Studies on acromioclavicular joint dislocations : clinical outcomes and radiological classification

Abstract

Background: Acromioclavicular joint (AC joint) dislocation is a common injury among young sports-active people. The injuries are divided into types I–VI based on the Rockwood classification. The classification system has been validated in multiple studies, but the results have not been consistent. Currently, AC joint dislocations Rockwood types I and II are treated non-operatively, treatment of type III is started non-operatively, and types IV–VI are treated operatively. There is no gold standard regarding type of surgical procedure.

Aims: The aim of this thesis was to improve knowledge on AC joint dislocations, outcome of treatment, and reliability of radiological classification.

Materials and methods: Studies I and II were retrospective studies, evaluating the outcome after operative treatment for chronic AC joint dislocation Rockwood types III–V, and timing of operative treatment for Rockwood type V. Study III was an RCT, evaluating the outcome after non-operative and operative treatment for acute Rockwood type III and V dislocations. These clinical studies were based on clinical examinations, validated PROMs, radiologic evaluation and questions regarding subjective satisfaction with shoulder and cosmesis. Study IV evaluated the inter-observer and intra-observer reliability of Rockwood’s classification in plain radiographs and plain radiographs in combination with computed tomography. A new simpler method for classification av AC joint injuries was evaluated.

Results: In Study I, patients operated with a modified Weaver-Dunn procedure augmented with PDS suture or with a hook plate had no significant differences in outcome, except for pain; the group with hook plate had more pain during movement. In Study II, patients with type V dislocation operated within the acute phase with a hook plate had significantly better outcome in almost all parameters than the group undergoing delayed treatment. In Study III, there were no significant differences in outcome after 2 years, between patients treated non-operatively and operatively, regardless of whether they had a type III or V dislocation. In Study IV the inter-observer and intra-observer reliability of the simple classification, when classifying Rockwood type III and V, was significantly better than the Rockwood classification using plain radiographs, but not significantly better than plain radiographs in combination with CT. Finally, the reliability of Rockwood classification using plain radiographs in combination with CT is significantly better than using plain radiographs alone.

Conclusions: A hook plate did not improve the results after operative treatment of chronic AC joint dislocation types III–V. If AC joint dislocation Rockwood type V was treated surgically, the results were better after acute surgery than after delayed surgery. Patients with acute Rockwood type III or V dislocations regained good shoulder function and subjective satisfaction with the result after 2 years, regardless of if they were treated non-operatively or operatively. Plain radiographs and CT in combination improved the reliability of the Rockwood classification type III and V, but the clinical relevance of this is unclear. The simple classification needs further investigation.