Year 2021
October 2021


Abstract Title
Frailty and Sarcopenia: Prevalence and Role in Prognostication in Asian COPD



Tan Tock Seng Hospital1

Background & Hypothesis

To determine the prevalence of frailty and sarcopenia among subjects with chronic obstructive pulmonary disease(COPD) and assess their association with mid-term survival.


A prospective multi-centre observational study. Subjects with stable disease of GOLD Classification Stages2 were recruited from specialist clinics and a pulmonary rehabilitation service. Data for disease severity, physical performance, frailty, sarcopenia and co-morbidity burden were collected using validated scores. The participants were followed-up at 6-monthly intervals for 2 years. The area under receiver-operating-characteristic curve(AUC) analysis was applied to evaluate association with 12-month mortality. We applied the Kaplan-Meier method to evaluate survival time from point of recruitment.


136 participants were enrolled. They were predominantly male(94.9%), of Chinese ethnicity(80.9%), with a mean(SD) age of 71(8.69) years. The mean predicted FEV1 was 43.0(14.8%). 52.2% had GOLD stage 3 and 18.4% had GOLD stage 4 COPD. Using the FRAIL questionnaire, 42.9% were pre-frail and 11% frail; using Fried’s Frailty Index (FFI), 54.4% were pre-frail and 27.9% frail. Based on the SARC-F questionnaire, 11.8% were sarcopenic. FRAIL(AUC=0.761), FFI(AUC=0.754) and SARC-F(AUC=0.746) had moderately high AUCs when predicting 12-month survival. With the exception of the modified Medical Research Council dyspnoea scale(AUC=0.814), conventional variables like FEV1(AUC=0.237), body mass index(AUC=0.247) and Charlson Comorbidity Index(AUC=0.506) had poorer prognostic ability. Significant differences in survival were observed among robust, pre-frail and frail groups, also between groups with and without sarcopenia.

Discussion & Conclusion

Frailty and sarcopenia were observed in a significant proportion of subjects and associated with shorter survival. Future studies should evaluate incorporating these measures into a composite index to improve prognostication.