Differences in growth trajectories in breastfed HIV-exposed uninfected and HIV-unexposed infants in Kenya: An observational cohort study.

backgroundChildren who are HIV-exposed and uninfected (CHEU) are at increased risk for poor growth compared to children who are HIV-unexposed (CHU). There are limited data on growth among CHEU in the era of preferred dolutegravir-based antiretroviral therapy (ART) for pregnant and breastfeeding women living with HIV (WLWH). We aimed to compare child growth outcomes in the first two years of life between breastfed CHEU and CHU, and to examine maternal HIV factors associated with growth in CHEU. METHODS AND

findingsWe enrolled pregnant women in Kenya and followed them with their child to age 24 months. We measured anthropometry within 7 days of birth, at 3 and 6 weeks, and months 3, 6, 9, 12, 18, and 24. We compared length-for-age Z-scores (LAZ), weight-for-age Z-scores (WAZ), weight-for-length Z-scores (WLZ), head circumference-for-age Z-scores (HCZ), and mid-upper arm circumference-for-age Z-scores (MUAC), and stunting (LAZ < -2), underweight (WAZ < -2), and wasting (WLZ < -2) between groups using linear mixed effects or modified Poisson regression models adjusted for maternal age, education, depression, anemia, household wealth index, time-varying breastfeeding, time-varying food insecurity, parity, and child sex. Among 333 mother-child pairs with at least two child visits (CHEU = 171; CHU = 162), mothers of CHEU were older, less educated, and had lower wealth than mothers of CHU. Birth characteristics were similar between groups, with 9% preterm births and 6% low birthweight. All WLWH were on ART, 89.5% on dolutegravir-lamivudine-tenofovir, 76.6% initiating ART preconception, and 91.2% virally suppressed. The duration of breastfeeding was significantly shorter for CHEU than CHU (median 15 versus 17 months). CHEU had significantly lower LAZ at birth, 18- and 24-months than CHU. In multivariable analysis, growth trajectories for WLZ and HCZ were lower among CHEU than CHU in the first 24 months (interaction p = 0.001 and p = 0.009, respectively). There was no difference in trajectory in LAZ, WAZ, and MUACZ between groups. By 24 months, 31.5% of CHEU were stunted, 9.3% underweight, and 2.4% wasted, versus 27.2%, 3.2%, and 0.6% of CHU, respectively; only the difference in underweight prevalence was statistically significant. CHEU had a higher risk of being underweight from 9- to 24 months than CHU (adjusted Relative Risk at 24 months, 2.99 [95% CI: 1.08, 8.30]; p = 0.034). Growth was associated with maternal education, wealth, and breastfeeding and was lower among male infants. Among CHEU, maternal preconception ART was not associated with growth. Important limitations of this study include the possibility of unmeasured confounding and limited generalizability to contexts with differing prevalence of malnutrition, access to and uptake of ART, or breastfeeding practices.

conclusionsDespite breastfeeding and optimal maternal dolutegravir-based ART, CHEU experienced growth deficits compared to CHU in the first two years of life. Continued monitoring of the expanding CHEU population is essential in the context of rapidly evolving guidelines and policies to optimize their health and to identify and prevent future health disparities and disease risks.