Background: We analyzed antiretroviral therapy (ART) regimens and pregnancy outcomes in naive and ART-experienced HIV-positive women from Italian Cohort Naive Antiretrovirals cohort and investigated frequency and predictors of detectable viral load (VL) at delivery.
Methods: All pregnancies resulting in live births were included. Based on ART at the beginning of pregnancy, pregnancies were allocated either to the ART-naive or ART-experienced group. Analyses were stratified according to calendar periods. Multivariate logistic regression was used to describe predictors of detectable VL at delivery.
Results: One hundred fifty-eight of 2862 women experienced 169 pregnancies (88 in naives and 81 in 70 ART-experienced women). ART regimens varied according to calendar periods; mono-dual combination regimens progressively decreased over time (P value for trend <0.0001). Protease inhibitor-including regimens were the most frequently used regimens at delivery (71.6% vs 63.0% in naives and in ART experienced, P = 0.2). VL was detectable in 35.6% of women at delivery; this was less likely with increasing calendar periods (adjusted odds ratio per 1-year longer: 0.8, 95% confidence interval: 0.7 to 0.9, P = 0.007) and more likely in women with HIV RNA >50 copies per milliliter at pregnancy ascertainment (adjusted odds ratio: 7.1, 95% confidence interval: 1.9 to 33.3, P = 0.006). Nevertheless, no cases of vertical transmission were diagnosed. Preterm birth rate of 17.3% (11.9% vs 22.6% naive and ART experienced, P = 0.1) was reported; this was not associated with ART duration or protease inhibitor-including regimens; 27.2% of infants had <2500 g birth weight.
Conclusions: Antiretroviral regimens prescribed during pregnancy changed over time according to guidelines. Although undetectable VL was not always achieved, no vertical transmission occurred; preterm delivery and low birth weight occurred in some cases and still remain key issues.