| 杨英,刘璐萍,韩康,等.脑卒中患者伴发偏瘫侧髋部骨折的危险因素分析.骨科,2026,17(4): 354-359. |
| 脑卒中患者伴发偏瘫侧髋部骨折的危险因素分析 |
| Analysis of risk factors for hip fracture on the hemiplegic side in stroke patients |
| 投稿时间：2025-10-31 |
| DOI：10.3969/j.issn.1674-8573.2026.04.009 |
| CN KeyWords: 脑卒中 偏瘫 髋部骨折 危险因素 |
| EN KeyWords: Stroke Hemiplegia Hip fracture Risk factors |
| Fund Project: |
|
| View Counts: 239 |
| PDF Download Counts: 67 |
| CN Abstract: |
| 目的 探讨脑卒中患者伴发偏瘫侧髋部骨折的危险因素。方法 回顾性分析2023年1月至2024年12月我院康复科及骨科收治的436例脑卒中患者，根据是否出现偏瘫侧髋部骨折，分为骨折组（28例）和非骨折组（408例），分别记录两组患者相关临床资料及评定结果，采用单因素和多因素Logistic回归分析脑卒中患者伴发偏瘫侧髋部骨折的危险因素，并建立回归模型绘制受试者工作特征（receiver operating characteristic，ROC）曲线。结果 单因素分析显示两组患者跌倒次数、Brunnstrom分期、Ashworth分级、Berg平衡量表（Berg Balance Scale，BBS）评分、美国国立卫生研究院卒中量表（National Institute of Health Stroke Scale，NIHSS）评分、Hold步态分级比较，差异有统计学意义（P<0.05）。多因素Logistic回归显示跌倒次数（OR=3.566，95% CI：1.471~8.642），P=0.005）、Brunnstrom分期（OR=0.007，95% CI：0.001~0.689，P=0.034）、BBS评分（OR=0.599，95% CI：0.489~0.734，P<0.001）、NIHSS评分（OR=1.457，95% CI：1.104~1.924，P=0.008）是脑卒中患者伴发偏瘫侧髋部骨折的独立危险因素。Ashworth分级、Hold步态分级可能与骨折风险存在关联趋势，但需更大样本量验证。模型预测灵敏度为85.7%，特异度为94.4%，Hosmer-Lemeshow检验显示χ2=2.988，P=0.935，ROC曲线下面积（area under curve，AUC）为0.963（P<0.001），提示拟合优度良好。Bootstrap内部验证（1 000次重复抽样）显示，验证后模型AUC=0.921，95% CI：0.886~0.957，灵敏度82.1%，特异度91.3%，提示模型具有较好的稳定性和预测一致性。结论 跌倒次数增加、下肢运动分期低下、BBS评分低、NIHSS评分高是脑卒中患者伴发偏瘫侧髋部骨折的危险因素。Ashworth分级、Hold步态分级可能与骨折风险存在关联趋势，但需更大样本量验证。 |
| EN Abstract: |
| Objective To explore the risk factors for hip fracture on the hemiplegic side in stroke patients. Methods A retrospective analysis was conducted on 436 stroke patients admitted from January 2023 to December 2024. The patients were divided into a fracture group (28 cases) and a non-fracture group (408 cases) based on whether they had a hip fracture on the hemiplegic side. The relevant clinical data and assessment results of the two groups were recorded. Univariate and multivariate Logistic regression analyses were used to identify the risk factors for hip fracture on the hemiplegic side in stroke patients. A regression model was established and a receiver operating characteristic (ROC) curve was drawn. Results Univariate analysis showed that significant differences were found between the two groups in the number of falls, Brunnstrom stage, Ashworth grade, Berg Balance Scale (BBS), National Institutes of Health Stroke Scale (NIHSS), and Hold gait function classification (P<0.05). Multivariate Logistic regression analysis indicated that the number of falls (OR=3.566, 95% CI: 1.471-8.642, P=0.005), Brunnstrom stage (OR=0.007, 95% CI: 0.001-0.689, P=0.034), BBS (OR=0.599, 95% CI: 0.489-0.734, P<0.001), and NIHSS (OR=1.457, 95% CI: 1.104-1.924, P=0.008) were independent risk factors for hip fracture on the hemiplegic side in stroke patients. Ashworth grade and Hold gait classification may be associated with fracture risk, but further verification with larger sample sizes is needed. The model's predictive sensitivity was 85.7%, and specificity was 94.4%. The Hosmer-Lemeshow test showed χ2=2.988, P=0.935, and the area under curve (AUC) was 0.963 (P<0.001), indicating a good fit. To verify the stability of the model, Bootstrap internal validation (1 000 repeated samplings) was performed. The results showed that the AUC of the validated model was 0.921, 95% CI: 0.886-0.957, with a sensitivity of 82.1% and a specificity of 91.3%, suggesting that the model has good stability and predictive consistency. Conclusion An increased number of falls, lower limb motor stage, low BBS score, and high NIHSS score are independent risk factors for hip fracture on the hemiplegic side in stroke patients. Ashworth grade and Hold gait classification may be associated with fracture risk, but further verification with larger sample sizes is needed. |
|
查看全文
Download PDF Reader |
| CLose |
|
|
|
