Diagnosing a chronic full-thickness supraspinatus tendon tear using ultrasound is challenging because tendon defects are usually obliterated with fibrous tissues, mimicking continuity of tendon. Among many direct and indirect ultrasonographic signs for full-thickness supraspinatus tendon tears, diagnostic values of indirect signs, especially sagging peribursal fat sign (SPFS), have been undervalued. To investigate whether the presence of SPFS would be suggestive of a full-thickness tear, the sensitivity, specificity and predictive value of SPFS were analyzed by regarding arthrosonography as a standard reference.

Hundred and four patients with shoulder pain for more than 3 months, who underwent arthrosonography under the suspicion of full-thickness supraspinatus tendon tear were included. Suspicion of full-thickness tear was determined by substantial loss (>50% of thickness) of fibrillary patterns in either the bursal or articular side with more than one of four indirect signs such as long biceps tendon sheath swelling, cortical irregularity, double cortex sign and SPFS. Patients with obvious tendon defect were excluded because diagnosis is too obvious to consider indirect sings. Arthrosonography was done by injecting 20ml of fluid into the glenohumeral joint capsule followed by careful sonographic evaluation on the subacromial subdeltoid bursal space to find any fluid leaking, which would confirm the existence of true full-thickness tear, serving as a communication channel between the joint capsule and superficial space overlying the rotator cuff tendons.

The SPFS was the most sensitive (88.4%) and specific (98%) to diagnose full-thickness tear among the indirect signs. When SPFS was combined with other indirect signs such as biceps tendon sheath swelling, cortical irregularity of humeral head, double cortex sign, the specificity and positive predictive value reached up to 100%.

These results highlight that SPFS can be served as a strong evidence of full-thickness supraspinatus tendon tear.