基于真实世界的深静脉血栓形成复发风险预测模型的构建与验证OA

Objective This study aimed to screen the risk factors of deep venous thrombosis(DVT)recurrence,establish a pre-liminary risk prediction model and verify it.Methods The real world research model and retrospective case-control study were used as the design method.The patients with first-time DVT who were hospitalized in the Affiliated Hospital of Shandong Univer-sity of Traditional Chinese Medicine from January 1,2019 to May 31,2022 were selected as the study objects,and their clinical data and follow-up information were used as the data sources.The grouping was based on whether there was recurrence of DVT.The risk factors of DVT recurrence were identified by univariate analysis and multiple Logistic regression screening.Lastly,a risk prediction model was built,a visual nomogram was drawn,and cinternal verification was carried out.Results A total of 1724 pa-tients with first-time DVT were included in this study,295 of whom relapsed,and the overall recurrence rate was about 17.11％.In the training dataset,207 patients were included in the recurrence group and the non-recurrence group,respectively.And in the validation dataset,88 patients were included in the recurrence group and the non-recurrence group,respectively.The clinical data of the two groups in the training dataset were first analyzed by univariate analysis,and variables with statistical differ-ences(P＜0.05)were included in the multivariate analysis of binary Logistic regression,and the results were as follows.Ten fac-tors,including age≥60 years old,clinical classification of the first DVT,classification of the first DVT inducing factors,iliac vein compression,pulmonary thromboembolism(PTE),post thrombotic syndrome(PTS),D-dimer level,venous revascularization rate upon discharge,standard anticoagulation/antiembolism and use of Chinese medicine and related preparations,were independent influencing factors for the recurrence of DVT(P＜0.05).According to the above analysis results,the prediction model formula of DVT recurrence risk was constructed,and the column graph was drawn.Internal verification showed that the model had good dif-ferentiation,calibration and clinical practicability.Conclusion Age ≥60 years old and the onset type of the first DVT was the cen-tral type and mixed type.The first DVT induction factors were classified into persistent inductions,combined inductions and no in-ductions,iliac vein compression,PTE,PTS and high D-dimer level were independent risk factors for the recurrence of DVT.High venous revascularization rate,standard anticoagulation/antithrombolysis and use of blood-activating Chinese medicine and related preparations are protective factors for the recurrence of DVT in the patients with first-time DVT.The DVT recurrence risk model containing the above 10 predictive factors has a good predictive performance,which can provide a certain basis for clinical screen-ing of high-risk groups of DVT recurrence.