文献速递血栓与止血

克利夫兰诊所出血模型的开发与内部验证:预测内科住院患者入院时大出血风险

Development and internal validation of the Cleveland Clinic Bleeding Model to predict major bleeding risk at admission in medical inpatients.
PubMed 文献 · J Thromb Haemost · 2024-07-11

研究基于克利夫兰诊所医疗系统10家医院2017至2020年收治的46314例内科住院患者,开发并内部验证了预测院内大出血的风险评估模型。最终模型含16个危险因素,判别能力优于IMPROVE模型(AUC 0.85 vs 0.70,P<.001),在相同灵敏度52%下将更少患者归为高风险(7.2% vs 11.8%,P<.001),校准良好(Brier评分=0.0057)。

要点

  • 克利夫兰诊所出血模型含16个危险因素,预测院内大出血的AUC为0.85。
  • 模型判别能力优于IMPROVE模型(AUC 0.85 vs 0.70,P<.001)。
  • 在灵敏度52%时,模型将7.2%患者归为高风险,IMPROVE为11.8%(P<.001)。

核心结论

  • 利用大样本内科住院患者及经病历核实的大出血事件,开发并内部验证了大出血风险评估模型,其性能优于IMPROVE模型。

研究设计与样本

  • 研究纳入克利夫兰诊所医疗系统10家医院2017至2020年收治的内科住院患者,共46314例次入院。
  • 大出血按国际血栓与止血学会标准、结合诊断代码与实验室值识别,并经病历审查确认。
  • 在训练集中拟合最小绝对收缩选择算子逻辑回归模型,在验证集中比较本模型与IMPROVE模型的判别度和校准度。

主要发现

  • 46314例次入院中,268例(0.58%)发生大出血。
  • 最终模型包含16个危险因素,其中既往出血(OR=4.83)、消化性溃疡(OR=3.82)、脓毒症病史(OR=3.26)和类固醇使用(OR=2.59)最强。
  • 克利夫兰诊所出血模型的判别能力优于IMPROVE模型(受试者工作特征曲线下面积=0.85 vs 0.70;P<.001)。
  • 在相同灵敏度(52%)下,本模型将更少患者归为高风险(7.2% vs 11.8%;P<.001)。
  • 校准良好(Brier评分=0.0057)。

关键数值与推荐

  • 最终模型含16个危险因素,包括既往出血(OR=4.83)、消化性溃疡(OR=3.82)、脓毒症病史(OR=3.26)、类固醇使用(OR=2.59)。
  • 模型判别能力:AUC=0.85,IMPROVE模型AUC=0.70(P<.001)。
  • 在灵敏度52%时,本模型高风险比例为7.2%,IMPROVE模型为11.8%(P<.001)。
  • 校准度:Brier评分=0.0057。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

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

BACKGROUND: Guidelines recommend pharmacologic VTE prophylaxis for acutely ill medical patients at acceptable bleeding risk, but only the International Medical Prevention Registry on Venous Thromboembolism (IMPROVE) model has been validated for bleeding risk assessment. OBJECTIVES: We developed and internally validated a risk assessment model (RAM) to predict major in-hospital bleeding using risk factors at admission and compared our model with IMPROVE. METHODS: We selected patients admitted to medical services at 10 hospitals in the Cleveland Clinic Health System from 2017 to 2020. We identified major bleeding according to the International Society on Thrombosis and Haemostasis criteria, using a combination of diagnostic codes and laboratory values, and confirmed events with chart review. We fit a least absolute shrinkage selection operator logistic regression model in the training set and compared the discrimination and calibration of our model with the IMPROVE model in the validation set. RESULTS: Among 46 314 admissions, 268 (0.58%) had a major bleed. The final RAM included 16 risk factors, of which prior bleeding (odds ratio [OR] = 4.83), peptic ulcer (OR = 3.82), history of sepsis (OR = 3.26), and steroid use (OR = 2.59) were the strongest. The Cleveland Clinic Bleeding Model had better discrimination than IMPROVE (area under the receiver operating characteristics curve = 0.85 vs 0.70; P < .001) and, at equivalent sensitivity (52%), categorized fewer patients as high risk (7.2% vs 11.8%; P < .001). Calibration was adequate (Brier score = 0.0057). CONCLUSION: Using a large population of medical inpatients with verified major bleeding events, we developed and internally validated a RAM for major bleeding whose performance surpassed the IMPROVE model.

来源:J Thromb Haemost,查看原文。版权归原作者所有。

引用

  1. Mittman BG, Sheehan M, Kojima L, et al. Development and internal validation of the Cleveland Clinic Bleeding Model to predict major bleeding risk at admission in medical inpatients. J Thromb Haemost. 2024. PMID: 39002732. doi:10.1016/j.jtha.2024.06.025

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。