目的 探究老年髋部骨折患者发生急性脑梗死（acute cerebral infarction，ACI）的危险因素并构建风险预测模型。 方法 回顾性选取2023年1月至2025年6月因髋部骨折就诊于天津医院创伤髋关节三科的患者603例，分为训练集（n = 400）和验证集（n = 203）。根据是否发生ACI分为发生组和未发生组，收集患者的临床资料，采用单因素和多因素logistic回顾分析，发现老年髋部骨折患者发生ACI的独立危险因素。基于这些独立危险因素，通过R语言构建列线图预测模型，并评价模型的预测效能。 结果 训练集中纳入的400例老年髋部骨折患者中，有27例患者出现ACI，发生率为6.75％。单因素及多因素logistic分析显示，脑梗死病史、淋巴细胞数、白蛋白、血同型半胱氨酸、胱抑素C、糖化血红蛋白、凝血酶原时间是老年髋部骨折患者合并ACI的独立危险因素（P < 0.05）。基于以上危险因素构建列线图预测模型，受试者工作特征曲线分析显示，训练集和验证集的曲线下面积和95%CI分别为0.896 （0.843 ~ 0.949）、0.830 （0.737 ~ 0.923），预测模型具有较好的区分度。校准曲线显示，预测模型发生与实际发生概率具有较好的一致性。临床决策曲线显示，模型具有较高的临床实用价值。 结论 脑梗死病史、淋巴细胞数、白蛋白、血同型半胱氨酸、胱抑素C、糖化血红蛋白、凝血酶原时间是老年髋部骨折患者发生ACI的独立危险因素，基于7个独立危险因素构建的列线图具有良好的区分度及一致性，有助于临床中在老年髋部骨折患者中识别ACI发生的高危人群。

Objective To identify independent risk factors for acute cerebral infarction （ACI） in elderly patients with hip fracture and to develop a clinically applicable nomogram for individualized risk prediction. Methods This retrospective study included 603 elderly patients with hip fracture admitted to the Third Department of Traumatic Hip Surgery， Tianjin Hospital， China， between January 2023 and June 2025. Patients were randomly allocated to a training cohort （n = 400） and a validation cohort （n = 203） and categorized based on the occurrence of ACI. Clinical and laboratory variables were collected. Univariate and multivariable logistic regression analyses were performed to identify independent predictors of ACI. A nomogram was constructed using R software based on these predictors， and its performance was evaluated in terms of discrimination， calibration， and clinical utility. Results In the training cohort， 27 of 400 patients （6.75%） developed ACI. Multivariable logistic regression identified seven independent predictors： prior cerebral infarction， lower lymphocyte count， lower albumin level， elevated serum homocysteine， higher cystatin C， elevated glycated hemoglobin （HbA_1c）， and prolonged prothrombin time （all P < 0.05）. The nomogram based on these factors demonstrated excellent discrimination， with an area under the ROC curve （AUC） of 0.896 （95%CI： 0.843 ? 0.949） in the training cohort and 0.830 （95%CI： 0.737 ? 0.923） in the validation cohort. Calibration plots showed good agreement between predicted and observed probabilities of ACI. Decision curve analysis confirmed the model's favorable clinical net benefit across a range of threshold probabilities. Conclusions Prior cerebral infarction， lymphopenia， hypoalbuminemia， hyperhomocysteinemia， elevated cystatin C， higher HbA1c， and prolonged prothrombin time are independent predictors of ACI in elderly hip fracture patients. The developed nomogram exhibits robust discrimination， calibration， and clinical utility， offering a practical tool to facilitate early identification and targeted prevention of ACI in this high-risk population.