Abstract:
This PhD dissertation reports data from the very early infant diagnosis (VEID) study that was conceptualized when universal birth HIV PCR testing at Kalafong Provincial Tertiary Hospital was mandated. Three research questions were addressed: (1) What factors are associated with early infant HIV acquisition?; (2) What are the growth outcomes of HIV exposed, uninfected (HEU) infants?; and (3) What is the 9-month infant HIV seroreversion rate in the context of PMTCT Option B+ when using different rapid HIV tests?
The first research chapter demonstrates that maternal HIV viral load detectable in the perinatal period, maternal combination antiretroviral therapy (cART) with a duration <1 month, and having a symptomatic infant at birth are significant predictors of early infant HIV acquisition. Small-for-gestational-age was included with the above three characteristics in multivariate analyses. Two-risk (maternal cART duration and viral load), three-risk (maternal cART duration, maternal viral load and symptomatic newborn), and four-risk (maternal cART duration, maternal viral load, symptomatic and SGA newborn) models for HIV acquisition were developed with a predictive probability score of a newborn PCR positive test of 0.28, 0.498, and 0.57 respectively. These findings could guide a targeted approach to infant HIV-testing at birth. However, using the three- and four-risk scores at a probability of 0.02 and 0.04, 20% and 24% of HIV infected infants will be missed respectively at birth compared with universal testing. Therefore, we support universal birth PCR testing within the South African PMTCT context.
The second research chapter describes growth outcomes of HEU infants in relation to maternal and infant birth characteristics. Mothers were mostly breastfeeding at birth and on universal lifelong cART. Maternal to child transmission of HIV after birth were less than 1% (0.31%) and occurred mostly by 6 weeks of age. HIV infection was associated with symptoms and signs of HIV-associated immunosuppression, such as growth faltering. The hospitalization rate was 41.3/1000 person-years. Longitudinal growth trends illustrated lower weight-for-age (WAZ) and length-for-age (LAZ) in males. HEU newborns that had one or more symptoms suggestive of HIV-associated immunosuppression had lower weight trends and newborns with a birth weight <2.5kg had significantly lower WAZ, LAZ and head circumference (HC) trends. The significance of poor LAZ trends, especially in male HEU infants, remains a concern. The impact on final height and body mass index (BMI) need careful follow up to ascertain if these growth trends will have a negative impact such as higher BMI values and possibly more obesity among HEU male adolescents/young adults.
Lastly, we describe seroreversion at 9 months amongst HEU infants whose mothers received cART. We observed that different rapid assays vary in performance, with specificity ranging between 45-97%. HIV ELISA testing did not document any seroreversion at nine months in our cohort. This finding highlights the need for further research to determine the age at seroreversion within the context of the PMTCT option B+ / universal maternal access to lifelong cART. It is possible that uninfected infants will remain seropositive beyond 18 months of age. Hence, future recommendations might include HIV PCR testing in infants up to 24 month of age.
