Background:
Despite tremendous achievement in reducing HIV mother-to-child-transmission (MTCT), 150,000 children were newly infected in 2020 globally. High maternal viral load (VL) at delivery is among the strongest risk factors for MTCT. Currently, the choice of postnatal antiretroviral prophylaxis (PNP) for HIV-exposed infants is based on WHO high-risk (HR) criteria. We aimed to estimate the contribution of point-of-care (PoC) maternal VL testing at delivery in profiling the risk of MTCT and its impact on standard PNP or enhanced PNP (ePNP) for HIV-exposed infants.
Methods:
The cluster-randomized LIFE trial was conducted at 28 health facilities in Tanzania and Mozambique. At delivery, the intervention arm A provided PoC maternal VL to aid MTCT HR assessment in addition to clinical criteria and antenatal care information as available in the control arm B only. In Tanzania both arms initiated ePNP based on maternal risk factors, including VL for arm A, while in Mozambique, ePNP was provided universally. We used mixed effects logistic regression models to estimate the effect of our intervention on the proportion of infants 1) identified as HR (Tanzania and Mozambique) and 2) HR infants receiving ePNP (Tanzania only). Standard errors were clustered to account for health facility and multiple births.
Results:
A total of 6512 mothers living with HIV were enrolled: 72% were diagnosed before the 2nd trimester, 99% were on ART, and 78% were virally suppressed at delivery. Of 6568 newborns, 781 (12%) were classified as HR (636 (19.5%) vs. 145 (4.4%) in arms A and B, respectively; p< 0.0001). In arm A, 512 (80.5%) HR infants were classified only based on maternal PoC VL at delivery. In arm B, overall 609 (18.4%) additional infants would have been identified as HR if their mothers would have received PoC VL assessment. In Tanzania, HR infants in arm A were significantly more likely to receive ePNP, with 67/112 (59.8%) vs. 16/51 (31.4%) receiving ePNP in arms A and B, respectively (OR 4.49, 95% CI: 1.23, 16.36). However, 40.2% in arm A and 68.6% in arm B did not receive ePNP despite available information for HR classification at delivery.
Conclusions:
PoC maternal VL testing at delivery significantly increased the proportion of infants identified as HR. Further, HR infants with maternal PoC VL at delivery were more often initiated on ePNP. However, linkage of HR infants to appropriate antiretroviral prophylaxis remains suboptimal, warranting consideration of universal ePNP irrespective of transmission risk.