Posterior Ossification of the Shoulder: The Bennett Lesion

Transcription

1 Posterior Ossification of the Shoulder: The Bennett Lesion Etiology, Diagnosis, and Treatment* James D. Ferrari, MD, Dudley A. Ferrari, MD, James Coumas, MD, and Arthur M. Pappas, MD From the Departments of Orthopedics and Physical Rehabilitation and Radiology, University of Massachusetts Medical Center, Worcester, Massachusetts ABSTRACT We report a series of ossific lesions of the posterior inferior glenoid in a group of elite baseball players. We hope to clarify the etiology, diagnosis, and treatment of the Bennett lesion. From August 1985 to August 1991, we identified six professional baseball pitchers and one college pitcher with evidence of ossification of the shoulder on plain radiographs, computed tomography, or magnetic resonance imaging. Arthroscopic examination was performed in all cases. All seven players had identifiable posterior labral injury on arthroscopic examination ; six of these seven also had varying degrees of undersurface posterior rotator cuff damage. No anterior tissue damage, anterior instability, or subacromial impingement was noted. No ossification was identified arthroscopically. Intraarticular labral and rotator cuff tears were debrided arthroscopically and patients underwent rehabilitation for 4 to 6 months after surgery. Six of the seven athletes returned to preinjury performance levels; however, one pitcher is no longer playing competitive baseball. The Bennett lesion is an extraarticular posterior ossification associated with posterior labral injury and posterior undersurface rotator cuff damage. It is not, however, a result of traction stresses in the region of the triceps insertion. Recognition is important for identification and treatment of the lesion and associated pathologic damage. mechanism, and lower body mechanics to throw a baseball at speeds approaching 95 mph.19 It is not surprising that injuries are common in professional baseball pitchers because any disruption of this exact kinesiology may lead to unbalanced stresses about the shoulder joint, producing pain and subsequent decreased performance. Posterior shoulder injuries in throwing athletes are a poorly understood and often debilitating phenomenon. In 1941, Bennett described a posterior inferior glenoid lesion similar to an osteoarthritic deposit that he thought was caused by traction stresses in the region of the long head of the triceps muscle. Pain was thought to be from the proximity of the calcific deposit to the axillary nerve with subsequent irritation. Radiographic visualization was performed with the arm abducted and externally rotated and the x-ray beam angled 5 cephalad. Bennett was able to remove the deposit through a posterior approach, although he did not note its exact intra- or extracapsular location. Years later, Bennett stated that operative removal was not necessary. 4 Little mention has been made of this lesion in the more than 50 years since Bennett s original description.3,12,15 Furthermore, since the advent of computerized tomography (CT), magnetic resonance imaging (MRI), and arthroscopy, no study has assessed the exact nature of the lesion and the associated pathologic damage. In this study, we report a series of elite baseball pitchers with posterior ossification and attempt to clarify the causes, diagnosis, and treatment of the Bennett lesion. Professional baseball pitchers rely on the precise synchrony of their shoulder girdle musculature, capsular *Presented at the 18th annual meeting of the AOSSM, San Diego, California, July taddress correspondence and reprint requests to: Dudley A. Ferrari, MD, Department of Orthopedics and Physical Rehabilitation, University of Massachusetts Medical Center, 55 Lake Avenue North, Worcester, MA No author or related institution has received any financial benefit from research in this study. 171 MATERIALS AND METHODS From August 1985 to August 1991, we identified seven elite baseball players who complained of shoulder symptoms and had radiographic evidence of crescentic mineralization emanating from the posteroinferior glenoid on CT scan examination. Bennett and Stryker notch views were not routinely taken. Because the standard radiographic views did not always show the ossification, CT or MRI scans were

