Level of Evidence: Therapeutic Level IV. See Instructions to Authors for a complete description of levels of evidence.

Transcription

1 1752 COPYRIGHT 2005 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED Arthroscopic Osseous Bankart Repair for Chronic Recurrent Traumatic Anterior Glenohumeral Instability BY HIROYUKI SUGAYA, MD, JOJI MORIISHI, MD, IZUMI KANISAWA, MD, AND AKIHIRO TSUCHIYA, MD Investigation performed at the Funabashi Orthopaedic Sports Medicine Center, Funabashi, Chiba, and the Department of Orthopaedic Surgery, Kawatetsu Chiba Hospital, Chiba, Japan Background: A chronic osseous Bankart lesion has traditionally been treated with soft-tissue repair and/or open bone-grafting for a large glenoid defect. We developed an arthroscopic method of osseous reconstruction of the glenoid without bone-grafting. The purpose of this study was to evaluate the postoperative outcomes of our technique for chronic recurrent traumatic anterior glenohumeral instability. Methods: A consecutive series of forty-two shoulders in forty-one patients with chronic recurrent traumatic glenohumeral instability underwent an arthroscopic osseous Bankart repair. All shoulders were evaluated preoperatively with three-dimensionally reconstructed computed tomography, which confirmed an osseous fragment at the anteroinferior portion of the glenoid. The average bone loss in the glenoid was 24.8% (range, 11.4% to 38.6%), and the average fragment size was 9.2% (range, 2.1% to 20.9%) of the glenoid fossa. In all shoulders, a displaced osseous fragment, firmly attached to the labroligamentous complex, was separated from the glenoid neck before reduction and fixation in the optimal position with use of suture anchors. All patients were assessed with use of the scoring systems of Rowe et al. and the University of California at Los Angeles preoperatively and at the final evaluation. Results: The mean duration of follow-up was thirty-four months. At that time, thirty-nine of the forty-two shoulders were rated as having a good or excellent result. The mean Rowe score improved from 33.6 points preoperatively to 94.3 points postoperatively (p < 0.01). The mean score on the University of California at Los Angeles system improved from 20.5 points preoperatively to 33.6 points at the final evaluation (p < 0.01). The average passive external rotation was 75 with the arm at the side and 93 with the arm at 90 of abduction. Two patients had a reinjury. Eventually, thirty-five of thirty-seven patients who were active participants in sports returned to the sport they had played before the injury. Conclusions: Arthroscopic osseous Bankart repair with use of suture anchors yields a successful outcome even in shoulders with a chronic large glenoid defect. Level of Evidence: Therapeutic Level IV. See Instructions to Authors for a complete description of levels of evidence. Fractures of the anterior glenoid rim can result in persistent instability of the glenohumeral joint 1,2. An osseous Bankart lesion is defined as a displaced fracture of the anterior glenoid rim associated with chronic anterior glenohumeral instability 3,4. The reported prevalence of glenoid osseous defects has ranged from 8% (eighteen of 226) to 73% (116 of 158) 5-9, possibly because of a lack of practical imaging techniques and quantification methods 10,11. Sugaya et al. reported that the prevalence of an osseous Bankart lesion, in which an osseous fragment from the anteroinferior quadrant of the glenoid was confirmed with use of preoperative threedimensionally reconstructed computed tomography, was 50% (fifty of 100) 11. An osseous Bankart lesion has traditionally been treated with only soft-tissue repair and/or open bonegrafting for a large glenoid defect 12,13. The osseous fragment of this lesion tends to be connected firmly with the adjacent labrum and capsule 4,14. We determined that these lesions can be reconstructed arthroscopically because it is possible to separate the osseous fragment from the glenoid neck together with the labroligamentous complex even in the chronic stage and to restore the glenoid morphology by reducing the fragment with use of suture anchors, similar to the methods described

