Are systemic progesterone levels in true natural cycle euploid frozen embryo transfers with luteal phase support predictive for ongoing pregnancy rates?

Abstract

Are serum progesterone (P4) levels on the embryo transfer (ET) day predictive of ongoing pregnancy (OP) following a single euploid blastocyst transfer in a natural cycle (NC) when luteal phase support is routinely given? In single euploid frozen ETs in NC, P4 levels on ET day are not predictive for OP, when luteal phase support (LPS) is routinely added after the ET. In an NC frozen embryo transfer (FET), P4 produced by the corpus luteum initiates secretory transformation of the endometrium and maintains pregnancy after implantation. There are ongoing controversies on the existence of a P4 cutoff level on the ET day, being predictive for the chance of OP as well as of the possible role of additional LPS after ET. Previous studies in NC FET cycles, evaluating and identifying P4 cutoff levels did not exclude embryo aneuploidy as a possible reason for failure. This retrospective study analyzed single, euploid FET in NC, conducted in a tertiary referral IVF centre between September 2019 and June 2022, for which measurement of P4 on the day of ET and the treatment outcomes were available. Patients were only included once into the analysis. Outcome was defined as OP (ongoing clinical pregnancy with heartbeat, >12 weeks) or no-OP (not pregnant, biochemical pregnancy, early miscarriage). Patients with an ovulatory cycle and a single euploid blastocyst in an NC FET cycle were included. Cycles were monitored by ultrasound and repeated measurement of serum LH, estradiol, and P4. LH surge was identified when a rise of 180% above the previous level occurred and P4 levels of ≥1.0 ng/ml were regarded as confirmation of ovulation. The ET was scheduled on the fifth day after P4 rise and vaginal micronized P4 was started on the day of ET after P4 measurement. Of 266 patients included, 159 (59.8%) patients had an OP. There was no significant difference between the OP- and no-OP-groups for age, BMI, and day of embryo biopsy/cryopreservation (Day 5 versus Day 6). Furthermore, P4 levels were not different between the groups of patients with OP (P4: 14.8 ng/ml (IQR: 12.0-18.5 ng/ml)) versus no-OP (P4: 16.0 ng/ml (IQR: 11.6-18.9 ng/ml)) (P = 0.483), and no differences between both groups, when P4 levels were stratified into categories of P4 levels of >5 to ≤10, >10 to ≤15, >15 to ≤20, and >20 ng/ml (P = 0.341). However, both groups were significantly different for the embryo quality (EQ), defined by inner cell mass/trophectoderm, as well as when stratified into three EQ groups (good, fair, and poor) (P = 0.001 and 0.002, respectively). Stratified EQ groups remained the only significant parameter influencing OP in the uni- and multivariate analyses (P = 0.002 and P = 0.004, respectively), including age, BMI, and P4 levels (each in categories) and embryo cryopreservation day. Receiver operator characteristic curve for the prediction of an OP revealed an AUC of 0.648 when age, BMI and EQ groups were included into the model. The inclusion of P4 measurement on ET day into the model did not add any benefit for OP prediction (AUC = 0.665). The retrospective design is a limitation. Monitoring serum P4 levels can be abandoned in NC FET cycles with routine LPS as they do not seem to be predictive of live birth. No external funding was used for this study. The authors state that they do not have any conflicts of interest. N/A.