| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology and Nuclear Medicine, Uniformed Services University of
the Health Sciences, 4301 Jones Bridge Rd., Bethesda, MD 20814-4799.
2 Department of Radiology, Mayo Clinic, 200 First St. S.W., Rochester, MN
55905.
3 Department of Radiology, Wilford Hall Medical Center, 59 MDW/MTRD, 2200
Bergquist Dr., Ste. #1, Lackland AFB, TX 78236.
4 Section of Biostatistics, Mayo Clinic, Rochester, MN 55905.
Received May 21, 2002;
accepted after revision August 27, 2002.
Address correspondence to D. P. Beall at 3711 Medical Dr., #611, San
Antonio, TX 78229.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. One hundred eleven patients underwent MR imaging for shoulder pain followed by arthroscopic or open shoulder surgery from January 1997 to December 2000. Patients were identified by a retrospective search, and all consecutive patients having undergone both MR imaging and surgery were included in the patient cohort. Official MR imaging interpretations were compared with operative reports, and all findings were recorded.
RESULTS. Twenty-three patients were identified with partial- or full-thickness tears of the long head of the biceps tendon. The sensitivity, specificity, and accuracy of unenhanced MR imaging of the shoulder for detecting these bicipital tears were 52%, 86%, and 79%, respectively. When a tear was present in the biceps tendon, the prevalence of supraspinatous, infraspinatus, and subscapularis tendon tears was 96.2%, 34.6%, and 47.1%, respectively. Patients with biceps tendon tears were significantly more likely to also have subscapularis tendon tears (p < 0.0001) and supraspinatous tendon tears (p < 0.008) than those patients who did not have biceps tendon tears. No significant relationship was found between the presence or absence of a biceps tendon tear and the presence or absence of a infraspinatus or teres minor tendon tear (p = 0.17).
CONCLUSION. Tears of the long head of the biceps tendon have a statistically significant association with tears of the anterior and superior rotator cuff and are highly correlated with tears of the supraspinatous and subscapularis tendons. When tears of these tendons are detected, specific attention directed toward the long biceps tendon is warranted to characterize the status of this structure that provides additional stability to the shoulder joint.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The biceps tendon functions to stabilize the anterosuperior portion of the rotator cuff by its position in the bicipital grove anteriorly and its attachment to the glenoid labrum superiorly. Tears of the rotator cuff, specifically, the subscapularis tendon, have been associated with medial subluxations or dislocations of the biceps tendon as well as with biceps tendinopathy [1, 2]. The relationship of tears of the subscapularis tendon with medial dislocation of the biceps tendon is well known, but the relationship between tears of the anterior and superior portions of the rotator cuff and tears of the biceps tendon is less well established and has largely been described as massive tears (involving two or more tendons or a tear that has a maximal diameter > 5 cm) of the anterosuperior rotator cuff [3].
Tears of the subscapularis are unusual (accounting for 2.0-3.5% of rotator cuff tears), are difficult to recognize on routine shoulder MR imaging, and may require a separate surgical approach for adequate repair [2]. Tears of the long head of the biceps tendons are also less optimally visualized by MR imaging than tears of the superior rotator cuff and have a direct effect on the operative approach to rotator cuff repair [4].
Establishing a relationship between the prevalence of biceps tendon tears in association with tears of the anterosuperior rotator cuff and characterizing the various patterns of injury may assist in improving the prospective accuracy of the MR imaging interpretation. We present data from 23 patients with surgically proven tears of the biceps tendons taken from 111 consecutive patients who underwent both MR imaging and surgery.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
Stability of the biceps tendon is primarily provided by the coracohumeral ligament that stabilizes the tendon in the intertubercular groove. The coracohumeral ligament strengthens the interval between the subscapularis and supraspinatus tendons, serving as the primary obstacle to dislocation of the biceps tendon in this region [5]. Cadaveric studies have shown that the main restraint to dislocation of the biceps tendon is the medial portion of the coracohumeral ligament near its insertion on the lesser tuberosity. The transverse humeral ligament has been found to be too weak to prevent medial dislocation of the biceps tendon [5].
