指南与共识消化、代谢与肾脏急症重症综合管理与镇静镇痛

基于机器学习共识聚类的高钠血症住院患者亚型及其生存差异

Hypernatremia subgroups among hospitalized patients by machine learning consensus clustering with different patient survival.
指南 / 共识 · J Nephrol · 2021-10-08
一句话结论入院高钠血症患者经机器学习聚类可分为三种临床表型,其住院及1年死亡风险显著不同,集群1风险最高。

本研究对922例入院血清钠>145 mEq/L的成人住院患者,基于人口学、主要诊断、合并症和实验室数据进行共识聚类分析。识别出三个集群:集群1(318例,34%)病情更重、高钠更严重且伴低钾性高氯性代谢性酸中毒;集群2(339例,37%)年龄更大、合并症更多、BMI更高及代谢性碱中毒;集群3(265例,29%)较年轻、合并症少、基线eGFR、血红蛋白和血清白蛋白更高。与集群3相比,集群1住院死亡OR 15.74(95%CI 3.75-66.18),1年死亡HR 6.25(95%CI 3.69-11.46);集群2住院死亡OR 6.51(95%CI 1.48-28.59),1年死亡HR 4.66(95%CI 2.73-8.59)。

要点

  • 入院高钠血症(血清钠>145 mEq/L)患者可分为三个临床表型,集群1病情最重且死亡风险最高。
  • 集群1特征:更危重、高钠更严重、低钾性高氯性代谢性酸中毒;住院死亡OR 15.74(95%CI 3.75-66.18) vs 集群3。
  • 集群2特征:年龄更大、合并症更多、BMI更高、代谢性碱中毒;住院死亡OR 6.51(95%CI 1.48-28.59) vs 集群3。
  • 集群3特征:较年轻、合并症少、基线eGFR、血红蛋白和血清白蛋白更高,死亡风险最低。
  • 1年死亡风险:集群1 HR 6.25(95%CI 3.69-11.46),集群2 HR 4.66(95%CI 2.73-8.59),均以集群3为参照。

适用人群与场景

  • 入院时血清钠>145 mEq/L的成人住院患者(共922例)。

诊断标准与关键阈值

  • 高钠血症定义:入院血清钠>145 mEq/L。

研究设计与关键结果

  • 研究类型:共识聚类分析(无监督机器学习),样本量922例成人住院患者。
  • 集群1(318例,34%):更危重、高钠更严重、低钾性高氯性代谢性酸中毒;集群2(339例,37%):年龄更大、合并症更多、BMI更高、代谢性碱中毒;集群3(265例,29%):较年轻、合并症少、基线eGFR、血红蛋白和血清白蛋白更高。
  • 与集群3相比,住院死亡OR:集群1 15.74(95%CI 3.75-66.18),集群2 6.51(95%CI 1.48-28.59)。
  • 与集群3相比,1年死亡HR:集群1 6.25(95%CI 3.69-11.46),集群2 4.66(95%CI 2.73-8.59)。

证据类型、等级与局限

  • 证据类型:观察性研究(共识聚类分析)。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 作者认为,聚类分析识别出三种临床表型,其死亡风险不同,可能有助于对入院高钠血症患者进行风险分层。

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: The objective of this study was to characterize hypernatremia patients at hospital admission into clusters using an unsupervised machine learning approach and to evaluate the mortality risk among these distinct clusters.

METHODS: We performed consensus cluster analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 922 hospitalized adult patients with admission serum sodium of > 145 mEq/L. We calculated the standardized difference of each variable to identify each cluster's key features. We assessed the association of each hypernatremia cluster with hospital and 1-year mortality.

RESULTS: There were three distinct clusters of patients with hypernatremia on admission: 318 (34%) patients in cluster 1, 339 (37%) patients in cluster 2, and 265 (29%) patients in cluster 3. Cluster 1 consisted of more critically ill patients with more severe hypernatremia and hypokalemic hyperchloremic metabolic acidosis. Cluster 2 consisted of older patients with more comorbidity burden, body mass index, and metabolic alkalosis. Cluster 3 consisted of younger patients with less comorbidity burden, higher baseline eGFR, hemoglobin, and serum albumin. Compared to cluster 3, odds ratios for hospital mortality were 15.74 (95% CI 3.75-66.18) for cluster 1, and 6.51 (95% CI 1.48-28.59) for cluster 2, whereas hazard ratios for 1-year mortality were 6.25 (95% CI 3.69-11.46) for cluster 1 and 4.66 (95% CI 2.73-8.59) for cluster 2.

CONCLUSION: Our cluster analysis identified three clinically distinct phenotypes with differing mortality risk in patients hospitalized with hypernatremia.

来源:J Nephrol,查看原文。版权归原作者所有。

参考来源

  1. Thongprayoon C, Mao MA, Keddis MT, et al. Hypernatremia subgroups among hospitalized patients by machine learning consensus clustering with different patient survival. J Nephrol. 2021. PMID: 34623631. doi:10.1007/s40620-021-01163-2

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。