Diagnostic Accuracy of Renal Angina Index in Predicting Acute Kidney Injury in Pediatric Patients with Sepsis: A Philippine Tertiary Hospital Experience

Abstract

Background: The coexistence of acute kidney injury (AKI) in sepsis contributes significantly to morbidity and mortality rates. Traditional diagnostic markers still pose variable limitations in early AKI prediction. The use of renal angina index (RAI) as a clinical predictive tool for AKI is an emerging concept. Objectives: To determine the diagnostic accuracy of RAI in predicting AKI in patients with sepsis Methodology: This is a five-year retrospective cohort study conducted at the Philippine General Hospital (PGH). Records of eligible patients with sepsis were reviewed. RAI was calculated based on the composite of risk factors and clinical evidence of injury on day 0 of admission stratifying subjects into two groups: RAI (-) and RAI (+) for those with scores ≥ 8. Prediction of AKI with the RAI was analyzed. Results: A total of 222 patients were enrolled. The RAI (+) group (score ≥ 8) consisted 95 patients (43%). AKI incidence rate was 40.5 % (90/222) and 87/90 patients (91.6%) were classified in the RAI (+) group. The use of RAI in predicting AKI has a sensitivity of 96.7%, specificity of 94.0%, positive predictive value (PPV) of 91.6%, negative predictive value (NPV) of 97.7%, positive likelihood ratio (LR) of 15.95, negative LR of 0.04 and area under the curve-receiver operating characteristic (AUC-ROC) of 0.953 (95% CI 0.92-0.98). Conclusions: RAI is a good screening tool in predicting sepsis-associated AKI among pediatric patients. It provides early recognition of AKI and is a practical method which can be used at bedside.