Neurodevelopmental outcomes following possible serious bacterial infection in early infancy in Karachi, Pakistan: a prospective cohort study

Background Pakistan reports a significant burden of neonatal mortality, with infections as one of the major causes. We aim to assess the long-term impact of early infancy infections on neurodevelopmental outcomes during later childhood.Methods We conducted a prospective follow-up study of the cohort enrolled at the Karachi site of the Aetiology of Neonatal Infection in South Asia (ANISA) during 2019-2020. Children with a possible serious bacterial infection (based on the WHO IMCI algorithm) at early infancy were assessed for neurodevelopment at 6-9 years of age and compared with healthy controls. The Ten Questions (TQS) questionnaire, Strengths and Difficulties Questionnaire (SDQ), and Parent's Evaluation of Developmental Stage Assessment Level (PEDS: DM-AL) neurodevelopmental assessment tools, were administered and scored by the research staff who were blinded to the child's exposure status. Generalized Structural Equation Modelling (GSEM) was employed to verify relationships and associations among developmental milestones, anthropometry, and sociodemographic variables.Results A total of 398 children (241 cases and 157 controls) completed neurodevelopmental and growth assessments. Cases had a significantly higher rate of abnormal TQS scores (54.5% vs. 35.0%, p-value 0.001), greater delays in motor milestones (21.2% vs. 12.1%, p-value 0.02), lower fine motor skills (78.4 +/- 1.4 vs. 83.2 +/- 1.5, p-value 0.02). The receptive language skills were well-developed in both groups. According to the logistic regression model, exposure to infection during the first 59 days of life was associated with delayed TQS milestones (beta = -0.6, 95% CI -1.2,-0.04), TQS hearing domain (beta = -0.3, 95% CI: -1.2 to 0.7), PEDS: DM-AL fine motor domain (beta = -1.3, 95% CI: -4.4 to 1.7), PEDS: DM-AL receptive language development (beta = -1.1, 95% CI: -3.7 to 1.4) and child anthropometric measurements such as weight and height (beta = -0.2, 95% CI: -0.4 to 0.01 and beta = -0.2, 95% CI: -0.4 to -0.01, respectively). Early pSBI exposure was positively associated with PEDS: DM-AL self-help domain (beta = 0.6, 95% CI: -1.2 to 2.4) and SDQ-P overall score (beta = 0.02, 95% CI: -0.3 to 0.3).Conclusion Children exposed to PSBI during early infancy have higher rates of abnormal development, motor delays, and lower fine motor skills during later childhood in Pakistan. Socioeconomic challenges and limited healthcare access contribute to these challenges, highlighting the need for long-term follow-ups with integrated neurodevelopment assessments.