2 1753 Materials and Methods Patient Selection rom May 2000 to April 2002, forty-three consecutive shoulders in forty-two patients underwent arthroscopic stabiliza- F tion by a single surgeon (H.S.) for recurrent traumatic anterior glenohumeral instability associated with an osseous Bankart lesion. All patients gave their informed consent to participate in this study prior to surgery. The criteria for inclusion in this study were repeated anterior subluxation or dislocation following an initial episode of anterior instability caused by a traumatic event, which had occurred at least six months before surgery, and an osseous fragment at the anteroinferior quadrant of the glenoid rim or glenoid neck, confirmed both by preoperative three-dimensionally reconstructed computed tomography of the glenoid 11 and during arthroscopy (Figs. 1 and 2). One patient was lost to follow-up. Therefore, this study includes forty-two shoulders in forty-one patients who were followed for a mean of thirty-four months (range, twenty-four to fifty-four months). All patients had experienced a minimum of two episodes of anterior subluxation and/or dislocation. There were thirty-seven male and four female patients, with an average age of 22.9 years (range, fifteen to thirty-six years) at the time of surgery. Thirty-eight shoulders were in thirty-seven patients who participated in sports, including those involving contact and/or collision activities (thirty patients) and overhead activities (four patients). Fig. 1 An osseous Bankart lesion in a three-dimensionally reconstructed computed tomography image of a glenoid with the humeral head eliminated. previously to treat acute fractures of the glenoid rim 4,14,15. The purpose of this study was to evaluate prospectively the outcome of an arthroscopic osseous Bankart reconstruction for the treatment of chronic recurrent traumatic anterior glenohumeral instability, including the so-called inverted pear type of large glenoid bone deficiency, for which open bonegrafting has been recommended 13,16. Quantification of Bone Loss Bone loss was calculated in two different ways with use of three-dimensionally reconstructed computed tomography with the humeral head eliminated. In one method, the percentage was calculated as the ratio of the fragment area to the area of the assumed lower circular portion of the glenoid, as described by Sugaya et al. 11. In the other method, the percentage was calculated as the ratio of the glenoid width to the diameter of the same circle (bone loss) regardless of fragment size, which is the three-dimensionally reconstructed computed tomography representation of the method described by Lo et al. 16 and Burkhart et al. 17 (Fig. 3). The mean fragment size was 9.2% (range, 2.1% to 20.9%), and the mean glenoid bone loss was 24.8% (range, 11.4% to 38.6%). Twenty-two glenoids were measured with >25% bone loss regardless of fragment size according to the latter method. Arthroscopic Evaluation All patients were placed in the beach-chair position under general anesthesia, and joint laxity was assessed in both shoulders prior to preparation for surgery. Anteroinferior instability of the affected shoulder was confirmed in all patients. A Fig. 2 An arthroscopic view of the right shoulder, as seen through the posterior portal, shows an osseous fragment with the labroligamentous complex separated from the anterior portion of the glenoid neck. The probe was inserted through the anterior portal. G = glenoid, and B = osseous fragment.

