Associations of relative fat mass and BMI with all-cause mortality: Confounding effect of muscle mass.

Abstract

The study objective was to examine associations of relative fat mass (RFM) and BMI with all-cause mortality in the Dutch general population and to investigate whether additional adjustment for muscle mass strengthened these associations. A total of 8433 community-dwelling adults from the PREVEND general population cohort (1997-1998) were included. Linear regression models were used to examine associations of RFM and BMI with 24-h urinary creatinine excretion, a marker of total muscle mass. Cox regression models were used to examine associations of RFM and BMI with all-cause mortality. The mean age of the cohort was 49.8 years (range: 28.8-75.7 years), and 49.9% (n = 4209) were women. In age- and sex-adjusted models, both RFM and BMI were associated with total muscle mass (24-h urinary creatinine excretion), and these associations were stronger with BMI (standardized beta [Sβ]_RFM : 0.29; 95% CI: 0.27-0.31 vs. Sβ_BMI : 0.38; 95% CI: 0.36-0.40; p_difference  < 0.001). During a median follow-up period of 18.4 years, 1640 deaths (19.4%) occurred. In age- and sex-adjusted models, RFM was significantly associated with all-cause mortality (hazard ratio per 1-SD [HR_RFM ]: 1.16; 95% CI: 1.09-1.24), whereas BMI was not (HR_BMI : 1.04; 95% CI: 0.99-1.10). After additional adjustment for muscle mass, associations of both RFM and BMI with all-cause mortality increased in magnitude (HR_RFM : 1.24; 95% CI: 1.16-1.32 and HR_BMI : 1.12; 95% CI: 1.06-1.19). Results were broadly similar in multivariable adjusted models. In the general population, a higher RFM was significantly associated with mortality risk, whereas a higher BMI was not. Adjusting for total muscle mass increased the strength of associations of both RFM and BMI with all-cause mortality.