Patient Data
From January 1997 to December 2000, 126 patients underwent MR imaging for
shoulder pain. These individuals were identified by a retrospective
investigation, and 111 consecutive patients who underwent both MR imaging and
shoulder surgery were included in the patient cohort. The 111 patients
included 70 men and 41 women ranging in age from 13 to 82 years (mean age,
50.33 years). The MR imaging included both routine shoulder examinations
(n = 103 patients) and MR arthrography of the shoulder (n =
8 patients). Study participants who underwent MR examination were screened for
the presence of intracranial surgical clips, cardiac pacemakers, or metallic
objects in the orbit. IV contrast agent was not administered. Intervals from
the time the patients underwent MR imaging of their shoulders to the time of
shoulder surgery ranged from 1 day to 679 days, with an average interval of
110 days (median, 47 days). By definition in this retrospective examination,
the decision to obtain MR imaging of the shoulder was at the discretion of the
ordering clinician. The subsequent shoulder surgery was performed by the
orthopedic surgeon caring for the patient, and the operative procedures after
MR imaging included both arthroscopic and open surgery. The decision as to the
appropriate surgical approach was made by the treating surgeon with full
consent of the patient. Institutional review board approval was granted for
all aspects of this study.
Imaging Techniques, Surgical Techniques, and Interpretation
Imaging studies consisted exclusively of MR imaging, and all studies were
performed on 1.5-T MR imaging units (two 1.5-T Signa MR/I Echospeed Plus units
and two 1.5-T Signa MRI units; General Electric Medical Systems, Milwaukee,
WI) with a dedicated shoulder coil (Mayo Clinic, Rochester, MN). Although
various shoulder MR imaging protocols were used, the following parameters were
used most often in our examination: coronal and sagittal oblique double-echo
spin-echo sequences with fat saturation (field of view, 16 cm; TR/first-echo
TE, second-echo TE, 2000/20, 60; number of excitations, 1; slice thickness, 4
mm; gap, 1 mm; matrix, 256 x 192), axial and coronal oblique fast
spin-echo sequences with fat saturation (field of view, 14 cm; TR/TE, 3500/45;
number of excitations, 2; echo-train length, 8; slice thickness, 4 mm; gap,
0.5 mm; matrix, 256 x 256), an axial gradient-echo sequence (field of
view, 16 cm; 250/12; flip angle, 30°; number of excitations, 3; slice
thickness, 3 mm; gap, 0 mm; matrix, 256 x 192). A frequency-selective
fat saturation pulse was used to reduce the high signal from surrounding
fat.
The following imaging protocol was used for the eight patients who underwent MR arthrography: axial, coronal oblique, and sagittal oblique T1-weighted spin-echo sequences (field of view, 14 cm; 700/10; number of excitations, 1; slice thickness, 4 mm; gap, 0.5 mm; matrix, 256 x 192), a coronal oblique fast spin-echo T2-weighted sequence (field of view, 14 cm; 3500/45; number of excitations, 2; echo-train length, 8; slice thickness, 4 mm; gap, 0.5 mm; matrix, 256 x 256), a coronal oblique spin-echo non-fat-saturated T1-weighted sequence (field of view, 14 cm; 700/10; number of excitations, 1; slice thickness, 4 mm; gap, 0.5 mm; matrix, 256 x 192). A frequency-selective fat saturation pulse was applied to all sequences (except as indicated previously) to reduce the high signal from surrounding fat.
Shoulder surgery included both arthroscopy (n = 81 patients) and open surgical techniques (n = 30 patients). Arthroscopy included analysis of the entire glenohumeral joint along with arthroscopic subacromial decompression if needed. In general, patients with signs of early impingement or partial tears were treated arthroscopically, whereas patients with full-thickness tears were treated with an open procedure using a deltoid-splitting approach. Patients with fraying or attenuation of the biceps tendon seen at surgery were treated by resecting the biceps tendon at its attachment on the superior glenoid and attaching the tendon to the bicipital groove, using a tenodesis procedure.
