脑卒中偏瘫患者继发骨质疏松症的影响因素及列线图预测模型构建

The influencing factors of secondary osteoporosis in stroke patients with hemiplegia and the construction of a predictive model of nomogram

目的探讨脑卒中偏瘫患者继发骨质疏松症的影响因素,构建列线图预测模型并进行评估。方法研究对象为2019年6月至2023年6月于临沂市中心医院就诊并治疗的110例脑卒中后偏瘫患者,通过检测骨密度分为骨质疏松症组和非骨质疏松症组。对患者临床资料和实验室指标进行收集。采用单因素分析及二元Logistic多因素回归分析筛选影响因素,采用R软件(R3.3.2)和软件包rms构建列线图预测模型。结果骨质疏松者52例,非骨质疏松者58例。骨质疏松组中女性、单侧前循环多发梗死及晒太阳较少者所占比例显著高于非骨质疏松组,差异有统计学意义(χ^(2)=8.27、14.77、6.96,P<0.05);收缩压(159.32±21.72 vs.151.67±19.52)、总胆固醇(4.29±0.50 vs.3.57±0.42)、低密度脂蛋白胆固醇(low density lipoprotein-cholesterol,LDL-C)(2.87±0.33 vs.2.04±0.31)、同型半胱氨酸(homocysteine,Hcy)(3.81±2.51 vs.112.33±2.47)水平显著高于非骨质疏松组,差异有统计学意义(t=5.23、8.38、7.98、5.63,P均<0.001);白蛋白(38.15±5.21 vs.33.26±5.000.73)显著低于非骨质疏松组,差异有统计学意义(t=4.90,P<0.05)。二元Logistic回归分析结果显示,女性、较高的总胆固醇、LDL-C、Hcy水平及较少晒太阳是脑卒中偏瘫患者继发骨质疏松症的独立危险因素(P<0.05)。列线图模型预测的准确性和有效性较好,具有较高的区分度和一致性。ROC曲线分析结果显示,构建模型预测脑卒中偏瘫患者继发骨质疏松症的曲线下面积为0.891(0.833~0.949),灵敏度和特异度分别为80.8%和79.3%。结论基于脑卒中偏瘫患者继发骨质疏松症的影响因素构建的列线图模型,可为脑卒中偏瘫发生骨质疏松症高危人群的识别和早期干预提供一定参考依据。

Objective To investigate the influencing factors of secondary osteoporosis in stroke patients with hemiplegia,and to construct a nomogram prediction model and evaluate it.Methods The study subjects were 110 patients with hemiplegia after stroke who were treated in our hospital from Jun.2019 to Jun.2023,and were divided into osteoporosis group and non-osteoporosis group by bone mineral density detection.Clinical data and laboratory indicators were collected.Single factor analysis and binary Logistic multiple factor regression analysis were used to screen the influencing factors.R software(R3.3.2)and software package rms were used to construct the nomogram prediction model.Results There were 52 patients with osteoporosis and 58 patients without osteoporosis.In the osteoporosis group,the proportion of female,unilateral anterior circulation multiple infarction and less sunlight was significantly higher than that in the non-osteoporosis group,with statistical significance(χ^(2)=8.27,14.77 and 6.96,respectively,P<0.05).Systolic blood pressure(159.32±21.72 vs.151.67±19.52),total cholesterol(4.29±0.50 vs.3.57±0.42),LDL-C(2.87±0.33 vs.2.04±0.31),Hcy(3.81±2.51 vs.The level of 112.33±2.47 was significantly higher than that of the non-osteoporosis group,and the difference was statistically significant(t was 5.23,8.38,7.98 and 5.63,respectively,P<0.001).The level of albumin(38.15±5.21 vs.33.26±5.73)was significantly lower than that of the non-osteoporosis group.The difference was statistically significant(t=4.90,P<0.05).Binary Logistic regression analysis showed that female,higher total cholesterol,LDL-C,Hcy levels and less sun exposure were independent risk factors for secondary osteoporosis in stroke patients with hemiplegia(P<0.05).ROC curve analysis results showed that the area under the curve of the established model to predict secondary osteoporosis in stroke hemiplegia patients was 0.891(0.833-0.949),and the sensitivity and specificity were 80.8%and 79.3%,respectively.Conclusion The categorization and consistency of the nomogram model based on the influencing factors of secondary osteoporosis in patients with stroke hemiplegia are good,which can provide a certain reference for the identification and early intervention of high-risk groups with stroke hemiplegia.