Early detection of acute kidney injury in critically ill children: Predictive value of renal arterial Doppler assessment

Abstract

ObjectiveRenal resistive index (RRI) and renal pulsatility index (RPI) are Doppler-based variables proposed to assess renal perfusion at the bedside in critically ill patients. This study aimed to assess the accuracy of such variables to predict acute kidney injury (AKI) in mechanically ventilated children.DesignProspective single-center observational studySettingPediatric intensive care unit of a quaternary care teaching hospital.Patients84 children under controlled ventilation (median age of 5.1 months and weight of 6.6 kg).InterventionsConsecutive children underwent renal Doppler ultrasound examination within 24 hours of invasive mechanical ventilation. Renal resistive index (RRI) and renal pulsatility index (RPI) were measured. The primary outcome was severe AKI (KDIGO stage 2 or 3) on day 3. Secondary outcomes included the persistence of severe AKI on day 5.ResultsOn day 3, 22 patients were classified as having AKI (any stage), of which 12 were severe. RRI could effectively predict severe AKI (area under the ROC curve [AUC] 0.94; 95% CI 0.86 – 0.98; p < 0.001) as well as RPI (AUC 0.86; 95% CI 0.76 – 0.92; p < 0.001). The AUC of the IRR was significantly greater than that obtained from the RPI (p = 0.023). The optimal cutoff for RRI was 0.85 (sensitivity, 91.7%; specificity, 84.7%; positive predictive value, 50.0%; and negative predictive value 98.4%). Similar results were obtained when the accuracy to predict AKI on day 5 was assessed. Significant correlations were observed between RRI and estimated glomerular filtration rate at enrollment (ρ = -0.495, p<0.001) and on day 3 (ρ = -0.467, p <0.001).ConclusionsRenal Doppler ultrasound may be a promising tool to predict AKI in critically ill children under invasive mechanical ventilation.