文献类型：期刊文献

中文题名：不同预测模型在结直肠癌患者术前深静脉血栓发生风险预测中的价值比较

英文题名：Comparative analysis of predictive models for preoperative deep vein thrombosis risk in colorectal cancer patients

作者：刘芳[1];王小英[2];李燕[2];王佳明[2];王红彦[2]

第一作者：刘芳

机构：[1]甘肃中医药大学,甘肃兰州730000;[2]甘肃省人民医院,甘肃兰州730000

第一机构：甘肃中医药大学

年份：2025

卷号：40

期号：12

起止页码：1297

中文期刊名：护士进修杂志

外文期刊名：Journal of Nurses Training

基金：甘肃省卫生健康行业科研计划项目(编号:GSWSHL2021-013);甘肃省人民医院项目(编号:21GSSYB-24)。

语种：中文

中文关键词：结直肠癌;深静脉血栓;预测模型;风险评估;护理

外文关键词：colorectal cancer;deep vein thrombosis;predictive models;risk assessment;nursing

摘要：目的 探讨并比较分析ABAD预测模型、鹿儿岛深静脉血栓评分(Kagoshima-deep reintroom bosis score,KDS)预测模型和COMPASS-CAT预测模型在结直肠癌(colorectal cancer,CRC)患者术前DVT发生风险预测中的应用价值。方法 便利选取2024年1—8月在甘肃省某三级甲等医院就诊并拟行手术治疗的425例CRC患者,应用受试者工作特征曲线(ROC)下面积(AUC)评价ABAD预测模型、KDS预测模型、COMPASS-CAT预测模型在CRC患者术前DVT发生风险预测中的区分度。采用Delong检验进行AUC之间的两两比较。采用Hosmer-Lemeshow拟合优度检验评价3种预测模型DVT发生风险预测的校准度。采用临床决策曲线分析,评估3种预测模型在不同DVT风险水平下对患者预防管理的获益程度。结果 ABAD预测模型、KDS预测模型、COMPASS-CAT预测模型的AUC值分别为0.849、0.860、0.822,3种预测模型间的AUC值差异无统计学意义(P>0.05)。Hosmer-Lemeshow拟合优度检验结果显示ABAD预测模型、KDS预测模型拟合较好(P>0.05),COMPASS-CAT预测模型拟合较差(P<0.05)。临床决策曲线分析表示在低中风险患者中KDS预测模型更受益,在高风险患者中COMPASS-CAT预测模型更受益,在极高风险患者中ABAD预测模型更受益。结论 在CRC患者术前DVT发生风险预测中,ABAD预测模型、KDS预测模型都表现出中等的区分度和较好的校准度。COMPASS-CAT预测模型表现出中等的区分度和较差的校准度。在低中风险患者中KDS预测模型更受益,在高风险患者中COMPASS-CAT预测模型更受益,在极高风险患者中ABAD预测模型更受益。
Objective To explore and compare the predictive value of the ABAD predictive model,Kagoshima-DVT score(KDS)predictive model,and the COMPASS-CAT predictive model in assessing the risk of preoperative deep vein thrombosis(DVT)in patients with colorectal cancer(CRC).Methods A total of 425 CRC patients scheduled for surgical treatment from January to August 2024 at a tertiary hospital in Gansu Province were conveniently selected.The discrimination performance of the ABAD predictive model,KDS predictive model,and COMPASS-CAT predictive model was evaluated using the area under the receiver operating characteristic curve(AUC).The Delong test was used to compare AUC s pairwise.The Hosmer-Lemeshow goodness-of-fit test was used to assess the calibration of each predictive model.Decision curve analysis was employed to assess the net clinical benefit of the three predictive models in the context of DVT prevention across stratified patient risk levels.Results The AUC s of the ABAD predictive model,KDS predictive model,and COMPASS-CAT predictive model were 0.849,0.860,and 0.822,respectively,with no statistically significant differences between them.The Hosmer-Lemeshow test indicated good calibration for the ABAD and KDS predictive models(P>0.05),but poor calibration for the COMPASS-CAT predictive model(P<0.05).Clinical decision curve analysis showed that the KDS predictive model was more beneficial for patients at low-to-moderate risk,the COMPASS-CAT predictive model was more beneficial for high-risk patients,and the ABAD predictive model was more beneficial for very high-risk patients.Conclusion Both the ABAD and KDS predictive models demonstrated moderate discrimination and good calibration in predicting preoperative DVT risk among CRC patients.The COMPASS-CAT predictive model showed moderate discrimination but poor calibration.The KDS predictive model may be more useful for patients at low-to-moderate risk,the COMPASS-CAT predictive model for high-risk patients,and the ABAD predictive model for very high-risk patients.