Predictors of mother to child transmission among women initiating HAART in pregnancy in a South African cohort

There are limited data from Africa describing mother to child transmission (MTCT) in mothers initiating HAART in pregnancy.

A poster authored by Risa Hoffman and coworkers from the Reproductive Health and HIV Research Unit of the University of Witwatersrand and the UCLA Program in Global Health, Los Angeles presented findings from a retrospective analysis from a cohort of women in an antenatal antiretroviral clinic at Johannesburg Hospital looking at factors associated with infant HIV infection.

In this study 689 women indicated for antiretroviral treatment (CD4 </=250 cells/mm3 or WHO stage 4) were referred to the antenatal clinic between August 2004 and February 2007. The women had a mean baseline CD4 of 154 cells/mm3 and 82% received d4T/3TC/NVP. 302 mothers completed 6 weeks postpartum follow up; of these15/302 (5%) infants had positive DNA PCR.

Using univariate analysis, the investigators found shorter duration of treatment (p=0.001) and lower CD4 baseline (p=0.03) to be associated with MTCT.

Analysis of variance (ANOVA) found a statistically significant difference in duration of gestational HAART in pregnancy among mothers whose infants were positive (n=15), negative (n=287) and of unknown status (n=376), p=0.0005.

Mothers with HIV-positive infants received HAART for a shorter duration than those with negative infants, 5.1 vs 11.2 weeks (OR,0.730, 95% CI 0.612-0.879), p=0.001. The investigators noted that for each additional week of HAART during gestation, the odds of transmission were reduced by 27%. The transmission rate for women receiving >7weeks HAART was 0.3%.

Lower CD4 baseline, 148 vs 106 cells/mm3 (OR, 0.991, 95% CI 0.982-0.999, based on change of 25 cells/mm3), p=0.03, was also predictive of MTCT in this analysis.

Viral load at baseline and follow up were not predictors of transmission in this analysis but the investigators suggest that this may be due to high variance in viral load and small numbers of women with complete viral load data.

Unsurprisingly, overall women receiving HAART in pregnancy in South Africa have low rates of transmission. The investigators wrote: Strategies are needed to facilitate earlier treatment of HIV-infected pregnant women with advanced disease.

Comment

In the cohort described in the poster, lower CD4 count (ie women with more advanced disease in need of treatment for their own health) was associated with an HIV-positive infant, and this is consistent with the literature, which shows this over and over again.

The authors highlight the need to initiate treatment earlier in pregnancy and this deserves emphasis both for the health of the mother and the babys HIV status.

The transmission rate of 5% at 6 weeks in this cohort of women with advanced disease (almost all with CD4 < 250) is far lower than the reported transmission rates for healthier women in South Africa who only have access to single dose NVP or other PMTCT regimens and breastfeed (18-25% at 6 weeks).

Black R, Hoffman R, Sugar C et al. Factors associated with MTCT in South African women with advanced immunosuppression initiated on HAART during pregnancy. 15th CROI, February 2008, Boston. Poster abstract 657.