| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 From the Departments of Radiology (A.L.S., B.B.F., D.L.J.) and Orthopedics (P.K., R.H.H.), University of British Columbia Hospital, 2211 Wesbrook Mall, Vancouver, British Columbia, Canada V6T 2B5. Received August 4, 1998; revision requested August 28; final revision received December 22; accepted July 1, 1999. Address reprint requests to B.B.F. (e-mail: formay@istar.ca).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Fifteen subjects who had undergone clinically successful rotator cuff repair were included in the study. All underwent functional testing of the affected shoulder and had good to excellent scores on the Constant scale. Standard MR imaging sequences were performed at 1.5 T, including oblique coronal fast spin-echo T2-weighted MR imaging with fat saturation.
RESULTS: Three (10%) of 30 supraspinatus and infraspinatus tendons had normal signal intensity, and 16 (53%) had mildly increased signal intensity on fast spin-echo T2-weighted fat-saturated images, compatible with tendonitis or tendinosis. Three partial and four complete tears of the supraspinatus tendon and two partial and two complete tears of the infraspinatus tendon were seen. Other findings included subacromial-subdeltoid effusion (10 subjects), joint effusions (five subjects), and bone marrow edema (six subjects).
CONCLUSION: Postoperative signal intensity changes consistent with tendonitis or tendinosis were common, and clinically "silent" partial and complete rotator cuff tears were seen. Such postoperative MR imaging findings should be interpreted with caution, and meticulous correlation with symptoms and clinical results is recommended.
Index terms: Shoulder, surgery, 41.45 • Tendinitis, 414.4813 • Tendons, injuries, 414.253 • Tendons, MR, 414.121411, 414.121415
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
–5 years previously. All subjects provided written informed consent, as required by our institutional review board, and were aware of the aim of the study and of the study procedure. The subjects had sustained an RC tear of less than 4 cm in greatest dimension, as identified during surgery. An open apical surgical approach was used in all subjects; the deltoid tendon was split longitudinally, and primary closure of the supraspinatus and infraspinatus tendons was achieved by using nonabsorbable suture material. The greater tuberosity of the humeral head was shaved with an osteotome to promote adherence of the rotator cuff tendons and to facilitate healing.
All subjects underwent thorough functional evaluation of the affected shoulder by an orthopedic surgeon on the same day as the MR imaging examination. The subjects were clinically assessed with the Constant score (10), a well-established clinical tool that combines both the subjective parameters of pain and performance of activities of daily living and the objective parameters of active range of motion and power. A 100-point score is used, with the subjective parameters accounting for 35 points and the objective parameters accounting for 65 points. A score of 80–100 points constitutes excellent results; a score of 65–79, good results; a score of 51–64, fair results; and a score of 0–50, poor results. All clinical scores in this subject cohort were good to excellent (70–80 points in four subjects, 81–90 points in five subjects, and >90 points in six subjects), with a range of 72–98 points and a mean of 86 points.
MR imaging was performed with a 1.5-T system (Horizon Echo-Speed; GE Medical Systems, Milwaukee, Wis) with a dedicated shoulder surface coil. All subjects underwent the following MR imaging protocol: coronal oblique T1-weighted spin-echo (600/14 [repetition time msec/echo time msec]); coronal oblique intermediate-weighted fast spin-echo (2,000/32 [repetition time msec/effective echo time msec]); coronal oblique T2-weighted fast spin-echo, both with and without frequency-selective presaturation of fat (4,000/85 [effective]); and sagittal oblique T2-weighted fast spin-echo (4,000/80 [effective]) sequences. The imaging parameters included 4-mm section thickness, 1-mm intersection gap, 15–16-cm field of view, and 256 x 192 matrix.
The images were interpreted independently by two radiologists (B.B.F., D.L.J.) experienced in musculoskeletal MR imaging, with differences resolved by means of consensus. The readers were aware that all subjects were asymptomatic but were blinded as to each subject's Constant score. The parameters assessed included tendon signal intensity, integrity of the RC and biceps tendons, presence of joint effusion, amount of subacromial-subdeltoid bursal fluid, continuity of the subacromial-subdeltoid fat, and other ancillary findings such as presence of bone marrow edema, subchondral cysts, and acromioclavicular osteophytes.
RC tendons were assessed for signal intensity abnormalities and integrity as follows (11,12): Normal tendon signal intensity was assigned if the tendon demonstrated normal thickness and low signal intensity on images obtained with all MR sequences. Type 1 tendon signal intensity was assigned if intermediate-weighted and T2-weighted images showed diffuse, mildly increased signal intensity (not equal to that of fluid) and an intact tendon. Type 2 tendon signal intensity was assigned if T2-weighted images showed that the tendon was intact but that focally increased signal intensity (equal to that of fluid) was present along the bursal or articular margin. Type 3 tendon signal intensity was assigned if T2-weighted images showed focally increased signal intensity (equal to that of fluid) that involved the entire thickness of the tendon, with or without tendon retraction.
