用简化临床数据算法识别以脂蛋白代谢失调为特征的脓毒症亚表型
该研究分析四项前瞻性研究(内部数据集)的386例脓毒症患者,构建1000棵树的随机森林分类器和logistic回归模型,用临床特征预测HYPO与NORMO亚表型。四个临床特征(肝脏SOFA、心血管SOFA、低LDL-C、低HDL-C)预测亚表型的AUC为0.86,敏感性0.771,特异性0.779。内部数据集28天死亡率HYPO 26% vs NORMO 15%,外部队列(法国三项研究281例)为30% vs 10%;HYPO患者LDL-C、HDL-C和总胆固醇均低于NORMO患者。作者认为该简化算法可能实现床旁识别脂质失调亚表型,但需外部验证。
要点
- 四个临床特征(肝脏SOFA、心血管SOFA、低LDL-C、低HDL-C)预测HYPO vs NORMO亚表型,AUC 0.86,敏感性0.771,特异性0.779。
- 内部数据集28天死亡率HYPO 26% vs NORMO 15%;外部队列(281例)HYPO 30% vs NORMO 10%。
- HYPO亚表型特征为脂蛋白水平更低、死亡率更高;NORMO为脂蛋白水平更高、死亡率更低。
- HYPO患者LDL-C、HDL-C和总胆固醇在内部与外部数据集中均低于NORMO患者。
- 作者结论:该简化临床数据算法可能允许床旁识别脂质失调亚表型,但需外部验证。
适用人群与场景
- 脓毒症患者,内部数据集来自四项前瞻性研究共386例(中位年龄63岁,46%女性),外部数据集来自法国三项研究共281例。
诊断标准与关键阈值
- 亚表型定义:HYPO为脂蛋白水平更低且死亡率更高的亚表型;NORMO为脂蛋白水平更高且死亡率更低的亚表型。
- 预测模型使用四个临床特征:肝脏SOFA、心血管SOFA、低LDL-C、低HDL-C。
研究设计与关键结果
- 研究类型:基于四项前瞻性研究内部数据集的分析,并应用外部数据集验证;内部队列386例,外部队列281例(来自法国三项研究)。
- 模型构建:1000棵树的随机森林分类器和logistic回归模型,用临床特征预测亚表型,并与机器学习模型导出的实际亚表型比较。
- 主要结果:四个临床特征预测HYPO vs NORMO的AUC为0.86,敏感性0.771,特异性0.779。
- 28天死亡率:内部数据集HYPO 26% vs NORMO 15%;外部队列HYPO 30% vs NORMO 10%。
- 脂蛋白水平:HYPO内部 vs 外部数据集LDL-C水平相似(P=0.99),HDL-C水平不同(P=0.02);NORMO内部 vs 外部数据集LDL-C(P=0.99)和HDL-C(P=0.12)水平相似。
- HYPO患者LDL-C、HDL-C和总胆固醇在内部和外部数据集中均低于NORMO患者。
证据类型、等级与局限
- 证据类型:基于四项前瞻性研究数据的模型开发与外部验证分析。
- 局限:外部验证是必要的,以核实这些结果。
对床旁实践的改变
- 作者认为,该简化临床数据算法可能允许床旁识别显示脂质失调亚表型的脓毒症患者。
- 作者指出,需外部验证以核实这些结果。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
Background: Cholesterol metabolism is dysregulated in sepsis contributing to patient heterogeneity. Subphenotypes displaying lower lipoprotein levels and higher mortality (previously subphenotyped hypolipoprotein phenotype [HYPO]) or higher lipoprotein levels and lower mortality (previously subphenotyped normolipoprotein phenotype [NORMO]) were described. We developed a simplified clinical algorithm for bedside subphenotype recognition. Methods: We analyzed data from four prospective studies (internal dataset), focusing on HYPO and NORMO subphenotypes. A 1,000-tree random forest classifier and logistic regression models were built, using clinical features to predict subphenotypes. Performance was evaluated by comparing predictions to actual subphenotypes derived from a machine learning model. The model was applied to an external dataset of 281 patients from three French studies. Results: The internal cohort consisted of 386 patients (median age, 63 years; 46% female). Four clinical features (hepatic SOFA, cardiovascular SOFA, low [low-density lipoprotein cholesterol {LDL-C}] and high-density lipoprotein cholesterol [high-density lipoprotein cholesterol {HDL-C}]) predicted HYPO versus NORMO subphenotypes with an area under the receiver operating characteristic curve of 0.86, a sensitivity of 0.771, and a specificity of 0.779. In the internal dataset, 28-day mortality for HYPO versus NORMO patients was 26% versus 15%, and in the external cohort, 30% versus 10%. HYPO internal versus external dataset LDL-C levels were similar ( P = 0.99), but HDL-C ( P = 0.02) levels were different. Median NORMO internal versus external dataset LDL-C ( P = 0.99) and HDL-C ( P = 0.12) levels were similar. HYPO patients had lower LDL-C, HDL-C and total cholesterol than NORMO patients in both internal and external datasets. Conclusions: Our simplified clinical data algorithm may allow for bedside recognition of septic patients displaying lipid dysregulation subphenotypes. External validation is needed to verify these results.
参考来源
- Labilloy G, Tanaka S, Black LP, et al. IDENTIFYING A SEPSIS SUBPHENOTYPE CHARACTERIZED BY DYSREGULATED LIPOPROTEIN METABOLISM USING A SIMPLIFIED CLINICAL DATA ALGORITHM. Shock. 2025. PMID: 40267485. doi:10.1097/SHK.0000000000002605
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。