Detection and quantification of glenohumeral joint effusion: reliability of ultrasound

To evaluate reliability of ultrasound for detection and quantification of glenohumeral joint effusion.

With institutional review board approval and informed consent ultrasound of 30 consecutive patients before and after MR arthrography of the shoulder was performed. Presence and width of any anechoic collection was noted within various locations (biceps tendon sheath, subscapular recess (neutral position and internal rotation), posterior glenohumeral joint recess (neutral position and external rotation)). Injected fluid (8-12 ml) into the glenohumeral joint served as gold-standard. Widths of anechoic collections were correlated (Spearman rank correlation) with injected fluid.

Glenohumeral anechoic collection was consistently seen in the posterior glenohumeral joint recess with the arm in external rotation (100%, 30/30), and in the biceps tendon sheath (97%, 29/30). Ultrasound was not sensitive at other locations (7%-17%). Mean width in anterior-posterior direction of anechoic collection in the posterior glenohumeral joint recess was 7 mm (range: 3-18 mm), 2 mm (range: 1-7 mm) in the biceps tendon sheath. Significant correlation (R = 0.390, p = 0.033) was found between width of anechoic collection and injected fluid in the posterior glenohumeral joint recess.

Glenohumeral joint effusion can be detected and quantified most reliably in the posterior glenohumeral joint recess with the arm in external rotation.