausal status and time of blood draw. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated using unconditional logistic regression models adjusting for matching factors and BMI.main results and the role of chance: After adjusting for BMI, there were no statistically significant associations between endometriosis and leptin [RR ¼ 1.2; 95% CI ¼ 0.7 -2.0; P-value, test for linear trend (P trend ) ¼ 0.72], adiponectin (RR ¼ 0.8; 95% CI ¼ 0.5-1.2; P trend ¼ 0.48) or the leptin to adiponectin ratio (RR ¼ 0.8; 95% CI ¼ 0.4 -1.4; P trend ¼ 0.14) when comparing the upper with the lower quartile. Results were unaltered when analyses were stratified by BMI or restricted to cases diagnosed ≥4 years after blood draw. To evaluate statistical significance and limit the role of chance to the gold standard of 5%, we present 95% CIs about the RRs, and for P-values calculated for linear tests of trend and tests of heterogeneity, we have set the a-level to be 0.05 (i.e. ,0.05 is considered to be statistically significant).limitations and reasons for caution: A limitation of this study is the inability to differentiate the time of endometriosis 'diagnosis' from the time of disease 'onset' due to the impossibility in identifying a precise time point at which the disease process was first initiated at a molecular or cellular level. Additional limitations include lack of information regarding stage of endometriosis and the possibility of asymptomatic disease in the control population.wider implications of the findings: The mean age at diagnosis of endometriosis in the study population is 41.7, 10 years older than the mean age of diagnosis in the general population. While this may limit the generalizability of the results, there is no reason to suspect that the association between adipokines and endometriosis risk should differ at a younger age of diagnosis in an adult population.

The Formidable yet Unresolved Interplay between Endometriosis and Obesity

“…Increased levels of leptin have been identified in serum and peritoneal fluid specimens from women with endometriosis for more than two decades [ 55 , 58 ]. The role of leptin in the establishment and propagation of endometriotic lesions has also been shown in experimental models [ 59 ], but case-control analysis within the large cohort of NHS II failed to show statistically significant associations between endometriosis and leptin, even after adjustment for BMI (RR 1.2, 955 CI 0.7–2.0, p 0.72) [ 60 ]. More recently, Rathore et al showed that women with endometriosis had higher peritoneal fluid concentrations of leptin as compared to nonendometriosis peers ( p 0.040) [ 54 ].…”

“…Initial studies showed that adiponectin secretion was suppressed in the context of endometriosis, both in the peritoneal fluid [ 64 ] and serum [ 65 ]. Nevertheless, the large NHS II study did not reveal any significant associations between endometriosis and adiponectin (RR 0.8, 95% CI 0.5–1.2, p 0.48) or leptin-to-adiponectin ratio (RR 0.8, 95% CI 0.4–1.4, p 0.14) after adjustment for BMI [ 60 ]. Similarly, the study of Choi et al did not yield any difference in expression of adiponectin between ovarian endometrioma and normal endometrial tissue samples (31.8% vs. 42.9%, p 0.29) [ 61 ].…”

Update on Biomarkers for the Detection of Endometriosis

“…Contrasting evidence exists on prolactin, leptin, luteinizing hormone (LH), and adiponectin levels in endometriosis patients versus controls, showing either no difference, an increase (prolactin, leptin LH) or a decrease (adiponectin) [ 23 , 75 , 76 ]. No consensus exists on changes in steroid hormone levels [ 23 ].…”

Body size during adulthood, but not in childhood, associates with endometriosis, specifically in the peritoneal subtype—population‐based life‐course data from birth to late fertile age

“…Interestingly, however, body size in childhood and from early to late adulthood seems to be inversely associated with endometriosis 5,6 . Moreover, a body fat distribution characterized by a low waist‐hip ratio (WHR) has more often been observed in affected women, 6,7 which is also supported by genetic studies 8 . These results, however, have been obtained from multiple study populations and with no distinction between endometriosis subtypes.…”