Cardiorespiratory fitness, fatness and incident diabetes.

Aims: Increases in prevalence have led to a diabetes pandemic. Obesity and low cardiorespiratory fitness (CRF) are considered to be central mechanisms. We investigated if the effect of CRF on diabetes risk was equivalent across levels of fatness among healthy men.

Methods: In total 4988 middle-aged Caucasian employed men free of cardiovascular disease, diabetes and cancer were included from the Copenhagen Male Study starting in 1970-71. CRF was assessed using a sub-maximal bicycle ergometer test and body mass index (BMI) was measured by height and weight. Their interaction and stratified associations with diabetes incidence were estimated in multivariable Cox-models including conventional risk factors and social class. Diabetes incidence was assessed through a national register.

Results: During 44 years of follow-up, 518 (10.4%) incident cases of diabetes occurred. In the multi-adjusted model, the obese had a significantly higher risk of diabetes compared to normal weight men (Hazard Ratio (HR):4.89; 95% CI: 3.62-6.61) and CRF was significantly inversely associated with diabetes (HR:0.86; 95% CI: 0.75-0.98 per 10-unit increase in ml/kg/min CRF). A significant multi-adjusted interaction between CRF, BMI and diabetes was found (p=0.009). The stratified multi-adjusted analyses on BMI showed a significantly stronger reduced risk of diabetes per 10-unit increase in ml/kg/min of CRF among the obese (HR:0.58; CI: 0.38-0.89), but a weaker association among overweight (HR:0.86; CI: 0.71-1.03) and normal weight (HR:0.97; CI: 0.76-1.23).

Conclusion: High CRF has a stronger protective effect on diabetes among obese than among normal weight men, supporting the recommendation of fitness-enhancing physical activity for preventing diabetes among the obese.