 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Magnetic Resonance Imaging for Evaluation of Failed Repairs of the Rotator Cuff. Relationship to Operative Findings*   , KANSAS CITY, MISSOURIInvestigation performed at the Department of Orthopaedic Surgery, University of Missouri, and the Magnetic Imaging Center of Kansas City, Missouri
We used magnetic resonance imaging to evaluate thirty shoulders in twenty-nine patients who had unacceptable pain and loss of function after at least one operative procedure for a lesion of the rotator cuff. A preoperative history was recorded and physical and radiographic examinations were performed before magnetic resonance imaging. All patients had a reoperation two to 156 months (average, thirty-two months) after the operation. The results of the magnetic resonance imaging and the findings at the most recent operation were then compared with regard to the integrity of the rotator cuff, the deltoid muscle, and the long head of the biceps tendon. The sensitivity and specificity of magnetic resonance imaging for the diagnosis of full-thickness tears, partial-thickness tears, and intact rotator cuffs were 84 and 91 per cent, 83 and 83 per cent, and 80 and 100 per cent, respectively. The positive and negative predictive values were 94 and 77 per cent, 56 and 95 per cent, and 100 and 96 per cent, respectively. Three shoulders had a clinically detached and retracted origin of the deltoid muscle that was identified correctly on magnetic resonance imaging, and this finding was confirmed operatively. A rupture of the long head of the biceps tendon was identified correctly in four of six shoulders. In one of the two remaining shoulders, the rupture was obscured on magnetic resonance imaging because of a so-called balloon artefact from a nearby metal suture anchor. Magnetic resonance imaging detected two unexpected lesions: a denudation of the articular cartilage in one shoulder and a ganglion cyst in the supraspinatus muscle in another.
Magnetic resonance imaging has become an increasingly useful means for evaluating lesions of the rotator cuff. Specific criteria for diagnosis with magnetic resonance imaging, corresponding to the three stages of impingement lesions12, have been described8, and the sensitivity and specificity of these criteria have been determined7,8,11,17,18. However, these studies have involved the imaging of shoulders that have not been operated on, for which diagnosis is based on the signal intensity and the morphology of the rotator cuff tendon, the status of the subdeltoid and subacromial fat planes, and the transmission of synovial or bursal fluid into or through the substance of the tendon. In shoulders that have been treated operatively, the physiology and morphology of the tendon are changed and therefore different diagnostic criteria are needed. To our knowledge, only one other report has documented the value of magnetic resonance imaging for evaluation of a symptomatic, painful shoulder postoperatively15. The purpose of the current study was to assess the ability of magnetic resonance imaging to detect recurrent tears of the rotator cuff and associated lesions after operative treatment for a lesion of the rotator cuff had failed.
We used magnetic resonance imaging to evaluate thirty shoulders in twenty-nine patients who had pain and loss of function after at least one operative procedure for a lesion of the rotator cuff. Twenty-five shoulders had had a repair of the rotator cuff; four, arthroscopic débridement of the cuff; and one, open acromioplasty alone. Twenty-one shoulders had had one operation, seven had had two, and two had had three. All but one of the procedures had been performed elsewhere. There were twenty-two men and seven women, and the average age was fifty-four years (range, twenty-one to seventy-three years). A history was recorded, a physical examination was performed, and anteroposterior radiographs with the shoulder in internal and external rotation, supraspinatus outlet radiographs, and axillary radiographs were made for all patients. Magnetic resonance imaging was performed with a 1.5-tesla Signa unit (General Electric, Milwaukee, Wisconsin). Oblique coronal T1 and T2-weighted images, oblique sagittal T2-weighted images, and axial multiplanar gradient recall images were made with the patient supine, the upper extremities at the side, and the humerus in neutral rotation. The images were interpreted preoperatively by a musculoskeletal radiologist (G. W. H.) who had no information on the patients. All shoulders were reoperated on by the same one of us (C. C. S.) two to 156 months (average, thirty-two months) after the most recent operative procedure. The results of the magnetic resonance imaging, which were known to the surgeon before the reoperation, were compared with the operative findings with regard to the integrity of the rotator cuff, the deltoid muscle, and the long head of the biceps tendon.
Tears of the Rotator Cuff
Retraction of the Deltoid Muscle
Rupture of the Long Head of the Biceps Tendon
Calculations Rupture of the long head of the biceps tendon, retraction of the deltoid muscle, and unexpected findings are reported without such determinations as their numbers were small.
