Published online before print May 5, 2008, 10.1148/radiol.2481071003
(Radiology 2008;248:185-193.)
© RSNA, 2008
Acute Traumatic Posterior Shoulder Dislocation: MR Findings1
Nadja Saupe, MD,
Lawrence M. White, MD,
Robert Bleakney, MD,
Mark E. Schweitzer, MD,
Michael P. Recht, MD,
Bernhard Jost, MD, and
Marco Zanetti, MD
1 From the Department of Medical Imaging, Mount Sinai Hospital and University Health Network, University of Toronto, Toronto, Ontario, Canada (N.S., L.M.W., R.B.); Departments of Orthopedic Surgery (M.P.R.) and Radiology (N.S., M.Z.), Orthopedic University Hospital Balgrist, Forchstrasse 340, CH-8008 Zurich, Switzerland; Department of Radiology, Hospital for Joint Disease, New York University Medical Center, New York, NY (M.E.S.); and Department of Radiology, Cleveland Clinic Foundation, Cleveland, Ohio (B.J.). Received June 11, 2007; revision requested August 20; revision received October 29; accepted December 28; final version accepted February 4, 2008.
Address correspondence to N.S. (e-mail: nadja.saupe{at}balgrist.ch).
 |
ABSTRACT
|
|---|
Purpose: To retrospectively evaluate the appearance of lesions of osseous and soft-tissue structures of the glenohumeral joint on magnetic resonance (MR) images after first-time traumatic posterior shoulder dislocation.
Materials and Methods: The study was institutional review board approved and HIPAA compliant, as appropriate, for the four institutions at which the involved patients were treated. Informed patient consent was obtained, were applicable. Thirty-six male patients (age range, 15–80 years; mean age, 40.2 years) with clinically documented first-time traumatic posterior shoulder dislocation were examined with MR arthrography (18 patients) or conventional shoulder MR imaging (18 patients). Causes of posterior shoulder dislocation were electric shock in one patient, seizure in one patient, and trauma in 34 patients. Hill-Sachs lesions, rotator cuff tears, biceps tendon abnormalities, posterior labrocapsular complex lesions, humeral head translation, and osseous glenoid version angle were evaluated. Spearman rank correlation and Student t test analyses were performed.
Results: In 31 (86%) of the 36 patients, a reverse Hill-Sachs lesion was found. Eleven (31%) patients had a reverse osseous Bankart lesion. Twelve full-thickness rotator cuff tears were seen in seven (19%) patients: four supraspinatus tendon, three infraspinatus tendon, and five subscapularis tendon tears. Six (17%) patients had biceps tendon abnormalities. Posterior labrocapsular complex tears were identified in 21 (58%) patients: 10 (48%) with posterior labral sleeve avulsions and 11 (52%) with reverse Bankart lesions. Twenty-seven (75%) patients had a retroverted scaphoglenoid angle (mean, 4.5°). The mean humeral translation distance relative to the osseous glenoid fossa was –4.8 mm; in 33 (92%) patients, this distance was translated posteriorly.
Conclusion: The MR appearance of traumatic posterior shoulder dislocation was characterized by reverse Hill-Sachs lesions in 86% of patients and posterocaudal labrocapsular lesions in nearly 60% of patients. Full-thickness rotator cuff tears were seen in approximately 20% of patients.
© RSNA, 2008
|
INTRODUCTION
|
|---|
Posterior shoulder dislocation accounts for 1%–4% of all shoulder dislocations (1–4). Acute traumatic posterior shoulder dislocation most commonly results from axial loading of the adducted internally rotated arm, violent muscle contraction, electric shock, or convulsive seizures (5). Acute posttraumatic posterior shoulder dislocation may also cause impaction fractures of the humeral head (6,7). In addition, posterior dislocation of the shoulder may be associated with fractures of the proximal humeral shaft, tuberosities, and glenoid and soft-tissue injuries (8,9). The soft-tissue injuries associated with posterior dislocation of the shoulder, unlike the osseous lesions, are rarely reported (10). Only sporadic case reports have described rotator cuff tears (11) or lesions of the posterior labrocapsular complex (12,13).
To our knowledge, the osseous and soft-tissue lesions seen on magnetic resonance (MR) images after acute traumatic posterior shoulder dislocation have not been characterized in a large series. Thus, the purpose of our study was to retrospectively evaluate the appearance of lesions of osseous and soft-tissue structures of the glenohumeral joint on MR images after first-time traumatic posterior shoulder dislocation.
|
MATERIALS AND METHODS
|
|---|
Patients
The data of patients with first-time acute traumatic posterior dislocation of the glenohumeral joint were included. These patients were from four university facilities: three in North America (University of Toronto, Cleveland Clinic Foundation, and Hospital for Joint Disease, New York University Medical Center) and one in Europe (Orthopedic University Hospital Balgrist). For the North American patients, the study was Health Insurance Portability and Accountability Act compliant and was approved by each responsible institutional review board, with the requirement for informed patient consent waived. The rights of the patients from Orthopedic University Hospital Balgrist were protected by a law that requires that patients be informed that their charts and images might be retrospectively reviewed for scientific purposes and grants them the opportunity to forbid such use of their data. All patients from this institution agreed to allow their data to be used. (The institutional review board waives approval and the requirement for patient consent in such instances.)
