Abstract
Aims/hypothesis: Lower limb amputation is a serious complication of diabetes mellitus. Understanding how amputation risk differs by population subgroups is crucial in terms of directing preventive strategies. In this study, we describe those factors that impact amputation risk in the entire prevalent diabetic population of New Zealand.
Methods: A national prevalent cohort of 217,207 individuals with diabetes in 2010 were followed up until the end of 2013 for lower limb amputations, and 2014 for mortality. Inpatient hospitalisation data were used to define lower limb amputation using ICD-10 codes. Cox proportional hazards models were used to describe relative hazard of amputation over the follow-up period.
Results: A total of 784 individuals (3.6 cases/1000 individuals) underwent a major (above-ankle) lower limb amputation during follow-up, while 1217 (5.6/1000) underwent a minor (below ankle) amputation. The risk of major and minor amputation was 39% and 77% greater for men than women, respectively (adjusted HR: major amputation 1.39, 95% CI 1.20, 1.61; minor amputation 1.77, 95% CI 1.56, 2.00). Indigenous Māori were at 65% greater risk of above-knee amputation compared with the European/Other diabetic population (HR 1.65, 95% CI 1.37, 1.97). Amputation risk increased with increasing comorbidity burden, and peripheral vascular disease conferred the greatest independent risk of all comorbid conditions. Prior minor amputation increased the risk of subsequent major amputation by tenfold (HR 10.04, 95% CI 7.83, 12.87), and increased the risk of another minor amputation by 20-fold (HR 21.39, 95% CI 17.89, 25.57). Death was common among the total cohort, but particularly among those who underwent amputation, with more than half of those who underwent a major amputation dying within 3 years of their procedure (57%).
Conclusions/interpretation: Using a large, well-defined, national prevalent cohort of people with diabetes, we found that being male, indigenous Māori, living in deprivation, having a high comorbidity burden and/or having a previous amputation were strongly associated with subsequent risk of lower limb amputation. The use of this prevalent cohort strengthens the value of our estimates in terms of applicability to the general population, and highlights the subgroups at greatest risk of lower limb amputation.