Sonography of the painful shoulder: role of the operator's experience

The aim of our study was to determine the role of the operator's experience in the sonographic evaluation of the painful shoulder and to validate assumptions about its technical performance in routine practice.

Two radiologists, respectively standard and expert sonographic operators, independently and prospectively scanned 65 patients with clinical suspicion of rotator cuff lesion. Magnetic resonance arthrography was the reference standard.

The sensitivity of the expert ultrasound operator was 95.3% for full-thickness rotator cuff tears (41/43), 70.6% for partial-thickness tears (12/17), 64.3% for intratendinous tears (9/14), 100% for abnormality of the long head of biceps tendon (seven of seven), 88.9% for supraspinatus tendinosis (16/18), 96.4% for subacromial bursa abnormalities (53/55), and 91.7% for acromioclavicular joint osteoarthritis (33/36). The two sonographic operators were in very good agreement about full-thickness rotator cuff tears (kappa = 0.90), supraspinatus tendinosis (kappa = 0.80), abnormalities of the long head of biceps tendon (kappa = 0.84), subacromial bursa abnormalities (kappa = 0.89), and acromioclavicular osteoarthritis (kappa = 0.81). The agreement was only moderate for partial-thickness tears (kappa = 0.63) and intratendinous tears (kappa = 0.57).

Our results show that in moderately experienced hands as in experts' hands, sonography has a low level of interobserver variability for full-thickness rotator cuff tears. Considering partial-thickness and intratendinous rotator cuff tears, our data suggest that interobserver variability is higher.