Arthrography was performed initially in the fluoroscopy suite with the patient supine and the arm in external rotation. After we obtained informed consent and using sterile technique, a 22-gauge needle was inserted into the anterior and inferior portion of the glenohumeral joint using fluoroscopic guidance. An intraarticular location of the needle tip was confirmed by injecting 1 mL of iodinated contrast (Omnipaque 300 [iohexol]; Nycomed, Aukland, NZ). Subsequently, 11-14 mL of a dilute gadopentatate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) and saline mixture (0.1 mL in 20 mL of sterile normal saline solution, a 1:200 ratio) was injected into the glenohumeral joint, and MR imaging of the shoulder was completed within 45 min of the intraarticular instillation of the contrast solution.
The official MR imaging interpretations were examined, and the findings recorded in a database. Retrospective reviews were not obtained, and the initial MR interpretation, on which the surgery was based, was the imaging evaluation report used to compare with the operative findings. Radiologists with a wide range of musculoskeletal expertise performed the MR imaging interpretations. Whereas all radiologists were fellowship-trained, two thirds of the examinations were interpreted by subspecialists in musculoskeletal radiology. The MR imaging findings were compared with the operative record, and all elements of information were documented and inserted into a database. The anatomic regions routinely evaluated on MR imaging of the shoulder included rotator cuff, glenoid labrum, glenohumeral ligaments, osseous structures, biceps tendon, musculature, soft tissues, joint capsule, glenohumeral joint, acromioclavicular joint, and other miscellaneous regions.
Rotator cuff tears were classified as either partial-thickness or full-thickness tears. The classification of full-thickness tears included those rotator cuff tears in which a T2 signal abnormality (seen either on the conventional spin-echo T2-weighted images or on the fast spin-echo T2-weighted images) that was as bright as fluid extended all the way across a cuff tendon. A region of increased T1 signal seen extending across the tendon was also considered to represent a full-thickness tear on MR arthrograms. The category of full-thickness tears included complete rotator cuff tears that had an associated tendon retraction in addition to their full-thickness tendon defect. Partial-thickness tears included tears involving either the articular and bursal surfaces. A tear was classified as partial thickness if a T2 signal abnormality as bright as fluid extended across a portion of the tendon but not through its entire thickness. Intratendinous tears were also identified as being present if a fluidlike signal abnormality that did not extend to either the articular or bursal surfaces was present in the tendon.
Biceps tendon tears were classified separately from biceps tenosynovitis, which was identified as fluid in the bicipital synovial sheath that did not extend into the biceps tendon itself. The diagnosis of biceps tenosynovitis was made if the inflammatory change and fluid were found exclusively around the tendon or were present around the tendon clearly out of proportion to the amount of fluid in the glenohumeral joint. Tears were also considered separately from biceps tendinopathy, which was categorized as a region of increased T2 signal in the substance of the biceps tendon (often associated with enlargement of the tendon) that was not as intense as fluid on conventional spin-echo T2-weighted images or on fast spin-echo T2-weighted images. The MR imaging depiction of biceps tendon tears was made when T2 signal abnormalities as bright as fluid were seen in the biceps tendon on either the conventional spin-echo T2-weighted images or the fast spin-echo T2-weighted images or if gadolinium was seen in the substance of the tendon on MR arthrography. If a portion of the tendon was involved, the tear was classified as partial, and if the entire thickness of the tendon was involved, it was classified as a full-thickness tear. A full-thickness tear was identified if the biceps tendon was noted to be absent in the bicipital groove or near its attachment to the supraglenoid tubercle. The diagnosis of a partial-thickness tear was also rendered when a sudden change in the cross-sectional diameter was noted, especially on the axial images.
Statistical Analysis
The sensitivity, specificity, accuracy, positive predictive value, and
negative predictive value of MR imaging for detecting biceps tendon tears were
calculated as well as these same values for the MR imaging of the rotator
cuff. The values for prevalence and positive predictive value were calculated
for biceps tendon injuries in association with tears of the supraspinatous,
infraspinatus, subscapularis, and teres minor tendons. The values for the
reverse scenario were also calculated (rotator cuff tears associated with
tears of the long head of the biceps tendon). Additionally, the statistical
significance (using the chi-square test) of the association of biceps tendon
tears with rotator cuff tears was calculated for each tendon of the rotator
cuff.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When a surgically confirmed tear was present in the biceps tendon, the prevalence of supraspinatous, infraspinatus, and subscapularis tendon tears was 96.2%, 34.6%, and 47.1%, respectively. Biceps tendon tears were not associated with tears of the teres minor tendon in our patient population. In patients with biceps tendon tears, the positive predictive values for a tear of the rotator cuff tendons were 95.8% for a tear of the supraspinatous tendon, 80% for a tear of the infraspinatus tendon, and 61% for a tear of the subscapularis tendon.
