An arthroscopic assessment of the shoulder was performed. At the time of surgery, only a very small osseous Bankart lesion was noted, and most of the bone seen on the CT images appeared to be heterotopic ossification. Arthroscopic stabilization was performed.
The patient returned to competitive Australian Rules football and had experienced no additional instability episodes at the 2-year follow-up.
Case 2. A 31-year-old rugby player dislocated the left shoulder while making a tackle. He was diagnosed with an anterior shoulder dislocation, and the shoulder was reduced in the emergency department. Postreduction radiographs revealed no osseous Bankart lesion (Fig. 5). He presented 2 years later with recurrent shoulder dislocation. Clinical examination revealed positive apprehension and relocation tests.
Following CT, a senior musculoskeletal radiologist reported a “large bony Bankart lesion measuring 16 mm in length.” However, additional review of the images suggested that no osseous Bankart lesion was present. The anterior glenoid appeared intact on both the axial (Fig. 6) and sagittal (Fig. 7) views. The circular geometry of the lower three-fifths of the glenoid remained intact. The ossification seen medial to the glenoid appeared external to the cortex, and the images showed “fluffy” amorphous bone that is consistent with heterotopic ossification rather than a displaced osseous Bankart lesion (Fig. 8).
The patient underwent arthroscopic stabilization. As shown in Figure 9, the arrow indicates the anterior glenoid margin that was noted to be intact without a substantial osseous Bankart lesion. The heterotopic ossification (outlined in yellow) was located in the soft tissue anterior to the glenoid, approximately 1.5 cm medial to the glenoid margin. This location is not consistent with an osseous Bankart lesion. An osseous Bankart lesion would normally be located in the region of the labrum (star on Fig. 9), but no bone was seen within the labral tissue.
An arthroscopic stabilization was performed. After mobilization of the labrum and the capsule, the heterotopic bone remained embedded in the elevated anterior tissue; the heterotopic bone was ignored, and a routine labral repair was possible. The patient returned to rugby and had experienced no additional instability episodes at the 3-year follow-up.
Case 3. A 19-year-old man presented with recurrent subluxation of the right shoulder following a traumatic injury 3 months previously. Clinical examination confirmed anteroinferior instability with positive apprehension and relocation tests. Additionally, he had a long history of pain over the acromioclavicular joint, and radiographs confirmed osteolysis of the distal aspect of the clavicle.
CT was performed to further investigate the subluxation episodes. A musculoskeletal radiologist reported an “osseous Bankart injury” with a “fracture fragment measuring 2.6 × 1.5 × 0.5 cm.” Examination of the images suggested that the findings were not consistent with an osseous Bankart lesion. The ossification appeared external to the glenoid, and the cortex of the glenoid appeared intact (Fig. 10). The ossification extended a considerable distance medially along the glenoid neck. The circular outline of the lower three-fifths of the glenoid remained intact without disruption of the anteroinferior margin, which would be seen with an osseous Bankart lesion (Fig. 11). A series of progressively more medial images confirmed that the ossification was external to the native glenoid (Fig. 12).
The patient underwent arthroscopic excision of the distal aspect of the clavicle. During surgery, a diagnostic arthroscopy of the glenohumeral joint also was performed. A small Bankart lesion was identified. The anteroinferior glenoid was intact without an osseous Bankart lesion. Although the patient did report instability symptoms, he chose to not undergo stabilization surgery.
In the 3 patients described above, senior musculoskeletal radiologists misinterpreted heterotopic ossification as an osseous Bankart lesion. I believe that this mistake is commonplace—not only among radiologists, but also among orthopaedic surgeons, particularly those who are less familiar with shoulder pathology. This has serious implications for surgical planning. If a surgeon incorrectly believes that an osseous Bankart lesion is present, he or she may perform open surgery when an arthroscopic procedure may be a more appropriate initial approach.
Osseous Bankart lesions have become a major indication for recommending open surgery rather than arthroscopic stabilization. In a landmark paper, Burkhart and De Beer noted a dramatic increase in failure rates in patients who underwent arthroscopic shoulder stabilizations with substantial osseous Bankart lesions1. Patients without such a lesion had a recurrence rate of 4%, but with substantial bone defects, they had a 67% recurrence rate (89% in contact athletes). The authors suggested that patients with substantial bone loss should have a bone-grafting procedure rather than arthroscopic stabilization.
