Abstract
Background: Anecdotally, upper truncal obesity and large breasts have often been associated with inferior outcomes from non-operative management of diaphyseal humerus fractures. However, this assertion is without basis in the literature. Aims: To produce radiographic measurements of chest wall soft tissue thickness (STT) and determine association with non-union in diaphyseal humerus fractures. Methods: Two hundred and seventeen consecutive non-operative humeral shaft fractures were included. Radiographic STT measurements were taken at three standardised points (upper, middle and lower) using a simple reproducible method, with ratios derived (dividing these figures by the mid-humerus diameter). Bivariate and multivariable analyses were used to assess association with non-union. Results: There were 58 (26.7%) cases of non-union. On multivariable analysis, the middle (odds ratio (OR) 1.39, p < 0.001) and lower (OR 1.23, p = 0.009) STT measurements were independently associated with non-union. Additionally, the middle (OR 1.85, p < 0.001) and lower (OR 1.47, p < 0.001) STT ratios were independently associated with non-union. A receiver operating characteristic curve determined a threshold value of a middle STT ratio of >= 3 (OR 3.73, p < 0.001, sensitivity 69.0%, specificity 62.3%), which was independently associated with non-union. Conclusion: Chest wall STT is independently associated with humeral shaft non-union. Threshold values can assist in decision making for these fractures.