The association between nutritional status and malaria in children from a rural community in the Amazonian region: a longitudinal study.

Abstract

BACKGROUND:The relationship between malaria and undernutrition is controversial and complex. Synergistic associations between malnutrition and malaria morbidity and mortality have been suggested, as well as undernutrition being protective against infection, while other studies found no association. We sought to evaluate the relationship between the number of malaria episodes and nutritional statuses in a cohort of children below 15 years of age living in a rural community in the Brazilian Amazon. METHODOLOGY/PRINCIPAL FINDINGS:Following a baseline survey of clinical, malaria and nutritional assessment including anthropometry measurements and hemoglobin concentration, 202 children ranging from 1 month to 14 years of age were followed for one year through passive case detection for malaria episodes. After follow-up, all children were assessed again in order to detect changes in nutritional indicators associated with malaria infection. We also examined the risk of presenting malaria episodes during follow-up according to presence of stunting at baseline. Children who suffered malaria episodes during follow-up presented worse anthropometric parameters values during this period. The main change was a reduction of the linear growth velocity, associated with both the number of episodes and how close the last or only malaria episode and the second anthropometric assessment were. Changes were also observed for indices associated with chronic changes, such as weight-for-age and BMI-for-age, which conversely, were more frequently observed in children with the last or only episode occurring between 6 and 12 months preceding the second nutritional assessment survey. Children with inadequate height-for-age at baseline (Z-score < -2) presented lower risk of suffering malaria episodes during follow-up as assessed by both the log-rank test (p =0.057) and the multivariable Cox-proportional hazards regression (Hazard Ratio = 0.31, 95%CI [0.10; 0.99] p=0.049). CONCLUSIONS:Malaria was associated with impaired nutritional status amongst children in an endemic area of the Western Brazilian Amazon where P. vivax predominates. Our data all supports that the association presents differential effects for each age group, suggesting distinct pathophysiology pathways. We were also able to demonstrate that undernourishment at baseline was protective to malaria during follow-up. These findings support an intriguing interaction between these conditions in the rural Amazon and the need for a more integrative approach by health systems in endemic areas.