Background: Despite the significant public health burden of lower- extremity amputations in diabetes mellitus, few data are available on the epidemiology of lower-extremity amputations in diabetes mellitus in the community setting. Methods: A retrospective incidence cohort study based in Rochester, Minn, was conducted. Results: Among the 2015 diabetic individuals free of lower-extremity amputation at the diagnosis of diabetes mellitus, 57 individuals underwent 79 lower-extremity amputations (incidence, 375 per 100 000 person-years; 95% confidence interval, 297 to 467). Among the 1826 patients with non-insulin-dependent diabetes mellitus, 52 underwent 73 lower- extremity amputations, and the subsequent incidence of lower-extremity amputation among these residents was 388 per 100 000 person-years (95% confidence interval, 304 to 487). Of the 137 insulin-dependent diabetic patients, four subsequently underwent five lower-extremity amputations (incidence, 283 per 100 000 person-years; 95% confidence interval, 92 to 659). Twenty-five years after the diagnosis of diabetes mellitus, the cumulative risk of one lower-extremity amputation was 11.2% in insulin- dependent diabetes mellitus and 11.0% in non-insulin-dependent diabetes mellitus. When compared with lower-extremity amputation rates for Rochester residents without diabetes, patients with non-insulin-dependent diabetes mellitus were nearly 400 times more likely to undergo an initial transphalangeal amputation (rate ratio, 378.8) and had almost a 12-fold increased risk of a below-knee amputation (rate ratio, 11.8). In this community, more than 60% of lower-extremity amputations were attributable to non-insulin-dependent diabetes mellitus. Conclusions: These population-based data document the magnitude of the elevated risk of lower-extremity amputation among diabetic individuals. Efforts should be made to identify more precisely risk factors for amputation in diabetes and to intervene in the processes leading to amputation.
ASJC Scopus subject areas
- Internal Medicine