Patient data were included on the basis of a review (by authors from each facility [N.S., M.Z., L.M.W., M.E.S., M.P.R.]) of the clinical charts and surgical records of patients with a clinical history of traumatic posterior shoulder dislocation between March 2000 and July 2006. Detailed inclusion criteria were (a) clinically diagnosed first-time traumatic posterior shoulder dislocation, (b) MR arthrography or conventional shoulder MR imaging performed no later than 2 months after the injury, (c) no previous shoulder surgery, and (d) causative traumatic factors. The data of patients with a recurrent (atraumatic) posterior shoulder dislocation or clinically documented multidirectional instability were excluded from evaluation.
Thirty-six male patients (mean age, 40.2 years; age range, 15–80 years) fulfilled the inclusion criteria. The time between traumatic posterior shoulder dislocation and MR imaging ranged between 1 and 30 days. All 36 patients were referred for MR imaging of the injured shoulder. MR arthrography of the shoulder was performed in 18 patients, and conventional nonarthrographic MR examination of the shoulder was performed in the remaining 18 patients. Gadopentetate dimeglumine (Magnevist; Bayer Schering-Pharma, Berlin, Germany) was injected for MR arthrography by five musculoskeletal radiologists (including M.Z. and M.P.R.), who had 2–10 years experience when this study began in March 2000.
The clinical diagnosis of traumatic posterior shoulder dislocation was based on the results of a physical examination performed by orthopedic surgeons who subspecialized in shoulder surgery and on a history of an adequate causative factor. The mechanism of injury was a fall on the outstretched hand during skiing or snowboarding in five patients, a motorbike accident in four, a bicycle accident in seven, a fall on the outstretched arm during the course of routine daily activities in five, a baseball or softball injury in four, and for one patient each a fall on an outstretched arm during soccer, an epileptic seizure, and an electrical injury. For eight patients, the traumatic mechanism of posterior shoulder dislocation was not further characterized in the clinical chart; only the diagnosis given by the orthopedic surgeon was reported. However, there was unequivocal evidence of trauma in the clinical chart.
MR Imaging
All MR examinations were performed by using 1.5-T MR imaging systems (Signa Excite-II, GE Healthcare, Milwaukee, Wis; or Symphony, Siemens Medical Solutions, Erlangen, Germany) and dedicated phased-array receive-only shoulder coils. The MR arthrographic examinations were performed after the injection of approximately 8–12 mL of gadopentetate dimeglumine at a concentration of 2 mmol/L. The intraarticular position of the needle was confirmed in all patients with fluoroscopic guidance by injecting less than 1 mL of iodinated contrast material (iopamidol, Iopamiro 200; Bracco Suisse, Mendrisio, Switzerland).
For all MR arthrographic examinations, the following sequences were performed: T1-weighted spin-echo imaging in the coronal-oblique, transverse, and sagittal-oblique planes with or without fat saturation (repetition time msec/echo time msec, 500–750/20) or three-dimensional T2-weighted true fast imaging with steady-state precession (9.2/3.2, 1.7-mm section thickness, 180 x 157-cm field of view [FOV], 512 x 256 matrix, one signal acquired), coronal-oblique T2-weighted fat-suppressed fast spin-echo imaging (4000–3000/80–50), and coronal-oblique intermediate-weighted fat-suppressed fast spin-echo imaging (2000–2500/20–40). For all spin-echo and fast spin-echo sequences, the FOV ranged between 140 x 140 mm and 160 x 160 mm, the matrix size ranged from 265 x 256 to 512 x 512, the section thickness was 3–4 mm without an intersection gap, and two signals were acquired. The echo train length used for fast spin-echo imaging ranged between seven and 11.
In 18 patients, a conventional MR shoulder examination was performed by using the following sequences: coronal-oblique intermediate-weighted fast spin-echo imaging (1800–2000/22–40), coronal-oblique and sagittal-oblique T2-weighted fast spin-echo imaging with fat saturation (3000–4000/85–100), and transverse T1-weighted (300–500/minimum) or transverse intermediate-weighted fat-saturated (2700/37) imaging. For all conventional MR sequences, the FOV ranged between 120 x 120 mm and 160 x 160 mm, the matrix size ranged between 256 x 160 and 256 x 512, the section thickness was 3–4 mm with no intersection gap, and two signals were acquired. The echo train length used for fast spin-echo imaging ranged between five and 12.
