To explore the predicting performance of renal resistive index (RRI), semi quantitative power Doppler ultrasound (PDU) score and serum cystatin C (Cys C) for acute kidney injury (AKI) in patients with cardiac failure or sepsis. A prospective, observational study was conducted. Critically ill patients with acute cardiac failure or sepsis admitted to the emergency intensive care unit (ICU) of Cangzhou Central Hospital from January 1st to December 31st in 2018 were enrolled. In addition to the demographic data, serum Cys C, RRI, and PDU score were measured within 6 hours after admission to ICU. Renal function was assessed on day 5 according to Kidney Disease: Improving Global Outcomes (KDIGO) criteria. Patients who proceeded to AKI stage 2 or 3 within 5 days from admission were defined as the AKI 2-3 group; other patients were classified into the AKI 0-1 group. The differences of each index were compared in all patients, cardiac failure patients and sepsis patients between the two groups. Multivariate binary Logistic regression was carried out to identify the independent risk predictors of AKI 2-3. Receiver operator characteristic (ROC) curves were plotted to examine the values of Cys C, RRI, PDU score, and RRI+PDU in predicting AKI 2-3. Thirty-seven patients with cardiac failure (11 with no AKI, 10 with AKI stage 1, 3 with AKI stage 2, and 13 with AKI stage 3) and 26 patients with sepsis (8 with no AKI, 2 with AKI stage 1, 7 with AKI stage 2, and 9 with AKI stage 3) were recruited. In all patients as well as the subgroup of cardiac failure, compared with the AKI 0-1 group, acute physiology and chronic health evaluation II (APACHE II) score, sequential organ failure assessment (SOFA) score, rate of continuous renal replacement therapy (CRRT), 28-day mortality, serum creatinine (SCr), Cys C and RRI were higher in AKI 2-3 group, and urine output, PDU score were lower; in the subgroup of sepsis, rate of CRRT, SCr, and Cys C were higher in AKI 2-3 group, and urine output was lower. Multivariate Logistic regression analysis found that Cys C and PDU score were independent risk factors for AKI 2-3 in all patients [Cys C: odds ratio (OR) = 11.294, 95% confidence interval (95%CI) was 2.801-45.541, P = 0.001; PDU score: OR = 0.187, 95%CI was 0.056-0.627, P = 0.007]; RRI and PDU score were independent risk factors for AKI 2-3 in patients with cardiac failure [RRI (×10): OR = 6.172, 95%CI was 0.883-43.153, P = 0.067; PDU score: OR = 0.063, 95%CI was 0.007-0.584, P = 0.015]; Cys C was the independent risk factor for AKI 2-3 in patients with sepsis (OR = 22.830, 95%CI was 1.345-387.623, P = 0.030). It was shown by ROC curve analysis that: in the subgroup of cardiac failure, the predictive values of RRI, PDU score and Cys C were well [area under the curve (AUC) and 95%CI was 0.839 (0.673-0.942), 0.894 (0.749-0.971), 0.777 (0.610-0.897), all P < 0.01]. RRI+PDU performed best in predicting AKI (AUC = 0.956, 95%CI was 0.825-0.997, P < 0.01), and the predictive value was higher than Cys C [AUC (95%CI): 0.956 (0.825-0.997) vs. 0.777 (0.610-0.897), P = 0.034]. In the subgroup of sepsis, the predictive value of Cys C was well (AUC = 0.913, 95%CI was 0.735-0.987, P < 0.01), however, the predictive value of RRI, PDU, RRI+PDU were poor. RRI and PDU score effectively predict AKI stage 2 or 3 in cardiac failure patients, but not in patients with sepsis. The predictive values of Cys C for AKI are similar in patients with cardiac failure or sepsis.
Read full abstract