Do patients with diabetes have an increased risk of impaired fracture healing? A systematic review and meta-analysis

Background The majority of the existing evidence showing an association between diabetes and impaired fracture healing comes from basic scientific research. This systematic review and meta-analysis aimed to summarize the current clinical literature that investigates fracture healing in patients with diabetes. Methods The outcome of interest was impaired fracture healing including non-union, delayed union and malunion. Studies that compared fracture healing outcomes between patients with and without diabetes were included in this study. Subgroup analyses regarding different fracture sites, types of fracture and classifications of diabetes were performed. Results A total of 14 studies involving 695 patients with diabetes and 4937 controls fulfilled the inclusion criteria. Diabetes was associated with an increased risk of impaired fracture healing (odds ratio (OR): 2.11, 95% confidence interval (CI) 1.33-3.37, P = 0.002). Subgroup analyses showed that diabetes was associated with a significantly higher incidence of impaired fracture healing in lower extremity fractures (OR 2.63, 95% CI 1.30-5.30, P = 0.007), short bone fractures (OR 2.64, 95% CI 1.35-5.20, P = 0.005), long bone fractures (OR 2.13, 95% CI 1.23-3.70, P = 0.007) and osteoporosis-unrelated fractures (OR 2.39, 95% CI 1.19-4.80, P = 0.01). Both insulin-dependent diabetes (OR 4.04, 95% CI 1.05-15.56, P = 0.04) and non-insulin-dependent diabetes (OR 5.83, 95% CI 1.73-19.58, P = 0.004) were associated with significantly higher risks of impaired fracture healing. Conclusions Patients with diabetes have an increased risk of impaired fracture healing when compared to patients without diabetes. Fracture healing in the lower extremities, short bones and osteoporosis-unrelated fractures is affected more severely by diabetes.