| BackgroundA thrombotic event in a cancer patient can have devastating consequences related to its treatment.Some of these consequences are:risk of bleeding,delays in delivering the chemotherapy,interactions with medications,high risk of recurrent thrombotic events,a decreased quality of life,and increased consumption of health care resources.It has been described that cancers associated with thrombotic events are more aggressive and usually associated with worse prognosis.Lung cancer has been related to VTE in 7.3-13.6%of patients.Due to its high prevalence,lung cancer may be respon-sible for the highest incidence of thrombotic events associated with cancer.This incidence is higher in patients with non-small cell lung cancer(NSCLC)compared to patients with small cell lung cancer(SCLC).Patients harboring an advanced disease and those receiving chemotherapy are at higher risk of a VTE.ObjectivesThis study examined the value of Khorana combined with Wells DVT model in predict the thrombus of patients with Central Venous Catheters.Materials and MethodsA retrospective analysis of 150 patients with postoperative lung cancer who underwent PICC catheterization from January 01,2012 to January 01,2015 in our hospital was divided into thrombosis group(18 cases)and non-thrombus group(132 cases)according to whether thrombosis occurred;t test was used to detect the difference between the two model variables;ROC was used to evaluate the efficacy of predicting thrombosis;grouping according to different model risk status,OS and PFS were analyzed by Log-rank test among different risk groups.Results1.Among the 150 patients with lung cancer in the whole group,a total of 18 patients had thrombosis of varying degrees during the follow-up period,including 7(38.9%)patients with symptomatic thrombosis,whose main symptoms were local redness,swelling and pain.Asymptomatic thrombosis occurred in 11 cases(61.1%).According to whether there is thrombus group,comparing thrombosis occurs and patients’ gender,age,pathological type,whether to have high blood pressure,high cholesterol,diabetes,the relationship between the results showed that the occurrence of thrombosis are closely associated with hyperlipidemia,the difference between different state of hyperlipidemia(x2=4.006,P=0.045),in different age,gender,lesion site has nothing to do,whether to have high blood pressure,diabetes(P>0.05).2.According to the variable assignment Khorana scale,and the whole group of patients with low-risk 0 cases,moderate in 63 cases,87 cases of high risk,according to the variable assignment of Wells DVT model,low-risk patients with 12 cases,moderate in 66 cases,72 cases of high-risk,T test,the application of two groups of variables were compared between thrombus group and no thrombosis Khorana score model and Wells DVT assignment differences,results suggest two models of the cumulative score in the different between thrombus group were statistically significant(t=1.231,P=0.026;t=2.583,P=0.008).3.ROC curve method was used to analyze the efficacy of Khorana model and Wells DVT model in predicting thrombosis.The results showed that the area under the ROC curve of Khorana score was 0.642,the sensitivity was 71.7%,the specificity was 61.8%,and the Youden index was>0.176.The area under the ROC curve of Wells DVT model was 0.594,the sensitivity was 62.7%,the specificity was 64.9%,and the Youden index was>0.143.The cumulative scores of the two evaluation models were included in Logistic binary regression analysis,and the results showed that both evaluation methods were significant in predicting thrombosis(P<0.05).After the combination of the two models,the ROC curve analysis showed that the area of the combined factors under the ROC curve was 0.812,the sensitivity was 80.4%,the specificity was 79.28.3%,and the Youden index was>0.305,respectively.It is suggested that the combined detection is more efficient than the single model.4.Up to the end of follow-up,patients in the whole group were followed up for 6 to 70 months,with an average survival time of 32.7±12.5 months,median survival time of 37.4 months,1-year survival rate of 81.6%,3-year survival rate of 43.9%,and 5-year survival rate of 30.4%.According to the final cumulative score grouping of Khorana model,log-rank test indicated that OS of patients in the medium-low risk group was significantly better than that in the high-risk group,and there was a statistical difference between the two groups(x2=7.489,P=0.006).There was no statistical difference in PFS between the risk groups(x2=0.422,P=0.516).In addition,Wells DVT model was assigned to groups,and the OS of patients in the medium-low risk group was significantly better than that in the high-risk group(x2=4.379,P=0.036).There was no statistical difference in PFS between different groups(x2=0.748,P=0.387).Conclusion1.The cumulative scores of Khorana model and Wells DVT model were statistically different between different thrombotic groups(P<0.05).2.Combined with Khorana model and Wells DVT model,the efficacy of predicting thrombosis in patients with lung cancer is better than that of any single model. |
PDF Full Text Request