MRI assessment of scaphoid fracture displacement: Does it predict union?

Abstract

Fifty patients with acute scaphoid waist fractures were treated non-operatively in a below elbow plaster cast for 8-13 weeks. Each underwent a MRI scan of the scaphoid within 2 weeks of injury and fracture displacement was assessed on these by two observers who were unaware of the outcome (union or nonunion) by measuring a variety of parameters. The only quantitative measurements of displacement which were reproducible were fracture gapping (distraction) and translation. Qualitative categorisation of the fractures into undisplaced (41 cases) and moderately (7 cases) and severely (2 cases) displaced groups according to their overall MRI appearances was equally reproducible. Three of 49 fractures (one undisplaced fracture lost to follow up) failed to unite: two of these were moderately displaced and one was severely displaced. All the undisplaced fractures united. We conclude that MRI assessment of fracture displacement probably affects the likelihood of union with non-operative treatment though our small number of nonunions weakens the power of this study. We thus recommend that all acute scaphoid fractures should be assessed with MRI scans or tomography and classified as undisplaced and displaced. Undisplaced fractures can be treated conservatively in plaster with a high chance of union, whereas open operative fixation could be considered for displaced fractures.