MR Analysis
All MR images were analyzed in consensus by two experienced musculoskeletal radiologists, who had 12 (L.M.W.) and 7 (R.B.) years experience in musculoskeletal radiology. One radiologist (N.S.) with 2 years experience in musculoskeletal MR imaging measured the glenoid version angle and the humeral translation distance in each patient at a picture archiving and communication system workstation (Software Read, version 5.2.1; Image Device, Idstein, Germany). The following MR findings were evaluated:
1. Reverse Hill-Sachs lesion, visualized as any loss of normal convexity in the anteromedial aspect of the humeral head. A reverse Hill-Sachs lesion, when present, was subclassified, on the basis of the humeral impression fracture size, as a small impression defect, which comprised less than 25% of the articular surface of the humeral head; a medium head defect, which comprised between 25% and 50% of the articular surface; or a large defect, which comprised more than 50% of the articular surface (10).
2. Reverse osseous Bankart lesion, defined as an avulsion of a fragment of the posterior osseous glenoid rim.
3. Fractures of the tuberosities and/or proximal part of the humerus.
4. Rotator cuff lesions. Lesions of the supraspinatus (SSP) tendon, infraspinatus (ISP) tendon, subscapularis (SSC) tendon, and teres minor muscle or tendon were evaluated by using a modification of the grading system of Zlatkin et al (14): Grade 0 (normal) indicated a tendon with completely homogeneous low signal intensity on all MR images; grade 1 (degeneration or tendinopathy), a tendon with diffuse increased signal intensity on T1- or intermediate-weighted images; grade 2 (partial-thickness tear), a tendon with a focal area of high signal intensity on T2-weighted images or with contrast material extending into the tendon from the bursal or articular side on MR arthrograms; and grade 3 (full-thickness tear), a tendon with an area of high signal intensity on T2-weighted images or with contrast material extending through the full articular-to-bursal thickness of the tendon on MR arthrograms. Fat infiltration of the musculature of the rotator cuff was graded by using the classification system of Goutallier et al (15). This classification is based on the amount of fat relative to the amount of muscle. Stage 0 indicates no fat infiltration; stage 1, some fat streaks; stage 2, less fat than muscle; stage 3, as much fat as muscle; and stage 4, more fat than muscle. This classification was initially designed for computed tomographic (CT) findings and was subsequently validated for MR findings by Fuchs et al (16). If fat infiltration was different in the superior and inferior parts of the SSC and ISP tendons, the area with the highest stage was used for classification.
5. Tendinopathy, partial- or full-thickness tear, or longitudinal splitting tear of the long head of the biceps (LHB) tendon and/or medial dislocation of the LHB tendon from the bicipital groove of the humerus. Tendinopathy was diagnosed on the basis of the visualization of increased intrasubstance signal intensity of the LHB tendon, in which the signal was isointense or hyperintense compared with the surrounding muscle on T1- and T2-weighted images. Partial-thickness tear was defined as an LHB tendon of abnormal caliber (thickened or narrowed with an abrupt change in caliber). Full-thickness tear was diagnosed when the tendon was absent or completely disrupted in the subacromial portion and/or descending portion of the bicipital groove (17). The biceps tendon was judged to be dislocated when there was a total loss of contact between the tendon and its normal position within the humeral bicipital groove (18).
6. Injuries of the posterior labrum, including reverse Bankart tears and posterior labrocapsular sleeve avulsions (POLPSAs). For evaluation and localization of posterior labral injuries, the labrum was subdivided into posterosuperior and posteroinferior quadrants. A posterior labral tear was diagnosed when linear intralabral signal intensity changes extending to the labral articular surface were present or when the labrum was detached, displaced, or fragmented. A posterior labral tear was diagnosed at MR arthrography when focal extension of intraarticular contrast material into the fibrous portion of the labrum was seen. A POLPSA lesion was diagnosed on the basis of the presence of a displaced posterior labrum or labral fragments continuous with an intact, partially detached sleeve of adjacent glenoid periosteum. A reverse Bankart lesion was diagnosed on the basis of the visualization of a posterior inferior labral tear associated with a disrupted posterior labral-glenoid periosteal or capsular attachment (19).
7. Glenoid version angle, which was evaluated on axial MR images at the midglenoid level (Fig 1). A reference line was drawn tangent to the anterior and posterior regions of the glenoid rim to represent the plane of the glenoid fossa (line a). A second reference line was drawn through the middle of the glenoid and medial rims of the scapular blade to represent the plane of the scapular body (line b). A third line (line c) was drawn at the level of the glenoid articular surface, perpendicular to line b. The angle formed between lines a and c represented the glenoid version. Positive angles represented retroversion, and negative angles represented anteversion (20).
