Background: We determined whether the torsion profiles of the radius and ulna could be reliably assessed with cross-sectional magnetic resonance imaging and whether these torsion profiles were comparable on the two sides of volunteers.

Methods: We assessed magnetic resonance imaging cross sections of the left and right forearms of twenty-four asymptomatic volunteers. The torsion profile of the ulna was defined as the angle formed between a line tangential to the volar cortical surface of the distal part of the humerus and a line connecting the center of the ulnar head and the center of the ulnar styloid. Use of paired proximal and distal landmarks resulted in five different methods of assessment of the radial torsion profile. Intrarater and interrater reliabilities and side-to-side variability were assessed.

Results: This method of assessment of the ulnar torsion profile had intraclass and interclass coefficients of 0.95 and 0.91, respectively. A method previously described by Bindra et al. had the best combined intrarater and interrater reliabilities for assessment of the radius. The mean differences between the right and left sides of the volunteers were the lowest with the use of these two methods; nevertheless, the maximum side-to-side difference was >30° with techniques.

Conclusions: Torsion-profile assessment with cross-sectional magnetic resonance imaging had high intrarater and interrater reliabilities. However, individual side-to-side variations in the radial and ulnar profiles are important considerations.

Clinical Relevance: Cross-sectional magnetic resonance imaging is currently the only available method to quantify rotational malunion of the radius and ulna. Its low side-to-side reliability warrants comparison between the imaging results and the clinical findings. A side-to-side difference in the rotation profile may serve as a reason to perform an axial osteotomy when the results of the clinical and magnetic resonance imaging assessments are consistent with each other.