More recent papers have suggested that even small osseous Bankart lesions are important. Balg and Boileau proposed a 10-point preoperative shoulder instability score3. When a score is >6 points, they recommend open rather than arthroscopic surgery. Great importance is placed on the presence of osseous Bankart lesions. In the component of the score relating to the glenoid, 2 points are allocated to patients with any osseous Bankart lesion (described as “disappearance of the normal sclerotic contour of the anteroinferior glenoid”).
With the recognition of the importance of osseous Bankart lesions, CT has become commonplace in the preoperative assessment of patients having stabilization surgery2. However, the 3 cases described above show that heterotopic ossification may occur in the region of the anteroinferior glenoid, which may be falsely reported as osseous Bankart lesions, even by experienced musculoskeletal radiologists. I propose that this heterotopic ossification is most likely due to subperiosteal ossification underneath an ALPSA lesion.
ALPSA lesions are produced when periosteum that is attached to the labrum is stripped off of the anterior glenoid4. This stripping can progress medially along the glenoid for a considerable distance. Heterotopic ossification underneath periosteal stripping is a common phenomenon; it is seen in situations such as callus formation in fractures and in “Codman triangles” in bone tumors. I believe that ossification external to the glenoid occurs underneath the periosteum of an ALPSA lesion in a similar manner. This ossification may mimic an osseous Bankart lesion.
It is important that heterotopic ossification is recognized to prevent potential misdiagnosis of a substantial osseous Bankart lesion when no such lesion is present. I have noted several features that help to determine that ossification (rather than an osseous Bankart lesion) is present:
- Axial CT shows ossification extending medially along the anterior glenoid (Figs. 1, 6, and 10).
- Native glenoid cortical bone remains intact, with ossification appearing to be external (Figs. 3, 4, 6, 8, 10, 11, and 12).
- The circular geometry of the lower three-fifths of the glenoid remains intact without loss of the anteroinferior glenoid margin (Figs. 7 and 11).
- The ossification often has the “fluffy” amorphous appearance that is typical of immature bone. This differs from the appearance of mature bone, which would be expected with an osseous Bankart lesion (Fig. 8).
- The outline of the heterotopic ossification is often smooth and rounded rather than having an irregular margin, which would be expected with an osseous Bankart lesion (Figs. 3, 4, 8, 11, and 12).
- Arthroscopic images show an intact anterior glenoid and ossification medial to the glenoid margin (Fig. 9).
In summary, heterotopic ossification under an ALPSA lesion may mimic an osseous Bankart lesion and lead to errors in diagnosis and surgical decision-making. In order to prevent these errors, surgeons should be familiar with this phenomenon and be able to recognize the radiographic features that can distinguish heterotopic ossification from a substantial osseous Bankart lesion. The information in this article should help orthopaedic surgeons to recognize this issue and may lead to avoidance of errors in preoperative planning.
1. Burkhart SS, De Beer JF. Traumatic glenohumeral bone defects and their relationship to failure of arthroscopic Bankart repairs: significance of the inverted-pear glenoid and the humeral engaging Hill-Sachs lesion. Arthroscopy. 2000 Oct;16(7):677-94.
2. Auffarth A, Mayer M, Kofler B, Hitzl W, Bogner R, Moroder P, Korn G, Koller H, Resch H. The interobserver reliability in diagnosing osseous lesions after first-time anterior shoulder dislocation comparing plain radiographs with computed tomography scans. J Shoulder Elbow Surg. 2013 Nov;22(11):1507-13. Epub 2013 Jun 20.
3. Balg F, Boileau P. The instability severity index score. A simple pre-operative score to select patients for arthroscopic or open shoulder stabilisation. J Bone Joint Surg Br. 2007 Nov;89(11):1470-7.
4. Neviaser TJ. The anterior labroligamentous periosteal sleeve avulsion lesion: a cause of anterior instability of the shoulder. Arthroscopy. 1993;9(1):17-21.
Supplemental Digital Content
Copyright © 2019 The Author. Published by The Journal of Bone and Joint Surgery, Incorporated. All rights reserved.