Incidence and predictors of acute kidney injury in children with severe malaria

Background Acute kidney injury (AKI) is an underrecognized complication of severe malaria and an independent risk factor for mortality among children. Objective To determine the incidence and factors predictive of AKI as defined by the pediatric risk, injury, failure, loss, and end-stage (pRIFLE) criteria in children with severe malaria and to assess inhospital mortality rates in malarial AKI (MAKI). Methods This was a prospective cohort study in 170 children aged 0.5 to 14 years with confirmed Plasmodium falciparum on peripheral blood smears and clinical and/or laboratory features of severe malaria. Serum creatinine was determined using the Jaffe method and glomerular filtration rate (eGFR) was estimated using the Schwartz equation. The primary outcome was the incidence of AKI as defined by the pRIFLE criteria. Secondary outcomes included in-hospital mortality comparison between AKI and non-AKI groups, as well as factors predictive of AKI. Results The incidence of MAKI was 61.2% (104/170) and was comparable between males (66.7%) and females (70.6%). Mean eGFR was lower among children with AKI than those without [42.00 (SD 22) vs. 98.7 (SD 3.9) mL/min/1.73m(2), respectively; P=0.005]. Children with MAKI were categorized as having risk (47/104; 45.2%), injury (33/104; 31.7%), or failure (24/104; 23.1%). Mortality rates in AKI and non-AKI subjects were comparable (4.8% vs. 4.6%; P=0.888). Predictors of MAKI were hemoglobinuria [adjusted OR (aOR) 3.948; 95%CI 1.138 to 8.030], deep acidotic breathing (aOR 2.991; 95%CI 3.549 to 66.898), and longer hospital stay (aOR 2.042; 95%CI 3.617 to 12.156). Children with MAKI were more likely to have a longer hospital stay by a mean of 2.5 days. Conclusion Acute kidney injury is a common complication in children with severe malaria. Children with MAKI have a mortality rate comparable to those with severe malaria hut without AKI. Hemoglobinuria, deep acidotic breathing. and longer hospital stay were predictive of MAKI.