Scientific Programme
Abstract
Register
Abstract
Year 2021
October 2021

SHBC1227

Abstract Title
SARC-F: Defining a Validated Cut-off for Screening of Sarcopenia in Community Dwelling Older Persons (A 3-year follow-up study)
Authors

JM CABRADILLA-ESCANLAR1, WS LIM1, A YEO1, S YEW1, N HAFIZAH1, JP LIM1

Institutions

Tan Tock Seng Hospital1

Background & Hypothesis

The Asian Working Group for Sarcopenia (AWGS) recommends using SARC-F≥4 cutoff to screen for sarcopenia.  Studies have shown that SARC-F≥4 has low sensitivity for case detection in the community. Using lower cutoffs of SARC-F, we compared diagnostic performance, concurrent validity and predictive validity in well older persons.

Methods

We studied 200 independent community dwelling older adults from GeriLABS study. We measured muscle mass, grip-strength and physical performance at baseline and 3 years. We performed Receiver operating characteristic (ROC) analysis to determine area under curve (AUC).  Concurrent validity was assessed by comparisons of baseline measures between non-sarcopenic and sarcopenic groups defined by different SARC-F cutoffs. We compared predictive validity of SARC-F cutoffs against AWGS criteria for incident falls and outcome measures at 3 years using regression analysis.

Results

Compared with SARC≥4 (AUC:0.509, sensitivity:0.5%, specificity:100%), sensitivity improved with lower cutoffs (SARC-F≥2:32.8%, SARC-F≥3:13.8%, SARC-F≥3:3.4%) with best diagnostic performance for SARC-F≥2  (AUC:0.530). SARC-F≥2 showed concurrent validity for Short Physical Performance battery (SPPB) (Non-sarcopenia vs sarcopenia:11.46 vs 9.79, p<0.001), gait speed (1.15 vs 1.03, p=0.021) and chair-stand (10.33 vs 11.53, p=0.072). For predictive validity,  AWGS criteria significantly predicted low muscle mass and grip strength, whereas  SARC-F≥1 or SARC-F≥2 predicted only lower SPPB at 3 years. Neither AWGS nor SARC-F predicted falls.

Discussion & Conclusion

Amongst healthy community-dwelling older adults, lowering SARC-F cutoff to 2 demonstrates concurrent validity for muscle function, predictive validity only for physical performance, and improves discriminatory ability for sarcopenia screening, although sensitivity remains low.

Top