生物标志物组合预测急性肾损伤幸存者肾脏病进展
该研究基于723例住院AKI患者的前瞻性队列,利用机器学习分析出院后3个月测量的75个候选预测因子,预测3年内主要不良肾脏事件(MAKE)。随机森林和LASSO模型的AUC分别为0.80和0.79,最一致选择的预测因子为白蛋白尿、可溶性TNF受体-1和利尿剂使用。八变量简化模型AUC为0.78,可将筛查效率提高55%。
要点
- 出院后3个月的白蛋白尿、可溶性TNF受体-1和利尿剂使用是预测MAKE的关键因子。
- 八变量简化模型对MAKE的判别力强(AUC 0.78),需筛查人数从3.55降至1.97。
- 四变量模型对肾功能下降或ESKD的预测AUC达0.82。
核心结论
- 结合临床数据和生物标志物能准确识别高危AKI患者,实现个性化 post-AKI 护理并改善结局。
研究设计与样本
- 前瞻性队列研究,纳入723例住院AKI患者,来自ASSESS-AKI研究。
- 使用机器学习分析75个候选预测因子,包括出院后3个月测量的生物标志物。
主要发现
- 平均年龄64±13岁,68%为男性,79%为白人。
- 随机森林和LASSO模型AUC分别为0.80(95%CI 0.69-0.91)和0.79(95%CI 0.68-0.90)。
- 最一致选择的预测因子为白蛋白尿、可溶性TNF受体-1和利尿剂使用。
- 八变量简化模型AUC为0.78(95%CI 0.66-0.90)。
关键数值与推荐
- MAKE定义为eGFR下降≥40%、发生ESKD或死亡。
- 八变量模型需筛查/随访人数从3.55降至1.97。
临床怎么用
- 可利用出院后3个月的白蛋白尿、可溶性TNF受体-1和利尿剂使用等指标构建预测模型,识别高危AKI幸存者以优化随访。
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
KEY POINTS: Clinical characteristics and biomarkers after hospital discharge can predict major adverse kidney events among AKI survivors. Clinical impact plots based on parsimonious prediction models illustrate the potential to optimize post-AKI care by identifying high-risk patients. BACKGROUND: AKI increases the risk of CKD. We aimed to identify combinations of clinical variables and biomarkers that predict long-term kidney disease risk after AKI. METHODS: We analyzed data from a prospective cohort of 723 hospitalized patients with AKI in the Assessment, Serial Evaluation, and Subsequent Sequelae of AKI study. Using machine learning, we investigated 75 candidate predictors including biomarkers measured at 3-month postdischarge follow-up to predict major adverse kidney events (MAKEs) within 3 years, defined as a decline in eGFR ≥40%, development of ESKD, or death. RESULTS: The mean age of study participants was 64±13 years, 68% were male, and 79% were of White race. Two hundred four patients (28%) developed MAKEs over 3 years of follow-up. Random forest and least absolute shrinkage and selection operator penalized regression models using all 75 predictors yielded area under the receiver-operating characteristic curve (AUC) values of 0.80 (95% confidence interval [CI], 0.69 to 0.91) and 0.79 (95% CI, 0.68 to 0.90), respectively. The most consistently selected predictors were albuminuria, soluble TNF receptor-1, and diuretic use. A parsimonious model using the top eight predictor variables showed similarly strong discrimination for MAKEs (AUC, 0.78; 95% CI, 0.66 to 0.90). Clinical impact utility analyses demonstrated that the eight-predictor model would have 55% higher efficiency of post-AKI care (number needed to screen/follow-up for a MAKE decreased from 3.55 to 1.97). For a kidney-specific outcome of eGFR decline or ESKD, a four-predictor model showed strong discrimination (AUC, 0.82; 95% CI, 0.68 to 0.96). CONCLUSIONS: Combining clinical data and biomarkers can accurately identify patients with high-risk AKI, enabling personalized post-AKI care and improved outcomes.
引用
- Menez S, Kerr KF, Cheng S, et al. Biomarker Panels for Predicting Progression of Kidney Disease in Acute Kidney Injury Survivors. Clin J Am Soc Nephrol. 2024. PMID: 39671257. doi:10.2215/CJN.0000000622
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。