下肢动脉硬化闭塞症患者发生溃疡的风险预测模型构建与验证及护理启示 [中文引用][英文引用]

目的  探讨下肢动脉硬化闭塞症（lower extremities arteriosclerosis obliterans，LEASO）患者发生溃疡的影响因素构建风险预测模型并进行验证，为早期识别高危人群提供借鉴。方法  采用便利抽样法，选取 2020 年 1 月— 2025 年 5 月内蒙古自治区某三级甲等综合医院 291 例 LEASO 患者作为研究对象。将 291 例 LEASO 患者按 7：3 比例分为训练集（204 例）和验证集（87 例），分别用于模型构建与验证。采用单因素分析和二分类 Logistic 回归分析，确定 LEASO 患者发生溃疡的影响因素，并基于影响因素构建预测模型。采用受试者工作特征（receiver operating characteristic，ROC）曲线的曲线下面积（area under curve，AUC）、校准曲线和决策曲线分析（decision curve analysis，DCA）评价模型的预测效能和临床实用性。结果  吸烟史、饮酒史、糖尿病、中性粒细胞计数和血红蛋白是 LEASO 患者发生溃疡的影响因素。ROC 曲线显示，训练集预测 LEASO 患者发生溃疡的AUC 为 0.912（95%CI：0.872~0.953），验证集 AUC 为 0.904（95%CI：0.841~0.967）。Hosmer-Lemeshow 检验，训练集（χ2=2.250，P=0.972）、验证集（χ2=4.636，P=0.796），校准曲线显示，该模型与实际概率具有良好的一致性。决策曲线显示模型具有良好的临床实用性。结论  本研究构建的风险预测模型具有识别 LEASO 发生溃疡高危人群、指导临床实践的价值。

Objective To investigate the risk factors for ulceration in patients with lower extremity arteriosclerosis obliterans （LEASO）and to develop a risk prediction model，thereby providing a reference for the early identification of high-risk individuals. Methods A retrospective analysis was conducted on the clinical data of 291 LEASO patients admitted to a Tier-IIIA hospital in the Inner Mongolia Autonomous Region between January 2020 and May 2025. The 291 LEASO patients were divided in a 7：3 ratio into a training set（204 patients）and a validation set（87 patients）for model development and validation. Univariate analysis and binary logistic regression were used to identify risk factors for ulcer development in LEASO patients，and a predictive model was developed based on the identified factors. Receiver operating characteristic（ROC）curves，area under the curve（AUC），calibration curves，and decision curve analysis（DCA）were used to evaluate the predictive performance and clinical utility of the model. Results Smoking，alcohol consumption，diabetes，neutrophil count，and haemoglobin levels were identified as risk factors for ulcer development in LEASO patients. The ROC curve analysis showed that the area under the curve（AUC）for predicting ulcer development in LEASO patients was 0.912 （95%CI：0.872-0.953）in the training set and 0.904（95%CI：0.841–0.967）in the validation set. The Hosmer-Lemeshow test results were as follows：training set（χ2 =2.250，P=0.972）and validation set（χ2 =4.636，P=0.796）. The calibration curve indicated that the model had a good agreement with the actual probabilities. DCA indicated that the model has a good clinical utility. Conclusion The developed risk prediction model is valuable for identifying high-risk populations and guiding clinical practice.