A comparative study of 6-week and 12-week Radiographic Union Scores for HUmeral fractures (RUSHU) as a predictor of humeral shaft non-union.

Background: Non-union in non-operatively managed humeral shaft fractures are associated with significant morbidity. Hence, developing a robust system that could help with early diagnosis is important. We aimed to evaluate the validity of the Radiographic Union Score for HUmeral fractures (RUSHU) at 6 weeks (RUSHU-6) and test whether a RUSHU at 12 weeks (RUSHU-12) would be a better predictor of non-union. Methods: We retrospectively reviewed all non-operatively managed humeral diaphyseal fractures from 2012 to 2018. Statistical analysis was used to determine the cut-off RUSHU-12 and evaluate the effect of RUSHU-6 and RUSHU-12 on non-union prediction. Results: In sum, 32 patients had radiographs at 6 weeks post-injury, 27 of which also had radiographs at 12 weeks. A RUSHU cut-off of 9 was the best predictor of non-union at 12 weeks. Only RUSHU-12 had a statistically significant influence predicting non-union (P = 0.011) and there was a significant correlation (P = 0.003) between score progression from RUSHU-6 to RUSHU-12 and the development of non-union. Discussion: A RUSHU-12 of <9 and a low score progression between 6 and 12 weeks suggest superior predictive value in determining the likelihood of non-union. Further validation in the form of a large multicentred study is however required.