View larger version (168K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1: Transverse T1-weighted spin-echo MR arthrogram (500/20; FOV, 140 x 140 mm) obtained in 46-year-old man with a retroverted glenoid angle shows glenoid version angle calculated at midglenoid level. A reference line (line a) is drawn through anterior and posterior portions of the glenoid rim to represent the plane of the glenoid fossa. A second reference line (line b) is drawn through the middle of the glenoid and medial rims of the scapular blade to represent the plane of the scapular body. A third line (line c) is drawn perpendicular to line b at the level of the glenoid articular surface. The angle between lines a and c represents the glenoid version. Positive-value angles represent retroversion, and negative-value angles represent anteversion.
|
|
8. Humeral position relative to the osseous glenoid fossa. The humeral position at the midglenoid level on axial MR images was calculated (Fig 2). A short line drawn to connect the anterior and posterior tips of the osseous glenoid fossa represented the length of the glenoid fossa. A second, longer line was drawn tangent to the ventral or costal surface of the scapular body (scapular body line), bisecting the glenoid line segment. The distance between this scapular body line, which extended through the humeral head, and the center of the humeral head represented the humeral translation distance. When the center of the humeral head was posterior to the scapular body line segment, the humeral position was recorded as a negative value, and when it was anterior to the scapular body line segment, the humeral position was recorded as a positive value (Fig 2) (12).
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2: Transverse T1-weighted MR arthrogram (500/20; FOV, 140 x 140 mm) obtained in 46-year-old man with negative humeral translation shows the humeral position relative to the osseous glenoid fossa. A short line drawn to connect the anterior and posterior tips of the osseous glenoid fossa (GL) represents the length of the glenoid fossa. A second, longer line is drawn tangentially to the ventral or costal surface of the scapular body and bisects the glenoid line segment. The distance between the scapular body line (SBL), which extends through the humeral head, and the center of the humeral head (+) is the humeral translation distance. If the center of the humeral head is posterior to the scapular body line segment, the humeral position is recorded as a negative value. If the center of the humeral head is anterior to this line segment, the humeral position is recorded as a positive value.
|
|
Statistical Analyses
To analyze the association between age and MR morphologic findings, Spearman rank correlation coefficients were calculated. To analyze the influence of intraarticular contrast material on the humeral translation, the Student t test was performed. This test was performed because Tung and Hou (12) reported a possible effect of arthrography on the physiologic humeral position. P < .05 was considered to indicate a significant difference. For all statistical calculations, SPSS, version 11.0, software (SPSS, Chicago, Ill) was used.
|
RESULTS
|
|---|
MR Findings
A reverse Hill-Sachs lesion (Fig 3) was identified in 31 (86%) of the 36 patients. In 21 (58%) patients, less than 25% of the circumference of the articular surface of the humeral head was involved; in six (17%), 25%–50% of the circumference was involved; and in four (11%), more than 50% of the circumference was involved (Fig 4).
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3: Histogram shows frequency of MR findings in 36 patients with acute traumatic posterior shoulder dislocation. SSP, ISP, and SSC rotator cuff tears were full-thickness tears.
|
|
View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a: Transverse (a, c) T1-weighted (500/20) and (b, d) intermediate-weighted (2700/37) MR arthrograms show reverse Hill-Sachs lesions. (a) Reverse Hill-Sachs defect (arrowheads) in 48-year-old man involves less than 25% of articular surface of the humeral head. Bone marrow signal intensity changes (ie, bone bruise [arrows]) caused by the acute traumatic event are also seen. (b) Medium-size defect (arrowheads) involving 25%–50% of articular surface of the humeral head in 60-year-old man. (c) Large defect (arrowheads) involving more than 50% of articular surface of the humeral head in 74-year-old man. (d) Medium-size acute, irreducible "locked" dislocation with reverse Hill-Sachs lesion (arrowheads) in 58-year-old man. Bone bruise (arrows) caused by the acute traumatic event is also seen.
|
|
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b: Transverse (a, c) T1-weighted (500/20) and (b, d) intermediate-weighted (2700/37) MR arthrograms show reverse Hill-Sachs lesions. (a) Reverse Hill-Sachs defect (arrowheads) in 48-year-old man involves less than 25% of articular surface of the humeral head. Bone marrow signal intensity changes (ie, bone bruise [arrows]) caused by the acute traumatic event are also seen. (b) Medium-size defect (arrowheads) involving 25%–50% of articular surface of the humeral head in 60-year-old man. (c) Large defect (arrowheads) involving more than 50% of articular surface of the humeral head in 74-year-old man. (d) Medium-size acute, irreducible "locked" dislocation with reverse Hill-Sachs lesion (arrowheads) in 58-year-old man. Bone bruise (arrows) caused by the acute traumatic event is also seen.
