ACCURACY OF MAGNETIC RESONANCE ARTHROGRAPHY WHEN COMPARED TO ARTHROSCOPIC FINDINGS IN SHOULDER INSTABILITY

Abstract

The aim of this study was to assess the accuracy of Magnetic Resonance Arthrography (MRA) in symptomatic shoulder joint instability. Data were collected prospectively from MR Arthrograms performed in 40 consecutive patients with recurrent symptomatic instability. MR Arthrograms included views in the stress ABER position of the shoulder. Subsequently all patients underwent an arthroscopic shoulder stabilisation and the arthroscopic and MR Arthrographic findings were correlated. In case of discrepancy the films and operative findings were reviewed.

There were thirty three male and seven female patients with a mean age of 28 years (range 18–40). MR Arthrography showed 37 anterior-inferior tears (22 displaced Bankart tears, 8 nondisplaced Bankart tears, 5 chronic ALPSA lesions and 2 AGL lesions) and 3 posterior lesions. There were 3 discrepancies of which 2 were cases of missed Bankart lesion on MRArthrogram and one case of reverse Perthes lesion present on MRArthrogram but not seen on arthroscopy. Despite a review of the films, the missed Bankart’s lesions were not visualised. In the present study, MR arthrography had a 95% sensitivity in detecting ligamentolabral pathology and a positive predictive value of 0.975 in diagnosing a lesion in recurrent shoulder instability.

The results show that MR Arthrogram is a highly useful tool for investigating recurrent shoulder instability with very high sensitivity and positive predictive value. Of the 40 patients who underwent arthroscopy there were only 2 cases where MR Arthrography did not demonstrate an arthroscopically detected abnormality.