Nonlinear Associations of Body Muscle‐to‐Fat Ratio With Long‐Term Prognosis in
MASLD
Zhe‐Kun Xiong, Ru‐Tao Lin, Yan‐Hua Lin, Xin Xin, Jian‐Dong Cao, Xiao‐Shan He, Yi‐Yang Hu, Li‐Ming Gan, Qin Feng ABSTRACT
Background
As weight‐loss therapies are increasingly incorporated into clinical practice for MASLD, assessing body composition has become increasingly important. Body muscle‐to‐fat ratio (BMFR) may provide an intuitive and integrated measure of body composition balance, but its association with long‐term prognosis in MASLD remains unclear.
Methods
We studied 83 006 UK Biobank participants with MASLD. BMFR was calculated as whole‐body muscle mass divided by whole‐body fat mass using bioelectrical impedance analysis. Restricted cubic spline, threshold‐effect and two‐piecewise Cox models assessed associations with liver‐related events (LREs), cardiovascular events (CVEs), cardiovascular mortality and all‐cause mortality, adjusting for BMI and other confounders.
Results
BMFR showed significant non‐linear associations with all outcomes. Sex‐specific inflection points were identified for each outcome. For LREs, the thresholds were 2.82 in men and 1.45 in women; below these thresholds, each 1‐SD increase in BMFR was associated with a lower risk of LREs (men: HR = 0.86, 95% CI: 0.75–0.98; women: HR = 0.72, 95% CI: 0.63–0.83), whereas above them it was associated with a higher risk (men: HR = 1.21, 95% CI: 1.07–1.36; women: HR = 1.33, 95% CI: 1.11–1.59). For CVEs, the thresholds were 2.03 in men and 1.48 in women; below these thresholds, each 1‐SD increase in BMFR was associated with a lower risk in both sexes (men: HR = 0.77, 95% CI: 0.70–0.85; women: HR = 0.85, 95% CI: 0.81–0.89), whereas above them the association was not significant in men but was associated with a higher risk in women (HR = 1.19, 95% CI: 1.09–1.30). Cardiovascular mortality and all‐cause mortality showed sex‐specific non‐linear patterns similar to those observed for CVEs.
Conclusions
In MASLD, BMFR was associated with liver‐related and cardiovascular outcomes. Associations were generally U‐shaped in women; in men, they were U‐shaped for LREs and closer to L‐shaped for cardiovascular outcomes and mortality. BMFR may provide complementary information on body composition beyond BMI. Further studies are needed to validate these exploratory findings.