Comparison of sarcopenia screening indices using serum creatinine and cystatin C in metabolic dysfunction-associated...

Metabolic dysfunction-associated steatotic liver disease (MASLD) and sarcopenia share underlying pathophysiological mechanisms and can bidirectionally influence the development and progression of each other. Diagnosing sarcopenia in individuals with MASLD is challenging due to overlapping symptoms and the frequent requirement for expensive, specialized equipment for muscle mass assessment. Therefore, accessible screening methods are crucial. Serum indices based on creatinine (Cr) and cystatin C (CysC), including Calculated Body Muscle Mass (CBMM), Sarcopenia Index (SI), and estimated glomerular filtration rate (eGFR) ratio, have emerged as potential biomarkers for sarcopenia. This study aimed to compare the performance of these serum indices in screening for low skeletal muscle index (SMI) among patients with MASLD.

This prospective observational study enrolled 146 participants with MASLD. Anthropometric and laboratory data were collected. The CBMM, SI, and eGFR ratios were calculated using serum Cr and CysC levels. Low SMI was determined using Bioelectrical Impedance Analysis (BIA) according to the Asian Working Group for Sarcopenia (AWGS) 2019 criteria. Linear regression analysis was used to assess the correlations between serum indices and SMI. Receiver Operating Characteristic (ROC) curve analysis was used to evaluate the discriminative ability of these serum indices for detecting low SMI. Furthermore, machine learning models (Linear Regression, Random Forest, and XGBoost), coupled with SHapley Additive exPlanations (SHAP) analysis, were employed to evaluate the importance of these indices in predicting low SMI.

Patients with higher fibrosis-4 (FIB-4) scores (≥2.67) had a significantly higher prevalence of low SMI. CBMM demonstrated the strongest correlation with SMI (R² = 0.4306, p < 0.0001) and the best discriminative ability for low SMI (Area under ROC: 0.9149 for males and 0.9444 for females) compared with SI and eGFR ratio. Machine learning models consistently identified CBMM as the most important feature for predicting SMI based on the SHAP analysis.

These findings suggest that CBMM, derived from readily available serum markers, could serve as a valuable initial screening tool for identifying MASLD patients at risk of sarcopenia who may benefit from further assessment and early interventions.