Glenohumeral joint effusions were categorized as absent, small, moderate, or large, by assessing the amount of fluid in the subcoracoid and axillary recesses and the biceps tendon sheath, as previously described by Schweitzer et al (13). Subacromial-subdeltoid bursal fluid was considered to be pathologic if it measured more than 2 mm in thickness. Metallic artifact also was assessed on oblique coronal T2-weighted fast spin-echo images obtained without fat saturation and was rated as absent, mild with involvement of the subcutaneous tissues alone, moderate with involvement of the capsular structures, or severe if the artifact precluded assessment of the RC.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
The mean age of the subjects with type 2 tendon signal intensity was 62 years, and that of the subjects with type 3 tendon signal intensity was 66 years, which, given the small sample size, were not markedly different from the mean age (60 years) of the entire study population. With respect to tendon signal intensity changes, there was complete agreement between the two observers in 12 of 15 subjects; the discrepancies occurred in the differentiation of type 1 from type 2 tendon signal intensity.
Subacromial-subdeltoid bursal fluid was noted in 10 (67%) of 15 subjects, and subacromial-subdeltoid fat was incomplete or absent in all subjects. Small joint effusions were noted in nine (60%) subjects, moderate effusions were noted in five (33%), and mild bone marrow edema was noted in six (40%). Other ancillary findings included subchondral cysts in nine (60%) subjects and acromioclavicular osteophytes in 11 (73%). Incidentally noted was a seroma in a supraspinatus muscle. Metallic artifact was absent in one subject, mild in seven, and moderate in seven. In no subject did metallic artifact preclude assessment of the supraspinatus or infraspinatus tendons. In one subject, the subscapularis tendon was not fully demonstrated due to magnetic susceptibility artifact.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The MR imaging assessment of RC repair is important because recurrent shoulder pain of various causes is common in such patients (7,8,14) and other diagnostic modalities such as shoulder arthrography are more invasive and may be difficult to interpret in the patient who has undergone surgery (15). Prior to the evolution of MR imaging of the shoulder, ultrasonography (US) was shown to be a useful, albeit operator-dependent, modality for the preoperative assessment RC tendon pathologic conditions (16). In the patient who has undergone surgery, the bone and soft-tissue US landmarks around the tendon are distorted or absent and the tendons are hyperechoic, which leads to further difficulty in the diagnosis of a recurrent RC tear (17–19).
To our knowledge, this is the first study to evaluate the MR imaging appearance of the successfully repaired RC in an asymptomatic population. All subjects in our study had demonstrated an excellent result at routine clinical follow-up after RC repair, continued to be asymptomatic when approached for inclusion in the study, and had a good to excellent score on the Constant scale, a well-accepted, sensitive, and accurately reproducible clinical tool (20). On the basis of conventional MR imaging criteria, only 10% of the supraspinatus and infraspinatus tendons were normal. Mildly increased signal intensity (type 1) was evident in 53% of the tendons. Without surgical and/or pathologic correlation, the relevance of this finding is uncertain, but it could indicate ongoing inflammation or scarring. Such signal intensity changes may represent part of the normal spectrum of the postoperative appearances of a tendon.
Type 2 and type 3 tendon signal intensity changes were seen in more than one-third of the supraspinatus and infraspinatus tendons in our study. It is presumed, on the basis of preoperative MR imaging criteria (11,12), that type 2 and type 3 signal intensity changes represented partial and full-thickness tendon tears, respectively; however, without surgical correlation, which is unavailable in this asymptomatic population, this cannot be proved. If we assume that a postoperative change in tendon signal intensity represents the same pathologic condition as a preoperative change in tendon signal intensity, four subjects had a complete tear of the supraspinatus or infraspinatus tendon, and one subject had complete tears of both tendons. Similar to results in a previous study (21), there was complete agreement in our study between two observers with regard to type 3 tendon signal intensity (full-thickness tear) and some interobserver variation with regard to type 1 and 2 signal intensity, although our sample size was too small for formal statistical analysis.
