Abstract
Introduction and Aims: The authors have previously reported the comparison between MRI and arthroscopic diagnosis in a blinded prospective study in patients with shoulder impingement syndrome. The purpose of this investigation was to determine whether the radiologist’s MRI report provided to an arthroscopist upon completion of a standard diagnostic shoulder arthroscopy would result in changes to the initial arthroscopic findings.
Method: Fifty-eight patients presenting with clinical signs and symptoms of shoulder impingement syndrome underwent an MRI one week prior to planned arthroscopic subacromial decompression. MRI scans were obtained following a standardised protocol. Images were read by one musculoskeletal radiologist. A standard diagnostic arthroscopy was performed. The anatomic and pathologic findings were documented intra-operatively by an independent observer. The arthroscopist was initially blinded to the MRI information until completion of the standard diagnostic arthroscopy. MRI results were then revealed to the surgeon. An arthroscopic re-evaluation was performed to resolve any discrepancies between MRI and the initial arthroscopic findings.
Results: The percentage discordance between MRI and initial arthroscopic findings for each structure of interest was calculated along with the percentage change in diagnosis based upon the arthroscopic re-evaluation and the consequence of the change.
Supraspinatus 55.2% discordance (n=32/58), 18.8% change in diagnosis (n=6/32), in one patient a change in the planned operation occurred from subacromial decompression to mini-open rotator cuff repair. Infra-spinatus 44.8% discordance (n=26/58), 3.8% change in diagnosis (n=1/26), and no consequence to planned treatment. Subscapularis 37.9% discordance (n=22/58), no change in diagnosis. Biceps tendon 62.3% discordance (n=33/53; five patients not adequately visualised on MRI), no change in diagnosis. Acromion type 50% discordance (n=26/52; five patients did not have a sub-acromial bursocopy and in one patient the acromion was not well visualised), 7.7% change in diagnosis (n=2/26) with no consequence to planned treatment. Acromioclavicular joint 22.5% discordance (n=9/40; in 18 patients the AC joint was not entered), 11.1% change in diagnosis (n=1/9) with no consequence to planned treatment.
Conclusion: Despite high percentage of discordance between MRI and arthroscopy, the MRI information modified the initial arthroscopic diagnosis in a much smaller percentage of cases. In only one patient, did the change in diagnostic information have an impact on the planned treatment.
These abstracts were prepared by Editorial Secretary, George Sikorski. Correspondence should be addressed to Australian Orthopaedic Association, Ground Floor, The William Bland Centre, 229 Macquarie Street, Sydney, NSW 2000, Australia.
At least one of the authors is receiving or has received material benefits or support from a commercial source.
