AIMS/HYPOTHESIS: Diabetes mellitus is associated with several changes in coagulation and fibrinolysis that may lead to a thrombogenic propensity. However, it is not known whether these perturbations actually cause increased risk of venous thromboembolism.
METHODS: In a retrospective population-based study we evaluated the medical records of all 302 adult patients who were admitted to the Umea University Hospital with verified deep vein thrombosis or pulmonary embolism during the years 1997 to 1999. The patients were classified as diabetic (n=56) and non-diabetic (n=246) according to clinical information. The total number of diagnosed diabetic patients in different age groups in the catchment area was obtained from computerised registries in the primary health care centres and the Umea University Hospital, and data on the background population were collected from the Swedish population registry.
RESULTS: The annual incidence rate of venous thromboembolism among diabetic patients in the population was 432 per 100,000 individuals (95% CI 375-496). In non-diabetic individuals it was 78 (95% CI 68-88). The age-adjusted incidence rate among the diabetic population was 274 (95% CI 262-286). The annual incidence rate of venous thromboembolism was elevated in type 1 and type 2 diabetic patients and the incidence rates were 704 (95% CI 314-1,566) and 412 (95% CI 312-544) respectively. The overall standardised morbidity ratio was 2.27 (95% CI 1.75-2.95), i.e. diabetic patients were more prone to venous thromboembolism after adjustment for age differences.
CONCLUSIONS/INTERPRETATION: These results suggest that the age-adjusted risk for venous thromboembolism is more than two-fold higher among diabetic patients than in the non-diabetic background population.
2005. Vol. 48, no 5, 1017-21 p.