首页|踝关节骨折术后深静脉血栓风险的列线图预测模型构建

踝关节骨折术后深静脉血栓风险的列线图预测模型构建

扫码查看
目的:分析踝关节骨折(AF)患者术后深静脉血栓(DVT)形成的影响因素,并构建术后深静脉血栓发生的列线图模型.方法:选取2021年1月至2023年1月收治的踝关节骨折患者348例,按照7:3比例随机分为建模组(244例)和验证组(104例).依据术后是否形成深静脉血栓分为深静脉血栓组和非深静脉血栓组.多因素Logistic回归模型分析踝关节骨折患者术后深静脉血栓危险因素,应用R软件绘制预测踝关节骨折患者术后深静脉血栓的列线图模型,并进一步对模型验证.结果:3 4 8例患者术后深静脉血栓发生率为20.69%.单因素分析显示,合并血管及神经损伤、年龄、D-二聚体(D-D)、高能量损伤与踝关节骨折患者术后深静脉血栓发生有关,差异有统计学意义(P<0.05).Logistic回归分析可知,年龄(OR=1.106,95%CI=1.045~1.171)、受伤机制(OR=10.212,95%CI=3.290~31.700)、合并血管及神经损伤(OR=4.805,95%CI=1.649~14.002)、D-二聚体(OR=13.007,95%CI=5.574~30.353)是影响踝关节骨折患者术后发生深静脉血栓的危险因素,差异有统计学意义(P<0.05).建模组与验证组校准曲线均拟合良好,曲线下面积(AUC)分别为0.932(95%CI=0.890~0.975)和 0.917(95%CI=0.872~0.962),校准曲线 Hosmer-Lemeshow(H-L)检验差异无统计学意义(x2=8.575,6.184;P=0.379,0.415),表明预测深静脉血栓发生概率与实际发生风险有良好一致性.结论:年龄、合并血管及神经损伤、D-二聚体、高能量损伤是影响踝关节骨折患者术后深静脉血栓发生的危险因素,该模型对踝关节骨折患者术后发生深静脉血栓有较好预测价值.
Construction of a Nomogram Prediction Model for the Risk of Deep Vein Thrombosis after Ankle Fracture
Objective:To analyze the influencing factors of postoperative deep vein thrombosis(DVT)in patients with an-kle fracture(AF),and to build a nomogram model of postoperative DVT.Methods:A total of 348 AF patients admitted from January 2021 to January 2023 were randomly divided into a modeling group(244 cases)and a validation group(104 cases)in a 7:3 ratio.And they were grouped into DVT group and non DVT group based on whether DVT was formed after surgery.Multivariate Logistic regression model was applied to analyze the risk factors of postoperative DVT in AF patients,R software was applied to draw a nomogram model for predicting postoperative DVT in AF patients and further validate the model.Results:The incidence of postoperative DVT in 348 patients was 20.69%.Univariate analysis showed that concomitant vascular and nerve injuries,age,D-Dimer(D-D),and high-energy injuries were associated with postopera-tive DVT in AF patients(P<0.05).Logistic regression analysis showed that age(OR=1.106,95%CI=1.045-1.171),injury mechanism(OR=10.212,95%CI=3.290-31.700),concomitant vascular and nerve injuries(OR=4.805,95%CI=1.649-14.02),and D-Dimer(OR=13.007,95%CI=5.574-30.353)were risk factors for postoperative DVT in AF patients(P<0.05).The calibration curves of the modeling and validation groups were well fitted,with area under curve(AUC)value of 0.932(95%CI=0.890-0.975)and 0.917(95%CI=0.872-0.962),respectively,the Hosmer-Lemeshow(H-L)test showed x2=8.575,6.184,P=0.379,0.415,indicated a good consistency between the predicted probability of DVT occurrence and the actual risk of occurrence.Conclusion:Age,concomitant vascular and nerve injuries,D-Dimer,and high-energy injuries are risk factors that affect the occurrence of postoperative DVT in AF patients.This model has good predictive value for postoperative DVT in AF patients.

ankle fracturedeep vein thrombosisrisk factorsnomogram

赵峰、周平、富银

展开 >

南京医科大学第四附属医院(南京,210031)

踝关节骨折 深静脉血栓 风险因素 列线图

2024

中国中医骨伤科杂志
中华中医药学会,湖北省中医药研究院

中国中医骨伤科杂志

CSTPCD
影响因子:0.732
ISSN:1005-0205
年,卷(期):2024.32(2)
  • 22