Predictors of early mortality and readmissions among dialysis patients undergoing lower extremity amputation

Objective: Patients receiving dialysis are at increased risk for lower extremity amputations (LEAs) and postoperative morbidity. Limited studies have examined differences in 30-day outcomes of mortality and health care use after amputation or the preoperative factors that relate to worsened outcomes in dialysis patients. Our objective was to examine dialysis dependency and other preoperative factors associated with readmission or death after LEA. Methods: A retrospective cohort study was conducted of dialysis-dependent and nondialysis patients undergoing major LEA in the 2012 to 2013 American College of Surgeons National Surgical Quality Improvement Program. Primary outcomes included death and hospital readmission within 30 days of amputation. Results: Of 6468 patients, 1166 (18%) were dialysis dependent. The dialysis cohort had more blacks (39% vs 23%), diabetes (76% vs 58%), below-knee amputations (62% vs 55%), and in-hospital deaths (8% vs 3%; all P <.001). The 30-day postoperative death rates (15% vs 7%) and readmission rates (35% vs 20% per 30 person-days; both P <.001) were higher in dialysis patients. Among the live discharges, the rate of any readmission or death within 30 days from amputation was highest in those aged ≥50 years (40% per 30 person-days). Multivariable analyses in the dialysis cohort revealed increased age, above-knee amputation, decreased physical status, heart failure, high preoperative white blood cell count, and low platelet count to be associated with death (P <.05; C statistic, 0.75). The only preoperative factor associated with readmission in dialysis patients was race (P =.04; C statistic, 0.58). Conclusions: Readmission or death after amputation is increased among dialysis patients. Predicting which dialysis patients are at highest risk for death is feasible, whereas predicting which will require readmission is less so. Risk factor identification may improve risk stratification, inform reimbursement policies, and allow targeted interventions to improve outcomes.