Relationship Between Hemoglobin A1c Value and Bone Healing in Diabetic Ankle Fractures Treated Operatively versus Non-operatively

Abstract

Category: Ankle Introduction/Purpose: Diabetic ankle fractures create unique challenges to fracture management, such as impaired wound / bone healing and increased infection rate. There is inconclusive evidence regarding a specific hemoglobin A1c (A1c) value or glucose level at which the risk of adverse outcomes increases. The aim of this study was to evaluate the A1c level at which the rate of nonunion and other major complications significantly increased in diabetic ankle fractures treated operatively (Op) and non-operatively (Nonop). Methods: This is a retrospective cohort study of diabetic ankle fractures treated Op or NonOp with at least 3 months follow up and a hemoglobin A1c value within 90 days of injury. A query of electronic medical records was performed to identify all patients 18 years of age or older with a diagnosis of diabetes and who underwent Op or NonOp management as the initial treatment of an ankle fracture between January 1, 2004 and December 31, 2014 within a single health system. Union was defined as callus formation at 3 of 4 cortices on radiographs, while nonunion was defined as lack of fracture healing at 6 months based on radiographs of the injured ankle. Multivariable logistic regression analysis was performed to explore demographic, injury-related, and management-related risk factors that influence nonunion and major complications. Results: A total of 243 ankle fractures (130 Op, 113 NonOp) were identified. Nine patients in the NonOp group failed nonoperative management and required surgical fixation. There were 51 patients that developed nonunion/malunions (19 Op, 32 NonOp). Patients with diabetic neuropathy or nephropathy had significantly higher risk of nonunion. In both groups, length of immobilization greater than 12 weeks was a significant predictor of nonunion. There was no A1c level nor preinjury glucose level at which there was significantly increased rate of nonunion or deep infections in either treatment group. However, A1c was a predictor of wound complications at 3 months postoperative in the Op group with an odds ratio of 1.26. There was no significant difference in wound complications at 3 months between treatment groups. Conclusion: While no specific A1c value or glucose level demonstrated a significant rise in fracture nonunion, this study found that diabetic neuropathy and diabetic nephropathy play a significant role in fracture healing. However, hemoglobin A1c was found to be a predictor of wound complications at 3 months postoperative. Additionally, the data suggested a longer duration of immobilization may negatively impact fracture union. These results highlight the complexity of treating diabetic ankle fractures and suggest prolonged immobilization may not be beneficial for this subpopulation.