3 1754 then created an anterior portal just superior to the subscapularis tendon as well as just lateral to the conjoined tendon using an outside-in technique, to facilitate instrument insertion without cannulae 18. Arthroscopy through the anterior portal confirmed an osseous defect of the anteroinferior quadrant of the glenoid, which was compatible with the findings on the preoperative three-dimensionally reconstructed computed tomography, in the majority of shoulders. An osseous fragment, which was covered by or embedded in the surrounding labroligamentous complex, was observed in all shoulders (Fig. 4-A). A detachment of the superior portion of the labrum was observed in twenty shoulders, and a capsular tear was observed in two shoulders. Fig. 3 The percentage of the glenoid involved by the osseous defect, regardless of the fragment size, is calculated as the glenoid defect rate with use of three-dimensionally reconstructed computed tomography on the en face view, with the equation: b/a 100%, where A = the diameter of the assumed outer fitting circle based on the inferior portion of the glenoid contour, and b = the width of the defect. 4-mm arthroscope was introduced through a standard posterior portal, and a diagnostic arthroscopy was performed. We Surgical Procedure After the cannula is removed, separation and mobilization of the labroligamentous complex together with the osseous fragment from the glenoid neck up to the six o clock position is performed with use of an elevator, rasps, scissors, shavers, and a radiofrequency instrument (VAPR; Mitek, Norwood, Massachusetts) through the anterior portal, which has no cannula (Figs. 2 and 4-B). Then, an anterosuperior portal is established as the second working portal, through an outside-in technique, at the anterosuperior portion of the rotator interval, which is right behind the biceps tendon as seen from the posterior portal. In shoulders with superior labral detachment, a lateral acromial portal, established just lateral to the mid-acromion through the muscle-tendon junction of the infraspinatus 19, is used instead of the anterosuperior portal. In total, three or four bioabsorbable suture anchors (Panalok; Mitek) loaded with number-2 permanent suture (Ethibond; Ethicon, Somerville, New Jersey) are inserted next on the edge of the glenoid with use of a drill-guide introduced through the anterior portal. Because this portal has no can- Fig. 4-A Figs. 4-A through 4-E Diagrams and arthroscopic image showing the insertion of the suture anchors in relation to the displaced osseous fragment. Fig. 4-A Illustrations showing preoperative images of an osseous Bankart lesion, with the coronal image (left) and the axial image (right) at the dotted line. The osseous fragment (dark area) is displaced medially and inferiorly and is embedded in the surrounding soft tissue.

4 1755 Fig. 4-B Fig. 4-C Fig. 4-B After separation of the osseous fragment (dark area), together with the surrounding complex from the glenoid neck, the fragment and complex can be reduced laterally and superiorly. Fig. 4-C Suture anchors are inserted on the edge of the glenoid, and all sutures are tied in a simple fashion. nula, obtaining the optimal angle to the glenoid is facilitated 18. After the first anchor insertion, the labrum adjacent to the inferior side of the osseous fragment is secured with use of a modified Caspari punch (Linvatec, Largo, Florida) loaded with a looped 2-0 nylon suture. A suture relay is performed intra-articularly 18. After completion of the suture relay, knottying following insertion of a 5-mm cannula into the anterior portal is performed with use of a self-locking sliding knot. To accomplish secure knot-tying, the complex, together with the fragment, is held upward and laterally on the glenoid surface by a grasper introduced through the accessory working portal to reduce tensile force on the suture. The next step is the suturing of the osseous fragment itself, either passing the suture through the fragment by penetrating it with use of bonepenetrating tools such as a Suture Leader (Mitek), IDEAL Suture Grasper (Mitek), or passing it around the fragment with use of a Suture Hook (Spectrum Tissue Repair System, CONMED; Linvatec) or Suture Leader (Mitek) 4,15. This procedure is facilitated when the osseous fragment is reduced and stabilized by grasping the labrum adjacent to the superior portion of the fragment with a grasper as described above during the first knot-tying. Although the number of suture anchors depended on the size and shape of the osseous fragment, normally one or two suture anchors are used in this process (Fig. 4-C). Knot-tying is performed after placing the fragment sutures. The final step is to suture the labrum adjacent to the superior side of fragment with use of an IDEAL Suture Grasper (Mitek) to augment the stability of the entire complex. Three to four suture anchors with simple sutures are used to reconstruct the entire lesion (Figs. 4-D and 4-E). Fig. 4-D The final appearance after knot-tying. In the coronal image (left) and the axial image at the dotted line (right), the displaced osseous fragment (dark area) is reduced and firmly fixed.

