Supra-Short Ultrasound Protocol for Rotator Cuff Tears in the Emergency Department: Pilot Study

Although ultrasound is readily available to emergency physicians and known to be very accurate for diagnosing rotator cuff tears, it is rarely used for this purpose. Our goal in this study was to develop and preliminarily assess the accuracy of a simplified shoulder ultrasound protocol (the "supra-short" protocol), designed to be used by emergency physicians for diagnosis of supraspinatus tears.

We performed a pilot diagnostic accuracy study in which we assessed the accuracy of the supra-short protocol as performed by minimally trained emergency physicians for identifying supraspinatus tears in volunteers. As a criterion standard, a sports medicine physician also performed a complete shoulder ultrasound on each volunteer. We determined the test characteristics of the supra-short protocol for supraspinatus tears, as well as the median time to complete a scan and the percentage of images deemed adequate by expert review.

Nine emergency physicians performed a total of 40 bilateral supra-short scans on six volunteers (two of whom were known to have shoulder pathology and four of whom had normal shoulders). Of the 80 shoulders scanned, there were 18 cases in which complete ultrasound performed by the sports medicine physician revealed a supraspinatus tear; 12 (66.7%) of those were identified by the novice sonographers using the supra-short protocol. Overall, the sensitivity of the supra-short protocol was 66.7% (95% CI 29.9-92.5%) and the specificity was 87.1% (95% CI 70.2-96.4%). The median time to completion of each shoulder was 133 seconds (interquartile range 88-182). Upon expert image review, 80.0% of the images were deemed adequate.

After minimal training, emergency physicians were able to quickly perform the supra-short US protocol but were only able to identify supraspinatus tears with moderate accuracy, suggesting the need for more extensive training before clinical use.