Patients with biceps tendon tears were also significantly more likely to have subscapularis tendon tears (p < 0.0001) and supraspinatous tendon tears (p < 0.008) than those patients who did not have a biceps tendon tear. No significant relationship was found between the presence or absence of a biceps tendon tear and the presence or absence of infraspinatus or teres minor tendon tears (p = 0.17).
When surgically confirmed rotator cuff tears were present, the positive predictive values for a tear of the long head of the biceps tendon were 57.9%, 34.5%, and 50% for tears of the supraspinatous, infraspinatus, and subscapularis, respectively. As expected, we found that patients with rotator cuff tears had a higher prevalence of biceps tendon tears than those patients without rotator cuff tears (p = 0.04). The prevalence of biceps tendon tears in patients with infraspinatus tears was 50%, followed by that of subscapularis tears (28%) and supraspinatous tears (22%).
Despite the higher prevalence of associated rotator cuff tears, the patients with tears of the long head of the biceps tendons did not have more prompt surgical intervention than those patients with either normal tendons or tendonopathy. Patients with biceps tendon tears had a mean interval of 140 days from their shoulder MR imaging to surgery. This interval compared with 102 days for those patients without biceps tendon tears; the difference between the two groups was not significant (p = 0.35).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
Many reports of MR imaging of the abnormal biceps tendon describe either bicipital tendonitis or the association of biceps tendon dislocation with tears of the subscapularis tendon [7]. Recently, there has been interest in the bicipital labral complex and its associated injury pattern in those patients with superior labral tears. To our knowledge, reports of the accuracy of MR imaging for evaluation of tears of the long head of the biceps tendon are few, and most of the reported data are from studies of MR arthrography and the evaluation of the bicipital labral complex. The long head of the biceps has previously been reported in approximately 7% of patients with supraspinatous tears and is abnormal in approximately one third [8]. Our reported sensitivity, specificity, and accuracy of MR imaging in the assessment for tears of the proximal biceps tendon are 52%, 86%, and 79%, with a negative predictive value of 87%. Although the sensitivity is less than optimal in detecting tears of the biceps tendon, the negative predictive value is adequate for excluding most abnormalities and approaches the negative predictive value of MR imaging of the rotator cuff. These values also represent the initial MR imaging findings on which the treatment decisions were based rather than a retrospective evaluation or a consensus review by a group of radiologists. This method of cataloging imaging findings is far less common than the group interpretation analysis technique but more accurately reflects the actual process of examination interpretation followed by treatment decisions that are partially or largely based on the initial imaging findings.
Our relatively low sensitivity for detecting biceps tendon tears was thought to be multifactorial and at least partly due to the fact that the original interpretations were used rather than a consensus opinion by subspecialist radiologists. Additional factors include the oblique orientation of the biceps tendon across the glenohumeral joint, which makes the tendon somewhat difficult to image, and the fact that most patients with biceps tendon tears also had other tendon tears and were more predisposed to move during the examination because of their shoulder discomfort. Also, unlike the sequences for the rotator cuff, those dedicated strictly to the evaluation of the biceps tendon are seldom used and were not used for this study.
We concluded that patients with partial or complete tears of the biceps tendons were significantly more likely to have associated tears of the supraspinatous and subscapularis tendons. This important distinction is most likely related to the anterosuperior stabilization contribution made by the biceps tendon. The adverse effect that acromiohumeral impingement has on the biceps tendon also affects the superior rotator cuff, so associated abnormalities in these regions should not be surprising.
