Construction of a risk prediction model for poor prognosis in elderly coronary heart disease patients with frailty
Wu Jing
Zhang Li
Tian Xiaomiao
Yang Ling
Zhang Wuqiong
Liang Min
Yao Na
Shi Yuan
Abstract：Objective To construct a risk prediction nomogram model for poor prognosis in elderly coronary heart disease(CHD)patients with frailty.Methods A total of 120 elderly CHD patients with frailty admitted to the First Affiliated Hos-pital of Xinjiang Medical University from July 2021 to July 2023 were enrolled.Based on unscheduled visits,major adverse cardiac and cerebrovascular events(MACCE),or all-cause death during follow-up,patients were assigned to poor prognosis(n=66)and good prognosis(n=54)groups.Clinical data were compared between groups.Logistic multivariate regression analysis was used to identify risk factors for poor prognosis.Based on these factors,a nomogram prediction model was con-structed using R Studio 3.6.5 software and the rms package.The model's calibration,predictive performance,and clinical net benefit were evaluated using the Hosmer-Lemeshow goodness-of-fit test,receiver operating characteristic(ROC)curves,and decision curve analysis(DCA),respectively.Results During follow-up,66 patients experienced poor prognosis events,with an incidence of 55.00%.The poor prognosis group had lower left ventricular ejection fraction(LVEF)levels,and higher Charlson Comorbidity Index,Fried Frailty Score,Mini Nutritional Assessment-Short Form(MNA-SF)score,homocysteine(Hcy)levels,and proportions of patients with New York Heart Association(NYHA)class Ⅲ-Ⅳ heart function,smoking history,and taking≥7 types of medication than the good prognosis group,with statistically significant differences(P<0.05).Logistic multivariate regression analysis showed that LVEF<60%,Charlson Comorbidity Index≥2,Fried Frailty Score≥4,MNA-SF score≥11,Hcy level≥15 μmol/L,NYHA class Ⅲ-Ⅳ,and smoking history were independent risk factors for poor prognosis(P<0.05).The calibration curve of the nomogram model based on these risk factors showed high agreement between predicted and observed curves within a prediction probability range of 19%-77%.Internal validation showed the model had an area under the curve(AUC)of 0.885 for predicting poor prognosis,with a sen-sitivity of 85.71%and a specificity of 88.24%.The Kolmogorov-Smirnov goodness-of-fit test indicated good model calibra-tion(χ2=3.268,R2=0.857,P>0.05).DCA results demonstrated that within a threshold probability range of 5%-92%,the model provided higher net benefit than both the"intervene-for-all"and"intervene-for-none"strategies.Conclusion LVEF<60%,Charlson Comorbidity Index≥2,Fried Frailty Score≥4,MNA-SF score≥11,Hcy level≥15 μmol/L,NYHA class Ⅲ-Ⅳ,and smoking history are independent risk factors for poor prognosis in elderly CHD patients with frail-ty.The constructed nomogram model demonstrates good discrimination,calibration,and clinical application value,provi-ding a reference for risk assessment and management strategy development in clinical practice for this patient population.
Keywords：Coronary heart diseaseFrailtyElderly patientsMajor adverse cardiac and cerebrovascular eventsRisk modelModel validation
Publication Date：2025-12-20
Online Publishing Date：2026-03-09(First online date of this platform, not the publication date of the document)
Pages：7( 1-7 )