Accuracy of ultrasound imaging of the rotator cuff in shoulders that are painful postoperatively.

BACKGROUND: Evaluation of the rotator cuff in a shoulder that has been persistently painful postoperatively could be challenging. Magnetic resonance imaging has been the preferred imaging method, but accuracy rates have been lower for those shoulders than for shoulders that have not been operated on; also, magnetic resonance imaging is susceptible to suture anchor artifact. The purpose of this study was to evaluate the diagnostic accuracy of ultrasound for evaluating the rotator cuff in shoulders that are painful postoperatively.

METHODS: The records of forty-four consecutive patients who had undergone both a high-resolution ultrasound examination and subsequent shoulder arthroscopy were retrospectively reviewed. Thirty-four patients underwent rotator cuff repair at the time of the arthroscopy, and ten had subacromial decompression or another procedure without cuff repair. The results of the ultrasound examination were compared with the intraoperative findings of the arthroscopic examination (the "gold standard").

RESULTS: Ultrasound led to a correct diagnosis for forty of the forty-four patients. Twenty-two recurrent rotator cuff tears and twenty-two intact rotator cuffs were found at surgery. Ultrasound correctly identified twenty of the twenty-two rotator cuff tears (true-positive results) and twenty of the twenty-two intact rotator cuffs (true-negative results). Two intact cuffs were identified as having a full-thickness tear (false-positive results), and two cuffs with a full-thickness rotator cuff tear were identified as being intact (false-negative results). The sensitivity and specificity of ultrasound for identifying rotator cuff integrity postoperatively were 91% and 86%, respectively. The accuracy was 89%.

CONCLUSIONS: Ultrasound is a highly accurate imaging study for evaluating the integrity of the rotator cuff in shoulders that have undergone an operation. Its accuracy for operatively treated shoulders appears to be comparable with that previously reported for shoulders that had not been operated on.