2 z 172 used to investigate the cause of pain. All patients underwent CT-arthrography or MRI and had arthroscopic surgery performed on the symptomatic shoulder. Patients underwent a period of rest and a rehabilitation program 18 and treatment with nonsteroidal antiinflammatory drugs for 3 to 4 months before arthroscopic surgery. The decision to perform arthroscopic surgery was based on the findings on imaging or intractability of the patient s symptoms. The patient s symptoms, physical examination findings, arthroscopic surgery findings, and length of followup are presented in Table 1. Average age at arthroscopy was 24 years and 10 months (range, 18 years and 7 months to 25 years and 11 months). Average followup was 41 months (range, 8 to 79). RESULTS Generalized nonspecific shoulder complaints were common, but all patients had posterior shoulder pain at one time. Two patients complained of the pain during the follow-through phase of the pitching motion, one during the late cocking phase, three in both the late cocking and follow-through phases, and one felt a nonspecific discomfort throughout the pitching motion. The only consistent physical examination finding was tenderness on palpation of the posteroinferior glenoid region. All other significant physical examination findings listed in Table 1 are related to physical maneuvers that exaggerate the various phases of the pitching motion in which the patients experienced symptoms. Two patients had slight posterior subluxation when examined in the supine position with the arm forward flexed and in 20 of abduction and force applied in a posterior direction. One complained of posterior pain with the maneuver. No patient exhibited any anterior instability or impingement. Plain radiographs were available in most cases. Specific Bennett views taken with the arm abducted and externally rotated and the x-ray beam angled 5 cephalad were not available. In some instances, the ossification could be visualized on plain film (Fig. 1). Computed tomography-arthrography was performed in all cases, and in all instances an extraarticular curvilinear, calcification originating from the posterior inferior glenoid and extending toward the humeral head was visualized (Fig. 2). All patients underwent arthroscopic surgery (Table 1). Six patients had posterior labral tears that were partially debrided. Five of the six tears were posterosuperior in location and one was posteroinferior. One posterosuperior labral tear was completely detached and associated with a posterior capsular tear not in the area of the ossification. Posterior labral thinning was noted in one patient. Four of the patients with posterior labral lesions also exhibited posterior rotator cuff fibrillation, and two others had posterior undersurface rotator cuff tears. All tears were grade I or II undersurface type lesions. Rotator cuff tears were debrided to viable, bleeding tissue. No anterior capsular injury was noted. In no case was there arthroscopically visible evidence of the posteroinferior glenoid ossification. At an average postoperative followup of 41 months, all patients but one (who had a completely detached posterosuperior labrum associated with a posterior capsular tear) were at or above their preoperative professional baseball class (Table 1). DISCUSSION The lesion we describe is an ossification of the posteroinferior glenoid in the proximity of the posterior band of the inferior glenohumeral ligament complex. Although it can TABLE 1 Patient data a All patients had pain on palpation of the posterior glenoid. b No patients had visualized evidence of the Bennett lesion. PI, posteroinferior; PS, posterosuperior; RC, rotator cuff.

3 173 Figure 1. Anteroposterior (A) abducted and AP (B) internally rotated plain film radiographs of the left shoulder show crescentic mineralization adjacent to the posteroinferior glenoid (arrows). be visualized with plain films (Fig. 1), this lesion is not easily seen unless Bennett and Stryker notch views are used. Magnetic resonance imaging and CT-arthrography may show calcification earlier, as well as the associated posterior labral tears, posterior undersurface rotator cuff tears, and subluxation. Since Bennett s original description of this lesion in 1941,6 there has been little mention of it in the literature. In 1977, Lombardo et al. 14 described four cases of professional baseball players who underwent operative removal of posterior capsular lesions to alleviate intractable shoulder symptoms. All patients had pain on palpation of the posterior glenoid as well as pain with both abductionexternal rotation and adduction-internal rotation. Plain films revealed ossification at the posteroinferior glenoid region that was clearly not contiguous with the glenoid ar- Figure 2. Computed tomography-arthrograms obtained at the onset of symptoms (A) and at 5 months after rehabilitation (B). In A, note the reactionary sclerosis of the posterior glenoid (arrowhead) and adjacent crescentic soft tissue