老年急性脑梗死后吞咽障碍患者发生衰弱的风险预测列线图模型构建与验证

Construction and validation of a risk-prediction nomogram model for frailty in elderly patients with dysphagia after acute cerebral infarction

目的探讨老年急性脑梗死(ACI)后吞咽障碍患者发生衰弱的的危险因素,并建立风险预测列线图模型。方法回顾性分析2020年2月至2022年3月苏州市中西医结合医院收治的195例老年ACI后吞咽障碍患者的临床资料,按2∶1的比例将其分为建模组(130例)和验证组(65例),并根据是否衰弱将建模组分为衰弱组(83例)和非衰弱组(47例)。采用SPSS 26.0统计软件进行数据分析。根据数据类型,分别采用t检验或χ^(2)检验进行组间比较。采用多因素logistic回归分析建模组老年ACI后吞咽障碍患者发生衰弱的危险因素,应用R3.6.1软件建立风险预测列线图模型。绘制受试者工作特征(ROC)曲线、校准曲线和决策性曲线分析(DCA)评估列线图模型的预测效能、准确度和临床效益,并使用验证组评估列线图的可行性。结果195例患者衰弱发生率为64.10%(125/195);建模组衰弱发生率为63.85%(83/130);验证组衰弱发生率为64.62%(42/65)。多因素logistic回归分析显示,年龄>75岁、吞咽障碍病程≥21 d、独居、自费、经济水平较差、合并症≥3种、服药种类≥3种、肢体运动障碍、抑郁、营养不良及低社会支持水平均是老年ACI后吞咽障碍患者发生衰弱的危险因素(OR=3.618,4.459,4.358,2.843,3.102,2.130,2.659,3.770,3.501,4.646,1.887;P<0.05);膳食指导是其保护因素(OR=0.570;P<0.001)。ROC曲线分析显示建模组、验证组列线图预测模型的曲线下面积(AUC)分别为0.853、0.844,灵敏度分别为78.31%、76.19%,特异度分别为85.11%、82.61%;建模组和验症组校准曲线的一致性指数分别为0.798、0.793,二者校准曲线均接近标准曲线,且两组DCA结果均显示列线图模型具有良好的净收益。结论年龄>75岁、吞咽障碍病程、独居、自费、经济水平较差、合并症≥3种、服药种类≥3种、肢体运动障碍、抑郁、营养不良、膳食指导及低社会支持水平均是老年ACI后吞咽障碍患者发生衰弱的影响因素,根据上述影响因素构建列线图模型有助于筛选高危患者,指导临床早期干预。

Objective To investigate the risk factors of frailty in the elderly patients with dysphagia after acute cerebral infarction(ACI)and to establish a risk-prediction nomograph model.Methods A retrospective analysis was conducted of the clinical data of 195 elderly patients with dysphagia after ACI,who were treated in Suzhou Hospital of Integrated Traditional and Western Medicine from February 2020 to March 2022.They were randomly divided into a modeling group(n=130)and validation group(n=65)in a ratio of 2∶1,and those in the modeling group were further divided into frailty group(n=83)and non-frailty group(n=47)according to the occurrence of frailty.Data statistical analysis was conducted using SPSS 26.0.t test andχ^(2)test was employed for inter-group comparisons based on the data type.Multivariate logistic regression was used to analyze the risk factors of frailty in the elderly patients with dysphagia after ACI in the modeling group,and a risk-prediction nomograph model was established using R3.6.1 software.The receiver operating characteristic(ROC)curve,calibration curve,and decision curve analysis(DCA)were drawn to evaluate the predictive efficiency,accuracy and clinical benefit of the nomogram model,and the validation group was used to evaluate the feasibility of the nomogram.Results The incidence of frailty in all 195 patients was 64.10%(125/195),63.85%(83/130)in the modeling group,and 64.62%(42/65)in the validation group.Multivariate logistic regression analysis showed that the age above 75 years,course of swallowing disorders≥21 d,living alone,self-financing,poor economic level,≥3 complications,≥3 medications,limb movement disorders,depression,malnutrition,and low social support level were risk factors for the frailty of elderly patients with dysphagia after ACI(OR=3.618,4.459,4.358,2.843,3.102,2.130,2.659,3.770,3.501,4.646,1.887;P<0.05),and that dietary guidance was a protective factor(OR=0.570;P<0.001).The ROC curve analysis showed that the area under the curve(AUC)of the prediction nomograph model was 0.853 for the modeling group and 0.844 for the validation group,with a sensitivity of 78.31%and 76.19%,and a specificity of 85.11%and 82.61%,respectively.The consistency index of the calibration curves was 0.798 for the modeling group and 0.793 for the validation group,and both calibration curves were close to the standard curve.The DCA results in both groups showed that the nomograph model had good net returns.Conclusion Age above 75 years,course of dysphagia,living alone,self-financing,poor economic level,≥3 complications,≥3 medications,limb movement disorder,depression,malnutrition,dietary guidance and low social support level are all influencing factors for the frailty in the elderly patients with dysphagia after ACI,and the nomograph model constructed based on the above factors is helpful to screen high-risk patients and guide early clinical intervention.