Although the sensitivity and specificity for the MR imaging diagnosis of a partial RC tear are lower than those for detection of a complete tear in subjects who have not undergone surgery, we used a fast spin-echo T2-weighted fat-saturated MR sequence, which has been shown (1) to have a higher sensitivity, to demonstrate type 2 signal intensity (partial tear) in five (17%) of 30 tendons. Overall, the finding of 11 (37%) of 30 tendons with partial or complete tear is comparable to results of a study by Sher et al (22) with an asymptomatic group that did not undergo surgery. They found a 34% overall prevalence of RC tear (15% complete tear, 20% partial tear) and, in individuals over 60 years of age, a 54% prevalence. Our findings are relevant for the postoperative assessment of the symptomatic RC in that they emphasize that correlation of clinical results with imaging findings is essential to ensure that the identified RC abnormality does, in fact, account for the ongoing symptoms.
Although controversial, substantial subacromial-subdeltoid bursal fluid and focal obliteration of subacromial-subdeltoid fat have been cited (23) as useful markers for RC tears in the population that has not undergone surgery. Subacromial-subdeltoid bursal fluid was a common postoperative finding in our study and did not correlate with full- or partial-thickness tears. It is not surprising that the subacromial-subdeltoid fat was universally abnormal in this group of subjects, because this area is traversed during surgery, and the fat often is resected. Therefore, the amount of subacromial-subdeltoid bursal fluid and the integrity of subacromial-subdeltoid fat cannot be used as ancillary postoperative findings of RC tear.
Small to moderate joint effusions were commonly found in our study group, and, of interest, mild bone marrow edema was found in 43% of subjects despite the passage of 1–5 years since surgery. Metallic artifact, as assessed on fast spin-echo T2-weighted images obtained without fat saturation, did not markedly interfere with assessment of the RC.
Our study was limited by the small number of subjects, and further MR imaging investigation of a larger asymptomatic group that has undergone surgery would be warranted to confirm our results. As previously mentioned, inherent in a study of asymptomatic individuals is the inability to obtain surgical and/or pathologic confirmation of the findings. A robust and clinically accepted method of functional assessment of the shoulder was performed by the same orthopedic surgeon on the same day as MR imaging, but any clinical evaluation is at least partly subjective. We did not perform MR arthrography, but this test may be unreliable after RC repair for similar reasons as those for radiographic arthrography: Contrast material may leak through a well-repaired but incompletely healed tendon or, owing to adherent scar tissue, may fail to leak into the subacromial-subdeltoid bursa through a tendon that has torn again (8).
On the basis of our preliminary results with standard postoperative MR imaging criteria, we conclude that in individuals with a good to excellent clinical result after RC repair, postoperative signal intensity changes consistent with tendonitis or tendinosis are common and that clinically "silent" partial and complete RC tears may be seen. Joint effusions, subacromial-subdeltoid bursal fluid, and focal obliteration of the subacromial-subdeltoid fat are common postoperative findings and are not reliable ancillary findings of an acute RC abnormality. Extrapolation of the accepted criteria for diagnosis of preoperative RC abnormalities to the patient that has undergone surgery must, therefore, be performed with care; in the symptomatic patient who has undergone surgery, the MR imaging findings must be interpreted with meticulous clinical correlation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Author contributions: Guarantors of integrity of entire study, B.B.F., A.L.S.; study concepts and design, B.B.F., A.L.S., R.H.H.; definition of intellectual content, B.B.F., A.L.S.; literature research, A.L.S., P.K.; clinical studies, R.H.H., P.K.; data acquisition, all authors; data analysis, B.B.F., A.L.S., D.L.J.; manuscript preparation, B.B.F., A.L.S.; manuscript editing and review, all authors.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. J. Probyn, L. M. White, D. C. Salonen, G. Tomlinson, and E. L. Boynton Recurrent Symptoms after Shoulder Instability Repair: Direct MR Arthrographic Assessment Correlation with Second-Look Surgical Evaluation Radiology, December 1, 2007; 245(3): 814 - 823. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Duc, B. Mengiardi, C. W. A. Pfirrmann, B. Jost, J. Hodler, and M. Zanetti Diagnostic Performance of MR Arthrography After Rotator Cuff Repair Am. J. Roentgenol., January 1, 2006; 186(1): 237 - 241. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Boileau, N. Brassart, D. J. Watkinson, M. Carles, A. M. Hatzidakis, and S. G. Krishnan Arthroscopic Repair of Full-Thickness Tears of the Supraspinatus: Does the Tendon Really Heal? J. Bone Joint Surg. Am., June 1, 2005; 87(6): 1229 - 1240. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Mengiardi, C. W. A. Pfirrmann, C. Gerber, J. Hodler, and M. Zanetti Frozen Shoulder: MR Arthrographic Findings Radiology, November 1, 2004; 233(2): 486 - 492. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. V. R. Mohana-Borges, C. B. Chung, and D. Resnick MR Imaging and MR Arthrography of the Postoperative Shoulder: Spectrum of Normal and Abnormal Findings RadioGraphics, January 1, 2004; 24(1): 69 - 85. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