mineralization (arrow). Note extraarticular location. In B, note the progressive maturation and healing of the ossification (arrow). ticular surface. The ossification was noted to be within the posterior capsule, and histologic analysis disclosed reactive new bone formation. The authors speculated that the lesion could be the result of posterior impingement of the humeral head on the glenoid during cocking, traction phenomenon during deceleration, or the wringing action that occurs during acceleration when the humeral head rapidly internally rotates after external rotation. The authors, however, did not associate their lesion with the Bennett lesion. Barnes and Tullos3 discussed a posterior capsular syndrome associated with the Bennett lesion in They described 24 patients who had pain during the cocking phase, follow-through phase, or both. Only one third of the patients had demonstrable lesions on Bennett view radiographs. Two patients underwent operative removal of the

4 174 ossification; one had recurrent pain and the other was placed on waivers, but there was no note as to whether this patient had recurrent pain. The authors noted that the lesion was a subperiosteal exostosis at the posteroinferior glenoid associated with posterior labral tears. The authors were unsure as to whether the lesion represented an actual posterior capsular tear with the ossification occurring secondarily, or if it was a response to posterior humeral head subluxation. O Brien et al.16 demonstrated that the posterior band of the inferior glenohumeral ligament complex (IGHLC) is consistently localized to the posteroinferior quadrant. This suggests that the posterior band of the IGHLC may play a role in the production of the avulsion lesion. Throughout the shoulder s range of motion, the posterior band acts to stabilize the humeral head, resisting inferior translation with the arm abducted and externally rotated and resisting posterior translation with the arm abducted and internally rotated. The location of the lesion in all cases was the posteroinferior quadrant of the glenoid. Since these two studies, however, major advances have been made in imaging and orthopaedic technology: CT, MRI, and arthroscopic surgery. These technologies have enabled us to gain additional insight into the Bennett lesion and its associated pathologic changes. Arthroscopy, CT, and MRI clearly demonstrate that the lesion is extraarticular, with the air and dye uniformly coating the glenoid surface in no continuity with the lesion (Fig. 2). The ossification was not visualized during arthroscopy in any of the cases. This is in agreement with previous reports of resection of the lesion. 1,14,15 The Bennett lesion is, however, associated with intraarticular injuries such as posterior labral tears and posterior undersurface rotator cuff tears. The posterior labral tears in all but one instance were located in the posterosuperior quadrant of the glenoid labrum, i.e., the infraspinatus region. Rotator cuff fibrillation or tearing always involved the infraspinatus region. Changes of the lesion were seen in two cases. In Case 2, CT scans obtained 5 months apart demonstrate maturation of the ossific lesion from the glenoid toward the humeral head (Fig. 2). In Case 4, the initial CT scan clearly showed the lesion, but the pitcher continued to throw (Fig. 3). A repeat CT-arthrogram done 8 months later not only verified progression of the lesion but a newly diagnosed posterior labral tear and subluxation. Since this is an ossific lesion, some degree of healing or maturation must be anticipated; how much time the lesion needs was not determined in this study. Once there was no tenderness on palpation, it was assumed that the lesion had healed. Rehabilitation did not begin until shoulder pain ceased, usually 1 to 2 months after surgery. From a review of these cases of the Bennett lesion, we believe that operative removal of the lesion is not necessary, but that arthroscopic treatment of the associated intraarticular pathologic tissue is warranted.2,17 Furthermore, arthroscopic removal of the Bennett lesion with a synovial resector, which has been reported, may jeopardize the integrity of the posterior capsule. It is essential, however, to recognize the lesion and its related problems in Figure 3. Computed tomography-arthrogram (A) shows a small focus of soft tissue mineralization (arrow) in addition to reactive sclerosis of the bony glenoid (arrowhead). Continued activity (B) despite persistent and progressive symptoms prompted a repeat examination that shows progressive soft tissue mineralization (arrow) and a new posterior labral tear (thin arrow), which was confirmed at arthroscopy. Computed tomography-arthrogram shows posterior subluxation of the humeral head with reference to the bony glenoid. any throwing athlete with posterior tenderness on palpation and pain during the cocking or follow-through phases of the throwing motion. Bennett or Stryker notch views should be used to locate the ossification, and CT or MRI scans can be taken if needed to determine intraarticular injuries. Just which lesion caused the pain that interfered with pitching was not determined. First, the ossific lesion was allowed to mature as seen in Figure 2, and no rehabilitation