5 1756 In shoulders with a capsular tear, capsular repair with use of two to three side-to-side stitches with number-2 Ethibond suture (Ethicon) is performed prior to the osseous Bankart repair. Furthermore, in shoulders with a superior labral detachment, arthroscopic reattachment is performed with use of the same suture anchors and a lateral acromial portal 19 instead of the anterosuperior portal, following the anterior reconstruction. Beginning in November 2001, we managed patients with relatively high-risk shoulders, such as athletes who played contact sports, young individuals with lax ligaments, and those with a large Hill-Sachs lesion with a large osseous defect (grade III according to the system of Rowe et al. 20 ), with closure of the rotator interval as an augmentation, by suturing the superior margin of the subscapularis tendon to the superior glenohumeral ligament, with the arm held at the side and in maximum external rotation, with use of number-2 permanent sutures Fig. 4-E The actual arthroscopic image of the completed osseous Bankart repair as seen through the posterior portal. G = glenoid, and H = humeral head. Postoperative Protocol The shoulders were immobilized for three weeks with use of a sling (UltraSling II; DonJoy, Carlsbad, California). After immobilization, passive and assisted-active exercises were initiated for forward flexion and external rotation, avoiding the provocation of pain. After six weeks, patients began strengthening exercises of the rotator cuff and scapular stabilizers. Three months after the operation, they were permitted to practice noncontact sports. Full return to throwing or contact sports was allowed after six months according to each individual s functional recovery. Fig. 5-A Fig. 5-B Figs. 5-A through 5-D Case 9. Preoperative and postoperative three-dimensionally reconstructed computed tomography images of a glenoid with a fragment size of 20.9% as well as bone loss of 27.3%, showing excellent reduction and union of the fragment without step formation. Fig. 5-A Preoperative en face view. Fig. 5-B Preoperative oblique view.

6 1757 TABLE I Postoperative Range of Motion of the Shoulders* Nonaffected Side Affected Side P Value Forward flexion 178 ± ± External rotation at side 80 ± ± External rotation at 90 of abduction 97 ± ± Internal rotation behind back (thoracic vertebral level) 7.3 ± ± *The values are given as the mean and the standard deviation. Outcome Measurement All patients were assessed with use of the rating scale of Rowe et al. 8 and the University of California at Los Angeles (UCLA) scoring system 24 preoperatively and at the final evaluation, which was an average of thirty-four months (range, twentyfour to fifty-four months) postoperatively. Postoperative range of motion, including forward elevation, external rotation with the arm at the side, external rotation at 90 of abduction, and the highest reach of the thumb at the back, were evaluated bilaterally at the final evaluation. Furthermore, athletic activity was evaluated subjectively at the final evaluation. Statistical Analysis The Student t test was used to compare the difference between the preoperative and postoperative scores of the Rowe and the UCLA rating scales and the difference in range of motion between the affected and the nonaffected side. The significance level was set at p = Postoperative Imaging Study Postoperative three-dimensionally reconstructed computed tomography was performed on the patients who agreed to receive the imaging study at the final evaluation. The quality of the reduction of the fragment was graded on the basis of the position of the fragment, with use of postoperative threedimensionally reconstructed computed tomography images with the humeral head eliminated. A grade of excellent indicated that there was sufficient superior and lateral repositioning, without a step-off between the fragment and the glenoid face; good, that there was sufficient superior repositioning but insufficient lateral repositioning with a step-off apparent between the fragment and the glenoid face; fair, that there was Fig. 5-C Fig. 5-D Fig. 5-C Postoperative en face view. Fig. 5-D Postoperative oblique view.

