天津医药 ›› 2026, Vol. 54 ›› Issue (5): 478-483.doi: 10.11958/20252907

• 临床研究 • 上一篇    下一篇

糖尿病足溃疡患者住院期间小截肢的预测模型构建及验证

黄斐1,2(), 王利航1,2, 孙官文2△(), 包呼和2, 杨鹏波3, 张雅兴3   

  1. 1 内蒙古科技大学包头医学院(邮编014040)
    2 内蒙古自治区人民医院创伤骨科
    3 内蒙古医科大学
  • 收稿日期:2025-09-18 修回日期:2026-01-15 出版日期:2026-05-15 发布日期:2026-05-13
  • 通讯作者: E-mail:928568391@qq.com
  • 作者简介:黄斐(1996),男,硕士在读,主要从事糖尿病足的外科综合治疗方面研究。E-mail:1375208747@qq.com
  • 基金资助:
    内蒙古自治区自然科学基金项目(2024MS08060);内蒙古自治区公立医院科研联合基金科技项目(2024GLLH0057);内蒙古自治区公立医院科研联合基金科技项目(2024GLLH0071)

Construction and validation of a prediction model for minor amputation in patients with diabetic foot ulcers during hospitalization

HUANG Fei1,2(), WANG Lihang1,2, SUN Guanwen2△(), BAO Huhe2, YANG Pengbo3, ZHANG Yaxing3   

  1. 1 Baotou Medical College, Inner Mongolia University of Science & Technology, Baotou 014040, China
    2 Department of Orthopedic Trauma, Inner Mongolia Autonomous Region People’s Hospital
    3 Inner Mongolia Medical University
  • Received:2025-09-18 Revised:2026-01-15 Published:2026-05-15 Online:2026-05-13
  • Contact: E-mail:928568391@qq.com

摘要:

目的 探讨糖尿病足溃疡(DFU)患者住院期间发生小截肢的独立危险因素,构建并验证个体化风险预测模型。方法 回顾性分析154例DFU患者的临床资料并根据患者住院期间是否行小截肢手术分为小截肢组(48例)和非截肢组(106例)。采用多因素Logistic回归分析筛选影响因素并构建预测模型。另外收集71例DFU患者的临床资料并将其作为外部验证组。在建模组及验证组中,通过受试者工作特征(ROC)曲线、校准曲线及决策曲线分析 (DCA)评估模型的区分度、校准度与临床适用性。结果 与非截肢组比较,小截肢组患者男性比例、溃疡持续时间、深溃疡及下肢动脉闭塞比例、空腹血糖(FPG)、肌酸激酶(CK)、血肌酐(Scr)、三酰甘油(TG)、D-二聚体(D-Dimer)水平升高,院前治疗比例降低(P<0.05)。多因素Logistic回归分析显示,男性、溃疡持续时间延长、深溃疡、下肢动脉闭塞、Scr和D-Dimer水平升高及下肢血管闭塞是DFU患者小截肢的独立危险因素(P<0.05);接受院前治疗是保护因素。ROC曲线分析结果显示,建模组预测小截肢的曲线下面积(AUC)为0.898(95%CI:0.842~0.953),敏感度为87.5%,特异度为82.1%;验证组AUC为0.887(95%CI:0.808~0.966),敏感度为92.3%,特异度为77.8%。校准曲线及DCA显示模型具有良好的校准度及临床实用性。结论 本研究构建的列线图模型能有效预测DFU患者住院期间的小截肢风险,其中院前治疗作为可干预因素,为早期识别高危患者并实施精准干预提供了实用工具。

关键词: 糖尿病足, 小截肢, 危险因素, 预测模型, 列线图

Abstract:

Objective To explore the independent risk factors for minor amputation during hospitalization in patients with diabetic foot ulcer (DFU), and to construct and validate an individualized risk prediction model. Methods A retrospective analysis was conducted on the clinical data of 154 patients with DFU. Patients were categorized into the minor amputation group (48) and the non-amputation group (106) based on whether they underwent minor amputation during hospitalization. Multivariate Logistic regression analysis was employed to identify risk factors and construct a prediction model. Additionally, clinical data from 71 DFU patients were collected as an external validation cohort. In both the modeling and validation cohorts, the discrimination, calibration and clinical utility of the model were evaluated using receiver operating characteristic (ROC) curves, calibration curves and decision curve analysis (DCA). Results Compared with the non-amputation group, higher proportion of male patients, longer ulcer duration, higher rates of deep ulcers and lower extremity arterial occlusion, elevated levels of fasting plasma glucose (FPG), creatine kinase (CK), serum creatinine (Scr), triglycerides (TG) and D-Dimer were found in the minor amputation group (all P < 0.05), while the proportion of pre-hospital treatment was significantly lower in the minor amputation group (P < 0.05). Multivariate Logistic regression analysis showed that male sex, prolonged duration of ulcer, deep ulcer, lower extremity arterial occlusion, elevated levels of Scr and D-Dimer, and lower extremity vascular occlusion were independent risk factors for minor amputation in patients with DFU (P<0.05). Receiving prehospital treatment was a protective factor. The area under the curve (AUC) for predicting minor amputation in the modeling group was 0.898 (95%CI: 0.842-0.953), with a sensitivity of 87.5% and a specificity of 82.1%. The AUC of the validation group was 0.887 (95%CI: 0.808-0.966), with a sensitivity of 92.3% and a specificity of 77.8%. The calibration curve and DCA showed that the model had good calibration accuracy and clinical utility. Conclusion The nomogram model constructed in this study can effectively predict the risk of minor amputation during hospitalization in patients with DFU. Among them, "pre-hospital treatment" as an interventional factor provides a practical tool for the early identification of high-risk patients and the implementation of precise intervention.

Key words: diabetic foot, minor amputation, risk factors, prediction model, nomogram

中图分类号: