Li An, Wei Gao, Liqun Ren, Jing Wang, Yao Wang
The developed nomogram incorporating nine accessible predictors demonstrated robust discrimination and clinical applicability for sarcopenia risk assessment in Chinese CKM patients, supporting its use in early screening and individualized intervention.
BACKGROUND: Sarcopenia research based on the China Health and Retirement Longitudinal Study (CHARLS) frequently employs an anthropometric equation to estimate appendicular skeletal muscle mass (ASM). Liu et al. recently questioned the reliability of this equation. We aimed to validate the equation against bioelectrical impedance analysis (BIA) measurements and to examine the impact of different cutoff values for low muscle mass used in CHARLS-based studies. METHODS: We retrospectively analyzed BIA data from 548 diabetic patients (age range 23-90 years, median 62 years; 41.06% women). ASM measured by BIA (bASM) was compared with ASM calculated by the equation (cASM). Spearman correlation, intraclass correlation coefficient (ICC), and Bland‑Altman analysis were performed. Two commonly used height‑adjusted ASM (ASM/Ht2) cutoff sets were tested against the AWGS 2019 reference standard. RESULTS: cASM correlated strongly with bASM (r = 0.94, p < .001). ICC demonstrated excellent agreement (0.933; 95% CI 0.919-0.945). Bland‑Altman plots showed acceptable limits of agreement. One cutoff set (men 7.05, women 5.63 kg/m2) showed almost perfect agreement with AWGS 2019 (Kappa = 0.909, McNemar p = 1.000), whereas the other set (men 6.79, women 4.90 kg/m2) showed only moderate agreement (Kappa = 0.492) and a significant level of discordance (McNemar p < .001). CONCLUSION: The ASM estimation equation remains valid for CHARLS-based sarcopenia research, including in diabetic populations. To avoid misclassification and ensure cross-study comparability, we recommend using AWGS 2019-aligned cutoffs.