Supplementary Material for: Central lean mass distribution and the risks of all-cause and cause-specific mortality in 40283 UK Biobank participants

Introduction
The purpose of this study is to investigate the association of central lean mass distribution with the risk of mortality.
Methods
This cohort study included 40283 UK Biobank participants. Cox proportional hazards regression models were used to estimate the association of central lean mass distribution, i.e., trunk-to-leg lean mass ratio, assessed by dual energy X-ray absorptiometry, with the risk of mortality.
Results
The median age of the participants was 65 years and 52% were women. During a median follow-up of 4.18 years, 674 participants died, of whom 366 were due to cancer, and 126 were due to cardiovascular causes. Compared with the lowest tertile of a trunk-to-leg lean mass ratio, the multivariable-adjusted (age, sex, ethnicity, lifestyle, comorbidities, body mass index, and appendicular muscle mass index) hazards ratios of the highest tertile of trunk-to-leg lean mass ratio were 1.55 (95%CI, 1.23 - 1.94), 1.69 (95%CI, 1.26 - 2.26), and 1.14 (95%CI, 0.72 - 1.80) for all-cause, cancer and cardiovascular mortality, respectively. Neutrophil-to-lymphocyte ratio mediated 9.3% (95% CI, 3.3%-40.4%) of the association of trunk-to-leg lean mass ratio with all-cause mortality. There was evidence for additive interactions of trunk-to-leg lean mass ratio with older age and poor diet quality for all-cause mortality.
Conclusion
Trunk-to-leg lean mass ratio, assessed by dual energy X-ray absorptiometry, was positively associated with the risks of all-cause and cancer mortality, independent of general obesity and central obesity, in UK middle-aged and older adults. Central lean mass distribution may interact synergistically with aging, and poor diet quality to further increase the risk of death.