基于超声征象建立股浅动脉支架再狭窄的风险预测评分模型

目的建立综合、有效的股浅动脉支架再狭窄风险预测模型,在术前预测支架再狭窄的可能,为手术方案的选择提供指导意见。方法回顾性纳入2016年1月至2018年1月在首都医科大学宣武医院接受股浅动脉支架治疗的患者328例(共381条患肢)作为建模队列。以超声作为评价手段,依据多因素Logistic回归分析筛选支架再狭窄的独立危险因素,按其权重赋值,建立支架再狭窄风险预测评分模型。以受试者工作特征(receiver operating characteristic,ROC)曲线验证模型的区分度,并根据其最佳界值将评分模型分为支架再狭窄低危组和高危组。结果多因素Logistic回归分析显示,8项危险因素被纳入评分系统并建立支架再狭窄风险预测模型,钙化斑块、腘动脉收缩期峰值流速<40 cm/s、膝下流出道评分≥4分、踝臂指数<0.5、女性均赋值1分,患者合并脑梗死、慢性肾病、病变总长度15.0~24.9 cm均赋值2分,病变总长度≥25.0 cm赋值3分,模型最高计12分。以数字减影血管造影(DSA)为金标准,经ROC曲线验证,提示该评分体系具有较高的预测价值ROC曲线下面积(AUC)=0.775,95%CI=0.727~0.824,P<0.001]和拟合优度(χ²=4.921,P=0.766),一致性检验Kappa值为0.609。将评分模型进一步分为再狭窄低危组(0~5分)和再狭窄高危组(6~12分),其敏感性为68.1%,特异性为74.6%,准确性为72.7%。结论股浅动脉支架再狭窄风险预测评分模型可在术前较准确地预测再狭窄的发生,为精准制定手术方案提供理论依据。

Establishment of a scoring model based on ultrasonic characteristics for predicting the restenosis after superficial femoral artery stenting

Objective To establish a comprehensive and effective scoring model based on ultrasonic characteristics for predicting the restenosis risk after superficial femoral artery stenting,in order to assess the possibility of in-stent restenosis and to provide guidance for the selection of therapeutic strategies.Methods A retrospective review of a database of 328 patients(381 limbs)undergoing superficial femoral artery stents in Xuanwu Hospital,Capital Medical University from January 2016 to January 2018 was made as a modeling group.In the modeling cohort,the multivariate logistic regression analysis was performed to screen independent risk factors for in-stent restenosis.A predictive scoring model of restenosis risk was established with weighted score of independent risk factors according to the odd ratio values.Based on the best cut-off value of the receiver operating characteristic(ROC)curves,the scoring table was divided into low-risk and high-risk groups of restenosis.Results Multivariate logistic regression analysis showed that 8 factors were included in the score system to establish the scoring model of in-stent restenosis risk prediction including calcified plaque,peak systolic velocity of popliteal artery<40 cm/s,runoff scores≥4,ankle-brachial index<0.5,female(1 point each);complicated stroke,complicated chronic renal disease,total lesion length 15.0-24.9 cm(2 points each);total lesion length≥25.0 cm(3 points),a total of 12 points in the model.The validation indicated that the scoring system had good predictive value(AUC=0.775,95%CI=0.727-0.824,P<0.001)and goodness of fit(Hosmer-Lemeshowχ²=4.921,P=0.766).The agreement with digital subtraction angiography(DSA)was good(Kappa value=0.609).The scoring system was further divided into the low-risk restenosis(0-5 points)and high-risk restenosis(6-12 points)according to the best cut-off value of 5.5,with a sensitivity of 68.1%,a specificity of 74.6%,and the accuracy of 72.7%.Conclusions The superficial femoral artery in-stent restenosis risk predicting score model based on ultrasonic characteristics may accurately predict the restenosis preoperatively.It provides a theoretical basis for the precise surgical plans.