The Contemporary Natural History of Minor Amputation Among Diabetic Patients with Peripheral Arterial Disease.

Abstract

Objectives: The growing prevalence of diabetes and concomitant PAD(DM/PAD) has led to an increase in patients at-risk for adverse limb events in current practice. Despite a widespread perception that minor amputation may result in both limb salvage and preserved functionality, the natural history of minor amputations remains unknown. Thus, we sought to quantify the rates of subsequent major amputation and survival among DM/PAD patients with any prior minor amputation.

Methods: We performed a retrospective cohort study using US Medicare claims from 2007-2019. We included patients with DM/PAD based on ICD-9 and ICD-10 diagnosis codes. We excluded patients lacking continuous fee-for-service coverage or with incomplete demographic data. The primary exposure was prior minor(below-ankle) amputation. The primary outcome was major(above-ankle) amputation. Statistical analyses were performed using the Kaplan-Meier method and Cox proportional hazards modeling.

Results: We identified 12,257,174 patients(age 73±11 years,48% male,76% White) with DM/PAD. Of these patients, 2.2%(N=274,225) underwent prior minor amputation. Patients with prior minor amputation were more likely to be male(63% versus 47%,p<0.0001), Black non-Hispanic(17% versus 13%,p<0.0001), and rural(25% versus 21%,p<0.0001) than those without prior minor amputation. The five-year Kaplan-Meier cumulative incidence of major amputation was 27%(N=58,613) of patients with prior minor amputation, compared to 1.4%(N=129,872) of patients without prior minor amputation. After risk-adjustment, patients with prior minor amputations were 6.1-fold more likely to require a subsequent major amputation(HR 6.11[6.04-6.18]) compared to those without prior minor amputations.

Conclusions: This contemporary claims-based analysis demonstrates that approximately 25% of Medicare beneficiaries with DM/PAD and prior minor amputation will necessitate a major amputation within five years. Prior minor amputation carries a risk of major amputation comparable to de novo tissue loss, and is a stronger predictor than any demographic or socioeconomic exposure. These results help to inform both clinical decision-making and anticipated real-world outcomes among those at greatest risk for limb loss.