Background. Severe malarial anemia (SMA) remains a major cause of pediatric illness and mortality in Sub-Saharan Africa. Here we test the hypothesis that prenatal exposures, reflected by soluble inflammatory mediators in cord blood, can condition an individual's susceptibility to SMA. Methods. In a Tanzanian birth cohort (n = 743), we measured cord blood concentrations of tumor necrosis factor (TNF), TNF receptors I and II (TNF-RI and TNF-RII), interleukin (IL)-1β, IL-4, IL-5, IL-6, IL-10, and interferon gamma (IFN-γ). After adjusting for conventional covariates, we calculated the hazard ratios (HR) for time to first SMA event with log(e) cytokine concentrations dichotomized at the median, by quartile, and per standard deviation (SD) increase. Results. Low levels of TNF, TNF-RI, IL-1β, and IL-5 and high levels of TNF-RII were associated statistically significantly and respectively with approximately 3-fold, 2-fold, 8-fold, 4-fold, and 3-fold increased risks of SMA (Hb < 50 g/L). TNF, TNF-RI, and IL-1β concentrations were inversely and log-linearly associated with SMA risk; the HR (95% confidence interval [CI]) per 1-SD increase were respectively 0.81 (.65, 1.02), 0.76 (.62,. 92), and 0.50 (.40,. 62). Conclusions. These data suggest that proinflammatory cytokine levels at birth are inversely associated with SMA risk and support the hypothesis that pediatric malarial disease has fetal origins.
