Abstract

Objective: To estimate the risks of non-traumatic lower extremity amputations (LEAs) in patients with type 1 diabetes mellitus (T1DM).

Research Design and Method: We identified 31,354 patients with T1DM (15,001 women and 16,353 men) in the Swedish Inpatient Register between 1975 and 2004. The incidence of non-traumatic LEAs was followed up until December 31, 2004, by cross-linkage in the Inpatient Register and linkage to the Death and Migration registers. Poisson regression modeling was used to compare the risks of non-traumatic LEAs during different calendar periods of follow-up, with adjustment both for gender and attained age at follow-up. Standardized incidence ratios (SIRs) was used to estimate the relative risks with the age-, sex-, and calendar-period-matched general Swedish population as reference. The cumulative probability of non-traumatic LEAs was calculated by the Kaplan-Meier method.

Results: In total, 465 patients with T1DM underwent non-traumatic LEAs. The risk was lower during the most recent calendar period (2000-2004) than the period prior to 2000 (relative risk = 0.6, 95% confidence interval [CI] 0.5-0.8). However, even in this most recent period, the risk for non-traumatic LEAs among these relatively young patients was 86-fold higher than that in the matched general population (SIR =85.8, 95% CI 72.9-100.3). By the age of 65, the cumulative probability of having a non-traumatic LEA was 11.0% for women with T1DM, and 20.7% for men with T1DM.

Conclusion: Although the risks appeared to have declined in recent years, patients with type 1 diabetes still run a very high risk for non-traumatic LEAs.