Abstract
BACKGROUND: Understanding the relationship between relative skeletal muscle mass and newly proposed steatotic liver disease (SLD) is crucial, but research gaps still exist. Based on a cohort study, we investigated the impact of relative skeletal muscle mass on incident SLD and its subtypes and explored potential mediators involved in these relationships. METHODS: We followed 1964 subjects aged 55-70 years (median age: 61.4 [58.4-65.1] years; 45.5% male participants). Appendicular skeletal muscle mass (ASM) was measured using bioelectrical impedance analysis and adjusted for height squared (ASM/height(2)), weight (ASM/weight) and body mass index (ASM/BMI) to quantify relative skeletal muscle mass. SLD was diagnosed using ultrasonography and classified into metabolic dysfunction-associated steatotic liver disease (MASLD), metabolic dysfunction and alcohol-associated liver disease (MetALD) and alcohol-associated liver disease (ALD). RESULTS: Over a mean 4.3-year follow-up period, 598 participants (30.4%) developed SLD: 539 MASLD (27.4%), 38 MetALD (1.9%) and 21 ALD (1.1%). Higher ASM/weight and ASM/BMI were associated with lower risks of SLD and MASLD (RR per SD [95% CI]: 0.71 [0.61-0.82], 0.59 [0.46-0.76]; 0.66 [0.58-0.76], 0.51 [0.40-0.64]; all p < 0.001). Associations of ASM/height(2) with SLD and MASLD shifted from positive to negative after adjustment for BMI (from 1.67 [1.51-1.84] to 0.77 [0.67-0.89]; from 2.30 [1.92-2.76] to 0.61 [0.47-0.79], all p < 0.001). ASM/height(2) and ASM/weight were negatively associated with MetALD (0.49 [0.26-0.94], p = 0.031; 0.50 [0.27-0.93], p = 0.029), whereas ASM/BMI was inversely associated with ALD (0.40 [0.18-0.88], p = 0.023). The effects of ASM/height(2), ASM/weight and ASM/BMI on incident MASLD were partially mediated by adiponectin (percentage mediated [95% CI]: 6.3% [2.5%-11.1%]; 9.4% [5.0%-14.6%]; 9.5% [5.1%-15.5%]), uric acid (4.7% [1.6%-8.9%]; 5.3% [2.6%-8.5%]; 5.3% [2.4%-8.8%]), triglyceride (7.1% [3.9%-11.1%]; 7.5% [4.4%-10.9%]; 8.7% [5.3%-13.4%]) and homeostasis model assessment of insulin resistance (13.9% [9.5%-20.4%]; 15.0% [10.0%-20.2%]; 14.5% [9.9%-20.7%]). The effects of ASM/weight and ASM/BMI on incident MASLD were mediated by cholinesterase (8.2% [3.6%-13.1%]; 10.5% [6.1%-16.3%]), prealbumin (6.2% [2.9%-9.8%]; 6.0% [3.0%-10.1%]), retinol-binding protein-4 (5.4% [3.0%-8.5%]; 4.6% [1.9%-8.5%]) and osteocalcin (2.1% [0.1%-4.5%]; 2.9% [0.6%-5.7%]). CONCLUSIONS: Relative skeletal muscle mass adjusted for weight or BMI, rather than height alone, better reflects protective effects against SLD. Mediation analysis reveals key metabolic factors linking muscle mass and liver health, offering insights into the pathogenic pathways involved in muscle-liver crosstalk.