Unexpected estradiol decline during ovarian stimulation monitoring affects cumulative live birth

Haixiao Chen¹Kezhen Huang¹Caihui Ma²Jie Geng¹lanlan Liu^1,3Zhenfang Liu¹Xuhong Na¹Xiaoming Jiang²Jiali Cai^1,3*Jianzi Ren^1*

• ¹Xiamen University Affiliated Chenggong Hospital, Xiamen, China
• ²Chenggong Hospital, Xiamen University, Xiamen, China
• ³Xiamen University, Xiamen, China

Background: E2 is important in follicular development. During monitoring of stimulated cycles, serum levels of E2 are expected to increase steadily with follicle growth until final maturation. Unexpected E2 decline before triggering is reported in monitored COS cycles, yet its clinical significance remains controversial. Methods: The retrospective study was carried out in 27,487 conventional COS cycles at Xiamen University Affiliated Chenggong Hospital between January 2013 and December 2021. The occurrence of E2 decline during the monitoring was defined as the observation of a lower E2 value than the previous visit. Propensity matching and multivariate generalized linear models were used to analyze the association between E2 decline and cumulative live birth rates (CLBRs). Results: A total of 2,863 (10.3%) patients with E2 decline during COS monitoring were identified. In both unmatched and matched cohorts, the CLBRs were significantly decreased (unmatched cohort: 66.3% versus 55%, P<0.001, adjusted OR 0.83, 95% CI: 0.76,0.91; matched cohort: 59% versus 55%, P=0.003, adjusted OR 0.84, 95%CI: 0.75,0.94). The E2 decline also decreased the oocyte yield and embryo yield, but the live birth following fresh transfer was not affected after matching. Mediation analyses showed that the decrease in CLBR was primarily due to decreased embryo yield in both unmatched (76.5% mediated, P=0.002) and matched cohorts (72.5% mediated, P=0.01). Subgroup analyses suggested that increasing the gonadotropin (Gn) dose did not improve CLBR (adjusted OR 0.91, 95% CI: 0.71,1.16). However, the patients with two consecutive declines in two visits may have worse outcomes (adjusted OR 0.72, 95% CI: 0.56,0.94). Conclusions: Although E2 is frequently monitored during COS, the value of routine E2 monitoring during COS has already been questioned. Our data suggest that the decline in E2 during COS monitoring is associated with the CLBR following a complete cycle, indicating it remains a critical biomarker in predicting the outcomes during COS. However, the overall size of the association is modest, and further attention should be paid to specific subgroups of patients, such as patients with consecutive E2 decline.