7 1758 insufficient superior and lateral repositioning with a step-off; and poor, that the fragment was not repositioned, remaining in the original preoperative location. Osseous union was determined by the extent of reduction of the boundary line between the main fragment and the glenoid. Results Outcome Assessment he mean postoperative shoulder scores were significantly T improved at the time of follow-up (p < 0.01). The scores on the Rowe rating scales improved from a mean (and standard deviation) of 33.6 ± 6.9 points (range, 20 to 40 points) preoperatively to 94.5 ± 13.9 (range, 40 to 100 points) (p < 0.01). The scores on the University of California at Los Angeles system improved from a mean of 20.5 ± 0.9 points (range, 18 to 24 points) preoperatively to 33.6 ± 2.5 points (range, 22 to 35 points) at the final evaluation (p < 0.01). According to both the Rowe and the UCLA scoring system, the result was rated as excellent or good for thirty-nine (93%) of the fortytwo shoulders, as poor for two (5%), and as fair for one (2%). Postoperative Imaging Study Postoperative three-dimensionally reconstructed computed tomography images with the humeral head eliminated were performed to evaluate the status of the reduction and the fusion of the fragment for twelve patients who provided informed consent to receive the extra postoperative imaging study at the final follow-up examination. The studies demonstrated excellent reduction in three patients (Figs. 5-A through 5-D), good reduction in seven, and fair reduction in two. Furthermore, excellent union of the fragment was established in all of these patients (see Appendix). Postoperative radiographic images were also made for all patients; however, the studies were not helpful in evaluating the anatomic position and osseous union of the fragments. Range of Motion The average forward flexion, external rotation with the arm at the side, external rotation at 90 of abduction, and the vertebral level of internal rotation was 176, 75, 93, and T8, respectively, on the affected side and 178, 80, 97, and T7 on the healthy side (Table I). Return to Sports Among thirty-eight shoulders in thirty-seven patients who actively participated in sports preoperatively, thirty-two (84%) were in patients who had returned to their sport at the preinjury level, four (11%) were in patients who had returned with minimal restriction, and two (5%) were in patients who experienced a redislocation while participating in the sport. Eventually, the overall rate of return to sports was 95% (thirty-six of thirty-eight shoulders). Failure Two shoulders (5%) had repeated instability, and the outcome was rated as poor because of redislocation during sports after the patients had returned to activity. One patient had a redislocation three months postoperatively while playing soccer before receiving permission to return to full activity. Revision surgery revealed that suture breakage and displacement of the osseous fragment before osseous union had been achieved were the cause of the failure. Another patient had a redislocation twelve months postoperatively while playing rugby. Both of these patients underwent initial surgery early in the series, before we began to use the rotator interval closure as an augmentation for high-risk individuals as described above. Complications No patient had a postoperative infection or other problems related to surgery. Discussion rthroscopic stabilization for recurrent anterior glenohumeral instability can yield satisfactory outcomes with A careful patient selection and adequate surgical technique 13,25,26. Many authors have mentioned that shoulders with a large osseous defect on the anteroinferior portion of the glenoid are not candidates for arthroscopic stabilization because the failure rate in such patients is very high 12,13,26. Lo et al. 16 defined the large osseous defect of the glenoid that requires bone-grafting as one that is a minimum of 25% to 27% of the entire width of the inferior portion of the normal glenoid. They wrote that both an osseous Bankart lesion and an impression (compression) lesion can lead to loss of the pear-shaped glenoid configuration. However, our investigations (unpublished data) have shown that the majority of such large glenoid defects, as quantitatively defined by Lo et al. 16, almost always have an osseous fragment of variable size lying somewhere along the anteroinferior portion of the glenoid neck 27. Furthermore, our data showed that the eroded