Underweight is Associated with Mortality Among Ugandan Children with Plasmodium falciparum Severe Malaria

Abstract

Background: The past two decades have witnessed a 60% decline in global malaria mortality. However, two thirds of all malaria deaths continue to occur among children <5 years, with a majority in the WHO African Region. Malnutrition is an important risk factor for malaria. Globally, wasting, stunting and being underweight are crucial indicators of malnutrition, and are associated with increased mortality in children <5. Those most vulnerable to malaria and malnutrition are children <5 living in Sub-Saharan Africa, particularly in rural areas often facing a higher burden of disease.Objective: The objective of this study was to assess the prevalence and persistence of nutritional abnormalities causing children to be underweight, stunted, or show signs of wasting, and the association of these abnormalities with in-hospital and post-discharge mortality, risk of repeat illness and long-term sequelae in Ugandan children with 5 different forms of severe malaria (SM) compared to community children (CC).Methods: We conducted a prospective observational study investigating neurocognitive outcomes at 12-months after severe malaria episode in 600 children with SM and 120 CC, aged 0.5-4 years, between 2014-2017 at 2 hospitals (Kampala and Jinja) in Uganda. Using age-adjusted scores from healthy CC, we calculated z-scores for weight-for-age (WAZ), height-for-age (HAZ), and weight-for-height (WHZ). We defined underweight, stunting, and wasting as 2SD below the WAZ, HAZ, and WHZ means. Results: At baseline, children with SM had significantly lower mean WAZ and HAZ compared to CC (-1.1 [1.1 SD] vs. -0.47 [1.1], p<0.001; -0.68 [1.1] vs. 0.30 [1.0], p<0.001, respectively), with no difference by site. By 12-month follow-up there were no significant differences in nutritional markers between SM and CC. During admission, 44 (7.3%) children with SM died. Higher baseline WAZ was associated with decreased risk of in-hospital death in children with SM across the two sites (OR [95% CI] = 0.70 [0.51, 0.95], p=0.02), with no significant interaction between WAZ and site. Baseline HAZ and WHZ were not significantly associated with in-hospital mortality (OR [95% CI]) = 0.84 [0.66, 1.06] and 0.77 [0.57, 1.02], respectively). During the post-discharge period, 23 (4.4%) children with SM and 1 (0.9%) CC died prior to 12-month follow-up. Nutritional marker z-scores were not significantly associated with risk of post-discharge mortality in children with SM, overall and by site. However, children in Kampala who were underweight had increased odds of post-discharge mortality compared to those who were normal weight (OR [95% CI] = 4.18 [1.36, 12.84], p=0.01). Among those who survived in Kampala, higher WAZ was associated with increased risk of returning to clinic for any cause within 12-month follow-up (HR [95% CI] = 1.16 [1.06, 1.28], p=0.002), but was not significant for malaria sick visits or readmission. HAZ and WHZ in Kampala and all nutritional markers in Jinja were not significantly associated with return sick visits or readmission. Conclusion: Underweight and stunting were worse in both sites among children with SM versus the controls at 1 month, and both of these nutritional parameters normalized by 12 months. In Kampala, low WAZ was associated with worse mortality outcomes after discharge, while high WAZ was associated with repeat clinic visits. In both sites, high WAZ was protective against in-hospital mortality. Chronic malnutrition and SM remain severe risk factors for mortality in Uganda. Weight status was found to have the most significant impact on mortality outcomes. The high incidence of mortality among children <5 with SM requires urgent intervention, and nutrition programs should be aimed at increasing weight, especially in the early months of disease.