Comparison of the application value of three diagnostic and staging criteria for acute kidney injury in patients with preexisting kidney dysfunction
10.3760/cma.j.cn114656-20240424-00298
- VernacularTitle:三种急性肾损伤诊断和分期标准在既往肾功能障碍患者中应用价值的比较
- Author:
Dejiang HONG
1
;
Ze CHEN
;
Guangju ZHAO
;
Jiang LUO
Author Information
1. 温州医科大学附属第一医院急诊科，温州 325000
- Keywords:
Pre-existing kidney dysfunction;
Acute kidney injury;
Kidney Disease;
Improving Global Outcomes;
Acute on preexisting kidney dysfunction;
Diagnostic and st
- From:
Chinese Journal of Emergency Medicine
2025;34(8):1106-1112
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To compare the diagnostic, staging, and prognostic performance of three AKI criteria (KDIGO, cROCK, and APKD) in patients with pre-existing kidney dysfunction (PKD).Methods:Clinical data from 570 PKD patients admitted to the ICU of the First Affiliated Hospital of Wenzhou Medical University (October 2020-October 2022) were analyzed. Patients were classified by three AKI diagnostic criteria. We compared AKI detection rates and mortality predictions using ROC analysis and logistic regression (reporting ORs with 95% CIs). Results:AKI detection rates differed significantly: APKD 80.7% (460/570), KDIGO 66.0% (376/570), and cROCK 47.0% (268/570) (both P<0.001 vs. APKD). Mortality rates were comparable across criteria (APKD 32.4%, KDIGO 33.0%, cROCK 33.6%; P>0.85). APKD showed superior predictive value for in-hospital mortality (AUC=0.736, 95% CI 0.691-0.781), with significantly higher sensitivity than KDIGO (0.738 vs. 0.663, P<0.001) and cROCK (0.738 vs. 0.613, P<0.001). KDIGO staging identified few AKI2 cases (1.1%), with mortality rates of 27.3% (AKI1), 0% (AKI2), and 37.0% (AKI3). APKD staging demonstrated graded mortality risks: 20.1% (AKI1), 28.1% (AKI2), and 40.9% (AKI3) (all P<0.05). Multivariate analysis revealed increasing mortality risk with APKD stage (AKI1: OR=2.34, 95% CI 1.09-5.31; AKI2: OR=4.50, 95% CI 2.05-10.46; AKI3: OR=7.03, 95% CI 3.59-15.01). APKD staging outperformed KDIGO in mortality prediction (AUC=0.756 vs. 0.724, P=0.02). Conclusions:The APKD criteria demonstrate superior AKI detection in PKD patients and better prognostic prediction compared to KDIGO and cROCK. APKD staging provides more clinically meaningful risk stratification than KDIGO for this population.