glenoids without an osseous fragment tended to have a smaller defect size than those with a related osseous fragment, when evaluated through three-dimensionally reconstructed computed tomography 28. In the present study, twenty-two of the forty-two glenoids had bone loss of >25% according to our quantification method using threedimensionally reconstructed computed tomography, which appears to be more accurate than, but similar to, the methods of arthroscopic measurement described by Burkhart and DeBeer 13 and Lo et al. 16. The patients should have been treated with an open bone-grafting procedure according to the patient selection criteria of those authors; nevertheless, the outcome in our patients was quite favorable. We believe this was because our patients had an osseous fragment at the anteroinferior portion of the glenoid neck, which can only be detected reliably with preoperative three-dimensionally reconstructed computed tomography with the humeral head eliminated 11 and because the glenoid morphology could be restored with use of our arthroscopic technique. Although postoperative three-dimensionally reconstructed computed tomography demonstrated excellent osseous union in all shoulders, the fragment position on these images was less ideal than we had expected, despite the fact that we had

8 1759 confirmed an excellent reduction of the fragment arthroscopically in all shoulders. The osseous fragment was consistently shown to be covered by or embedded in the surrounding labroligamentous complex. This surrounding soft tissue tended to be harder and less elastic than the soft tissue in shoulders without an osseous fragment, even in shoulders with a relatively small osseous fragment. Thus, these pathologic soft-tissue changes may have contributed to a discrepancy between the arthroscopic findings and the postoperative three-dimensionally reconstructed computed tomography image, which exclusively demonstrated osseous tissue. One of the concerns with regard to the use of this procedure was whether osseous union can be achieved after arthroscopic repair of a chronic lesion. In this series, every osseous fragment was firmly attached to the labroligamentous complex, was displaced, and had malunited to the glenoid neck. Therefore, a blood supply to the fragment through the surrounding soft tissue can be expected even in the chronic stage. In fact, Fujii et al. 29 examined these osseous fragments in patients with chronic traumatic instability histologically and concluded that they were viable. However, as seen in our two patients in whom the procedure failed, osseous union is not easily achieved because the bone-tissue quality on both the fragment and the glenoid side is not optimal and the time to union may be longer than that needed for fresh fractures. Therefore, we believe that increasing the initial fixation strength, with use of a stronger permanent suture or double-loaded suture anchors, as well as augmenting the repair with a rotator interval closure for highrisk patients are important measures to take to avoid failure before osseous union can be achieved in this patient population. In conclusion, arthroscopic osseous Bankart repair can provide a successful outcome in terms of recurrence rate and function in shoulders with chronic recurrent anterior glenohumeral instability with a medium-to-large osseous defect of the glenoid. This study suggests that this technique may obviate the need for open bone-grafting in patients with a large glenoid defect and an osseous fragment. Appendix A table containing clinical and radiographic data on the twelve patients who underwent a postoperative threedimensional reconstructed computed tomography scan is available with the electronic versions of this article, on our web site at jbjs.org (go to the article citation and click on Supplementary Material ) and on our quarterly CD-ROM (call our subscription department, at , to order the CD-ROM). Hiroyuki Sugaya, MD Joji Moriishi, MD Izumi Kanisawa, MD Akihiro Tsuchiya, MD Shoulder and Elbow Service, Funabashi Orthopaedic Sports Medicine Center, Hazama, Funabashi, Chiba , Japan. address for H. Sugaya: hsugaya@nifty.com The authors did not receive grants or outside funding in support of their research or preparation of this manuscript. They did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated. doi: /jbjs.d References 1. Aston JW Jr, Gregory CF. Dislocation of the shoulder with significant fracture of the glenoid. J Bone Joint Surg Am. 1973;55: Ideberg R. Fractures of the scapula involving the glenoid fossa. In: Bateman JE, Welsh RP, editors. Surgery of the shoulder. Philadelphia: BC Decker; p Bigliani LU, Newton PM, Steinmann SP, Connor PM, McIlveen SJ. Glenoid rim lesions associated with recurrent anterior dislocation of the shoulder. Am J Sports Med. 1998;26: Porcellini G, Campi F, Paladini P. Arthroscopic approach to acute bony Bankart lesion. Arthroscopy. 