The association of biceps tendon abnormalities with those of the rotator cuff should be of value not only in preoperative planning but also for interpretation purposes. Whereas supraspinatous tendon tears are accurately depicted on routine MR imaging, subscapularis tendon tears are not [2]. Tears of the subscapularis tendon are difficult to diagnose, and some studies have shown that they are missed most of the time on routine MR imaging [2]. Despite this inaccuracy, the knowledge that subscapularis tendon tears may be confined to the superior portion of the tendon and that biceps tendon tears may be associated with tears of the anterior rotator cuff should increase the interpreter's sensitivity for these lesions (Fig. 3A, 3B, 3C). Additionally, the more easily detected tear of the supraspinatous tendon should heighten the suspicion for a biceps tendon tear because we found the prevalence of biceps tendon tears in those patients with a supraspinatous tear to be 22%. The positive predictive values for the presence of a biceps tendon tear when tears of the supraspinatous and subscapularis tendons were present were also relatively high at 57.9% and 50%, respectively.
|
|
|
A well-described relationship between abnormalities of the rotator cuff and disorders of the long head of the biceps tendon is the association between a subscapularis tendon tear and a medial biceps tendon dislocation (Fig. 4A, 4B, 4C). Although once thought to be related to the rupture of the transverse humeral ligament, dislocation of the biceps tendon is probably due to disruption of the coracohumeral ligament. The intraarticular portion of the biceps tendon is stabilized by the coracohumeral ligament. This ligament strengthens the interval between the subscapularis and supraspinatus tendons, bridges the tuberosities, and serves as the primary obstacle to dislocation of the biceps tendon in this region (Fig. 5A, 5B). Cadaveric studies have shown that the main restraint to dislocation of the biceps tendon is the medial portion of the coracohumeral ligament near its insertion on the lesser tuberosity rather than the transverse humeral ligament [5].
|
|
|
|
|
Although the coracohumeral ligament cannot be reliably identified on every MR imaging examination, this ligament and the structures forming the anterosuperior portion of the shoulder capsule all contribute to normal function and may be injured together [9, 10, 11]. The long head of the biceps, the bicipital labral complex, the superior subscapularis tendon, the anterior supraspinatous tendon, the superior glenohumeral ligament, the coracohumeral ligament, and the transverse ligament are all intimately associated with each other in the region of the rotator cuff interval. Given the close anatomic and functional relationship, we expect that injuries in this region may affect multiple structures.
Despite this association, to our knowledge, there is relatively little literature on the association of tears of the biceps tendon with tears of the anterosuperior rotator cuff. This association is most often described in the scenario of a massive tear of the anterosuperior portion of the rotator cuff, but reports of this type of tear pattern are relatively few [4]. This scarcity is probably partially due to the fact that tears of the subscapularis tendons are relatively unusual, accounting for approximately 2-4% of all tears of the rotator cuff [2, 12, 13, 14]. Also, as mentioned previously, tears of the subscapularis tendon are also difficult to detect. These factors combined with the moderate to low sensitivity of unenhanced MR imaging for detecting biceps tendon tears provide a plausible explanation as to the paucity of literature reporting biceps tendon tears in association with tears of the rotator cuff.
In our patient cohort, both partial- and full-thickness tears of the biceps tendons were more commonly associated with isolated tears of the supraspinatous tendon than with massive tears of the rotator cuff involving the supraspinatous and infraspinatus tendons or tears involving these tendons along with the subscapularis tendon. Partial-thickness tears of the biceps tendon were somewhat more frequently associated with rotator cuff tears than full-thickness tears of the biceps tendon, but this difference was negligible and likely due to the overall number of partial tears (n = 14) compared with the number of full-thickness tears (n = 9). Although massive rotator cuff tears were fewer (n = 30 patients), they had a stronger association with biceps tendon tears when present. Isolated tears of the supraspinatous tendon were more frequent (n = 61 patients) and had a greater overall number of associated biceps tendon tears (Fig. 6A, 6B), but isolated tears were less consistently associated with biceps tendon abnormalities than were massive tears of the rotator cuff.
|
|
Rotator cuff abnormalities in patients without biceps tendon tears showed a distribution similar to previously published data on the distribution of rotator cuff tears [15, 16, 17]. As we would expect, there were proportionally fewer multi-tendon tears and more normal rotator cuffs than in the group of patients with biceps tendon tears. There were also proportionally fewer full-thickness tears (compared with the number of partial-thickness tears) in this patient group.
