Relation of Exercise Capacity to Risk of Development of Diabetes in Patients on Statin Therapy (the Henry Ford Exercise Testing Project).

High exercise capacity (EC) has been associated with a lower risk of incident diabetes, whereas statin therapy has been associated with a higher risk. We sought to investigate whether the association between EC and diabetes risk is modified by statin therapy. This retrospective cohort study included 47,337 patients without diabetes or coronary artery disease at baseline (age 53 ± 13 years, 48% women, 66% white) who underwent clinical treadmill stress testing within the Henry Ford Health System from January 1, 1991, to May 31, 2009. The patients were stratified by baseline statin use and estimated peak METs achieved during exercise testing. Hazard ratios for incident diabetes were calculated using Cox proportional hazards models adjusted for demographic characteristics, co-morbidities, pertinent medications, and stress test indication. We observed 6,921 new diabetes cases (14.6%) over a median follow-up period of 5.1 years (interquartile interval of 2.6 to 8.2 years). Compared with the statin group, the no-statin group achieved higher mean METs (8.9 ± 2.7 vs 9.6 ± 3.0, respectively; p <0.001). After adjustment for covariates, a higher EC was associated with a lower risk of incident diabetes, irrespective of statin use (p-interaction = 0.15). Each 1-MET increment was associated with an 8%, 8%, and 6% relative risk reduction in the total cohort, the no-statin, and the statin groups, respectively (95% confidence interval, 0.91 to 0.93, 0.91 to 0.93, and 0.91 to 0.96, respectively; p <0.001 for all). We conclude that a higher EC is associated with a lower risk of incident diabetes regardless of statin use.