Introduction: Previous work has shown that uncontrolled diabetes mellitus is associated with adverse surgical outcomes. The purpose of the present study was to establish if a high peri-operative random blood sugar (RBS) concentration among patients with diabetes with non-traumatic lower-extremity amputation (LEA) is a decisive factor behind post-operative outcomes (re-amputation/mortality) within three months after the first amputation.
Methods: In this retrospective cohort study, the independent sample t-test, Pearson's chi-squared test and a Cox proportional hazards model were used.
Results: A total of 270 patients underwent non-traumatic LEA of whom 105 had diabetes, whereas 81 patients were included for this study. The mean age was 71 years (standard deviation: ± 11.8). Mortality was 27% and 16% were re-amputated within three months after their first amputation.The median pre-operative RBS level was 8.6 mmol/l (range: 4.6-18.7 mmol/l) with tertile ranges as follows: Q1 4.0-7.0 mmol/l; Q2 7.1-11.0 mmol/l; Q3 > 11.0 mmol/l. For the Q3 tertile, the age-adjusted hazard ratio for re-amputation was 0.77 (95% confidence interval (CI): 0.16-3.62) and for mortality it was 1.90 (95% CI: 0.50-7.22), with the Q1 tertile as the reference group.
Conclusions: This study does not confirm that a high peri-operative RBS level can predict increased mortality or re-amputation among patients with diabetes who undergo non-traumatic LEA. Furthermore, based on our results, we cannot inform clinical decision-making about whether to delay or to avoid elective surgery in patients with a high RBS preoperatively. Further investigation is warranted.
Funding: none.
Trial registration: This trial was registered with the Danish Data Protection Agency (record no. 01975 HVH-2012-053).