三种衰弱评估工具对老年慢性心力衰竭患者不良结局预测能力的比较

Comparison of three frailty assessment tools in predicting adverse outcomes in elderly patients with chronic heart failure

目的 比较FRAIL量表(FRAIL scale,FS)、Fried衰弱表型(Fried frailty phenotype,FFP)、埃德蒙顿衰弱量表(Edmonton frail scale,EFS)3种衰弱评估工具对老年慢性心力衰竭(chronic heart failure,CHF)患者的衰弱患病率情况和不良结局的预测性能,并确定研究方法的可行性。方法 便利抽样法选取2020年1月至2022年3月在黄山市人民医院老年医学科及心内科住院的老年CHF患者146例,根据FS评分将研究对象分为FS非衰弱患者(0~2分,81例)和FS衰弱患者(3~5分,65例);根据FFP评分将研究对象分为FFP非衰弱患者(0~2分,65例)和FFP衰弱患者(≥3分,81例);根据EFS评分将研究对象分为EFS非衰弱患者(0~7分,77例)和EFS衰弱患者(≥8分,69例)。出院后进行1年随访,随访终点事件包括全因死亡和(或)再住院,采用logistic回归分析终点事件的影响因素。结果 FS衰弱检出率为44.52%,FFP衰弱检出率为55.48%,EFS衰弱检出率为47.26%。随访1年,FS非衰弱患者与FS衰弱患者、FFP非衰弱患者与FFP衰弱患者、EFS非衰弱患者与EFS衰弱患者发生再住院、全因死亡、终点事件比较,差异有统计学意义(P<0.01)。单因素logistic回归分析显示,年龄、年龄校正Charlson合并症指数、多重用药、过去1年住院、过去1年急诊治疗、体质量指数、血红蛋白、估算肾小球滤过率对CHF患者随访1年终点事件有影响(P<0.05,P<0.01)。在校正相关变量后,多因素logistic回归分析显示,FS、FFP、EFS评价的衰弱仍是发生终点事件的危险因素(OR=8.247,95%CI:2.649~25.674,P=0.001;OR=5.573,95%CI:2.005~15.489,P=0.001;OR=4.743,95%CI:1.612~13.953,P=0.005)。结论 FS、FFP、EFS均可用于CHF住院患者的衰弱评价,且对CHF住院患者1年内的不良结局有良好的预测价值。

Objective To compare the frailty prevalence and the predictive performance of adverse outcomes in elderly patients with chronic heart failure(CHF) with three frailty assessment tools,FRAIL Scale(FS),Fried Frailty Phenotype(FFP) and Edmonton Frail Scale(EFS),and to determine the feasibility of our study method.Methods A total of 146 elderly CHF patients who were hospitalized in Department of Geriatrics and Department of Cardiology of Huangshan People’s Hospital from January 2020 to March 2022 were selected as the study objects by convenience sampling.According to FS,they were divided into non-frail patients(score:0-2,n=81) and frail patients(score:3-5,n=65).Based on FFP,the subjects were divided into non-frail patients(score:0-2,n=65) and frail patients(score:≥3,n=81).Depending on EFS,the patients were divided into non-frail patients(score:0-7,n=77) and frail patients(score:≥8,n=69).All patients were followed up for 1 year after discharge.The endpoint events of the follow-up included all-cause death and/or rehospitalization.Influencing factors of endpoint events was indentified by using logistic regression analysis.Results The prevalence of frailty assessed by FS,FFP,and EFS was 44.52%,55.48% and 47.26%,respectively.After 1 year of follow-up,there were significant differences in rehospitalization,all-cause death,and endpoint events between FS non-frail patients and FS frail patients,FFP non-frail patients and FFP frail patients,and EFS non-frail patients and EFS frail patients(P<0.01).Univariate logistic regression analysis showed that age,age-adjusted Charlson comorbidity index,polypharmacy,hospitalization in the past year,emergency treatment in the past year,body mass index(BMI),hemoglobin and eGFR had an impact on the endpoint events of CHF patients during the 1 year follow-up(P<0.05,P<0.01).After adjusting for relevant variables,multivariate logistic regression analysis indicated that frailty evaluated by FS,FFP and EFS was still a risk factor for the occurrence of endpoint events(OR=8.247,95%CI:2.649-25.674,P=0.001;OR=5.573,95%CI:2.005-15.489,P=0.001;OR=4.743,95%CI:1.612-13.953,P=0.005).Conclusion All FS,FFP and EFS can be used to evaluate the frailty of CHF inpatients,and they have good predictive value for the adverse outcome of CHF patients within 1 year after discharge.