Histopathologic evaluation of dichorionic twin placentas in unassisted and in vitro fertilized pregnancies affected by preeclampsia

Pregnancies achieved with assisted reproductive technology have an increased risk of multiple gestations, preeclampsia, and placental morphologic abnormalities. Inflammatory processes affect dichorionic twin pregnancies disproportionately more than singleton gestations and have been associated with adverse pregnancy outcomes, such as fetal growth restriction and preeclampsia. Our objective is to investigate the placental morphology of dichorionic twin pregnancies complicated by preeclampsia conceived with in vitro fertilization (IVF) versus unassisted.

This is a retrospective analysis of placentas from dichorionic twin pregnancies affected by preeclampsia conceived with IVF versus without assistance from 2010 to 2016 at a tertiary care university hospital. Placental pathology findings were analyzed both independently and in aggregate stratified into composite outcome scores using a modified placental synoptic framework. Individual placental abnormalities were grouped into composite categories based on the site of origin: anatomic placental abnormalities; maternal vascular malperfusion; placental villous maldevelopment; fetal vascular malperfusion; chronic utero-placental separation; maternal-fetal interface disturbance; inflammation of infectious etiology; and inflammation of idiopathic etiology. Placental histopathological statistical analysis was performed using Fisher's exact test. Demographic variables and pregnancy outcomes were compared between groups using the Student's t test or Mann-Whitney U test, where appropriate. p < .05 defined statistical significance.

Of 117 dichorionic twin pregnancies, 60 resulted from IVF (Group A) and 57 were conceived without assistance (Group B). Patients in Group A were older (36 [29-37] vs. 33 [32-38] respectively; p = .042) and less parous (18.3% vs. 38.6% percent parous in Group A and Group B, respectively p = .009) than Group B, respectively. No differences were found between groups regarding mode of delivery, gestational age at delivery, placental weight/birthweight, fetal growth restriction, and discordance of fetal growth. There were significantly more inflammatory changes of unknown etiology and composite inflammatory abnormalities in Group A versus Group B (26.7% vs. 10.5%, p = .02). The cumulative number of inflammatory abnormalities per patient had a significantly different distribution among groups (p = .005), and Composite Chronic Inflammation and Infection were found to be significantly more abundant in Group A versus Group B (p = .02). The distribution of placental composite anatomic placental abnormalities, maternal vascular malperfusion, placental villous maldevelopment, fetal vascular malperfusion, chronic utero-placental separation, or maternal-fetal interface disturbance was not statistically different between groups. The distribution of placental abnormalities was not different between groups for any individually analyzed pathological condition. Due to the relatively small sample size, adjustment for potential confounders was not performed.

Dichorionic twin pregnancies affected by preeclampsia are associated with more placental inflammatory abnormalities if conceived with IVF versus unassisted. Further research is needed to ascertain the underlying mechanisms of these observed differences.