STUDY QUESTION

Is there an association of progesterone (P) on the day of trigger with live birth in autologous ART transfer cycles on day 5 versus day 6?

SUMMARY ANSWER

P had a greater negative effect on live birth in day 6 fresh transfers compared to day 5 fresh transfers.

WHAT IS KNOWN ALREADY

Premature P elevation is associated with lower live birth rates in fresh autologous ART cycles, likely due to worsened endometrial-embryo asynchrony. Few studies have evaluated whether the effect of an elevated P on the day of trigger is different on live birth rates with a day 5 compared to a day 6 embryo transfer.

STUDY DESIGN SIZE, DURATION: This was a retrospective cohort study with autologous IVF cycles with fresh embryo transfers on day 5 and day 6 from 2011 to 2014. A total of 4120 day 5 and 230 day 6 fresh autologous embryo transfers were included. The primary outcome was live birth, defined as a live born baby at 24 weeks gestation or later.

PARTICIPANTS/MATERIALS, SETTING, METHODS: Patients from a large private ART practice were included. Analysis was performed with generalized estimating equations (GEE) modeling and receiver operating characteristic (ROC) curves.

MAIN RESULTS AND THE ROLE OF CHANCE

Day 6 transfers were less likely to have good quality embryos (73% versus 83%, P < 0.001) but the cohorts had similar rates of blastocyst stage transfer (92% versus 91%, P = 0.92). Live birth was less likely in fresh day 6 versus day 5 embryo transfers (34% versus 46%, P = 0.01) even when controlling for embryo confounders. In adjusted GEE models, the effect of P as a continuous variable on live birth was more pronounced on day 6 (P < 0.001). Similarly, the effect of P > 1.5 ng/ml on day of trigger was more pronounced on day 6 than day 5 (P < 0.001). Day 6 live birth rates were 8% lower than day 5 when P was in the normal range (P = 0.04), but became 17% lower when P was > 1.5 ng/ml (P < 0.01). ROC curves for P predicting live birth demonstrated a greater AUC in day 6 transfers (AUC 0.59, 95% CI 0.51-0.66) than day 5 (AUC 0.54, 95% CI 0.52-0.55). Interaction testing of P × day of embryo transfer was highly significant (P < 0.001), further suggesting that the effect of P was more pronounced on day 6 embryo transfer. In fresh oocyte retrieval cycles with elevated P, a subsequent 760 frozen-thaw transfers did not demonstrate a difference between embryos that were frozen after blastulation on day 5 versus 6.

LIMITATIONS REASONS FOR CAUTION

Limitations include the retrospective design and the inability to control for certain confounding variables, such as thaw survival rates between day 5 and day 6 blastocysts. Also, the data set lacks the known ploidy status of the embryos and the progesterone assay is not currently optimized to discriminate between patients with a P of 1.5 versus 1.8 ng/ml.

WIDER IMPLICATIONS OF THE FINDINGS

This study suggests further endometrial-embryo asynchrony when a slow growing embryo is combined with an advanced endometrium, ultimately leading to decreased live births. This suggests that premature elevated P may be a factor in the lower live birth rates in day 6 fresh embryo transfers. Further studies are needed to evaluate if a frozen embryo transfer cycle can ameliorate the effect of elevated P on the day of trigger among these slower growing embryos that reach blastocyst staging on day 6.

STUDY FUNDING/COMPETING INTERESTS: No external funding was received for this study. There are no conflicts of interest to declare.

TRIAL REGISTRATION NUMBER

Not applicable.