Effect of probable neonatal sepsis on the development of infants in Eastern Uganda (ENON): a cohort study

BACKGROUND: Neonatal sepsis is a common cause of mortality and morbidity during the first 28 days of life, particularly in resource-limited settings. Of those infants who survive sepsis, there is a lack of clarity as to the likelihood of developmental difficulties in these children. This study aimed to understand the impact of probable neonatal sepsis on the development of 2-year-old children in Uganda. METHODS: We assessed 404 infants aged between 18 and 36 months using the WHO's Global Scales for Early Development (GSED) Long Form (directly observed) and Short Form (parent report). This included 100 infants treated for probable neonatal sepsis and were part of a large community infection prevention trial (BabyGel). We then recruited 304 age-matched children from the same communities who took part in the parent trial but had no probable neonatal sepsis. We used linear regression and inverse probability of treatment weighting analyses to calculate measures of association and effect. We adjusted for the clustering effect, trial arm or other confounders. RESULTS: The mean age (SD) of children in our cohort was 23.3 months (4.4) in both groups. GSED development for age Z-scores were significantly lower for the children with a probable sepsis diagnosis compared with those without for both the GSED Short Form -0.22, 95% CI -0.47 to -0.04 and the GSED Long Form -0.33, 95% CI -0.60 to -0.05 when adjusted. An additional model adjusting for HIV exposure as a confounder (only available 373/404) provided similar results with adjusted mean GSED Short Form developmental age Z-scores difference of -0.29, 95% CI -0.53 to -0.04 and -0.33, 95% CI -0.59 to -0.07 for GSED Long Form. CONCLUSION: Neonatal sepsis may be associated with poorer developmental outcomes in resource-limited settings such as Uganda. To ensure better outcomes for these children, it is vital to explore more effective strategies for follow-up, monitoring and early interventions in infants with probable neonatal sepsis.