The acutely injured acromioclavicular joint - which imaging modalities should be used for accurate diagnosis? A systematic review.

Background: The management of acute acromioclavicular (AC) joint injuries depends on the degree of injury diagnosed by the Rockwood classification. Inadequate imaging and not selecting the most helpful imaging protocols can often lead to incorrect diagnosis of the injury. A consensus on a diagnostic imaging protocol for acute AC joint injuries does not currently exist. Therefore we conducted a systematic review of the literature considering three diagnostic parameters for patients with acromioclavicular (AC) joint injuries: 1) Assessment of vertical instability; 2) Assessment of horizontal instability; 3) Benefit of weighted panoramic views.

Methods: Internet databases were searched in March 2016 using the terms ("AC joint" OR "acromioclavicular joint") AND (MRI OR MR OR radiograph OR X-ray OR Xray OR ultrasound OR "computer tomography" OR "computed tomography" OR CT). Diagnostic, prospective, retrospective, cohort and cross- sectional studies were included to compare their use of different radiological methods. Case reports, cadaveric studies, and studies concerning chronic AC injuries and clinical outcomes were excluded.

Results: This search returned 1359 citations of which 1151 were excluded based on title, 116 based on abstract and 75 based on manuscript. 17 studies were included for review and were analyzed for their contributions to the three parameters of interest mentioned above. The inter- and intra-observer reliability for diagnosing vertical instabilities of the clavicle using x-ray alone show a high level of reproducibility while for horizontal instabilities the values were much more variable. In general, digitally measured parameters seem to be more precise and reliable between investigators than visual classification alone. Currently, evidence for the value of weighted views and other additional diagnostic imaging to supplement standard x-rays is controversial.

Conclusion: To date there is no consensus on a gold standard for diagnostic measures needed to classify acute AC joint injuries. The inter- and intra-observer reliability for diagnosing vertical instabilities of the clavicle using bilateral projections show a high level of reproducibility while for horizontal instabilities the results are much more inconsistent. There is currently no clear consensus on a protocol for image-based diagnosis and classification of acute AC joint injuries, leading to a lack of confidence in reproducibility and reliability.