Operatively, nineteen shoulders were found to have a full-thickness tear of the rotator cuff; six, a partial-thickness tear; and five, no tear (Table I). Of the nineteen full-thickness tears, sixteen were diagnosed correctly with magnetic resonance imaging and three were diagnosed incorrectly as partial-thickness tears. No rotator cuff that had a full-thickness tear was identified incorrectly as being intact on magnetic resonance imaging. Of the six partial-thickness tears, five were diagnosed correctly with magnetic resonance imaging and one was diagnosed incorrectly as a full-thickness tear. No cuff that had a partial-thickness tear was identified incorrectly as being intact on magnetic resonance imaging. Of the five intact rotator cuffs, four were identified correctly with magnetic resonance imaging and one was diagnosed incorrectly as having a partial-thickness tear on magnetic resonance imaging. No cuff that was intact was diagnosed incorrectly as having a full-thickness tear on magnetic resonance imaging.
Magnetic resonance imaging had a sensitivity and specificity of 84 and 91 per cent in the diagnosis of full-thickness tears, with positive and negative predictive values of 94 and 77 per cent; a sensitivity and specificity of 83 per cent each in the diagnosis of partial-thickness tears, with positive and negative predictive values of 56 and 95 per cent; and a sensitivity and specificity of 80 and 100 per cent in the identification of intact cuffs, with positive and negative predictive values of 100 and 96 per cent (Table II).
In three shoulders, a clinically detached and retracted origin of the deltoid muscle was also seen on magnetic resonance imaging and was confirmed operatively (Table III).
Rupture of the long head of the biceps tendon was found operatively in six shoulders and was diagnosed correctly with magnetic resonance imaging in four (Table III). One biceps tendon that had a rupture was incorrectly interpreted as being intact on magnetic resonance imaging. In an additional shoulder, a rupture of the biceps tendon was obscured on magnetic resonance imaging because of a balloon artefact from a nearby metal suture anchor (Fig. 2). The long head of the biceps tendon was found to be intact operatively in twenty-three shoulders; this was identified correctly with magnetic resonance imaging in twenty-two shoulders. In one patient in whom the biceps tendon was intact, the presence of metal artefact made the assessment of the integrity of the biceps tendon impossible. No patient who had an intact biceps tendon was diagnosed incorrectly as having a rupture on magnetic resonance imaging.
Two unexpected lesions—a denudation of the articular cartilage in one shoulder and a ganglion in the supraspinatus muscle (Fig. 3) in another—were found on magnetic resonance imaging and were confirmed operatively.
Although it is uncommon, some patients remain symptomatic or have recurrent symptoms after operative treatment of a lesion of the rotator cuff. The cause of these symptoms is often multifactorial and can be difficult to determine on clinical examination10,12,14,16. Although a recurrent tear of the rotator cuff is a component of the pathological condition in many patients after a repair of the cuff has failed, many others have an intact cuff16. Consequently, an imaging modality that can be used to evaluate not only the status of the rotator cuff but also other potential sources of symptoms is advantageous. We found that magnetic resonance imaging reliably defined the status of the rotator cuff, the deltoid muscle, and the long head of the biceps tendon. Magnetic resonance imaging was also able to demonstrate the vascularity of the humeral head and the presence of abnormalities in the surrounding soft tissues, such as the ganglion found in one of our patients. Arthrography has been used in other series to determine the cause of persistent symptoms after a repair of the rotator cuff4-6,14,18; however, this modality may be unreliable, as contrast medium can leak through the interstices of a well repaired cuff or can fail to leak from the joint in the presence of a major scar despite a tear of the cuff. Calvert et al. performed double-contrast arthrography in nineteen asymptomatic shoulders six to seventy-eight months after a repair of the cuff; they found leakage of contrast medium into the subacromial space in seventeen shoulders. Crass et al. found that arthrography of persistently symptomatic shoulders that had had a repair of the rotator cuff had a sensitivity of 66 per cent and a specificity of 50 per cent for the diagnosis of a recurrent tear. DeOrio and Cofield reported on twenty-seven patients who had had a second repair after a previous repair had failed. An arthrogram was made for fifteen of these patients, but it was false-negative for two. Arthrography also has the disadvantage of being invasive, although it is less expensive than magnetic resonance imaging. Ultrasound has been used at some centers to evaluate shoulders postoperatively2,5,9. Although their study included only ten patients in whom the shoulder had been explored, Crass et al. found that ultrasound had a sensitivity of 85 per cent and a specificity of 100 per cent for the diagnosis of recurrent tears of the rotator cuff. Mack et al., in a study of twenty-seven shoulders, found a sensitivity of 100 per cent and a specificity of 90 per cent for the diagnosis of a recurrent tear. Brandt et al. reported three false-positive sonographic diagnoses of a recurrent tear and one correct identification of an intact cuff in four patients who also had operative findings. In revealing the status of the rotator cuff in shoulders that have not been treated operatively, the accuracy of ultrasound has been quite variable and inferior to that of magnetic resonance imaging3. The diagnostic disadvantages of ultrasound include its limited contrast resolution in the anatomy of the cuff and its relative inadequacy for the assessment of