Background: Few resource-limited, high HIV prevalence settings produce data on national 18-24 month infant ''mother-to-child transmission of HIV (MTCT)-or-death'' - the gold standard measurement of programme impact. We studied South African national MTCT and ''MTCT-or-death'' when health policy provided infant nevirapine during breastfeeding (Option A) and changed to triple antiretroviral therapy for all HIV infected women during pregnancy and lactation (Option B).
Methods: A nationally representative cross-sectional survey was conducted to estimate early (4-8 weeks postpartum) MTCT. Facilities (n=580) were randomly selected following multistage probability proportional to size sampling methodology. Consenting caregivers of systematically or consecutively sampled infants (4-8 weeks old) receiving their six week immunisation, were interviewed. Infant dried blood spot specimens (iDBS) were drawn and tested for HIV exposure and, if positive, for infection (total nucleic acid polymerase chain reaction - TNA PCR). Then, all HIV exposed infants (antibody, or maternal self-reported positive) were invited for facility-based follow-up at 3, 6, 9, 12, 15 and 18 months. At each follow-up visit caregivers were interviewed and infants were tested for HIV infection). Analysis was weighted for sample ascertainment, population live births, consent to follow-up (if eligible) and loss to follow-up.
Results: Analysis of 9120 iDBS at 4-8 weeks revealed 33.1% infant HIV exposure (95% Confidence Interval, 31.8-34.3%) and 2.6% (2.0-3.2) MTCT. 1880 (71%) HIV-exposed infants were followed up at 18 months. Cumulative MTCT and ''MTCT-or-death'' by 3, 6, 9, 12, 15, months was 2.7% (2.6-12.6) and 2.8% (2.6-19); 3.5% (3.1-4.4) and 4.2% (3.5-5.4); 3.7% (3.2-4.6) and 5.1% (4.4-6.2); 3.9% (3.4-4.7) and 5.7% (5.0-6.8); 4.1% (3.5-4.8) and 6.0% (5.2-7.0), and at 18 months, 4.3% (3.7-5.0) and 6.2% (5.5-7.3) respectively. 81% of MTCT and 67% of ''MTCT-or-death'' occurred by 6 months postpartum. Maternal receipt of CD4-cell-count result and avoiding breastfeeding protected against MTCT (Adjusted hazard ratio HRa, 0.3 [0.2-0.6], and 0.3, [ 0.07-0.9],respectively). Mixed feeding and infant nevirapine did not significantly increase MTCT-or-death (HRa 1.4 [0.8-2.4] and 2.1 [0.8-5.4], respectively).Having a refrigerator significantly protected against MTCT-or-death (HRa 0.5 [0.3-1.0], respectively).
Conclusions: The first 6 months postpartum is a critical period for following up HIV-exposed infants and providing regular HIV testing.