文献速递血栓与止血

西方HA-VTE风险模型在东亚人群中的前瞻性多中心验证

Performance of Western HA-VTE risk models in East Asian populations: a prospective multicenter study.
PubMed 文献 · Blood Adv · 2026-01-01

该研究前瞻性验证了5种医院相关静脉血栓栓塞症(HA-VTE)风险模型在2948例中国成人住院患者中的表现,90天HA-VTE发生率为2.54%,所有模型区分度中等(AUC 0.726-0.759)。Caprini敏感性最高(0.83)但42.6%患者被归为高风险,Padua和Intermountain特异性较好(0.75-0.81)但敏感性较低(0.64-0.67),外科患者中Caprini表现最佳但输入变量过多。

要点

  • 5种西方HA-VTE模型在中国住院人群中区分度中等(AUC 0.726-0.759)
  • Caprini敏感性最高但高风险分类比例达42.6%,Padua和Intermountain特异性更优
  • 外科患者中Caprini表现最佳(AUC 0.804),但超过40个输入变量限制实用

核心结论

  • 西方HA-VTE模型在东亚住院人群中具有中等预测价值。
  • 模型性能在不同亚组间存在差异,需要本地化重校准和开发更简单的人群特异性评分。

研究设计与样本

  • 前瞻性多中心队列研究,纳入中国17家医院的2948例成人住院患者。
  • 验证5种HA-VTE风险评分(Caprini、Padua、IMPROVE、Kucher、Intermountain),随访3个月记录症状性VTE事件。

主要发现

  • 90天HA-VTE发生率为2.54%。
  • 所有5种模型区分度中等,AUC为0.726-0.759。
  • Caprini敏感性最高(0.83),但42.6%患者被归为高风险;Padua和Intermountain特异性为0.75-0.81,敏感性为0.64-0.67。
  • 外科住院患者中Caprini表现最佳(AUC 0.804,敏感性0.90,阴性预测值1.00),但超过40个输入变量限制实际使用。

关键数值与推荐

  • 90天HA-VTE发生率为2.54%。
  • 5种模型AUC范围为0.726-0.759。
  • Caprini敏感性0.83,高风险分类比例42.6%;Padua和Intermountain特异性0.75-0.81,敏感性0.64-0.67。
  • 外科患者中Caprini的AUC为0.804,敏感性0.90,阴性预测值1.00。
  • 预防亚组中Caprini的AUC为0.682。

临床怎么用

  • 临床选择模型时应根据敏感性与实用性的具体平衡来决定。
  • 外科住院患者可考虑Caprini模型,但需注意其输入变量较多。
  • 在东亚人群中应用西方模型时需考虑本地化重校准。

局限性

  • 模型性能在不同亚组间存在差异,预防亚组中性能衰减。
  • Caprini模型输入变量超过40个,限制实际使用。

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

Hospital-associated venous thromboembolism (HA-VTE) causes significant morbidity, but risk prediction models are inconsistently validated, especially in non-Western populations. We prospectively validated 5 HA-VTE risk scores (Caprini, Padua, IMPROVE [International Medical Prevention Registry on Venous Thromboembolism], Kucher, and Intermountain) in a prospective multicenter cohort of 2948 adult inpatients from 17 Chinese hospitals. Symptomatic VTE events were recorded through 3 months of follow-up. The 90-day HA-VTE incidence was 2.54%. All 5 models showed moderate discrimination (area under the receiver operating characteristic curve [AUC], 0.726-0.759). Caprini, Padua, and Intermountain models did not differ significantly (DeLong test, P> .05). Caprini had the highest sensitivity (0.83) but classified 42.6% of patients as being at high risk. Padua and Intermountain favored specificity (0.75-0.81) with lower sensitivity (0.64-0.67). Clinical model selection should align with specific balances between sensitivity and utility. In surgical inpatients, Caprini performed best (AUC, 0.804; sensitivity, 0.90; negative predictive value, 1.00), although its more than 40 input variables limit practical use. Calibration was generally acceptable. Performance varied across subgroups. The prophylaxis subgroup showed attenuated performance, with reduced discrimination (eg, Caprini AUC decreased to 0.682) and less stable calibration. Western HA-VTE models have moderate predictive value in East Asian inpatient populations. Performance differs across subgroups. Local recalibration and development of simpler, population-specific scores are needed.

来源:Blood Adv,查看原文。版权归原作者所有。

引用

  1. Li M, Duan Y, Chen Z, et al. Performance of Western HA-VTE risk models in East Asian populations: a prospective multicenter study. Blood Adv. 2026. PMID: 42480218. doi:10.1182/bloodadvances.2026019933

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。