Accuracy of High-Resolution Ultrasonography in the Diagnosis of Articular-Sided Partial Thickness Rotator Cuff Tears

Abstract

Background The present study aimed to assess the accuracy of high-resolution ultrasonography in the detection of articular-sided partial thickness rotator cuff tears. Methods Two-hundred and forty-six shoulders of 245 patients underwent ultrasonography and subsequent shoulder arthroscopy. All scans were performed by an experienced specialist musculoskeletal radiologist and arthroscopies were carried out by a single surgeon. Results Fifty-eight partial thickness tears (of which 56 were articular-sided) were found at arthroscopy. Of the remaining shoulders, 90 had full thickness tears and 98 had intact rotator cuffs. Ultrasonography correctly identified 21 of 56 articular-sided tears as partial thickness tears. It had a sensitivity of 7%, a specificity of 98% and an accuracy of 74% for the diagnosis of articular-sided partial thickness tears. If a full thickness tear was considered as a true positive, the sensitivity increased to 89%, specificity to 98% and accuracy to 96%. Discussion Ultrasonography is not sufficiently accurate to diagnose articular-sided partial thickness tears. A higher accuracy can be achieved if a full thickness tear is considered a positive finding. Ultrasonography should not be considered a first line investigation for a clinically suspected partial thickness rotator cuff tear. An articular-sided partial thickness rotator cuff tear should be considered in all patients undergoing an arthroscopic rotator cuff procedure.