Limitations of this study include having a patient population of only those individuals who underwent both MR imaging and shoulder surgery. Although all patients included in this category had shoulder pain as part of their clinical presentation, this group likely represents a segment of the population with shoulder symptoms and MR imaging findings severe enough to warrant surgery. Therefore a much higher rate of musculoskeletal shoulder abnormalities, not representative of the population at large, are to be expected in this group. The retrospective nature of this study also limits specific analysis of various anatomic regions and relies only on the recorded spectrum of findings rather than on a prospective evaluation of the regions of interest. Additionally, both open surgery and arthroscopic surgery function as gold standards for this study despite the recognized differences in the accuracy of the two techniques for detecting various types of shoulder abnormalities. Because of these limitations, it would be optimal for our findings to be corroborated by a prospective study evaluating the associations between biceps tendon tears and rotator cuff abnormalities.
In conclusion, in our patient cohort, tears of the long head of the biceps tendon had a statistically significant association with tears of the supraspinatous and subscapularis components of the rotator cuff. Most often an isolated tear of the supraspinatous tendon was present when a tear of the biceps tendon was recognized, but tears involving two or more tendons of the rotator cuff were also frequently identified. The overall pattern of injury in our findings supports the previously described relationship between acromiohumeral impingement and the adverse effects on both the biceps tendon and superior rotator cuff.
Our overall sensitivity of MR imaging for detecting biceps tendon tears was moderate, but the negative predictive value for the biceps tendon evaluation was sufficient to exclude most biceps tendon tears in patients presenting with shoulder pain. These values also reflect the original review rather than a retrospective consensus and are an accurate representation of the imaging information presented to the clinicians before their formulation of the treatment strategy.
Given that the clinical presentations of biceps tendon abnormalities can be quite similar, we believe that imaging will likely continue to play an important role in identifying specific abnormalities associated with biceps tendon injury. When injuries involving the anterosuperior rotator cuff or the long head of the biceps tendon are identified, all other possible associated injuries should be considered, and those structures closely evaluated. This attention could potentially improve the notoriously low sensitivity of MR imaging for detecting subscapularis tendon tears. The reverse of the biceps tendon-rotator cuff relationship may also be true, and more easily identified isolated tears of the supraspinatous tendon should warrant examining the integrity of the long head of the biceps tendon.
The information in this study should contribute to the understanding of the injury pattern seen in the anterosuperior portion of the rotator cuff and will better define the relationship between rotator cuff tears and tears of the long head of the biceps tendon. A recognition of these associations should improve the accuracy of MR imaging interpretations in patients with shoulder pain specifically related to biceps tendon abnormalities.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Namdari, R. F. Henn III, and A. Green Traumatic Anterosuperior Rotator Cuff Tears: The Outcome of Open Surgical Repair J. Bone Joint Surg. Am., September 1, 2008; 90(9): 1906 - 1913. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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 Ostlere and H Marmery Imaging the shoulder Imaging, September 1, 2007; 19(3): 191 - 200. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Hunt, Y. W. Kwon, and J. D. Zuckerman The Rotator Interval: Anatomy, Pathology, and Strategies for Treatment J. Am. Acad. Ortho. Surg., April 1, 2007; 15(4): 218 - 227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C U Schulz, H Anetzberger, and C Glaser Coracoid tip position on frontal radiographs of the shoulder: a predictor of common shoulder pathologies? Br. J. Radiol., November 1, 2005; 78(935): 1005 - 1008. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. P. Krief MRI of the Rotator Interval Capsule Am. J. Roentgenol., May 1, 2005; 184(5): 1490 - 1494. [Full Text] [PDF]
|
|
|
|
|
|
S Ostlere Imaging the shoulder Imaging, December 1, 2003; 15(4): 162 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Robin Goodfellow (42-8) Rheumatology, August 1, 2003; 42(8): 1028 - 1028. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP |
