Ovarian response is associated with anogenital distance in patients undergoing controlled ovarian stimulation for IVF

STUDY QUESTION: Is the length of the anogenital distance (AGD) a biomarker of ovarian reserve and response to controlled ovarian stimulation (COS)? SUMMARY ANSWER: Shorter AGD is associated with presence of poor ovarian response. WHAT IS KNOWN ALREADY: Organ development during prenatal life is influenced by the prevailing intrauterine environment, and it has been suggested that nutritional, environmental and toxic factors could affect ovarian reserve set prenatally. AGD is a biomarker of prenatal-hormonal environment and observational studies have shown an association between its length and reproductive parameters in both sexes. STUDY DESIGN, SIZE, DURATION: This was a prospective cohort study of 437 women treated with IVF/ICSI conducted in a tertiary-care university hospital between January and December 2016. PARTICIPANTS/MATERIALS, SETTING, METHODS: All women underwent their first COS for IVF/ICSI and reached criteria for oocyte retrieval. Based on the number of oocytes obtained, patients were divided into three groups: poor responders (<= 3 oocytes) (n=50), normoresponders (4-15 oocytes) (n=332) and high responders (> 15 oocytes) (n=55). Before retrieval, the following patient data were recorded: age, body mass index (BMI), ovarian reserve markers (anti-Mullerian hormone [AMH], antral follicle count [AFC] and follicular stimulation hormone [FSH]), cause of infertility, total doses of gonadotropins used and ovarian sensitivity index (OSI). Patients with previous pregnancies, polycystic ovary syndrome (PCOS), endometriosis and previous ovarian or genital surgery were excluded. Anthropometric biomarkers of AGD(AC) (anus-clitoris) and AGD(AF) (anus-fourchette) were measured in all patients under sedation on the day of retrieval and before proceeding to oocyte pick-up. Multiple linear regression analyses were used to examine the association between both AGD and ovarian reserve markers, the total units of gonadotropins used, the number of oocytes obtained and the OSI. Logistic regression was used to predict poor response in COS for IVF/ICSI, while accounting for confounders such as age and BMI. MAIN RESULTS AND THE ROLE OF CHANCE: Baseline FSH, AMH, AFC and age were significantly different among the three groups of ovarian response, as were the units of gonadotropin used, and the ovarian sensitivity index (OSI) (P<0.001). Both AGD(AC) and AGD(AF) measurements were positively correlated with AMH levels (r=0.38 and r=0.21; P<0.05), AFC (r=0.41 and r=0.20; P<0.05), the OSI (r=0.24 and r=0.19; P<0.05) and the number of oocytes retrieved (r=0.29 and r=0.28, respectively; P<0.05). Conversely, there was a negative correlation between both AGD measurements and the doses of gonadotropins used (r=-0.19 and r=-0.15; P<0.05). The area under the curve (AUC) for prediction of poor response of AGD(AC) was 0.70 (95% CI 0.66, 0.75), which was comparable to the classic ovarian reserve markers, such as AFC and AMH. AGD(AF) showed a significantly worse predictive capacity for poor ovarian response (AUC 0.60 [95% CI 0.55, 0.60]) than AMH and AFC. LIMITATIONS, REASONS FOR CAUTION: The population used for the study was a highly selected group of infertile women who underwent COS for IVF, so the findings of this research may not be applicable for general population. Besides, measurement or selection biases might have been possible and must be considered. WIDER IMPLICATIONS OF THE FINDINGS: The findings of this study suggest that in utero exposure to certain hormonal environments could affect the ovarian reserve set prenatally. STUDY FUNDING/COMPETING INTEREST(S): None. The authors have no competing interests to declare.