CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Results: Of the 31,988 infants included in the analyses, 16.6%were preterm, 15.3%were SGA and 7.3%were LBW. Proportion of preterm and SGA births was higher among the HIV infected women. Preterm birth was associated with an increased risk of neonatal and infant mortality, with further increased risk for early preterm. Moderate SGA was associated with an increased risk of mortality, and the risk was even greater for infants born with severe SGA. Compared to term-appropriate-for-gestational-age (AGA) infants, infants born both preterm and SGA had a greater risk of neonatal mortality (HR 5.43, 95% CI: 2.01-14.63) than preterm-AGA infants (HR 2.40 95% CI: 1.89-3.05). Maternal HIV status modified the risk of mortality associated with preterm birth in the post-neonatal and overall infant period (p for interaction 0.004). Compared to full term infants, early preterm infants born to HIV-positive women were 2.08 times (95% CI: 1.36-3.18) more likely to die during infancy, while early preterm infants born to HIV-negative women were 4.86 times more likely to die during infancy (95% CI: 3.52-6.71). The lower relative risk of death among preterm infants in the HIV infected cohort was driven by the higher mortality among term infants. Conclusions: Preterm and SGA significantly increased the risk of mortality during the neonatal period, and the risk extends well beyond the neonatal period in both HIV exposed and unexposed infants. 804 HBV and HCV Infections in HIV-Infected Pregnant Women: Obstetrical Outcomes Valerie Benhammou 1 ; RolandTubiana 2 ; Sophie Matheron 3 ; Pierre Sellier 4 ; Laurent Mandelbrot 5 ; Jérôme Le Chenadec 6 ; Emannuelle Marel 7 ; Babak Khoshnood 8 ; JosianeWarszawski 7 ; for the ANRS CO1/CO11 - EPF French Perinatal Cohort Study Group 1 INSERM 1153, PARIS, France; 2 Sorbonne Univs, Paris, France; 3 Hosp Bichat-Claude Bernard, Paris, France; 4 Groupe Hospier Saint-Louis, Lariboisiere, Fernand Widal, AP-HP, Paris, France; 5 Hosp Louis Mourier, Colombes, France; 6 INSERM, Le Kremlin-Bicêtre, France; 7 CESP, INSERM U1018, Le Kremlin Bicêtre, France; 8 INSERM 1154, Paris, France Background: Data on the impact of chronic hepatitis infections on the immunovirological response to antiretroviral therapy (ART) and obstetric outcome in HIV-infected pregnant women are scarce and conflicting. Methods: We analyzed data from all HIV-1 infected women included in the national ANRS-CO1 French Perinatal Cohort between 2005 and 2013. Prenatal testing for HBV and HCV infections was performed in most cases (95%). HBV/HIV and HCV/HIV co-infected mothers were compared with those infected only with HIV; the rare mothers with all three infections were excluded. Bivariate and multivariate analyses were performed. Results: Among 6548 pregnancies, the overall prevalences of HCV (RNA + ) and HBV (HBsAg + ) infections were 3.2% [95%CI: 2.9-3.8] and 6.9% [6.2-7.5], respectively. As expected, HCV infection was strongly associated with a history of drug use, whereas HBV infection was six times more frequent in women originating from Sub-saharan Africa compared with those frommainland France. HIV viral load, CD4 cell count at pregnancy initiation and HIV care were similar in co- and mono-infected HIV mothers except, for ART, with 90% of HBV/HIV co-infected women receiving tenofovir and /or 3TC or FTC, with potential to efficiently decrease HBV viral load. No efficient treatment against HCV was prescribed in the HCV/HIV group. HCV coinfection was significantly associated with poorer HIV immunovirological status during the third trimester (Fig), and higher risks of gestational diabetes (OR=1.9 [1.0-3.7], p=0.05), cholestasis (OR=8.9 [4.9 – 16.3], p<0.001) and preterm delivery (OR=3.3 [1.9-5.5], and OR=4.2 [2.3-7.9], p<0.001) for moderate and very preterm delivery, respectively: p<0.001). The association with prematurity remained significant after adjustment for known risk factors, HIV viral load and antenatal complications (aOR=2.3 [1.1-5.0], p=0.03). In HBV/HIV women no association was found with any of these outcomes. Conclusions: In HIV-infected pregnant women, chronic HBV infection, efficiently treated, had no major impact on mother health during pregnancy. In contrast, HCV co-infected mothers, without any efficient treatment against HCV, showed a poorer HIV immunovirological response to ART and higher risk of antenatal complications and prematurity. This suggests that efficient control of HCV activity, before conception, as likely to be obtained in HBV infection, may limit the deleterious impact of co-infection.

Poster Abstracts

805 Maternal ART and Hospitalization or Death Among HIV-Exposed Uninfected Infants Scott Dryden-Peterson 1 ;Tatiana Ramos 2 ; Roger L. Shapiro 3 ; Shahin Lockman 1 1 Brigham and Women’s Hosp, Harvard Med Sch, Boston, MA, USA; 2 Massachusetts General Hosp, Boston, MA, USA; 3 Harvard Sch of PH, Boston, MA, USA

Background: Excess deaths among HIV-exposed uninfected infants (HEU) are a major contributor to under-5 mortality in Africa. Antenatal maternal ART nearly eliminates the risk of mother-to-child transmission (MTCT), but its effect on health outcomes of HIV-uninfected infants is unknown. We examined the effect of maternal ART and risk of HEU infant hospitalization and/or death by 6 months of age. Methods: We pooled data from 2 randomized trials and 2 observational studies that enrolled HIV-infected mothers and their infants from four sites in Botswana from 2001 to 2012. Studies utilized harmonized covariate and endpoint assessments. Analysis excluded HIV-infected infants and stillbirths. Feeding method at maternity discharge defined exposure to formula/mixed or exclusive breastfeeding. Infants born to mothers who continued or initiated ART during pregnancy were considered ART exposed. We developed a Cox marginal structural model to estimate the causal effect of maternal ART. Inverse probability weights were used to adjust for maternal age, antenatal CD4 cell count, socioeconomic status, year of delivery, and participation in a randomized trial or observational study. Results: Following exclusion of 109 HIV-infected infants, 2665 HIV-EU infants were analyzed including 1460 (54.8%) exposed to maternal ART and 1205 (45.2%) exposed to maternal zidovudine or no antenatal antiretrovirals. Most (85.6%) maternal ART was initiated in pregnancy. Median maternal CD4 was lower for ART-exposed infants (354 cells/ μL, IQR 219-497) than ART-unexposed (390 cells/μL, IQR 279-538), p<0.001. The majority (62.5%) of infants were breastfed. A total of 291 (10.9%) infants were hospitalized or died—152 ART-exposed (10.4%) and 139 ART-unexposed (11.5%). Ninety-eight infants died (3.7%)—52 ART-exposed (3.6%) and 46 ART-unexposed (3.8%). In adjusted analyses, no effect of maternal ART was detected on rate of hospitalization or death (HR 1.01, 95%CI 0.65-1.56) or mortality (HR 0.98, 95%CI 0.56-1.71). The effect of maternal ART on HEU mortality appeared to be different between breastfed and formula fed infants, although did not reach significance (P=0.075 for interaction)—breastfeeding was protective among ART-unexposed infants (HR 0.40, 95%CI 0.17-0.93) but not among ART-exposed infants (HR 1.23, 95%CI 0.50-3.00).

336

CROI 2016