5 175 was started until posterior pain on palpation subsided. Second, intraarticular debridement of labral tears and rotator cuff tears was done based on CT and MRI findings. The combination of nonoperative treatment for the ossific lesion and debridement of the intraarticular pathologic tissue followed by rehabilitation allowed all but one pitcher to return to his previous level of baseball. The one pitcher no longer throwing exhibited the most tissue damage, incurring a posterior labral and capsular tear and increasing size of the lesion as shown in Figure 3B. Magnetic resonance imaging contributes to the understanding of the Bennett lesion as well. Gradient echo and spin echo sequences in the axial plane clearly demonstrate that the lesion is composed of dense calcification of equal signal intensity to cortical bone and contains no bone marrow, which one would expect if the lesion was an osteoarthritic growth as originally proposed by Bennett.5 6 The lesion s density on CT suggests that it is calcific material, a result of posterior capsular microtearing that produces bleeding and subsequent reactive new bone formation. This is in agreement with the histologic findings of Lombardo et al.14 Furthermore, MRI may demonstrate a paramagnetic effect suggesting prior hemorrhage. This study did not determine the cause of the Bennett lesion. The lesion was not related to triceps attachment as Bennett originally suggested. A posteriorly directed force is suggested by the presence of pain during cocking and posterior subluxation on physical examination and on CTarthrography (Fig. 3B) in a few cases. The posterior location of the associated intraarticular injuries suggests that the humeral head is forced posteriorly while in maximum external rotation during the cocking and extension phase. This is supported by Howell et al.ll The absence of any anterior injury further suggests that the anterior capsule is a buttress forcing the humeral head posteriorly during external rotation.&dquo; Fronek et a1.9 also demonstrated that posterior subluxation is associated with posterior capsular calcification and posterior glenoid erosions, and that these patients had pain in the follow-through phase of throwing. The lesion may possibly be produced during the followthrough phase of the pitching motion, when deceleration forces of -500,000 deg/sec/sec are present,&dquo; and the majority of the forces are absorbed by the posterior shoulder musculature.8, 12, 13 A traction injury could easily be produced during this phase since anterior translation occurs during cross-body movement of the arm,ll straining the posterior capsule, and the internal rotation of the humeral head in follow-through tightens the posterior band of the IGHLC. Furthermore, the rotator cuff tears were undersurface, traction-type lesions. A combination of the above mechanisms may be responsible for the production of the Bennett lesion. Pain in both the follow-through and cocking phases implies a combination of mechanisms, but it cannot be determined which is cause and which is effect. SUMMARY The Bennett lesion is an extraarticular ossification in the posteroinferior quadrant of the glenoid associated with past subluxation, posterior labral tears, and posterior undersurface rotator cuff tears. The lesion may be a source of pain in throwing athletes who present with posterior pain in the cocking phase, follow-through phase, or both, and have posterior tenderness on palpation. Bennett or Stryker views or MRI and CT-arthrograms should be considered, with the latter showing calcification earlier along with intraarticular injuries. Operative removal is not warranted and the lesion should be given time to mature as shown by absence of pain on palpation or by radiographic demonstration. The prognosis for return to pitching is related to the degree of associated intraarticular injury and if the lesion is associated with posterior subluxation. The Bennett lesion may be caused by traction on the posterior band of the IGHLC produced by posterior