Heparin-Binding Protein Stratifies Mortality Risk Among Ugandan Children Hospitalized With Respiratory Distress

Abstract

Abstract Background Current prognostic tools do not reliably and objectively identify children with pneumonia at risk of a severe or life-threatening episode. Heparin-binding protein (HBP) is a host immune protein that is released in response to infection. We hypothesized that measuring HBP concentrations at hospital admission could help risk-stratify children with pneumonia and identify those at higher risk of an adverse prognosis. Methods We evaluated the prognostic accuracy of HBP for predicting in-hospital mortality among children with respiratory distress, and whether HBP could improve the accuracy of validated composite clinical severity scores. Results Of 778 Ugandan children under 5 years of age and presenting with clinically defined pneumonia, 60 (7.7%) died during hospital admission. HBP concentrations at presentation were significantly higher in children with fatal outcomes (median, 76 ng/mL [interquartile range {IQR}, 41–150]) compared to children who survived (median, 31 ng/mL [IQR, 18–57]) (P < .001). Children with HBP >41 ng/mL on admission had an elevated risk of death (hazard ratio, 5.3 [95% confidence interval {CI}, 2.9–9.5]; P < .0001). In receiver operating characteristic (ROC) curve analysis, HBP concentrations distinguished between fatal and nonfatal outcomes (area under the ROC curve, 0.75 [95% CI, .66–.84]) and significantly improved the prediction provided by the Respiratory Index of Severity in Children, a composite clinical severity score (P = .0026). Conclusions Measuring HBP at presentation could help identify children at risk of severe and fatal pneumonia. Adding HBP to clinical scores could improve the recognition and triage of children with pneumonia at risk of death.