LAUSR

BACKGROUND The availability and use of frozen embryos after ovarian hyperstimulation for assisted reproduction has increased with improvement in vitrification techniques and the rise of pre-implantation genetic testing. However, there is conflicting data regarding whether obstetric outcomes differ between fresh and frozen embryo transfer cycles. OBJECTIVE To compare placental pathology from live births arising from fresh and frozen embryo transfer cycles. STUDY DESIGN A cohort of 1140 live births with placental pathology arising from autologous in vitro fertilization cycles with fresh or frozen programmed transfer performed at MGH Fertility Center between 2004 and 2017 was retrospectively reviewed. An experienced placental pathologist categorized the reported placental pathology as anatomic, infectious, inflammatory, or vascular/thrombotic. Our primary outcomes were differences in these placental pathologies between the two groups. Patient demographic, cycle, and birth outcomes were compared with chi square tests, Student's t-test, or nonparametric tests, as appropriate. Multivariate logistic regression models were used to compare placental pathology between the fresh and frozen transfer groups. RESULTS Of the 1140 cycles included in our analysis, 929 arose from fresh embryo transfers (81.3%) and 211 arose from programmed frozen embryo transfers (18.5%). For both transfer types, the average age of the women at time of treatment was 35 years; mean BMIs were within the normal range (23.6 kg/m2 for fresh transfers and 23.2 kg/m2 for frozen transfers, p = 0.26); and mean day 3 FSH was 7.1 and 7.0 IU/L (p = 0.44), respectively. Deliveries occurred on average at 37.5 and 38.0 weeks gestational age (p = 0.037) in the fresh vs. frozen transfer group, with similar rates of obstetric complications. However, frozen transfers were more likely to be associated with marginal cord insertion [aOR 1.87 (CI 1.21, 2.91); p = 0.01], accessory lobe formation [aOR 2.96 (CI 1.12, 7.79); p = 0.03], subchorionic thrombi [aOR 3.72 (CI 1.8, 7.71); p < 0.001], and fetal vascular malperfusion (FVM) characteristics with cord anomalies [aOR 2.34 (CI 1.22, 4.46); p = 0.01]. These trends persisted when we analyzed day 5 transfers alone, and single frozen embryo transfers remained associated with increased rates of subchorionic thrombi compared to single fresh embryo transfers. CONCLUSIONS Pregnancies arising from frozen embryo transfers demonstrated more anatomic and vascular placental pathology than those from fresh transfers in our cohort of patients, despite similar maternal outcomes. More research is needed to explore how these differences in pathology may influence obstetric and perinatal outcomes.