bone, cartilage, labrum, and other surrounding soft tissues of the shoulder. Practical disadvantages include a steep learning curve and the fact that the usefulness of the study depends greatly on the skill of the technician. Although ultrasound is relatively inexpensive as well as non-invasive, its use as the primary imaging modality for the shoulder has been advocated by few experienced investigators2,3. Magnetic resonance imaging has been shown to be very useful for the assessment of lesions in shoulders that have not been operated on1,3,7,8,10,17,18. Iannotti et al. found that, in such shoulders, magnetic resonance imaging had 100 per cent sensitivity and 95 per cent specificity for the diagnosis of full-thickness tears of the rotator cuff and 82 per cent sensitivity and 85 per cent specificity for the diagnosis of partial-thickness tears. Other studies have confirmed the diagnostic superiority of magnetic resonance imaging as compared with arthrography in the evaluation of shoulders that have not been treated operatively3,10,18. There is a paucity of data on the use of magnetic resonance imaging in shoulders that have been operated on. In a study of thirty-one patients who had persistent symptoms after an operation for a lesion of the rotator cuff, Owen et al. used magnetic resonance imaging to make a diagnosis of either an intact or a torn rotator cuff. The category of a partial-thickness tear was not used in their study. At reoperation, they found that magnetic resonance imaging had a sensitivity of 86 per cent and a specificity of 92 per cent for the diagnosis of a full-thickness tear15. Our findings of 84 per cent sensitivity and 91 per cent specificity for the diagnosis of a full-thickness tear are consistent with those of Owen et al. In both studies, the same criteria were used for the diagnosis of a full-thickness tear on magnetic resonance imaging. These criteria are narrower than those used by Iannotti et al. for the diagnosis of a full-thickness tear in a shoulder that has not been operated on. We believe that the high sensitivity and specificity achieved in both studies, while lower than the values for shoulders that have not been treated operatively8, validate the diagnostic criteria. Our study included the diagnosis of partial-thickness tears. Despite the fact that the magnetic resonance imaging criteria for this diagnosis are more restrictive for a shoulder that has been operated on than for one that has not, our findings of 83 per cent sensitivity and specificity are comparable with those for shoulders that have not had operative treatment8. There was, however, an overdiagnosis of partial-thickness tears with magnetic resonance imaging, primarily in patients who had a full-thickness tear. This resulted in magnetic resonance imaging having a poor positive predictive value for the diagnosis of partial-thickness tears. Diagnosis of a partial-thickness tear with magnetic resonance imaging in a shoulder that has been operated on relies on visualization of focal thinning of the tendon as well as an invagination of fluid at the articular or bursal surface. On a T2-weighted image, a true partial-thickness tear cannot be absolutely distinguished from advanced edematous degeneration (fluid around a suture defect) of the tendon. This is the reason why magnetic resonance imaging had a poor positive predictive value for the diagnosis of partial-thickness tears in the current series. We believe, however, that this is an important diagnosis, and we were able to make it with relatively high rates of sensitivity and specificity. Despite the relatively low positive predictive value, we believe that the criteria for this diagnosis with magnetic resonance imaging remain valid. Interpretation of the postoperative anatomy with magnetic resonance imaging is compromised by the presence of scar and artefacts from foreign bodies. Artefacts did not seem to interfere with the differentiation between a recurrent tear and an intact cuff, but it did contribute to the incorrect diagnosis of a partial-thickness tear in a patient who had an intact cuff (Fig. 1-B). In that patient, a non-absorbable suture resulted in an image similar to that of fluid in a tendon. A metal artefact obscured the image in two shoulders (Fig. 2), compromising the interpretation of the integrity of the biceps tendon. Metal suture anchors cause a balloon artefact, which can obscure the visualization of adjacent structures on magnetic resonance imaging. Artefacts from dystrophic calcification or microscopic metal shavings from the passage of needles through tissue do not involve such a large area and therefore do not hinder visualization to a great extent. As most patients have a successful result after repair of the rotator cuff, the possibility that an atypical lesion is responsible for persistent symptoms should be considered. In the current series, an intramuscular ganglion cyst in the supraspinatus muscle was identified in one patient (Fig. 3). An additional patient, for whom previous plain radiographs had shown a normal glenohumeral joint space, had substantial loss of articular cartilage as revealed on magnetic resonance imaging. In summary, our findings suggest that magnetic resonance imaging is useful for the evaluation of patients who have pain and loss of function after a repair of the rotator cuff. Although the history and physical examination remain the primary determinants of therapeutic decisions, magnetic resonance imaging can be useful for the preoperative confirmation of a suspected lesion or the detection of an unsuspected lesion. We found magnetic resonance imaging to be helpful for the identification of full-thickness tears, intact cuffs, ruptures of the biceps tendon, and detachments of the origin of the deltoid muscle. It was less effective for the identification of partial-thickness tears.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Technorati What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |