Treatment of long bones non-union: Factors associated with clinical outcome

Objectives: Non-union is a common complication of long bone fracture with an approximate incidence of 2–30%. Several factors have been implicated in developing non-union, including patient and surgeon-related factors. The treatment approach depends on whether the non-union is hypertrophic or atrophic. Recently, there has been interest in identifying factors that may be associated with clinical outcomes following surgically treated non-union and the study aimed to study these factors. Methods: This is a retrospective study spanning 5 years during which all information necessary for the study and data analysis were retrieved. The outcome was the healing of the surgically treated non-union, while other variables were considered as covariates. Initial bivariate logistic regression was done and only variables with P < 0.5 were further considered in the multiple logistic regression analysis (MLRA) to identify factors that significantly predict the healing of surgically treated non-union. Results: The results of the bivariate logistic regression showed that the use of bone graft, duration of non-union, initial care given, and implant used had a P < 0.5, and these were the variables included in the MLRA. The result of the MLRA showed that plate and screw implant is the only independent predictor of healing following surgical treatment of long bone non-union (odds ratio = 14.36, z = 2.34, confidence interval: 1.725–172.433, P < 0.0194). Conclusion: Plate and screw fixation predicts healing of long bone non-union. Future prospective studies are needed to validate the findings of this present study.