Cardiorespiratory Fitness and Cardiovascular Disease Mortality in Men Within Clinically Established Obesity Categories

Abstract

Earlier we reported on the association between cardiorespiratory fitness (CRF) or adiposity with cardiovascular disease (CVD) mortality. We now have 3 times the number of CVD deaths and are now able to evaluate the joint associations of these exposures within clinically established obesity categories. PURPOSE: To determine the prospective relation between CRF, adiposity, and CVD mortality in men. METHODS: Participants were 29,539 men (mean age 43.7 ± 9.4 years) who had no history of physician diagnosed CVD or cancer, no abnormal electrocardiogram, examined at the Cooper Clinic between 1974-2003. Low CRF was defined as the lowest fifth of the age-specific distribution of maximal treadmill exercise test duration. Body mass index (BMI), waist circumference (WC), and percent body fat (%fat) were grouped according to clinical guidelines. Mortality surveillance was completed through December 31, 2003. Hazard ratios (HR) were computed using Cox regression analyses. RESULTS: During a mean (SD) follow-up period of 15.0 (6.8) years, 1,459 deaths (436 CVD) occurred. Multivariate-adjusted HR (95% confidence intervals) of CVD mortality were: 1.00, 1.41 (1.13-1.75) and 2.42 (1.82-3.21) across BMI groups of 18.5-24.9, 25.0-29.9, and ≥30 kg/m2, respectively (linear trend P<0.001); 1.00 and 1.81 (1.45-2.25) for normal and high WC (≥102 cm), respectively (P<0.001); 1.00 and 1.43 (1.17-1.75) for normal and high %fat (≥25%), respectively (P<0.001); and 1.00, 0.71 (0.54-0.95), 0.53 (0.39-0.71), 0.48 (0.35-0.65) and 0.37 (0.27-0.52) across incremental CRFquintiles, respectively (linear trend, P<0.001). Further adjustment for CRF eliminated the significant association in CVD mortality risk across %fat categories and attenuated the trend in risk across BMI and WC categories. Adjustment of CRF for adiposity measures had little effect on CVD mortality risk. When grouped into categories of fit and unfit (upper 80% and lower 20% of CRF distribution, respectively) CVD mortality rates were significantly lower in fit compared with unfit men within each stratum of BMI, WC and %fat. CONCLUSIONS: Even moderate levels of CRF are associated with lower CVD mortality risk in men, independent of their clinical adiposity status. These findings underscore the importance of CRF in addition to adiposity, as a determinant of CVD mortality in men.