不同风险评估模型对妇科肿瘤患者下肢深静脉血栓形成预测价值

Application of different risk assessment models in deep venous thrombosis formation in gynecological tumor patients

目的分析比较Caprini、Khorna、Autar、Wells风险评估模型在妇科肿瘤患者下肢深静脉血栓形成(DVT)中的预测价值。方法采用回顾性病例对照研究,收集2019-01-01-2020-12-31山东第一医科大学附属肿瘤医院确诊为DVT的168例妇科肿瘤患者作为病例组(DVT组),按照1∶2的配比,选取同期、同科室住院、年龄相近的336例非DVT患者作为对照组。根据Caprini、Khorna、Autar、Wells风险评估模型对2组患者进行DVT风险评估。采用logistic回归模型分析DVT发生的影响因素,并绘制受试者工作特征曲线(ROC),分析不同风险评估模型对妇科肿瘤患者DVT风险的预测价值。结果DVT组Caprini评分为(4.38±1.26)分,对照组为(4.17±1.20)分,差异无统计学意义,t=1.821,P=0.069;DVT组Khorna、Autar、Wells评分分别为(1.26±0.82)、(11.44±3.07)、(1.30±0.50)分,对照组分别为(0.78±0.46)、(10.67±1.83)、(1.01±0.07)分,差异均有统计学意义,t值分别为7.055、2.995、7.481,P值分别为<0.001、0.003、<0.001。Logistic回归模型分析结果显示,D-二聚体水平(OR=4.780,95%CI为2.787~8.198,P<0.001)、血小板计数(OR=3.204,95%CI为2.005~5.119,P<0.001)为影响DVT发生的独立因素,D-二聚体水平升高、血小板计数上升会明显增加DVT的发生风险。Caprini、Khorna、Wells、Autar风险评估模型的ROC曲线下面积(AUC)分别为0.558、0.702、0.632、0.541。结论对于妇科肿瘤患者,Khorna风险评估模型对DVT发生有一定的辅助预测价值,但仍需多中心的大样本研究支持。

Objective To compare the predictive value of Caprini,Khorna,Wells,and Autar risk assessment model in patients with gynecological cancer patients with deep vein thrombosis(DVT).Methods A retrospective case-control study was used to collect 168 patients with gynecological tumors diagnosed with DVT of the lower extremities in Shandong Cancer Hospital from January 1,2019 to December 31,2020 as the DVT group.A total of 336 patients with non-lower extremity DVT who were hospitalized in the same department and were of similar age were assigned in a 1∶2 ratio to the control group during the same period.According to the Caprini,Khorna,Autar and Wells DVT risk assessment models,the two groups of patients were assessed for the risk of DVT of the lower limbs.Logistic regression was used to analyze the influencing factors of DVT,and the receiver operating characteristic curve(ROC)was drawn to analyze the predictive value of different risk assessment models for DVT risk in patients with gynecological cancer.Results There was no statistically significant difference in the Caprini risk scores between the two groups(4.38±1.26 vs 4.17±1.20,t=1.821,P=0.069),but there were significant differences in the Khorna(1.26±0.82 vs 0.78±0.46,t=7.055,P<0.001),Autar(11.44±3.07 vs 10.67±1.83,t=2.995,P=0.003),and Wells(1.30±0.50 vs 1.01±0.07,t=7.481,P<0.001)risk scores.Logistic regression model analysis showed that D-dimer level(OR=4.780,95%CI:2.787-8.198,P<0.001),platelet count(OR=3.204,95%CI:2.005-5.119,P<0.001)are independent factors affecting the occurrence of DVT.The increase of D-dimer level,and platelet count will significantly increase the risk of DVT.The area under the ROC curve of Caprini,Khorna,Wells and Autar risk assessment models were 0.558,0.702,0.632,0.541,respectively.Conclusion For patients with gynecological tumors,the Khorna risk assessment model has certain auxiliary predictive value for the occurrence of DVT,but it still needs the support of multi-center large sample research.