|
|
View larger version (169K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c: Transverse (a, c) T1-weighted (500/20) and (b, d) intermediate-weighted (2700/37) MR arthrograms show reverse Hill-Sachs lesions. (a) Reverse Hill-Sachs defect (arrowheads) in 48-year-old man involves less than 25% of articular surface of the humeral head. Bone marrow signal intensity changes (ie, bone bruise [arrows]) caused by the acute traumatic event are also seen. (b) Medium-size defect (arrowheads) involving 25%–50% of articular surface of the humeral head in 60-year-old man. (c) Large defect (arrowheads) involving more than 50% of articular surface of the humeral head in 74-year-old man. (d) Medium-size acute, irreducible "locked" dislocation with reverse Hill-Sachs lesion (arrowheads) in 58-year-old man. Bone bruise (arrows) caused by the acute traumatic event is also seen.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4d: Transverse (a, c) T1-weighted (500/20) and (b, d) intermediate-weighted (2700/37) MR arthrograms show reverse Hill-Sachs lesions. (a) Reverse Hill-Sachs defect (arrowheads) in 48-year-old man involves less than 25% of articular surface of the humeral head. Bone marrow signal intensity changes (ie, bone bruise [arrows]) caused by the acute traumatic event are also seen. (b) Medium-size defect (arrowheads) involving 25%–50% of articular surface of the humeral head in 60-year-old man. (c) Large defect (arrowheads) involving more than 50% of articular surface of the humeral head in 74-year-old man. (d) Medium-size acute, irreducible "locked" dislocation with reverse Hill-Sachs lesion (arrowheads) in 58-year-old man. Bone bruise (arrows) caused by the acute traumatic event is also seen.
|
|
In three patients (8%), fractures of the tuberosities and/or proximal part of the humerus were present. One (3%) patient had a subcapital humerus fracture, and two (6%) had an avulsion fracture of the lesser tuberosity.
In terms of reverse osseous Bankart lesions, an avulsion of a fragment of the posterior osseous glenoid rim was identified in 11 (31%) patients (Fig 5).
View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5: Consecutive transverse T1-weighted MR arthrogram sections (500/20; FOV, 140 x 140 mm) obtained in 39-year-old man show a reverse osseous Bankart lesion with an irregular margin of the posterior glenoid rim and a bone fragment (arrowheads).
|
|
Twelve full-thickness rotator cuff tears—four SSP, three ISP, and five SSC tendon tears—were identified in seven (19%) patients. A total of 15 (42%) patients in our study group had a rotator cuff lesion (partial- or full-thickness tear) (Table 1, Fig 3). An additional 11 (31%) patients had at least tendinopathy of one or more tendons. Fat infiltration of the rotator cuff musculature was mild. No grade 3 or grade 4 fat infiltration of the musculature was found in the SSP, ISP, or SSC tendons of any patient. Grade 4 fat infiltration into the teres minor muscle was seen in two (6%) patients. Regarding the SSP tendon, 30 (83%) patients had grade 0 fat infiltration and each of three (8%) patients had grade 1 and grade 2 infiltrations. Regarding the ISP tendon, 31 (86%) patients had grade 0 fat infiltration, four (11%) had grade 1 infiltration, and one (3%) had grade 2 infiltration. Regarding the SSC tendon, 31 (86%) patients had grade 0 fat infiltration, two (6%) had grade 1 infiltration, and three (8%) had grade 2 infiltration. Regarding the teres minor muscle, 32 (89%) patients had grade 0 fat infiltration; one (3%) patient, grade 1 infiltration; and no patients, grade 2 or 3 infiltration.
In terms of biceps tendon abnormalities, tendinopathy of the LHB tendon was identified in six (17%) patients. No full-thickness tears were observed. A longitudinal splitting of the LHB tendon was identified in one (3%) patient. Five (14%) patients had a medial dislocation of the tendon from the bicipital groove of the humerus (Fig 6). Posterior dislocation of the long biceps tendon was not seen in any patient.
View larger version (123K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6: Transverse T2-weighted three-dimensional true fast imaging with steady-state precession MR arthrogram (9.2/3.2; FOV, 180 x 157 mm) obtained in 58-year-old man shows partial-thickness tear of SSC tendon (arrowheads) and medial dislocation of LHB tendon (arrow) from the bicipital groove.
|
|
Of the 21 (58%) patients with posterior labral complex injuries (Fig 7), 10 (48%) had POLPSAs (Fig 8) and 11 (52%) had reverse Bankart lesions (Fig 9). In five patients, the labral lesions were in the posterosuperior quadrant, and in 17 patients, the lesions were posteroinferiorly located. In one patient, the labral lesion involved the posterosuperior and posteroinferior quadrants.
View larger version (12K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7: Histogram shows 21 (58%) of the 36 patients had a posterior labrocapsular complex lesion. Ten (48%) of these 21 patients had a POLPSA, and 11 (52%) had a Bankart lesion. In five patients, the posterior labral lesions were posterosuperiorly located, and in 17 patients, the lesions were posteroinferiorly located. In one patient, the lesion involved the posterosuperior and posteroinferior quadrants.
|
|
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8: Transverse conventional intermediate-weighted fat-saturated MR image (2700/37; FOV, 120 x 120 mm) obtained in 42-year-old man shows a POLPSA, which was diagnosed on the basis of findings of a detached posterior glenoid rim periosteum (arrowheads), displaced posterior labrum with increased signal intensity (arrow), and intact posterior joint capsule.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9: Transverse conventional T1-weighted MR image (500/minimum; FOV, 140 x 140 mm) obtained in 48-year-old man shows a reverse Bankart lesion, which was diagnosed on the basis of findings of a linear signal intensity change (arrow) and detachment of the labrum. A negative humeral translation and a glenoid-sided cartilage defect (arrowheads) also are present.
|
|
The mean glenoid version angle was 4.5° ± 0.75 (standard error) (range, 0°–20°) of retroversion. No patient had an anteversion of the glenoid angle. In 27 (75%) patients, the measured glenoid version angle was between 1° and 20° of retroversion. In nine (25%) patients, the glenoid version angle was 0°.
