Associating antiretroviral therapy initiated before or during pregnancy with placenta pathology in HIV infected women.

Antiretroviral therapy (ART) is universally recommended for all HIV-infected pregnant women to improve survival and to reduce mother-to-child HIV transmission. With widespread use of ART in areas of high HIV prevalence, there is concern that prolonged in-utero ART exposure may result in adverse birth outcomes, such as preterm delivery and low birthweight infants. Underlying drivers of these effects may be found in the placenta. We examined whole placentas from live births to seek an association between timing of ART initiation and pregnancy outcome. This was a nested sub-study in a larger cohort of HIV-infected women recruited in a large primary care antenatal clinic in Cape Town, South Africa. Whole placentas (n=130) were collected at delivery and examined for histopathology from two ART groups. The stable group (n=53) initiated ART prior to pregnancy and the initiating group (n=77) started ART at a median of 15 weeks gestation. Adverse birth outcomes were defined by preterm delivery (PTD n=9), small-for-gestational age (SGA n=12) and low birthweight (LBW n=11) in the two groups. Wilcoxon rank-sum and t-tests were used to measure differences in placental histopathology by ART group. Binomial and univariate linear regression models, chi-squared or Fishers exact test were used to quantify associations between placenta histopathology and maternal factors including ART timing and the association with pregnancy outcome. Women in the stable group were significantly older (median [IQR] 33 years [28 - 37] vs 28 [25 - 32]; p=0.003), had higher CD4 cell counts 455 cells/ml [381 - 602] vs 369.5 [251 - 534.5], p=0.04) and were less likely to be prehypertensive or hypertensive at first antenatal care visit (p=0.03) than women in the initiating group. Overall, 119 (91%), women were on a fixed dose regimen of Tenofovir (TDF) + Lamivudine (3TC) + Efavirenz (EFV) and only 3 (2%; all in the stable group) were on a protease inhibitor (PI)-based regimen with no difference between the groups. Overall, nearly a third of placentas were small, although the fetal-placenta weight ratio was in the normal range for most women. Meconium exposure and chorioamnionitis was seen in 19% and 15% of placentas, respectively. Apart from maternal vascular malperfusion (MVM) and decidual arteriopathy, histopathology observations did not differ significantly between stable and initiating groups. Placentas from the stable group showed increased MVM (39.6% vs 19.4%, p=0.01) and decidual arteriopathy (11.3% vs 1.3%, p=0.02), with a trend of decreased placental weight (392g vs 422g, p=0.09). MVM, in turn, was significantly associated with PTD and LBW (p=0.002 and p<0.0001) and association with SGA was borderline significant (p=0.09). ART initiation prior to pregnancy was associated with an increased risk of maternal vascular malperfusion and decidual arteriopathy, suggestive of placental dysfunction. The association between MVM with PTD and LBW suggests that a placenta-mediated mechanism may link the putative association between ART and adverse birth outcomes.