老年营养风险指数对脓毒症相关急性肾损伤患者28天死亡风险有预测作用

张英.老年营养风险指数对脓毒症相关急性肾损伤患者28天死亡风险有预测作用[J].内科急危重症杂志,2025,31(4):343-347

DOI：10.11768/nkjwzzzz20250411

英文关键词:

基金项目:甘肃省卫生健康行业科研计划项目（GSWSKY2022-51）

作者	单位	E-mail
张英	中国人民解放军联勤保障部队第九四〇医院	chunshxi@sina.com

摘要点击次数: 10

全文下载次数: 12

中文摘要:

摘要目的：探讨老年营养风险指数(GNRI)对脓毒症相关急性肾损伤(SAKI)患者28d死亡风险的预测价值。方法：回顾性收集美国重症监护医学信息数据库（MIMIC-IV）及联勤保障部队第九四〇医院的SAKI患者资料。采用logistic回归分析确定SAKI患者28d死亡风险因素。采用受试者工作特征（ROC）曲线、净重分类改进（NRI）与综合判别改进（IDI）确定GNRI对SAKI患者28d死亡风险的附加预测价值。结果：从MIMIC-IV数据库中共提取1843例患者资料，其中GNRI判定存在营养风险946例。多因素Logistic回归分析显示，GNRI判定存在营养风险是SAKI患者28d死亡的独立危险因素(中营养风险，P=0.007；高营养风险，P＜0.001)。限制性立方样条分析显示，GNRI与28d死亡呈现非线性相关。ROC曲线、NRI与IDI分析均表明，联合GNRI后预测模型的预测能力显著提高（P均＜0.001）。来自国内的验证队列同样证实了GNRI对SAKI患者28d死亡的预测价值。结论：GNRI是SAKI患者28d死亡风险的潜在预测因子，将GNRI纳入预测模型中可显著提高其预测能力。

英文摘要:

Abstract Objective: To investigate the prognostic value of geriatric nutritional risk index (GNRI) for the 28-day mortality risk of patients with sepsis-associated acute kidney injury (SAKI). Methods: The data of SAKI patients included in the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database and the 940th Hospital of Joint Logistics Support Force were retrospectively collected. Logistic regression analysis was used to determine the risk factors of 28-day mortality in SAKI patients. The additional predictive value of GNRI for 28-day mortality in SAKI patients was determined by receiver operating characteristic (ROC) curve, net reclassification improvement (NRI) and integrated discrimination improvement (IDI). Results: A total of 1843 patients were extracted from the MIMIC-IV database, of which 946 were at nutritional risk according to GNRI. Multivariate Logistic regression analysis showed that the nutritional risk according to GNRI was an independent risk factor for 28-day mortality in SAKI patients (Moderate risk: P= 0.007; Severe risk: P< 0.001). Restricted cubic spline analysis showed a non-linear relationship between GNRI and 28-day mortality. ROC curve, NRI and IDI analysis all showed that the predictive ability of the combined GNRI model was significantly improved (all P< 0.001). A validation cohort from China similarly confirmed the predictive value of GNRI for 28-day mortality in SAKI patients. Conclusion: GNRI is a potential predictor of 28-day mortality in SAKI patients, and the addition of GNRI to prediction models could improve the predictive accuracy.