Caprini及Padua风险评估模型评估内科住院患者静脉血栓栓塞症风险有效性研究

Effectiveness of Caprini and Padua Risk Assessment Models for Assessing Risk of Venous Thromboembolism in Internal Medicine Inpatients

摘要: 目的：评估Caprini及Padua两种风险评估模型在内科住院患者静脉血栓栓塞症风险的有效性，同时探讨内科住院患者发生VTE的危险因素。方法：采用回顾性病例对照研究，将2019年1月至2019年12月青岛大学附属医院确诊VTE的115例患者作为VTE组，同时选取同时期同科室的出院诊断非VTE的115例内科住院患者作为对照组，收集2组患者既往病史、实验室检查结果等临床资料，分别对2组患者进行Caprini评分、Padua评分，比较2种风险评估模型评分，并用二元logistic回归模型对内科住院患者VTE危险因素进行分析。结论：Caprini评分和Padua评分中VTE组均高于对照组，其中Caprini评分中VTE组和对照组分别为(6.783 ± 2.481, 3.357 ± 1.650) (t值为12.332，P < 0.05)；在Padua评分中，两组评分为(4.452 ± 1.846, 1.748 ± 1.835) (t值为11.143，P < 0.05)。Caprini评分诊断VTE的ROC曲线下面积分别为(0.890 ± 0.0216) (95% CI为0.842~0.927)，Padua评分诊断VTE的ROC曲线下面积为(0.855 ± 0.0257) (95% CI为0.802~0.898)，比较二者曲线下面积，差异有统计学意义(Z = 3.961, P = 0.0001)，提示Caprini评分敏感性高于Padua评分。多因素logistic回归提示BNP、红细胞压积及肺动脉压力为内科住院患者发生VTE的独立危险因素。

Abstract: Objective: To evaluate the effectiveness of the Caprini risk assessment model and Padua prediction score in internal medicine inpatients with VTE. To explore risk factors for VTE in internal medicine inpatients. Methods: A retrospective case-control study was performed among inpatients with VTE from January 2019 to December 2019 at the Affiliated Hospital of Qingdao University. A total of 115 inpatients with definite VTE were recruited. And 115 controls were randomly selected from the inpatients without VTE admitted into the same departments within the same period. The risks and scores of both cases and controls were retrospectively assessed and compared with the Caprini and Padua prediction score. In addition, the clinical data of the two groups were compared, and the risk factors of VTE in internal medicine inpatients were summarized by Binary logistic regression model. Results: The VTE group was higher than the control group in both Caprini score and Padua score. The Caprini scores were (6.783 ± 2.481) in the VTE group and (3.357 ± 1.650) in the control group (t = 12.332, P < 0.05), and the Padua scores were (4.452 ± 1.846) and (1.748 ± 1.835) (t = 11.143, P < 0.05). The areas under the ROC curves for the Caprini score in diagnosing VTE were (0.890 ± 0.0216) (95% CI 0.842~0.927) and for the Padua score in diagnosing VTE were (0.855 ± 0.0257) (95% CI 0.802~0.898), respectively. Comparing the areas under the ROC curves of the two scores, there was a significant difference (z = 3.961, P = 0.0001), suggesting that the Caprini score is more sensitive than the Padua score. Multivariate logistic regression suggested BNP, hematocrit, and pulmonary artery pressure as independent risk factors for VTE in medical inpatients.