Unexpected poor responder versus unexpected normal responder in the subsequent stimulation cycle, a cohort study

OBJECTIVE: To explore who shows a better outcome in the next stimulation cycle, a previous poor ovarian response (POR) being unexpected or a previous normal ovarian response being unexpected? MATERIALS AND METHODS: This was a single-center retrospective cohort study performed in Peking University Third Hospital. We included women who underwent their first and second stimulation cycles from January 2018 to December 2022. The primary outcome was the presence of POR in the second stimulation cycle, defined as the oocyte collecting number lower than 4. All participants were systematically categorized into four distinct groups based on their ovarian response predictor, which included either anti-müllerian hormone (AMH) or antral follicle count (AFC), as well as their actual ovarian response during the initial stimulation cycle. (I) expected normal response, (II) unexpected normal response, (III) expected poor response, and, (IV) unexpected poor response. For our multivariable analysis, we pre-specified several covariates for adjustment, including female age, female BMI, and the presence of tubal factor or ovulatory dysfunction as factors related to infertility. Furthermore, sensitivity analyses were carried out in women of younger age (35 years or less). RESULTS: The analysis included 9,408 women who had completed their initial two consecutive stimulation cycles. Women with an unexpected poor response had higher AMH values and a higher AFC, and a lower oocyte collecting number than women with an unexpected normal response in the first stimulation cycle (2.0 vs 0.8 ng/ml, 9 vs 6, 2 vs 6, respectively). In the second stimulation cycle, oocyte collecting number was higher in women with unexpected poor response than in women with unexpected normal response (8 vs 6), with a poor response rate of 13 % and 24 % (adjusted OR (95 %CI): 0.52 (0.36-0.75)), respectively. The percentage of women without an available embryo for transfer was significantly lower in women with a previous unexpected poor response than in women with unexpected normal response (12 % vs 21 %). Sensitivity analysis conducted in women with young age (<=35 years) showed the results were stable. Women with unexpected poor response showed significantly higher rate of biochemical pregnancy, clinical pregnancy, and multiple pregnancy than women with unexpected normal response. There was no significant difference on miscarriage rate between the two groups. CONCLUSIONS: Women experiencing an unexpected POR in the first stimulation cycle showed a reduced risk of poor ovarian response in the second stimulation cycle when compared to those who had an unexpected normal response.