脑梗死后下肢功能障碍老年患者发生骨质疏松的预测模型构建

Construction of a predictive model for osteoporosis in elderly patients with lower limb dysfunction after cerebral infarction

摘要:

目的 探究脑梗死后下肢功能障碍老年患者发生骨质疏松的影像因素并构建预测模型。方法 回顾性分析2021年1月—2024年1月南京医科大学附属脑科医院纳入的138例脑梗死后下肢功能障碍老年患者，根据是否发生骨质疏松分为骨质疏松组（31例）和非骨质疏松组（107例），比较两组的一般资料，采用多因素一般Logistic回归分析，构建脑梗死后下肢功能障碍老年患者发生骨质疏松的预测模型并构建列线图。结果 骨质疏松组的年龄、脑梗死病程、吞咽障碍率、睡眠状况不良率均高于非骨质疏松组（P <0.05），Fugl-Meyer运动功能评定量表评分、Holden步行功能分级、骨密度、骨钙素、维生素D水平均低于非骨质疏松组（P <0.05）。多因素一般Logistic回归分析结果表明：年龄大［O^R=1.272（95% CI：1.149，1.407）］、脑梗死病程长［O^R=1.755（95% CI：1.100，2.801）］、Fugl-Meyer运动功能评定量表评分低［O^R=0.739（95% CI：0.574，0.952）］、吞咽障碍［O^R=5.199（95% CI：1.343，20.126）］、睡眠状况不良［O^R=5.174（95% CI：1.315，20.353）］、Holden步行功能分级低［O^R=0.234（95% CI：0.098，0.560）］、骨密度低［O^R=0.196（95% CI：0.057，0.670）］、骨钙素低［O^R=0.885（95% CI：0.802，0.977）］、维生素D水平低［O^R=0.771（95% CI：0.624，0.952）］均为脑梗死后下肢功能障碍老年患者发生骨质疏松的危险因素（P <0.05）。多因素一般Logistic回归模型预测的曲线下面积为0.985（95% CI：0.970，1.000），敏感性为96.8%（95% CI：0.833，0.999），特异性为98.1%（95% CI：0.934，0.998）。结论 脑年龄大、脑梗死病程长、下肢运动功能差、吞咽障碍、睡眠状况不良、步行功能差、骨密度低、骨钙素低、维生素D水平低的梗死后下肢功能障碍老年患者更易发生骨质疏松，为早期识别和干预骨质疏松提供了科学依据，并为脑梗死后老年患者的综合管理提供了新思路。

Abstract:

Objective To evaluate a predictive model for osteoporosis in elderly patients with lower limb dysfunction after cerebral infarction.Methods A total of 138 elderly patients with lower limb dysfunction after cerebral infarction admitted to Nanjing Brain Hospital from January 2021 to January 2024 were enrolled. They were divided into an osteoporosis group (n = 31) and a non-osteoporosis group (n = 107) based on the presence of osteoporosis. General data were compared between the two groups. A predictive model for osteoporosis in elderly post-stroke patients with lower limb dysfunction was constructed based on the multivariable logistic regression analysis, and a nomogram was developed.Results The osteoporosis group had higher age, longer duration of cerebral infarction, and higher rates of dysphagia and poor sleep (P < 0.05), while exhibiting lower Fugl-Meyer Assessment of Lower Extremity scores, Functional Ambulation Classification scores, bone mineral density (BMD), osteocalcin levels, and vitamin D levels compared to the non-osteoporosis group (P < 0.05). Multivariable logistic regression analysis revealed that advanced age [O^R=1.272 (95% CI: 1.149, 1.407) ], prolonged duration of cerebral infarction [O^R=1.755 (95% CI: 1.100, 2.801) ], low Fugl-Meyer Assessment of Lower Extremity scores [O^R=0.739 (95% CI: 0.574, 0.952) ], dysphagia [O^R=5.199 (95% CI: 1.343, 20.126) ], poor sleep [O^R=5.174 (95% CI: 1.315, 20.353) ], low Functional Ambulation Classification scores [O^R=0.234 (95% CI: 0.098, 0.560) ], low BMD [O^R=0.196 (95% CI: 0.057, 0.670) ], low osteocalcin levels [O^R=0.885 (95% CI: 0.802, 0.977) ], and low vitamin D levels [O^R=0.771 (95% CI: 0.624, 0.952) ] were all risk factors for osteoporosis in elderly post-stroke patients with lower limb dysfunction (P < 0.05). The predictive model based on the multivariable logistic regression analysis achieved an area under the curve of 0.985 (95% CI: 0.970, 1.000), a sensitivity of 96.8% (95% CI: 0.833, 0.999), and a specificity of 98.1% (95% CI: 0.934, 0.998).Conclusion Elderly patients with lower limb dysfunction after cerebral infarction are more prone to developing osteoporosis if they are older, have a longer disease course, poor lower limb motor function, swallowing disorders, poor sleep quality, impaired walking ability, low BMD, low osteocalcin levels, and reduced vitamin D levels. These findings provide a scientific basis for the early identification and intervention of osteoporosis and offer new insights for the comprehensive management of elderly patients following cerebral infarction.