Abstract
Category:
Ankle; Diabetes
Introduction/Purpose:
Diabetes poses a significant challenge as a prognostic factor following acute ankle fractures. Surgeons often hesitate to proceed with necessary surgery due to anticipated complications. Our analysis compared complication rates among diabetic and non-diabetic patients treated by all surgeons within our unit. Additionally, we evaluated the efficacy of long-segment fixation methodologies as potential factors for successful outcomes.
Methods:
We conducted a retrospective review of ankle fractures managed in our department from 2014 to 2019, cross-referencing patient records with departmental databases. Patients were then observed prospectively for a minimum 5-year follow-up period. Radiographs taken before and after treatment were independently reviewed.
We identified patients with diabetic ankle fractures and recorded their HbA1c levels, utilizing the Charlson Comorbidity Index (CCI) to profile patient comorbidity. Our study included age, gender, and follow-up matched non-diabetic controls, as well as conservatively managed diabetic and surgically treated groups, focusing on rigid fixation. Fractures were categorized, and surgeries were classified accordingly.
Univariate and multivariate logistic regression analyses assessed predictors of negative outcomes (fixation failure, early wound complications, fracture-related infections, and Charcot arthropathy), while the Cox-proportional hazards model analyzed predictors of five-year morbidity and mortality. Univariate predictors assessed included age, sex, type of diabetes, and CCI score. Ethical approval was obtained in accordance with institutional policy.
Results:
We compared 152 diabetic ankle fracture patients with a control group. 74 received conservative treatment, and 78 underwent operative fixation, including 31 with rigid-fixation. Both diabetic groups had higher complication risks than control (n=180), with relative-risks ranging from 3.1-3.4(P< 0.002). Forty-five-patients had CCI>5(6.03±1.86), with increased diabetes complications: neuropathy (RR=5.9,p< 0.003), higher HbA1c levels (RR=4.6,p< 0.004). Risks post-surgery decreased with prolonged-immobilization (RR=0.86) and/or rigid-fixation (RR=0.65).
Complication risks were similar between conservative and surgical management (RR=4.6 vs. 5.1). CCI>5 score correlated with morbidity(r=0.43, p=0.03). Multivariate-logistic-regression showed increased morbidity with standard-fixation (OR=1.48,95%CI= 0.97–3.26, p=0.033). Comparing rigid-long-segment fixation, each unit increase in CCI within the non-rigid fixation group raised morbidity risk by 29.5% (HR=1.295,95%CI=0.937–1.789,p=0.033). The area under the receiver operating characteristic curve was 0.764, validating the index's use in our prediction model.
Conclusion:
We agree diabetics have a higher risk for complications than non-diabetics, however the risk is not as great as previously reported. There is little difference in surgical or conservative treatment but strong evidence indicating rigid long-segment-fixation, and prolonged-protected imobilisation could improve the risk benefit ratio when compared to non- operative management.