2002;18: Hovelius L, Eriksson K, Fredin H, Hagberg G, Hussenius A, Lind B, Thorling J, Weckstrom J. Recurrences after initial dislocation of the shoulder. Results of a prospective study of treatment. J Bone Joint Surg Am. 1983;65: Norlin R. Use of Mitek anchoring for Bankart repair: a comparative, randomized, prospective study with traditional bone sutures. J Shoulder Elbow Surg. 1994;3: Palmer I, Widen A. The bone block method for recurrent dislocation of the shoulder joint. J Bone Joint Surg Br. 1948;30: Rowe CR, Patel D, Southmayd WW. The Bankart procedure: a long-term endresult study. J Bone Joint Surg Am. 1978;60: Symeonides PP. The significance of the subscapularis muscle in the pathogenesis of recurrent anterior dislocation of the shoulder. J Bone Joint Surg Br. 1972;54: Itoi E, Lee SB, Amrami KK, Wenger DE, An KN. Quantitative assessment of classic anteroinferior bony Bankart lesions by radiography and computed tomography. Am J Sports Med. 2003;31: Sugaya H, Moriishi J, Dohi M, Kon Y, Tsuchiya A. Glenoid rim morphology in recurrent anterior glenohumeral instability. J Bone Joint Surg Am. 2003;85: Itoi E, Lee SB, Berglund LJ, Berge LL, An KN. The effect of a glenoid defect on anteroinferior stability of the shoulder after Bankart repair: a cadaveric study. J Bone Joint Surg Am. 2000;82: 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;16: Sugaya H, Kon Y, Tsuchiya A. Arthroscopic repair of glenoid fractures using suture anchors: technical note with cases series. Arthroscopy. 2005;21:635.e1-635.e Wilson F, Hinov V, Adams G. Arthroscopic repair for anterior shoulder instability with a Bigliani type I glenoid rim fracture. Arthroscopy. 2002;18:E Lo IK, Parten PM, Burkhart SS. The inverted pear glenoid: an indicator of significant glenoid bone loss. Arthroscopy. 2004;20: Burkhart SS, Debeer JF, Tehrany AM, Parten PM. Quantifying glenoid bone loss arthroscopically in shoulder instability. Arthroscopy. 2002;18: Sugaya H, Kon Y, Tsuchiya A. Arthroscopic Bankart repair in the beachchair position: a cannulaless method using an intra-articular suture relay technique. Arthroscopy. 2004;20 Suppl 2: Sugaya H, Kon Y, Matsuki K, Tsuchiya A. Arthroscopic reconstruction of SLAP lesion of the shoulder utilizing lateral acromial portal. Kansetsukyo (Arthroscopy). 2002;27: Japanese.

9 Rowe CR, Zarins B, Ciullo JV. Recurrent anterior dislocation of the shoulder after surgical repair. Apparent causes of failure and treatment. J Bone Joint Surg Am. 1984;66: Karas SG. Arthroscopic rotator interval repair and anterior portal closure: an alternative technique. Arthroscopy. 2002;18: Treacy SH, Field LD, Savoie FH. Rotator interval capsule closure: an arthroscopic technique. Arthroscopy. 1997;13: Gartsman GM, Taverna E, Hammerman SM. Arthroscopic rotator interval repair in glenohumeral instability: description of an operative technique. Arthroscopy. 1999;15: Ellman H, Hanker G, Bayer M. Repair of the rotator cuff. End-result study of factors influencing reconstruction. J Bone Joint Surg Am. 1986;68: Gartsman GM, Roddey TS, Hammerman SM. Arthroscopic treatment of anterior-inferior glenohumeral instability. Two to five-year follow-up. J Bone Joint Surg Am. 2000;82: Kim SH, Ha KI, Cho YB, Ryu BD, Oh I. Arthroscopic anterior stabilization of the shoulder: two to six-year follow-up. J Bone Joint Surg Am. 2003;85: Sugaya H, Maeda K, Moriishi J, Kanisawa I, Tsuchiya A. The Inverted-Pear glenoid in recurrent anterior glenohumeral instability. Presented at the Annual Meeting of the Arthroscopy Association of North America, 2005 May 12-15, Vancouver, British Columbia, Canada. 28. Sugaya H, Maeda K, Moriishi J. Osseous morphology in recurrent anterior glenohumeral instability in relation to inherent joint laxity. Presented at the 9th International Congress on Surgery of the Shoulder Meeting; 2004 May 2-5, Washington, DC. 29. Fujii Y, Yoneda M, Miyazaki Y, Obata M, Wakitani S. Histological study of bony Bankart lesion in recurrent dislocation and subluxation of the shoulder. The Shoulder Joint (Katakansetsu). 2000;24: Japanese.