The mean humeral translation distance was –4.8 mm ± 0.75 (range, –1 to –15 mm) posterior to the plane of the scapular body. No patient had a humeral translation anterior to this plane. In 33 (92%) patients, the humeral translation distance was posterior to the plane of the scapular body (range, –1 to –15 mm), and in three (8%) patients, it was 0 mm. We observed no significant difference (P = .9, Student t test) between the patients who received and those who did not receive intraarticular contrast material.
Age and MR Abnormalities
Reverse Hill-Sachs, SSP tendon, SSC tendon, and biceps tendon lesions were significantly more common in the older patients (Table 2). We did not observe a significant association between the other MR findings and patient age (Table 2).
|
DISCUSSION
|
|---|
In our study, 86% of patients had reverse Hill-Sachs lesions and 31% had reverse osseous glenoid rim fractures. Our study results also indicate that soft-tissue abnormalities are not rare after acute posterior shoulder dislocation and that MR imaging is potentially useful in such cases. The frequency of posterocaudal labrum capsule tears was 58% in our study. Only occasional labral tears in patients after acute posterior shoulder dislocation are reported in previously published studies (10,13). The posterocaudal labrocapsular tears observed in our study were seen in approximately the same distribution as those described in previous studies: as either reverse Bankart lesions—characterized by tearing or shredding of the posterior labrum (8,13)—or POLPSA lesions (19). POLPSA lesions are characterized by an avulsed posterior labrum attached to an intact periosteal sleeve of the posterior scapular glenoid neck.
Our results also show that rotator cuff tendon abnormalities in cases of acute traumatic posterior shoulder dislocation are not as rare as previously thought (11,21,22). To our knowledge, only three documented cases of rotator cuff tears following posterior shoulder dislocation are described in the orthopedic literature (11,21,22). In our study, we identified full-thickness rotator cuff tears involving the SSC tendon in 14% of patients, the ISP tendon in 8% of patients, and the SSP tendon in 11% of patients. A possible explanation for the relatively high frequency of SSC tendon tears may be extreme tension sustained by the SSC tendon at the time of posterior humeral dislocation relative to the glenoid fossa, which, according to Matsen et al (5), similarly may result in avulsion of the lesser tuberosity.
Early MR examination of patients with posterior traumatic shoulder dislocations may help avoid delays in recognizing possible underlying rotator cuff tears. Delayed recognition and treatment of posttraumatic rotator cuff tears may lead to irreversible fat infiltration of the muscles. As expected, severe fat infiltration after acute posterior shoulder dislocation was rare in our study. We identified only two (6%) patients with grade 4 fat infiltration of the teres minor muscle. Although isolated fat-induced atrophy of the teres minor muscle is commonly discussed in association with quadrilateral space syndrome and potential entrapment of the axillary nerve after trauma (23,24), the findings in these two patients probably were incidental. Entrapment of the axillary nerve occurs with quadrilateral space syndrome, but injury is most commonly associated with humeral fracture or dislocation (25). None of our study patients had a humeral fracture, and all patients were imaged a short time (within 2 months) after the injury, so a posttraumatic axillary nerve lesion with subsequent teres muscle atrophy was unlikely.
Acute posterior shoulder dislocation has to be distinguished from recurrent (atraumatic) posterior shoulder instability. The underlying mechanism of atraumatic posterior instability is completely different from that of acute posterior dislocation and is not related to trauma but rather to laxity of the supporting structures and/or the shape of the osseous glenoid fossa or labrum (5). Reverse Hill-Sachs lesions are rarely seen in patients with recurrent (atraumatic) posterior shoulder instability (20,26,27). Similarly, avulsion fractures of the posteroinferior glenoid rim (reverse osseous Bankart lesions) are seen only occasionally with recurrent (atraumatic) posterior shoulder instability (26,27). However, we identified such lesions in 31% of patients with acute traumatic posterior glenohumeral joint dislocation. This difference in the frequency of reverse osseous Bankart and Hill-Sachs lesions may be helpful in differentiating the features of traumatic versus atraumatic dislocation or instability in cases in which the etiologic mechanism or history is unclear. Evaluation of the glenoid version angle also may be helpful in distinguishing between these two conditions.
