Age-specific and sex-specific associations of visceral adipose tissue mass and fat-to-muscle mass ratio with risk of mortality

BackgroundLimited studies have explored the association between visceral adipose tissue (VAT) mass and fat-to-muscle mass ratio (FMR) and mortality. We aimed to evaluate the sex-specific association of VAT and FMR with all-cause and cause-specific mortality by age. MethodsA total of 438 896 participants (49.8% men, mean age +/- standard deviation: 57 +/- 8 years for men; 56 +/- 8 years for women) were included from the UK Biobank cohort. The nature of VAT was predictive, as obtained by sex-stratified, non-linear prediction models. Fat and muscle mass were estimated using a bioelectrical impedance assessment device. FMR was calculated as the fat mass divided by the muscle mass in the whole body. VAT and FMRs were divided into quintiles in ascending order, and the 3rd quintile was used as the reference. Cox regression analyses were used to estimate the associations between VAT, FMR and mortality. ResultsDuring a median of 12.4 years of follow-up, we documented 29 903 deaths. After adjusting for various covariates, the individuals in the highest quintiles of VAT and FMR had the highest hazard ratios (HRs) of all-cause mortality [1.24 (95% confidence interval: 1.17-1.33) for VAT and 1.24 (1.17-1.31) for FMR in men; and 1.11 (1.03-1.21) for VAT in women], except that the 1st quintile of FMR in women had the greatest HR [1.18 (1.09-1.27)]. Significant interactions were observed in both sexes according to age category (P for interaction < 0.05). Among men <50 years, participants in the 1st and 5th quintiles of VAT and FMR had significantly higher risks of mortality [1.30 (1.02-1.66) and 1.67 (1.27-2.19) in VAT; 1.25 (0.99-1.56) and 1.41 (1.11-1.79) in FMR, respectively]; in women, this phenomenon was observed in the >= 60 age group [1.16 (1.06-1.27) and 1.19 (1.08-1.31) in VAT; 1.18 (1.08-1.29) and 1.11 (1.01-1.22) in FMR, respectively]. VAT showed a linear positive association with mortality in women <60 years and a J-shaped association from respiratory disease in both sexes >= 60 years. FMR showed a linear positive association with mortality from cancer in men <60 years and a J-shaped association with mortality from cause-specific mortality in both sexes >= 60 years, except for mortality from cardiovascular disease in men. ConclusionsMost associations of VAT and FMR with all-cause mortality were J-shaped and were significantly modified by age status (<50, 50-59 and >= 60 years). The clinical implication is that regarding body composition and VAT mass, different health strategies may be adopted for people of different sexes and ages.