Abstract:

Objective　To analyze the influencing factors of death within half a year after hip fracture surgery in the elderly and construct an individualized prediction nomogram model. Methods　The clinical data of 458 elderly patients with hip fracture who underwent surgical treatment in Xi'an Honghui Hospital from May 2020 to August 2022 were retrospectively analyzed, including 214 males and 244 females, aged 65-94 (75.89±7.72) years. The patients were divided into a death group (43 cases) and a survival group (415 cases) according to whether they survived within half a year after surgery. The clinical data of the two groups were collected, and the influencing factors of death within half a year after hip fracture surgery were analyzed by multivariate logistic regression analysis. R software was used to draw the nomogram model, and the consistency of the nomogram model was verified by the calibration curve. The receiver operating characteristic curve (ROC) was used to evaluate the prediction value and discrimination of the model. Independent sample t test, Mann-Whitney U test, and χ² test were used for statistical analysis. Results　The mortality rate of elderly patients within half a year after hip fracture surgery was 9.39% (43/458). Multivariate logistic regression analysis showed that age (OR=1.167, 95%CI: 1.088-1.250) and type of comorbidities (OR=1.986, 95%CI: 1.143-3.452) were risk factors for death within half a year after hip fracture surgery in elderly patients, and skeletal muscle content (OR=0.814, 95%CI: 0.715-0.925) and hemoglobin (OR=0.923, 95%CI: 0.868-0.981) were protective factors (all P<0.05). The area under the curve was 0.899 (95%CI: 0.888-0.923) of the nomogram model in the prediction of death within half a year after hip fracture surgery in elderly patients, with good discrimination. The slope of calibration curve of the nomogram model for the prediction of death within half a year after hip fracture surgery in elderly patients was close to 1, and the Hosmer-Lemeshow goodness-fit test showed good consistency (χ²=4.091, P=0.849). Conclusions　This study constructed a nomogram model based on four influencing factors: age, type of comorbidities, skeletal muscle content, and hemoglobin, to predict the mortality risk within half a year after hip fracture surgery in elderly patients. It had relatively good discriminability and consistency, and could provide a reference basis for developing personalized intervention measures in clinical practice.

Key words:

Hip fracture, Elderly, Half a year after surgery, Death, Influencing factors, Nomogram

摘要：

目的　分析老年髋部骨折术后半年死亡的影响因素并构建个体化预测列线图模型。方法　回顾性分析西安市红会医院2020年5月至2022年8月行手术治疗的458例老年髋部骨折患者的临床资料，其中男214例、女244例，年龄65～94（75.89±7.72）岁。根据术后半年内是否存活分为死亡组（43例）和存活组（415例）。收集两组临床资料，采用多因素logistic回归分析老年髋部骨折术后半年死亡的影响因素；采用R软件绘制列线图模型，并通过校准曲线验证列线图模型的一致性，采用受试者操作特征曲线（ROC）评估模型预测价值及区分度。采用独立样本t检验、Mann-Whitney U检验、χ²检验进行统计学分析。结果　老年髋部骨折患者术后半年死亡率为9.39%（43/458）。多因素logistic回归分析显示，年龄（OR=1.167，95%CI：1.088～1.250）、合并症种类（OR=1.986，95%CI：1.143～3.452）为老年髋部骨折术后半年死亡的危险因素，骨骼肌含量（OR=0.814，95%CI：0.715～0.925）、血红蛋白（OR=0.923，95%CI：0.868～0.981）为保护因素（均P<0.05）。构建的列线图模型预测老年髋部骨折术后半年死亡的曲线下面积为0.899（95%CI：0.888～0.923），区分度较好。列线图模型预测老年髋部骨折术后半年死亡的校准曲线斜率趋近于1，且Hosmer-Lemeshow拟合优度检验χ²=4.091，P=0.849，一致性较好。结论　本研究基于年龄、合并症种类、骨骼肌含量、血红蛋白4项影响因素构建预测老年髋部骨折术后半年死亡风险的列线图模型，区分度及一致性较好，能够为临床制定个体化干预措施提供参考依据。

关键词:

髋部骨折, 老年, 术后半年, 死亡, 影响因素, 列线图