Early prediction and decision curve analysis of venous thrombosis risk in patients after acute stroke
10.3969/j.issn.1009-0126.2025.07.020
- VernacularTitle:急性脑卒中患者静脉血栓发生风险早期预测及决策曲线分析
- Author:
Wei SU
1
;
Jianzhong XIONG
1
;
Yaqun LI
1
;
Wenmiao JIE
1
Author Information
1. 337000 萍乡市人民医院神经科
- Publication Type:Journal Article
- Keywords:
stroke;
venous thrombosis;
logistic models;
decision curve
- From:
Chinese Journal of Geriatric Heart Brain and Vessel Diseases
2025;27(7):925-930
- CountryChina
- Language:Chinese
-
Abstract:
Objective To construct a risk prediction model for venous thromboembolism(VTE)in patients with acute stroke and analyze its predictive value with decision curve.Methods Clinical data of 265 patients with acute stroke admitted to Department of Neurology of Pingxiang People's Hospital between May 2022 and May 2024 were collected,and 185 of them(69.8％)were assigned to a training set and 80 cases(30.2％)into a validation set.Based on the results of color Doppler ultrasound examination,the patients in the training set were divided into the VTE group(72 cases)and the non-VTE group(113 cases).Multivariate logistic regression analysis was used to identify the risk factors for VTE occurrence in the patients.A prediction model was constructed and a nomegram was developed to visually present the logistic regression prediction model.ROC curve and decision curve analyses were applied to evaluate the predictive performance of the logis-tic regression model.Results The VTE group had significantly larger intracranial hemorrhage volume,higher fibrinogen(FIB),D-dimer and C-reactive protein(CRP)levels,and lower Glasgow Coma Scale(GCS)score than the non-VTE group(P＜0.01).Multivariate logistic regression analysis indicated that intracranial hemorrhage volume(OR=1.243,95％CI:1.015-1.522),FIB(OR=3.161,95％CI:1.711-5.841),CRP(OR=1.595,95％CI:1.283-1.981),and D-dimer(OR=6.234,95％CI:2.693-14.434)were independent risk factors for the occurrence of VTE in the patients with acute stroke,while GCS score(OR=0.587,95％CI:0.413-0.834,P=0.003)was an independent protective factor against the occurrence.A VTE prediction model based on these influencing factors obtained a consistency index of 0.978.The calibration curve results showed that the observed values were in good agreement with the predicted values.ROC curve analysis indicated that the AUC value of the prediction model was 0.978(95％CI:0.945-0.994)in the training set and 0.959(95％CI:0.890-0.991)in the validation set.Decision curve analysis revealed that the model achieved higher net benefit when the threshold probability was between 20％and 80％.Conclusion Intracranial bleeding volume,GCS score,CRP,Didimer and FIB are related influencing factors of VTE in patients with acute stroke,and the prediction model of deci-sion curve based on the above factors is of high prediction efficiency.
