Factors Associated with Lack of Viral Suppression at Deliver: IMPAACT P1025
Ingrid Katz*1,2, E Leister3, R Tuomala1,2, M Hughes3, D Kacanek3, A Bardeguez4, E Livingston5, A Stek6, and D Shapiro3
1Brigham and Women`s Hosp, Boston, MA, US; 2Harvard Med Sch, Boston, MA, US; 3Harvard Sch of Publ Hlth, Boston, MA, US; 4Univ of Med and Dentistry of New Jersey, Newark, US; 5Duke Univ Sch of Med, Durham, NC, US; and 6Univ of Southern California, Los Angeles, US
Background: High maternal plasma viral load is a major risk factor for MTCT. In the US, many women fail to achieve complete viral suppression at delivery. We investigated risk factors associated with viral load >400 copies/mL at delivery among pregnant women enrolled in International Maternal Pediatric Adolescent AIDS Clinical Trials Group (IMPAACT) P1025, a prospective cohort study, from 2002 through 2010.
Methods: We analyzed data on 1070 HIV-1+, treatment-naive pregnant women who initiated HAART during pregnancy. Logistic regression methods were used to identify socio-demographic factors, HIV disease characteristics, and treatment characteristics associated with viral load >400 copies/mL at delivery. Selected variables were chosen to predict outcomes relevant to the initial clinical presentation. Only co-variates with p <0.05 once adjusted remained in the final model.
Results: The median age of women was 27 years, and most (59%) were black. Protease inhibitor (PI)-based HAART was initiated in 76%, 8% took a NNRTI-containing regimen, and 16% took 3+NRTI only. At delivery, 188 women (18%) had viral load >400 copies/mL. In univariate analyses, characteristics associated with viral load >400 copies/mL at delivery were: black race, not completing high school, multiparous, pre-treatment viral load ≥1000 copies/mL, and pre-treatment CD4 <200. Women who initiated HAART in the third trimester, and more specifically after 32 weeks’ gestation, had higher odds of viral load >400 copies/mL. In a multivariate model, factors independently associated with viral load >400 copies/mL at delivery included: black race, not completing high-school, multiparous, pre-treatment viral load ≥1000 copies/mL, initiating HAART after 32 weeks’ gestation, and type of regimen. NNRTI-based regimens appeared to be the most protective, while unboosted PI-based regimen appeared to be least protective, when compared with boosted-PI. Pre-treatment CD4 count was not independently associated with viral load >400 at delivery (see the table).
Conclusions: Nearly 1 of 5 women who initiated HAART during pregnancy had a viral load >400 copies/mL at delivery. Maternal risk factors independently associated with unsuppressed viremia at delivery included: black race, not completing high school, multiparous, and having more advanced disease prior to HAART initiation (viral load ≥1000 copies/mL). Women who started HAART later in pregnancy (>32 weeks gestation) were also at significantly higher odds of having viral load >400 copies/mL at delivery. Future interventions should focus on initiating HAART early in pregnancy, particularly for women with high viral load.