下肢动脉硬化闭塞症患者合并膝骨关节炎的危险因素分析及预测模型构建  

Risk factors and construction of a prediction model of knee osteoarthritis in patients with lower extremity atherosclerotic occlusive diseaseb

作　　者：丁宁宁 尹翠林 崔玉玲 刘军[1] 强永乾[1] 杨健[1] 周丽 Ding Ningning;Yin Cuilin;Cui Yuling;Liu Jun;Qiang Yongqian;Yang Jian;Zhou Li(Department of Radiology,the First Affiliated Hospital of Xi'an Jiaotong University,Xi'an 710061,China;Department of Radiology,Tangdu Hospital,Air Forced Medical University,Xi'an 710038,China)

机构地区：[1]西安交通大学第一附属医院医学影像科,西安710061 [2]空军军医大学唐都医院放射科,西安710038

出　　处：《中华解剖与临床杂志》2024年第4期211-215,共5页Chinese Journal of Anatomy and Clinics

基　　金：国家自然科学基金(12226007);中华国际医学交流基金(Z-2014-07-1912-04)。

摘　　要：目的探讨下肢动脉硬化闭塞(ASO-LE)患者合并膝骨关节炎(KOA)的危险因素,并建立ASO-LE合并KOA的风险预测模型。方法回顾性队列研究。纳入2019年1月—2021年12月西安交通大学第一附属医院确诊为ASO-LE的139例患者的临床及影像资料,其中男107例、女32例,年龄47~96(68.7±9.1)岁。139例患者均行双下肢动脉CT血管成像检查,采用解剖血流分数(ARS)评分评估下肢动脉狭窄程度,采用外周动脉钙化评分系统(PACSS)评估下肢动脉钙化程度。139例ASO-LE患者根据是否合并KOA分为2组,其中ASO-LE合并KOA组47例(33.8%)、ASO-LE不合并KOA组92例(66.2%)。观察指标:(1)对2组患者的性别、年龄、病程、内科基础疾病史等临床基线资料,以及下肢动脉CTA检查ARS评分、PACSS评分进行单因素分析。(2)选择单因素分析P<0.2的变量进行多因素二元logistic回归分析,依据ASO-LE合并KOA的独立危险因素建立ASO-LE合并KOA的风险预测模型,并应用受试者操作特征(ROC)曲线检测预测模型的灵敏度、特异度。结果ASO-LE合并KOA组高脂血症占比、下肢动脉钙化PACSS评分≥5分占比均高于ASO-LE不合并KOA组,差异均有统计学意义(P值均<0.05)。多因素二元logistic回归分析显示:高脂血症(X_(1))、下肢动脉钙化PACSS评分≥5分(X_(2))是ASO-LE合并KOA的独立危险因素[OR(95%CI)为3.92(1.66~9.27)、2.92(1.32~6.46),P值均<0.05]。依据独立危险因素建立ASO-LE合并KOA的预测模型:logit(P)=-1.409+1.366X_(1)+1.072X_(2)。该模型ROC曲线下面积为0.756,灵敏度为74.5%,特异度为63.0%。结论高脂血症和下肢动脉钙化PACSS评分≥5分是ASO-LE患者合并KOA的独立危险因素。根据独立危险因素构建的ASO-LE合并KOA的预测模型,可为ASO-LE患者提供个性化的KOA风险评估。Objective This work aimed to explore the risk factors and construct a predictive model for knee osteoarthritis(KOA)in patients with atherosclerotic occlusive disease of lower extremities(ASO-LE).Methods This was a retrospective cohort study.The clinical and imaging data of 139 patients diagnosed with ASO-LE in the First Affiliated Hospital of Xi'an Jiaotong University between January 2019 and December 2021 were recorded.Among them,there were 107 males and 32 females,with an age of 47−96(68.7±9.1)years old.The patients underwent lower extremity CT angiography(LE-CTA).The degree of lower extremity arterial stenosis was evaluated according to the anatomical blood flow fraction(ARS)score,and the degree of lower limb artery calcification was evaluated according to the score of the peripheral arterial calcification scoring system(PACSS).The 139 patients with ASO-LE were divided into two groups according to whether they combined with KOA or not:47 patients(33.8%)were in the ASO-LE with KOA group,and 92 patients(66.2%)were in the ASO-LE without KOA group.The observation indicators were as follows:Univariate analysis was conducted for clinical baseline data such as gender,age,course of disease,history of underlying medical diseases,and LE-CTA score(i.e.,ARS and PACSS score)between the with and without KOA in ASO-LE groups.(2)Variables with P<0.2 at univariate analysis were selected for multivariate binary logistic regression analysis.An ASO-LE combined with KOA risk prediction model was established according to the independent risk factors,and the sensitivity and specificity of the prediction model were evaluated with receiver operating characteristic(ROC)analysis.Results Binary logistic regression analyses showed that hyperlipidemia(X_(1))and PACSS≥5(X_(2))were independent risk factors for ASO-LE with KOA(OR=3.92,2.92;95%CI:1.66−9.27,1.32−6.46;all P values<0.05).The ASO-LE with KOA prediction model on the basis of the independent risk factors was logit(P)=−1.409+1.366X_(1)+1.072 X_(2).The area under the ROC cu

关 键 词：动脉闭塞性疾病 下肢动脉硬化闭塞 膝骨关节炎 危险因素 预测模型 

分 类 号：R543.5[医药卫生—心血管疾病] R684.3[医药卫生—内科学]