Is biochemical pregnancy loss associated with embryo or endometrium? A retrospective cohort study in frozen single embryo transfer of own and donated oocytes.

ABSTRACT

STUDY QUESTION Does the use of preimplantation genetic testing for aneuploidies (PGT-A), personalized embryo transfer with endometrial receptivity assay (pET-ERA), or the use of donated oocytes modify the incidence of biochemical pregnancy loss (BPL) in frozen single embryo transfer (FSET)? SUMMARY ANSWER Following FSET, BPL incidence does not differ between own and donated oocytes, and the use of PGT-A with euploid embryo transfer or pET-ERA results in a similar incidence of BPL compared to cycles without embryo or endometrial analysis. WHAT IS KNOWN ALREADY BPL occurs frequently after IVF, and many factors have been associated with its incidence. The etiology of BPL is not well known, but the most probable cause seems to be either a low-quality embryo or impaired endometrial maintenance. The impact of techniques diagnosing embryonic ploidy or endometrial receptivity on BPL incidence and the BPL incidence between own and donated oocytes have not been analyzed. STUDY DESIGN, SIZE, DURATION This is a retrospective cohort study analyzing the incidence of BPL over 3741 cycles of FSET derived from own (2399 cycles) and donated (1342 cycles) oocytes between January 2013 and January 2022 in 1736 of which PGT-A, pET-ERA, or both were applied. PARTICIPANTS/MATERIALS, SETTING, METHODS: We defined BPL as a pregnancy diagnosed only by serum ß-hCG > 10 UI/l followed by a decrease that does not result in a clinical pregnancy. Clinical pregnancy was defined as the presence of gestational sac on transvaginal ultrasound. We compared BPL rates among patients undergoing 2399 FSETs from own oocytes, which comprised 1310 cycles of embryos analyzed by PGT-A, 950 cycles of untested embryos, 30 cycles of untested embryos with pET-ERA, and a subgroup of 109 cycles analyzed by both PGT-A and pET-ERA. We also included a total of 1342 FSET cycles from donated oocytes comprising 132, 1055, 140, and 15 cycles in the same groups, respectively. MAIN RESULTS AND THE ROLE OF CHANCE In FSET from own oocytes, the overall BPL rate per embryo transfer was 8.2% (95% CI [7.09-9.33]). In untested embryo transfers, the BPL rate was 7.5% [5.91-9.37]. In the PGT-A group, the BPL rate was 8.8% [7.32-10.47]. In the pET-ERA group, the rate was 6.7% [0.82-22.07]. In the PGT-A+ERA group, the rate was 6.5% [2.65-12.90]. No significant differences were found (P = 0.626). A multivariate analysis considering clinically meaningful variables that were significantly different among groups, taking the untested embryos/endometrium group as a reference, showed comparable incidences among groups. For PGT-A, the adjusted odds ratio (AdjOR) was 1.154 [0.768-1.735] (P = 0.49) and for PGT-A+ERA 0.885 [0.330-2.375] (P = 0.808). Because of a low number of registered cases in the pET-ERA group, and to prevent statistical errors and convergence issues, this group was excluded from further analysis. In FSET of donated oocytes, the overall BPL rate per embryo transfer was 4.9% [3.76-6.14]. In the PGT-A group, the BPL rate was 6.8% [3.16-12.55]. In the pET-ERA group, the rate was 5.0% [2.03-10.03]. In untested embryo transfers, the rate was 4.7% [3.46-6.10]. No cases occurred in the PGT-A+ERA group, and no significant differences were found (P = 0.578). The multivariate analysis showed comparable incidences among groups. For PGT-A the AdjOR was 1.669 [0.702-3.972] (P = 0.247) and for pET-ERA 1.189 [0.433-3.265] (P = 0.737). The PGT-A+ERA group was eliminated from the model to prevent statistical errors and convergence issues because no BPL cases were registered in this group. In the multivariate analysis, when the sources of oocytes were compared, own versus donated, no significant differences were found in the incidence of BPL. LIMITATIONS, REASONS FOR CAUTION This was a retrospective cohort study with potential biases. In addition, we were unable to control differences among groups due to modifications in medical or laboratory protocols during this long time period, which may modify the relationships being addressed. Factors previously associated with BPL, such as immunological conditions other than thyroid autoimmunity, were not considered in this study. Limited sample sizes of some groups may limit the statistical power for finding differences that can be present in the general population. WIDER IMPLICATIONS OF THE FINDINGS: BPL may be related to a mechanism not associated with the chromosomal constitution of the embryo or the transcriptome of the endometrium. More studies are needed to explore the factors associated with this reproductive issue. STUDY FUNDING/COMPETING INTEREST(S) No specific funding was available for this study. None of the authors have a conflict of interest to declare with regard to this study. TRIAL REGISTRATION NUMBER This trial was registered at clinicaltrials.gov (NCT04549909).