Association between markers of female adiposity and live birth among patients undergoing fertility treatment or attempting unassisted conception.

study questionBeyond BMI, are there better predictors of the impact of high female adiposity on reproductive outcomes in patients undergoing fertility treatment or attempting unassisted conception? SUMMARY ANSWER: Though BMI remains a predictor of fertility outcomes, alternative markers of adiposity, such as percent body fat, provide distinct information and may be more strongly associated with outcomes than BMI. WHAT IS KNOWN ALREADY: Elevated BMI is associated with a lower probability of live birth, though randomized trials have not consistently demonstrated the efficacy of weight loss for increasing live birth among patients utilizing infertility treatment. STUDY DESIGN, SIZE, DURATION: This was a secondary analysis of data gathered from 2013 to 2017 during the Folic Acid and Zinc Supplementation Trial (FAZST). Participants in FAZST included 2370 heterosexual couples seeking infertility care at four US fertility centers. Couples were followed for 9 months while undergoing fertility treatments or attempting unassisted conception, with up to 9 additional months of follow-up if pregnancy occurred. PARTICIPANTS/MATERIALS, SETTING,

methodsFor inclusion in the present study, female participants must have had at least one marker of adiposity measured at their baseline visit for FAZST. The primary exposure was high adiposity (defined by commonly used cutoffs in the literature) by each of five markers: BMI, percent body fat measured by dual-energy X-ray absorptiometry (DXA), serum leptin, serum adiponectin/leptin ratio, and waist circumference. Of the participants in FAZST, BMI was available for 99.6%, percent body fat for 7.3% (DXA only offered to 218 participants at two study sites between 2016 and 2017), leptin for 89.7%, adiponectin/leptin ratio for 89.7%, and waist circumference for 90.9%. Generalized linear models including age, race, parity, education, physical activity, male partner BMI ≥30 kg/m2, and Healthy Eating Index were used to estimate the relative risk of live birth. MAIN RESULTS AND THE ROLE OF CHANCE: High adiposity by BMI was associated with decreased probability of live birth (adjusted relative risk [aRR] 0.85, 95% CI 0.74-0.98). The other markers demonstrated similar associations, though a stronger effect size was seen with percent body fat (aRR 0.34, 95% CI 0.22-0.55). In an analysis by tertile, even moderately elevated percent body fat was associated with a decrease in live birth. When stratifying by infertility treatment status, associations were attenuated for most markers in the group utilizing infertility treatment, though percent body fat remained significantly associated with live birth. However, this marker was only available in a subset of participants. LIMITATIONS, REASONS FOR CAUTION: Only a subset of participants underwent DXA scans and had data on percent body fat, limiting the generalizability of the finding that this marker was most strongly associated with live birth. There were few participants with low BMIs, limiting the ability to draw conclusions on how low adiposity may affect reproductive outcomes. Findings may not be generalizable to the non-infertility population. WIDER IMPLICATIONS OF THE

findingsThe findings support prior data that high adiposity is associated with a lower probability of live birth. While most markers of adiposity performed similarly to BMI, there may be a role for percent body fat as an alternative assessment of adiposity, particularly among patients utilizing infertility treatment. STUDY FUNDING/COMPETING INTEREST(S): The FAZST and Impact of Diet, Exercise, and Lifestyle studies were supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland (contracts HHSN275201200007C, HHSN275201500001C, HHSN275201300026I/HHSN27500008, and HHSN275201300026I/HHSN27500018). There are no conflicts of interest to declare. TRIAL REGISTRATION NUMBER: NCT00467363 (secondary analysis).