PURPOSE

To compare the prognostic performance of the Asian Working Group for Sarcopenia (AWGS) 2025 diagnostic constructs-specifically height- versus body mass index (BMI)-indexed muscle mass-for discriminating functional outcomes in patients with stroke.

METHODS

From a cohort of 1246 consecutive post-acute stroke admissions, we analyzed 864 inpatients (median age 76 years; 47.2% women) who met the inclusion criteria. Sarcopenia was classified according to AWGS 2025 criteria: possible sarcopenia (low handgrip strength) and confirmed sarcopenia (low strength plus low skeletal muscle mass).

Skeletal muscle mass was indexed by height-squared (SMI/height2) and BMI (SMI/BMI). Multivariable linear regression analysis, stratified by age (50-64 vs. ≥ 65 years) and sex, assessed the associations with the Functional Independence Measure motor subscale (FIM-motor) at discharge.

RESULTS

Confirmed sarcopenia defined by SMI/height 2 was independently and consistently associated with lower discharge FIM-motor scores across all age and sex strata (men ≥ 65 years: B - 3.51, P = 0.038; women ≥ 65 years: B - 3.08, P = 0.024; men 50-64 years: B - 4.59, P = 0.044; women 50-64 years: B - 9.19, P = 0.045).

In contrast, SMI/BMI-based definitions showed poor prognostic utility, with no significant associations in men. Possible sarcopenia demonstrated limited, inconsistent associations.

CONCLUSION

AWGS 2025 confirmed sarcopenia utilizing height-indexed muscle mass provided superior prognostic discrimination for discharge ADL compared with BMI-indexed or possible sarcopenia constructs in post-stroke inpatients.

These findings support prioritizing height-indexed confirmed sarcopenia for functional risk stratification in this setting.