Short-term outcomes and factors associated with mortality in children admitted with severe malaria in a tertiary health facility, South-East Nigeria

Severe malaria remains a life-threatening emergency and a leading cause of childhood morbidity and mortality in sub-Saharan Africa. Despite global efforts to reduce the malaria burden, Nigeria continues to bear a high proportion of cases and deaths, particularly among children under 5 years. This study aimed to determine the short-term outcomes and identify independent sociodemographic and clinical factors associated with mortality among adequately powered cohort of children admitted with severe malaria at a tertiary hospital in South-East Nigeria. A hospital-based retrospective cohort study was conducted, at the Imo State University Teaching Hospital from August 2019 to August 2024. Medical records of children aged 2 months to 12 years admitted with severe malaria were reviewed. Diagnosis was based on the 2015 World Health Organization (WHO) criteria confirmed by microscopy of a thick blood film smear. Data on sociodemographic characteristics, clinical presentations and outcomes were extracted and analyzed using SPSS version 23 and Firth’s penalized logistic regression addressed sparse data. Bivariate and multivariable analyses were performed to identify factors associated with mortality ( p < 0.05). Of 2384 paediatric admissions, 434 (18.2%) met the inclusion criteria for severe malaria. The majority 275 (63.4%) were under 5 years, with male-to-female ratio 1.8:1. Among the 422 children with known outcomes (excluding 12 discharged against medical advice), mortality was 53/422 (12.6%). Mortality was higher among children under 5 years who accounted for 44 of 53 (83.0%) deaths; (aOR = 3.17, 95% CI: 1.51–6.69; p = 0.002), 41 of 53 (77.4%) dead children were urban residents with aOR = 2.60, 95% CI: 1.33–5.11; p = 0.005) and 43 of 53 (81.1%) of the dead children presenting after more than 5 days of illness with aOR = 3.28, 95% CI: 1.60–6.72; p = 0.001. Highest case fatality rates occurred in children with shock 3 of 7 (42.9%), multiple convulsions 20 of 97 (20.6%), impaired consciousness 40 of 204 (19.6%) and respiratory distress 27 of 152 (17.8%). Independent associations with mortality were, age <5 years (aOR = 3.17, 95% CI: 1.51–6.69), urban residence (aOR = 2.60, 95% CI: 1.33–5.11), shock (aOR = 18.54, 95% CI: 3.21–107.08), impaired consciousness (aOR = 4.07, 95% CI: 2.11–7.86), multiple convulsions (OR = 2.81, 95% CI: 1.30–4.40), respiratory distress (OR = 2.13, 95% CI: 1.19–3.80) and delayed presentation (aOR = 3.28, 95% CI: 1.60–6.72). Severe anaemia was inversely associated with mortality (aOR = 0.29, 95% CI: 0.15–0.55). Severe malaria was associated with high mortality (12.6%) in hospitalized children in South-East Nigeria. Mortality is independently associated with young age (<5 years), urban residence, delayed presentation (>5 days of illness) and the presence of shock, impaired consciousness, multiple convulsions or respiratory distress at admission. This study provides the first robust region-specific evidence from South-East Nigeria using Firth’s penalized logistic regression to address sparse data. Our findings quantified the mortality burden and clinical spectrum of severe malaria in an understudied region. Urban residence was identified as an under-recognized, factor associated with mortality. Severe anaemia although common was demonstrated not to be independently associated with mortality. Strengthening community early health seeking behaviors, emergency triage and improving capacity of clinicians would likely reduce deaths. Prospective cohort studies are needed to establish temporal relationships and evaluate the effectiveness of targeted interventions.