Risks and Risk Factors for Contralateral Amputation in Patients who have Undergone Amputation for Chronic Limb Threatening Ischaemia

Objective: The aim of this study was to determine the cumulative incidence of, and the risk factors associated with, contralateral amputation in patients with chronic limb threatening ischaemia (CLTI). Methods: This was a retrospective cohort study of patients with incident unilateral transmetatarsal (TM), transtibial (TT), or transfemoral (TF) amputation secondary to CLTI, identified from the National Veterans Affairs Surgical Quality Improvement Program database (2004 - 2014). Thirteen potential pre-operative risk factors for contralateral amputation were considered. A competing risk analysis to estimate the cumulative incidence of contralateral amputation was performed using a Fine-Gray subdistribution hazard model. The effect of risk factors on contralateral amputation was estimated by computing subdistribution hazard ratios (sub-HR) with 95% confidence intervals (CI). Results: From the database, 7 360 patients met the inclusion criteria. The contralateral amputation risk was 7.7% and was greatest in those who underwent a TF amputation (9.7%), followed by TT (7.4%) and TM amputation (6.6%) (p < .001). Among the 588 contralateral amputations, 50% were at the TF level, 34% at the TT level, and 16% at the TM level. The adjusted risk of contralateral amputation was greater in those who underwent an incident TF amputation or were Black or Hispanic. The factor that contributed to risk of contralateral amputation to the greatest extent was dialysis (sub-HR, 2.3; 95% CI 1.7 - 3.0; p < .001) while those who were obese (compared with underweight) were at lowest risk (0.67; 95% CI 0.46 - 0.97; p = .030). Conclusion: The one year risk of contralateral amputation in patients with CLTI is related to incident amputation level, medical comorbidities, correlates with race/ethnicity, and body mass index at the time of the incident amputation. The identified risk factors are largely not modifiable; however, they can be used to help identify populations at elevated risk.