Context: The impact of older age and duration of diabetes mellitus on macrovascular complications is unclear.
Objective: We tested the hypothesis that in older men, diabetes duration predicts incident cardiovascular events and death, differently from prior myocardial infarction (MI) or stroke.
Design, setting, and participants: This was a longitudinal cohort study of 11 728 community-dwelling men aged ≥ 65 years in Perth, Western Australia, recruited in 1996-1999.
Main outcome measures: We assessed all-cause mortality, and deaths or hospital admissions with MI or stroke between recruitment and December 2010, analyzing age-specific hazard and adjusting for smoking, education, alcohol, exercise, BMI, hypertension, and hypercholesterolemia.
Results: Among 1433 (12.2%) men with diabetes, 208 (14.5%) reported age of onset of diabetes < 55 years, 451 (31.5%) 55-64 years, 679 (47.4%) 65-74 years with 95 (6.6%) > 74 years. Diabetes independently predicted increased all-cause mortality with hazard ratio (HR) of 1.37 (95% confidence interval [CI] = 1.15-1.62) for a duration of 5-9 years, 1.35 (1.18-1.55) for 10-14 years, 1.42 (1.22-1.66) for 15-19 years, and 1.75 (1.45-2.11) for 20-24 years. Mortality from MI was increased for diabetes duration up to 25 years, while stroke-specific mortality increased progressively with diabetes duration. Prior MI or stroke predicted increased risk of subsequent events peaking after 10-20 years.
Conclusions: In older men, increasing duration of diabetes predicts stable increases in all-cause and MI-related mortality and a progressively higher risk of stroke deaths. Prior MI was associated with increased risk of subsequent MI, and prior stroke with subsequent stroke, particularly in the 10-20 years following the first event. Diabetes is a duration-dependent risk factor for cardiovascular events which influences outcomes differently from prior vascular disease.