Abstract Study question Does the endometrial preparation approach for an euploid-frozen embryo transfer have an impact on the incidence of an ectopic pregnancy? Summary answer The natural cycle endometrial preparation approach seems to have a lower incidence of ectopic pregnancies compared to a hormonal replacement therapy cycle. What is known already Ectopic pregnancy (EP) is an uncommon complication of pregnancy and is associated with significant morbidity. IVF pregnancies have been implicated as a risk factor, but this finding is mostly confounded by multiple embryo transfer and complicated medical history of women undergoing IVF treatment (tubal disease, surgery, endometriosis, etc.). Recent studies have suggested that frozen embryo transfer may reduce the risk of EP, but it is not yet known whether embryo ploidy status and different endometrial preparation regimens affect the risk of EP. Study design, size, duration Retrospective cohort of 1517 pregnancies from 2607 single euploid frozen embryo transfers. Comparator estimates were estimated using multiple large cohort studies in the literature including 58,203,239 pregnancies from unselected cohorts and 220,167 pregnancies from IVF-treated women with the transfer of non-tested embryos. Participants/materials, setting, methods The rate of ectopic per pregnancies were compared to ectopic rate derived from multiple cohorts of IVF-treated women and population level reports of unselected pregnancies (natural conception and IVF). Comparator estimates were obtained with proportion meta-analyses using generalized linear mixed models. Factors associated with ectopic pregnancies were investigated with mixed-effects logistic regression analyses. Main results and the role of chance EP rate in our cohort was 10/1000 pregnancies (95% CI: 5 to 16 per 1000) resulting from single euploid embryo transfers (1.1%, 16/1517). The observed prevalence of EP was 12/1000 pregnancies (95% CI: 6.8 to 21 per 1000) in unselected cohorts and 14/1000 pregnancies (95% CI: 13 to 15 per 1000) after IVF-treatment. Ectopic rate in our study was not significantly different than rates reported in the literature from unselected and IVF pregnancies (P = 0.579 and 0.223, respectively). There were no significant differences between women with ectopic pregnancies and those with clinical pregnancies in terms of age (33.6±5.5 vs. 35.1±5.2 years, P = 0.290), BMI (26.7±4.8 vs. 25.8±4.6 kg/m2, P = 0.475) and serum AMH levels (3.3±2.9 vs. 3.4±3.9 ng/mL, P = 0.854), nor between the quality of embryo transferred between top, good, fair and poor (1.8% vs 0.7% vs 1.9% vs 0.9%, p = 0.088, 0.997 and 0.508 respectively). Ectopic rates were also similar between women with history of cesarean section (1.1% vs 1.0%, P = 0.847) or isthmocele (1.0 vs 1.1%, P = 0.857). There were less ectopic pregnancies following transfers with natural endometrial preparation (0.5 vs 1.5%, natural vs programmed) and the effect showed statistical significance in multivariable analyses (OR: 0.28, 95% CI: 0.06 – 0.88, P = 0.049). Limitations, reasons for caution Small sample size as well as infrequent occurrence of the studied event reduces the power of our study to detect small but clinically meaningful effects with statistical significance. Wider implications of the findings We did not observe a lower than reported prevalence of EP among euploid embryo transfer. However, rate of EP was significantly lower in natural endometrial preparation cycles compared to preparation with exogenous hormones, indicating that a natural cycle might have a protective effect against the development of an ectopic pregnancy. Trial registration number Not applicable
Read full abstract