O-244 Androgen levels and their predictive value in women with different ovarian responses. (30th June 2022)

• Record Type: Journal Article
• Authors: Weber, M; Capp, E; Strowitzki, T; Rehnitz, J
• Is Part Of: Human reproduction. Volume 37(2022)Supplement 1
• Publication Date: 2022-06-30
• Subjects: Human reproduction -- Periodicals; 618
• DOI: 10.1093/humrep/deac106.026
• Languages: English
• ISSNs: 0268-1161

Abstract

Abstract: Study question: Do androgen levels impact oocyte yield during ovarian stimulation and how predictive are they for distinct ovarian response? Summary answer: Testosterone positively correlates with the number of oocytes retrieved in all patients. DHEA-S appears more specific by only correlating significantly in poor responding women. What is known already: Androgens are involved in the initiation of primordial follicle recruitment and early development. Testosterone and DHEA-S increase the number of growing follicles by reducing follicular atresia. A reduction in androgen levels can be observed with diminished ovarian reserve and advanced age. To standardize the definition of POR Bologna criteria have been elaborated including age, laboratory or sonographic ovarian reserve and low ovarian response in previous stimulations. The mentioned markers have been reviewed concerning its predictive power for POR by the POSEIDON analysis and divided into four subgroups. To date androgen levels are not raised to stratify POR in these classifications. Study design, size, duration: All women were recruited in the division for reproductive medicine at the University Hospital Heidelberg. The patients' files were retrospectively analyzed for androgen levels before ovarian stimulation and outcome. A classification between poor and normal response was performed according to Bologna criteria. Participants/materials, setting, methods: All patients provided written informed consent. Blood samples were analyzed from serum drawn on day 2 to 5 of the menstrual cycle before ovarian stimulation. Statistical analysis was performed with SPSS; statistical significance was set to p < = 0.05 . Main results and the role of chance: Valid data for analysis was available in 335 cycles, 238 were scaled in normal (NOR), 97 in poor ovarian response (POR). Statistics in both subgroups were performed concerning age, AMH, count of retrieved oocytes, DHEA-S and testosterone with significant lower AMH as well as androgen levels, oocyte yields and a higher median age in the POR group. An expected significant positive correlation between AMH and retrieved oocytes was confirmed in both subgroups ( p =.000 and .004). Increasing age negatively correlates with the number of retrieved oocytes after ovarian stimulation in NOR ( p =.003), but not in POR (p=.441) patients suggesting age as less powerful predictive marker for oocyte yield in those patients. In addition, a positive correlation for testosterone and retrieved oocytes merely reached significance in all analyzed cycles independently of different ovarian responses. On the contrary DHEA-S levels and oocyte counts cohered with a marginal significance (p=.05) only in POR patient cycles. Serum levels for DHEA-S positively correlated with the variable "oocyte count ≥ 3" in the POR group (OR 1.006, 95% CI, 0.999- 1.014 Limitations, reasons for caution: The results are based on a limited sample size. Therefore, results need to be proven in further studies including further aspects affecting ovarian response. Wider implications of the findings: The significant correlation between pre-stimulatory DHEA-S levels in POR for oocyte yield on the contrary to NOR, supports the experimental supplementation with DHEA in this subgroup, aiming to improve outcomes during ovarian stimulation. Testosterone is less specific for estimating ovarian response in both subgroups. Trial registration number: n.A.