The duration of pre-ovulatory serum progesterone elevation before hCG administration affects the outcome of IVF/ICSI cycles.

STUDY QUESTION During controlled ovarian stimulation (COS), does the duration of premature serum progesterone (P) elevation before administration of hCG affect the outcomes of IVF/ICSI embryo transfer (-ET) cycles? SUMMARY ANSWER The duration of the premature serum P elevation is inversely related to the clinical pregnancy rate of IVF/ICSI-ET cycles. WHAT IS KNOWN AND WHAT THIS PAPER ADDS The majority of the previous studies only considered a single serum P measurement made on the day of hCG administration and the results of attempts to relate this to IVF/ICSI-ET outcomes were controversial. However, the effect of the duration of premature serum P elevation before the hCG administration on the outcomes of IVF/ICSI-ET cycles has not been studied well. Here we demonstrate that the duration of premature serum P elevation has a more significant inverse correlation than the absolute serum P concentration on the day of hCG administration with IVF/ICSI-ET outcomes. DESIGN It is a retrospective, single-centre cohort study. A total of 1784 IVF and/or ICSI-ET cycles were included from October 2005 to June 2011. PARTICIPANTS AND SETTING A total of 1784 patients underwent their IVF and/or ICSI-ET cycles in a university hospital IVF unit. The inclusion criteria include (i) age between 20 and 42 years and (ii) eligible indications for COS before IVF/ICSI. MAIN RESULTS AND THE ROLE OF CHANCE The duration of premature serum P elevation to >1 ng/ml is significantly inversely associated with the probability of clinical pregnancy (odds ratio = 0.773, 95% confidence interval: 0.660-0.891, P < 0.001), after adjustment for possible confounders with multivariate logistic regression analysis. However, the significance of inverse correlation between the absolute serum P concentration on the day of hCG administration with clinical pregnancy rate decreased after adjustment. BIAS, CONFOUNDING AND OTHER REASONS FOR CAUTION The cutoff value we chose to define premature serum P elevation (P > 1.0 ng/ml) might not be able to be applied to different immunoassay kits and study population. The retrospective nature of this study inevitably might be influenced by some selection bias. GENERALIZABILITY TO OTHER POPULATIONS Older patients (>42 years) are excluded from our study.