Slight retroversion of the glenoid angle with regard to the scapular body is a normal finding (28). In our study, the mean angle of glenoid retroversion was 4.5° ± 0.75, which corresponds to values reported for stable shoulders (mean angle, 4.0° ± 3.4) (20). In contrast, a markedly increased degree of retroversion has been demonstrated in patients with atraumatic shoulder instability (mean angle, 9.8° ± 5.9) compared with the degree of retroversion in individuals with stable shoulders (20).
Clinical posterior instability has been associated with excessive posterior humeral translation (12). Tung and Hou (12) calculated a mean posterior humeral translation distance of –6.2 mm ± 0.8 in 12 patients with posterior shoulder instability, whereas in patients with stable shoulders, the humeral head was within 1 mm of the plane of the scapular spine. These authors were uncertain as to whether intraarticular contrast material had accentuated the degree of dorsal humeral translation. In our study, we calculated a mean posterior humeral translation of –4.8 mm ± 0.75 (range, –1 to –15 mm), which was unrelated to the presence or absence of intraarticular contrast material (P = .9, Student t test).
Our study had several limitations. We included patients from four institutions with various imaging protocols, including MR examinations performed with and without intraarticular contrast material. Moreover, the data had to be gathered for a relatively long time (6 years), during which the protocols varied slightly. Follow-up and surgical correlations were not available for all patients. Therefore, the data of patients without such data were not included in our study. On the other hand, collecting data from four large orthopedic institutions enabled us to review and present in the literature what is, to our knowledge, the largest imaging series of such injuries to date. We assumed that all of the rotator cuff abnormalities were related to a single dislocation event and had not been present before; however, this may not have been the case for some of the patients whose data we studied. In addition, our study was limited by the lack of an external reference standard; however, MR imaging is an established method of diagnosing shoulder dislocation.
In summary, we believe that the described findings in 36 patients demonstrate that the role of MR imaging in the examination of patients with acute posterior shoulder dislocation in cases in which CT or CT arthrography is more commonly used may be more important than previously thought (10). The frequency of posterior labrocapsular lesions (58%) was much higher than that previously suspected. The frequency of rotator cuff lesions (42%) in our study should also be considered. The acutely traumatized shoulder is difficult to assess clinically, and MR information could influence the treatment plan.
|
ADVANCES IN KNOWLEDGE
|
|---|
- Reverse Hill-Sachs lesions (86% of patients), followed by posterior labrocapsular complex lesions (reverse Bankart and posterior labrocapsular sleeve avulsion lesions) (58% of patients) and posterior osseous glenoid rim fractures (31% of patients), were the most commonly seen abnormalities with acute traumatic posterior shoulder dislocation.
- Rotator cuff lesions were identified at shoulder MR imaging in 42% of patients with a traumatic posterior shoulder dislocation.
|
IMPLICATION FOR PATIENT CARE
|
|---|
- MR imaging after traumatic posterior shoulder dislocation revealed characteristic lesions (reverse Hill-Sachs lesions and posterocaudal labrocapsular tears) and, unexpectedly, rotator cuff tears.
|
FOOTNOTES
|
|---|
Abbreviations: FOV = field of view • ISP = infraspinatus • LHB = long head of the biceps • POLPSA = posterior labrocapsular sleeve avulsion • SSC = subscapularis • SSP = supraspinatus
Author contributions: Guarantors of integrity of entire study, N.S., L.M.W., M.Z.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, N.S., B.J.; clinical studies, N.S., L.M.W., R.B., M.E.S., M.P.R., M.Z.; statistical analysis, M.Z.; and manuscript editing, N.S., L.M.W., M.E.S., M.P.R., B.J., M.Z.
Authors stated no financial relationship to disclose.
|
References
|
|---|
- Cooper A. On the dislocations of the os humeri upon the dorsum scapulae, and upon fractures near the shoulder joint. Guys Hosp Rep 1839;4:265–284.
- Rowe CR. Prognosis in dislocations of the shoulder. J Bone Joint Surg Am 1956;38-A:957–977.[Abstract/Free Full Text]
- Samilson RL, Prieto V. Posterior dislocation of the shoulder in athletes. Clin Sports Med 1983;2:369–378.[Medline]
- Schulz TJ, Jacobs B, Patterson RL Jr. Unrecognized dislocations of the shoulder. J Trauma 1969;9:1009–1023.[Medline]
- Matsen FA, Thomas SC, Rockwood CA, Wirth MA. Glenohumeral instability. In: Rockwood, CA, Matsen FA, eds. The shoulder. Philadelphia, Pa: Saunders, 1998; 611–689.
- Blasier RB, Burkus JK. Management of posterior fracture-dislocations of the shoulder. Clin Orthop Relat Res 1988;232:197–204.[Medline]
- Hawkins RJ, Neer CS 2nd, Pianta RM, Mendoza FX. Locked posterior dislocation of the shoulder. J Bone Joint Surg Am 1987;69:9–18.[Abstract/Free Full Text]
- Norris T. Fractures of the proximal humerus and dislocations of the shoulder. In: Browner BD, Jupiter JB, Levine AM, eds. Skeletal trauma. Philadelphia, Pa: Saunders, 1992; 1201–1290.
