DXA-BIA discordance in low muscle mass and sarcopenia among older Korean adults under the AWGS 2025 criteria.

PURPOSE: To compare dual-energy X-ray absorptiometry (DXA)- and bioelectrical impedance analysis (BIA)-based classifications of low muscle mass and sarcopenia under the Asian Working Group for Sarcopenia (AWGS) 2025 criteria. METHODS: We analyzed nationally representative 2024 Korea National Health and Nutrition Examination Survey data from 1178 adults aged ≥65 years with both DXA and BIA measurements (survey-weighted mean age, 72.1 years; 54.3% women). Sarcopenia analyses included 1121 participants with handgrip-strength data, including 131 with low handgrip strength. Low muscle mass was defined using modality-specific height-adjusted, body mass index (BMI)-adjusted, and combined height-or-BMI criteria. Survey weights were used for prevalence estimates, whereas agreement statistics were unweighted. RESULTS: Under the combined criterion, low-muscle-mass prevalence was 83.0% by DXA and 55.5% by BIA; observed agreement was 65.4%, and Cohen's kappa was 0.246. Observed agreement was highest for the BMI-adjusted criterion (50.6% vs 41.1%; agreement, 79.8%; kappa, 0.598) and lowest for the height-adjusted criterion (70.8% vs 28.8%; agreement, 56.5%; kappa, 0.258). Among participants with low handgrip strength, combined low-muscle-mass prevalence was 93.1% by DXA and 80.3% by BIA, with 18 DXA-positive/BIA-negative and 4 DXA-negative/BIA-positive classifications. Sarcopenia prevalence was 10.0% by DXA and 8.7% by BIA (kappa, 0.895); 990 participants with normal handgrip strength were concordantly non-sarcopenic. CONCLUSION: DXA and BIA yielded method-dependent low-muscle-mass classifications under AWGS 2025. Similar overall sarcopenia prevalence largely reflected the shared handgrip-strength requirement, whereas discordance persisted among participants with low handgrip strength.