subluxation during cocking, posterior decelerative forces during follow-through, or a combination of the two. REFERENCES 1. Andrews JR, Angelo RL: Shoulder arthroscopy in the throwing athlete. Tech Orthop 3: 75-81, Andrews JR, Kupferman SP, Dillman CJ: Labral tears in throwing and racquet sports. Clin Sports Med 10: , Barnes DA, Tullos HS: An analysis of 100 symptomatic baseball players. Am J Sports Med 6: 62-67, Bennett GE: Elbow and shoulder lesions of baseball players. Am J Surg 98: , Bennett GE: Shoulder and elbow lesions distinctive of baseball players. Ann Surg 126: , Bennett GE: Shoulder and elbow lesions of the professional baseball pitcher. J Am Med Assoc 117: , Bowen MK, Warren RF: Ligamentous control of shoulder stability based on selective cutting and static translation experiments. Clin Sports Med 10: , DiGiovine NM, Jobe FW, Pink M, et al: An electromyographic analysis of the upper extremity in pitching. J Shoulder Elbow Surg 1: 15-25, Fronek J, Warren RF, Bowen MK: Posterior subluxation of the glenohumeral joint. J Bone Joint Surg 71A: , Harryman DT, Sidles JA, Clark JM, et al: Translation of the humeral head on the glenoid with passive glenohumeral motion. J Bone Joint Surg 72A: , Howell SM, Galinat BJ, Renzi AJ, et al: Normal and abnormal mechanics of the glenohumeral joint in the horizontal plane. J Bone Joint Surg 70A: , Jobe FW, Moynes DR, Tibone JE, et al: An EMG analysis of the shoulder in pitching: A second report. Am J Sports Med 12: , Jobe FW, Tibone JE, Perry J, et al: An EMG analysis of the shoulder in throwing and pitching: A preliminary report. Am J Sports Med 11: 3-5, Lombardo SJ, Jobe FW, Kerlan RK, et al: Posterior shoulder lesions in throwing athletes. Am J Sports Med 5: , Norwood LA, Terry GC: Shoulder posterior subluxation. Am J Sport Med 12:25-30, O Brien SJ, Neves MC, Arnoczky SP, et al: The anatomy and histology of the inferior glenohumeral ligament complex of the shoulder. Am J Sports Med 18: , Pappas AM, Goss TP, Kleinman PK: Symptomatic shoulder instability due to lesions of the glenoid labrum. Am J Sports Med 11: , Pappas AM, Zawacki RM, McCarthy CF: Rehabilitation of the pitching shoulder. Am J Sports Med 13: , Pappas AM, Zawacki RM, Sullivan TJ: Biomechanics of baseball pitching. A preliminary report. Am J Sports Med 13: , 1985 DISCUSSION Russell F. Warren, MD, New York, New York: This paper is interesting and informative. The lesion that the authors describe appears consistent with that noted by Bennett in Unfortunately, they have not noted it on routine films as described by Bennett, but they present to us a

6 176 lesion seen on the CT scan in all of their patients. This lesion, noted in throwers, is seen at the posteroinferior corner of the glenoid. For some time we have recommended using an instability series to evaluate shoulder pain in young athletes. This consists of an AP view in internal rotation, a West Point view, and a Stryker notch view. This latter view approximates that described by Bennett with the beam tilted cephalad while aiming at the coracoid. It was also able to detect about half of the Hill-Sachs lesions, although they may not show up on your standard internal rotation review. Their patients presented with pain in either the cocking phase or the final phase of throwing. Posterior instability was noted in only two patients. I have seen this lesion on a number of occasions with posterior subluxation and even, on occasion, anterior subluxation with a seemingly traction spur forming posteriorly. The lesion was thought by Bennett to be a traction lesion of the long head of the triceps. It appears usually to be a traction injury occurring within the capsule near the posterior insertion of the inferior ligament. The authors believe that arthroscopic debridement, ignoring the bony lesion, will result in a painless shoulder in most patients. I do not think that they have proven this point. Concerning those with clear-cut labral injury and the head subluxating posteriorly, we have debrided the, area and reattached the loose capsule. This has resulted in moderate success. It is still unclear in my mind whether debridement of the deposit plays any role in treating these patients if there is no instability. We have tended toward removing it and repairing the defect. However, it may be, as they have suggested, that without instability we only need to wait, debride the joint, and place them on a rehabilitation program. Overall, I will expect that these traction lesions will have some degree of instability as their basic underlying cause. This is an informative paper that draws our attention to an interesting sports medicine problem.