- Rowe CR, Zarins B. Chronic unreduced dislocations of the shoulder. J Bone Joint Surg Am 1982;64:494–505.[Abstract/Free Full Text]
- Cicak N. Posterior dislocation of the shoulder. J Bone Joint Surg Br 2004;86:324–332.[CrossRef][Medline]
- Steinitz DK, Harvey EJ, Lenczner EM. Traumatic posterior dislocation of the shoulder associated with a massive rotator cuff tear: a case report. Am J Sports Med 2003;31:1010–1012.[Free Full Text]
- Tung GA, Hou DD. MR arthrography of the posterior labrocapsular complex: relationship with glenohumeral joint alignment and clinical posterior instability. AJR Am J Roentgenol 2003;180:369–375.[Abstract/Free Full Text]
- Hottya GA, Tirman PF, Bost FW, Montgomery WH, Wolf EM, Genant HK. Tear of the posterior shoulder stabilizers after posterior dislocation: MR imaging and MR arthrographic findings with arthroscopic correlation. AJR Am J Roentgenol 1998;171:763–768.[Abstract/Free Full Text]
- Zlatkin MB, Iannotti JP, Roberts MC, et al. Rotator cuff tears: diagnostic performance of MR imaging. Radiology 1989;172:223–229.[Abstract/Free Full Text]
- Goutallier D, Postel JM, Bernageau J, Lavau L, Voisin MC. Fatty muscle degeneration in cuff ruptures: pre- and postoperative evaluation by CT scan. Clin Orthop Relat Res 1994;304:78–83.[Medline]
- Fuchs B, Weishaupt D, Zanetti M, Hodler J, Gerber C. Fatty degeneration of the muscles of the rotator cuff: assessment by computed tomography versus magnetic resonance imaging. J Shoulder Elbow Surg 1999;8:599–605.[CrossRef][Medline]
- Zanetti M, Weishaupt D, Gerber C, Hodler J. Tendinopathy and rupture of the tendon of the long head of the biceps brachii muscle: evaluation with MR arthrography. AJR Am J Roentgenol 1998;170:1557–1561.[Abstract/Free Full Text]
- Walch G, Nove-Josserand L, Boileau P, Levigne C. Subluxations and dislocations of the tendon of the long head of the biceps. J Shoulder Elbow Surg 1998;7:100–108.[CrossRef][Medline]
- Yu JS, Ashman CJ, Jones G. The POLPSA lesion: MR imaging findings with arthroscopic correlation in patients with posterior instability. Skeletal Radiol 2002;31:396–399.[CrossRef][Medline]
- Weishaupt D, Zanetti M, Nyffeler RW, Gerber C, Hodler J. Posterior glenoid rim deficiency in recurrent (atraumatic) posterior shoulder instability. Skeletal Radiol 2000;29:204–210.[CrossRef][Medline]
- Moeller JC. Compound posterior dislocation of the glenohumeral joint: case report. J Bone Joint Surg Am 1975;57:1006–1007.[Free Full Text]
- Schoenfeld AJ, Lippitt SB. Rotator cuff tear associated with a posterior dislocation of the shoulder in a young adult: a case report and literature review. J Orthop Trauma 2007;21:150–152.[CrossRef][Medline]
- Cothran RL Jr, Helms C. Quadrilateral space syndrome: incidence of imaging findings in a population referred for MRI of the shoulder. AJR Am J Roentgenol 2005;184:989–992.[Abstract/Free Full Text]
- Wilson L, Sundaram M, Piraino DW, Ilaslan H, Recht MP. Isolated teres minor atrophy: manifestation of quadrilateral space syndrome or traction injury to the axillary nerve? Orthopedics 2006;29:447–450.[Medline]
- Perlmutter GS. Axillary nerve injury. Clin Orthop Relat Res 1999;368:28–36.[Medline]
- Bigliani LU, Pollock RG, McIlveen SJ, Endrizzi DP, Flatow EL. Shift of the posteroinferior aspect of the capsule for recurrent posterior glenohumeral instability. J Bone Joint Surg Am 1995;77:1011–1020.[Abstract/Free Full Text]
- Pollock R, Bigliani LU. Recurrent posterior shoulder instability. In: Warner JJP, Ianotti JP, Gerber C, eds. Complex and revision problems in shoulder surgery. Philadelphia, Pa: Lippincott-Raven, 1997; 167–174.
- Gerber C, Terrier F, Zehnder R, Ganz R. The subcoracoid space: an anatomic study. Clin Orthop Relat Res 1987;215:132–138.[Medline]
This article has been cited by other articles:
TOP
ABSTRACT
INTRODUCTION
