High coverage of combination antiretroviral therapy (cART) among pregnant and breastfeeding women with HIV (WHIV) has resulted in an expanding population of HIV-exposed uninfected (HEU) children globally.¹ HEU have been shown to have poorer growth and health outcomes than HIV unexposed uninfected children (HUU).^2–6 Limited evidence also suggests that HEU may be at increased risk of microcephaly,^7,8 which may have long-term consequences on neurocognitive development. Underlying mechanisms for these differences remain unclear. Studies evaluating in-utero ART exposure and child growth have mixed results and vary by age.^9–11 HIV exposure has been associated with immunological vulnerabilities and lower birthweight, both of which adversely affect growth. Household and socioeconomic differences adversely affect child growth, and these inequalities may play a larger role in children born to WHIV, particularly in low and middle-income countries.¹² Population-level data on child head circumference, length, and weight in high HIV burden settings is warranted.

Of 2,457 infants surveyed, 456 (18.6%) were HEU, of whom 50.7% attended clinic for 6-week and 49.3% for 9-month immunizations. HEU at 6-weeks had lower mean birthweight and fewer were currently breastfeeding at 6-weeks and 9-months compared to HUU. WHIV were older and had completed less education than mothers without HIV at both timepoints (Table 1).

Forty-three percent of WHIV were on cART prior to the pregnancy. Sixty-four percent received cART and 22.4% received ARVs during pregnancy. Overall, 13.2% of WHIV did not receive ART or ARVs during pregnancy. Most women on ART were on non-nucleoside reverse transcriptase inhibitors-based (NNRTI) regimens. Most (97.8%) HEU received nevirapine for PMTCT and 96.9% of HEU attending 9-month immunizations had received cotrimoxazole prophylaxis.

In this population-level analysis, HEU had poor weight and length growth compared to HUU, and stunting was twice as prevalent in HEU at 6-weeks and 9-months. At 9-months, HEU had higher prevalence of microcephaly and lower mean head circumference than their HUU peers.

These findings are consistent with other studies in sub-Saharan Africa reporting poor early growth in HEU compared to HUU.^{2,4,5,15–17} Independent of HIV status, breastfeeding was associated with 72% lower prevalence of underweight at 6-weeks and prior hospitalization was associated with underweight at 6-weeks and 9-months. HEU have greater risk of low birthweight^2,15,18–20 and diarrhea, pneumonia, and hospitalization in early infancy than HUU,²¹ which impacts WAZ and LAZ. Our data are consistent with longitudinal studies showing greater variability in weight velocity during infancy and suggest other factors may play a more pivotal role in weight growth. Child stunting has been associated with socioeconomic, dietary, health, and environmental factors²² and increased morbidity and mortality.²³ While large-scale interventions to strengthen infant and young child feeding practices have led to reductions in early stunting,^24,25 nutritional interventions alone may not be able to prevent or reverse stunting.¹⁶ Understanding relevant socioeconomic and biological mechanisms for early length growth are important to refine interventions to reduce stunting in HEU.

Among children screened at 9-months, HEU had lower mean head circumference and microcephaly prevalence was twice as common. While the limited discussion on microcephaly in HEU has focused on maternal cART, we found that shorter maternal height and lower education were independently associated with microcephaly irrespective of HIV exposure. Both a pre-ART study in Zimbabwe and a study of HEU born to mothers on cART in Nigeria reported higher prevalence of microcephaly among HEU than HUU.^4,8 Recently, the U.S.-based SMARTT study reported lower neurodevelopmental scores in HEU with microcephaly.⁷ Our results showed no association between cART prior to or during pregnancy and poor head growth. Further studies are needed to better understand the population prevalence and drivers of microcephaly in early childhood.

Maternal characteristics, including lower height, BMI, and education have been associated with poor growth,^26,27 which is consistent with our findings. Strategies to reduce poverty and promote maternal health and education are needed to improve child growth. Strengthening relationships between health facilities and community programs could help to address some of the socioeconomic and psychosocial challenges faced by families of HEU.

HEU may be at higher risk of growth faltering due to immune activation,²⁸ alterations in the gut microbiome, and systemic inflammation due to in-utero or postnatal exposure to HIV and prolonged exposure to ART despite their HIV negative status.^9–11,29 We found that pre-pregnancy maternal ART was associated with lower length, lower weight, underweight, and stunting at 9 months, suggesting that peri-conception ART may have long-term impact on growth in HEU. It is likely this may differ by regimen; further studies evaluating the role of pre-conception ART on long-term growth are warranted.

This evaluation enrolled mother-infant pairs in 140 facilities across Kenya, allowing for a better understanding of growth among children in urban and rural settings. A sensitivity analysis including data from only the primary survey had similar findings (Supplementary Table 2). While mother-infant pairs not presenting to clinic were not captured, UNICEF estimated that 95% of living children received their first dose of Diphtheria-Tetanus-Pertussis vaccine immunization (6-week) and 94% of living children receive their first dose of measles vaccine (9-month) in 2013.³⁰ This study was conducted at facilities, and HIV-affected families could have higher engagement in care, resulting in a potential underestimation of differences between HEU and HUU children. In this study, mothers were on NNRTI regimens – further investigation on the impact of integrase inhibitor-based regimens, including dolutegravir, on growth among HEU is warranted, as dolutegravir has been associated with increased BMI^31–34 and higher pregnancy BMI has been associated with adverse infant outcomes.³⁵ While this study was conducted prior to WHO recommendations for universal ART, the majority of mothers received cART and recent studies continue to show poor growth in HEU.^2–5 In a sensitivity analysis restricted to only HEU with mothers on cART, we showed similar growth deficits in HEU.

Findings from this population-level comparison of HEU and HUU support existing evidence that HEU have poor length and weight growth in early childhood and add to the limited data on poor head growth in these children. Strategies to optimize breastfeeding and improve nutritional practices in HEU may improve long-term growth outcomes in this growing and vulnerable population.