Does Perioperative Hemoglobin A1c Level Affect the Incidence, Pattern and Mortality of Lower Extremity Amputation?

Background: We hypothesized that perioperative HbA(1c) influenced the pattern and outcomes of Lower Extremity Amputation (LEA). Methods: A retrospective analysis was conducted for all patients who underwent LEA between 2000 and 2013. Patients were categorized into 5 groups according to their perioperative HbA(1c) values [Group 1 (<6.5%), Group 2 (6.5-7.4%), Group 3 (7.5-8.4%), Group 4 (8.5-9.4%) and Group 5 (>= 9.5%)]. We identified 848 patients with LEA; perioperative HbA(1c) levels were available in 547 cases (Group 1:18.8%, Group 2:17.7%, Group 3:15.0%, Group 4:13.5% and Group 5:34.9%). Major amputation was performed in 35%, 32%, 22%, 10.8% and 13.6%, respectively. Results: The overall mortality was 36.5%; of that one quarter occurred during the index hospitalization. Mortality was higher in Group 1 (57.4%) compared with Groups 2-5 (46.9%, 38.3%, 36.1% and 31.2%, respectively, p=0.001). Cox regression analysis showed that poor glycemic control (Group 4 and 5) had lower risk of mortality post-LEA [hazard ratio 0.57 (95% CI 0.35-0.93) and hazard ratio 0.46 (95% CI 0.31-0.69)]; this mortality risk persisted even after adjustment for age and sex but was statistically insignificant. The rate of LEA was greater among poor glycemic control patients; however, the mortality was higher among patients with tight control. Conclusion: The effects of HbA(1c) on the immediate and long-term LEA outcomes and its therapeutic implications need further investigation.