| 张皓媛,田罡,刘劲松,等.糖尿病足截肢患者死亡危险因素分析及预测模型构建.骨科,2026,17(4): 360-365. |
| 糖尿病足截肢患者死亡危险因素分析及预测模型构建 |
| Analysis of mortality risk factors and establishment of predictive model for diabetic foot amputees |
| 投稿时间：2025-12-27 |
| DOI：10.3969/j.issn.1674-8573.2026.04.010 |
| CN KeyWords: 糖尿病足 截肢 死亡 危险因素 预测模型 |
| EN KeyWords: Diabetic foot Amputation Mortality Risk factors Prediction model |
| Fund Project:中国康复研究中心科研基金项目（2023ZX-10） |
| 作者 | 单位 | E-mail | | 张皓媛 | 首都医科大学康复医学院,北京 100068；四川大学华西第二医院儿童康复科,成都 610041 | | | 田罡 | 首都医科大学康复医学院,北京 100068；中国康复研究中心北京博爱医院骨科,北京 100068；北京截肢者康复会诊中心,北京 100068 | | | 刘劲松 | 北京截肢者康复会诊中心,北京 100068；中国康复研究中心北京博爱医院假肢矫形部,北京 100068 | | | 张倩 | 北京截肢者康复会诊中心,北京 100068；中国康复研究中心北京博爱医院假肢矫形部,北京 100068 | | | 黄秋晨 | 首都医科大学康复医学院,北京 100068；中国康复研究中心北京博爱医院理学疗法科,北京 100068 | | | 闵红巍 | 首都医科大学康复医学院,北京 100068；中国康复研究中心北京博爱医院骨科,北京 100068 | | | 唐鹏 | 首都医科大学康复医学院,北京 100068；中国康复研究中心北京博爱医院骨科,北京 100068 | | | 顾蕊 | 首都医科大学康复医学院,北京 100068；中国康复研究中心北京博爱医院骨科,北京 100068；北京截肢者康复会诊中心,北京 100068 | jj6567@icloud.com | | 刘克敏 | 首都医科大学康复医学院,北京 100068；中国康复研究中心北京博爱医院骨科,北京 100068 | |
|
| View Counts: 290 |
| PDF Download Counts: 66 |
| CN Abstract: |
| 目的 分析糖尿病足截肢患者的流行病学特点及临床资料，探讨可能导致其死亡的危险因素并构建列线图预测模型。方法 选取2017年4月至2022年4月于中国康复研究中心北京博爱医院骨科住院治疗的65例糖尿病足截肢患者的临床资料进行回顾性分析。其中男48例，女17例，年龄（66.52±11.45）岁（34~87岁）。收集患者一般临床资料、住院资料、术后资料与康复情况，根据术后3年是否死亡将患者分为生存组和死亡组。对两组数据进行组间分析和多因素Logistic回归分析，筛选出患者术后3年内死亡的独立危险因素，建立糖尿病足患者截肢术后3年内死亡风险的预测模型并绘制对应的列线图，通过受试者工作特征（receiver operating characteristic，ROC）曲线、曲线下面积（area under the curve，AUC）、校准曲线分析、Bootstrap重复抽样方法及决策曲线分析（decision curve analysis，DCA）对预测模型的性能进行评价。结果 纳入的65例患者术后3年内有19例死亡，死亡率为29.23%。多因素Logistic回归分析结果显示，出院日常生活活动能力（activities of daily living，ADL）评分较低（OR=1.412，95% CI：1.178~1.672，P=0.023）、糖尿病病程较长（OR=1.230，95% CI：1.012~1.495，P=0.037）、未装配假肢（OR=1.009，95% CI：1.002~1.312，P=0.035）、存在焦虑抑郁（OR=2.001，95% CI：1.468~3.144，P=0.016）是死亡的独立危险因素。预测模型的AUC值为0.932，95% CI：0.880~0.948，提示该模型区分性能良好，预测概率与理想线匹配良好。拟合优度检验结果为χ2=2.317，P=0.381，校准曲线进行验证后显示模型校准度较好、预测概率与实际概率具有良好的一致性。DCA显示模型有效范围大，具有较好的临床有效性。结论 糖尿病足患者截肢术后3年内死亡率为29.23%，出院ADL评分较低、糖尿病病程较长、未装配假肢、存在焦虑抑郁是死亡的独立危险因素。据此构建的预测模型预测能力较好，有助于早期识别并控制糖尿病足截肢患者死亡的危险因素，可有效改善其生活质量及身心健康，降低死亡率。 |
| EN Abstract: |
| Objective To analyze the epidemiological characteristics and clinical data of diabetic foot amputees, explore the risk factors that may lead to their death, and construct a Nomogram prediction model. Methods The clinical data of 65 amputees with diabetic foot who were hospitalized in the Department of Orthopedics of Beijing Bo'ai Hospital from April 2017 to April 2022 were selected for retrospective analysis. Among them, there were 48 males and 17 females, with an age of 66.52±11.45 years old (ranging from 34 to 87 years). The general clinical data, inpatient data, postoperative data and rehabilitation conditions of the patients were collected. All patients were divided into the survival group and the death group according to whether they died 3 years after the operation. Intergroup analysis was conducted on the two groups of data, and multivariate Logistic regression was performed to identify independent risk factors for mortality within 3 years after surgery. Establish a risk prediction model for death within 3 years after amputation in patients with diabetic foot and draw the corresponding nomogram. Through the receiver operating characteristic (ROC) curve and the area under the curve (AUC), the performance of the prediction model was evaluated by AUC, calibration curve analysis, Bootstrap resampling and decision curve analysis (DCA). Results Among the 65 included patients, 19 died within 3 years after the operation, with a mortality rate of 29.23%. The results of multivariate Logistic regression analysis showed that a lower activities of daily living (ADL) score at discharge (OR=1.412, 95% CI: 1.178-1.672, P=0.023), a longer duration of diabetes (OR=1.230, 95% CI: 1.012-1.495, P=0.037), lack of prosthetic fitting (OR=1.009, 95% CI: 1.002-1.312, P=0.035), and the presence of anxiety and depression (OR=2.001, 95% CI: 1.468-3.144, P=0.016) were independent risk factors for death. The nomogram showed good discriminative ability, AUC=0.932, 95% CI: 0.880-0.948, and was well-calibrated (goodness-of-fit, χ2=2.317, P=0.381). DCA shows that the model has a wide effective range and good clinical effectiveness. Conclusion The 3-year mortality after diabetic foot amputation was 29.23%. A lower ADL score at discharge, a longer course of diabetes, no prosthetic limb fitting, and the presence of anxiety and depression are independent risk factors for death. The predictive model constructed based on this has a good predictive ability and is helpful for early identification and control of the risk factors for death in patients with diabetic foot amputation. It can effectively improve their quality of life and physical and mental health, reduce the mortality rate. |
|
查看全文
Download PDF Reader |
| CLose |
|
|
|
