Pregnancy outcomes in women using non-AZT HAART in Europe
Polly Clayden, HIV i-Base
AZT is the only antiretroviral licensed for use in pregnancy. In resource rich settings AZT-containing regimens are becoming less and less common and in turn increasing numbers of women are becoming pregnant while receiving non AZT-containing HAART, or initiating non-AZT-containing HAART in pregnancy.
A poster at authored by Shema Tariq and colleagues reported findings from an investigation into the risk of detectable maternal viral load (VL) at delivery, congenital abnormality and mother to child transmission (MTCT) in pregnancies among women receiving HAART containing and not containing AZT. 
This analysis combined data from the National study of HIV in Pregnancy and Childhood (NSHPC) and the European Collaborative Study (ECS).
All live singleton births from 2000 to 2009 with >14 days of HAART in pregnancy were included. The investigators used logistic regression to estimate adjusted odds ratios (AOR).
A total of 7353 (6310, NSHPC and 1263, ECS) pregnancies were included, of which 1199 (15.8%) were exposed to non-AZT HAART. Of this group, 23.2% and 71.3% were on HAART prior to conception in the AZT and non-AZT groups respectively. Exposure to non-AZT HAART increased over time: 20002002 19.6% vs 17.1%, 20062009 41.4 vs 58.7% women received AZT vs non-AZT HAART respectively, p<0.01.
Tenofovir and abacavir was the most commonly used non-AZT drug in this cohort; approx 45% and 35% respectively.
In multivariate analysis the investigators found no evidence of associations between non-AZT HAART and detectable viral load at delivery, risk of congenital abnormalities (including in a sub-analysis of pregnancies with 1st trimester exposure or rate of MTCT, see Table 1.
Table1. Maternal and infant outcomes AZT vs non-AZT HAART multivariate analysis, AOR (95% CI)
|Detectable maternal VL at delivery||Congenital abnormalities (all pregnancies)||Congenital abnormalities (1st trimester HAART
|AZT HAART (ref. group)||1||1||1||1|
|Non-AZT HAART||0.90 (0.73-1.11); p=0.33||0.95 (0.641.41); p=0.80||0.76
|1.81 (0.774.26); p=0.18|
The investigators noted that information on maternal VL at delivery was missing for 45% of pregnancies however the proportions of missing data were similar for both groups. Additionally HIV status was not yet reported for 20% of infants, mainly those born recently. They suggested that this is unlikely to introduce bias.
The investigators described these results as reassuring and that continued monitoring of pregnancy outcomes and longer term consequences of in utero exposure to these antiretroviral drugs is required.
Ref: Tariq S et al. Pregnancy outcomes in HIV-infected women using non-zidovudine HAART in Europe: 2000 to 2009. 17th CROI, San Francisco, 2010. Poster abstract 895.