血清肌酐/胱抑素C比值诊断假性急性肾损伤的初步研究
本研究纳入239例患者(197例真性AKI,42例假性AKI),评估血清肌酐/胱抑素C比值(CrCyR)鉴别假性AKI与真性AKI的诊断性能。假性AKI组中位CrCyRpost显著高于真性AKI组(1.89 vs 0.78 l/dl,P<0.001),AUC为0.97(95%CI 0.95-1.00),探索性阈值>1.11 l/dl时灵敏度95%、特异度91%。
要点
- 假性AKI患者CrCyRpost中位数显著高于真性AKI(1.89 vs 0.78 l/dl,P<0.001)。
- CrCyRpost诊断假性AKI的AUC为0.97(95%CI 0.95-1.00)。
- 探索性阈值>1.11 l/dl时灵敏度95%、特异度91%。
核心结论
- CrCyRpost对鉴别假性AKI与真性AKI初步显示出良好的诊断准确性,仍需前瞻性研究验证。
研究设计与样本
- 研究纳入南京医科大学第一附属医院2023年5月至2024年8月确诊AKI(排除横纹肌溶解)患者及1994-2025年文献综述识别的假性AKI病例,共239例(真性AKI 197例,假性AKI 42例),采用ROC曲线下面积(AUC)评估CrCyRpost诊断性能。
主要发现
- 假性AKI组CrCyRpost中位数1.89(IQR 1.32-3.28)l/dl,显著高于真性AKI组的0.78(0.64-0.94)l/dl(P<0.001)。
- CrCyRpost的AUC为0.97(95%CI 0.95-1.00),年龄和性别调整后AUC为0.98(0.96-1.00)。
- 各假性AKI亚型AUC:检测干扰、尿液渗漏和肌酐生成增加均为1.00,肌酐排泄减少为0.94。
- 使用Salzburg Intensive Care数据库真性AKI患者进行跨中心稳健性评估,欧洲人群AUC为0.94(0.90-0.98)。
关键数值与推荐
- 探索性阈值CrCyRpost > 1.11 l/dl诊断假性AKI的灵敏度95%、特异度91%。
- 假性AKI机制分布:肌酐排泄减少18例,检测干扰14例,尿液渗漏7例,肌酐生成增加3例。
临床怎么用
- CrCyRpost可作为鉴别假性AKI与真性AKI的初步辅助工具,探索性阈值>1.11 l/dl时具有较高灵敏度和特异度。
局限性
- 本研究为概念验证性研究,假性AKI病例来自文献综述,样本量有限,阈值仅为探索性,需前瞻性研究进一步验证。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
INTRODUCTION: We conducted a proof-of-concept study to evaluate the diagnostic performance of the serum creatinine/cystatin C ratio (CrCyR) in distinguishing pseudo-acute kidney injury (AKI) from true AKI. METHODS: This study included patients with confirmed AKI (excluding rhabdomyolysis) at the First Affiliated Hospital of Nanjing Medical University (May 2023-August 2024) and pseudo-AKI cases identified through a literature review (1994-2025). CrCyR values (unit: l/dl) measured before (CrCyRpre) and during AKI onset (CrCyRpost) were collected. Diagnostic performance of CrCyRpost was assessed using the area under the receiver-operating-characteristic curve (AUC). RESULTS: Of the 239 patients, 197 had true AKI and 42 had pseudo-AKI (mechanisms: reduced creatinine excretion [n = 18], assay interference [n = 14], urine leakage [n = 7], and increased creatinine production [n = 3]). Median CrCyRpost was significantly higher in the pseudo-AKI group than in the true AKI group (1.89 [Inter Quartile Range {IQR}: 1.32-3.28] vs. 0.78 [0.64-0.94] l/dl; P < 0.001). CrCyRpost showed a preliminary AUC of 0.97 (95% confidence interval [CI]: 0.95-1.00); an exploratory threshold of > 1.11 l/dl yielded 95% sensitivity and 91% specificity for diagnosing pseudo-AKI. After adjustment for age and sex, the covariate-adjusted AUC remained high at 0.98 (0.96-1.00). Diagnostic utility was consistently high across pseudo-AKI subtypes (AUCs: 1.00 for assay interference, urine leakage, and increased creatinine production; 0.94 for reduced creatinine excretion). A cross-center robustness assessment using patients with true AKI from the Salzburg Intensive Care database suggested consistent diagnostic utility (AUC: 0.94 [0.90-0.98]) in Europeans. CONCLUSION: CrCyRpost preliminarily demonstrates promising diagnostic accuracy for distinguishing pseudo-AKI from true AKI. Further validation in prospective studies is warranted.
引用
- Jiang R, Huang Y, Deng L, et al. Pilot Study of Diagnostic Performance of Serum Creatinine/Cystatin C Ratio for Pseudo-AKI. Kidney Int Rep. 2026. PMID: 42058115. doi:10.1016/j.ekir.2026.106512
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。