Accuracy of office-based ultrasonography of the shoulder for the diagnosis of rotator cuff tears

Background: This prospective multi-institutional study was designed to define the accuracy of ultrasonography, when performed in an orthopaedic surgeon's office, for the diagnosis of rotator cuff tears. Methods: An anatomic diagnosis and a treatment plan were made on the basis of office-based shoulder ultrasonography, physical examination, and radiographs for ninety-eight patients (ninety-nine shoulders) with a clinical diagnosis of a rotator-cuff-related problem. The results of the ultrasonographic studies were then compared with the results of magnetic resonance imaging and the operative findings. Results: Office-based ultrasonography led to the correct diagnosis for thirty-seven (88%) of forty-two shoulders with a full-thickness rotator cuff tear or both full and partial-thickness tears, twenty-six (70%) of thirty-seven shoulders with a partial-thickness rotator cuff tear only, and sixteen (80%) of twenty shoulders with normal tendons. In no case was the surgical approach (open or arthroscopic) that had been planned on the basis of the ultrasonography altered by the operative findings, but the operative finding of a full-thickness tear resulted in an arthroscopic cuff repair in four shoulders. Magnetic resonance imaging led to the correct diagnosis for forty (95%) of forty-two shoulders with a full-thickness rotator cuff tear or both full and partial-thickness rotator cuff tears, twenty-seven (73%) of thirty-seven shoulders with only a partial-thickness tear, and fifteen (75%) of twenty shoulders with normal tendons. There were no significant differences between magnetic resonance imaging and ultrasonography with regard to the correct identification of a full-thickness tear or its size. The sensitivity of ultrasonography for detecting tear size in the anterior-posterior dimension was 86% (95% confidence interval, 71% to 95%), and that of magnetic resonance imaging was 93% (95% confidence interval, 81% to 99%) (p = 0.26). The sensitivity of ultrasonography for detecting tear size in the medial-lateral dimension was 83% (95% confidence interval, 69% to 93%), and that of magnetic resonance imaging was 88% (95% confidence interval, 74% to 96%) (p = 0.41). Conclusions: A well-trained office staff and an experienced orthopaedic surgeon can effectively utilize ultrasonography, in conjunction with clinical examination and a review of shoulder radiographs, to accurately diagnose the extent of rotator cuff tears in patients suspected of having such tears. Errors in diagnosis made on the basis of ultrasonography most often consist of an inability to distinguish between partial and full-thickness tears that are approximately 1 cm in size. In this study, such errors did not significantly affect the planned surgical approach. Level of Evidence: Diagnostic Level I. See Instructions to Authors for a complete description of levels of evidence.