Printer Friendly

Management of humeral and glenoid bone loss associated with glenohumeral instability: results with anatomical bone grafting.

Glenohumeral instability is a relatively common problem and may be complicated by either humeral or glenoid bone loss, or both. Recent studies using advanced imaging techniques, such as three-dimensional (3D) computed tomography (CT) reconstruction and magnetic resonance imaging (MRI), suggest that glenoid bone loss occurs in up to 90% of cases of traumatic anterior glenohumeral instability (1); in terms of presentation, in up to 76% of first-time traumatic anterior glenohumeral instability and in up to 100% with recurrent instability. (2,3) Rarely, however, is the bone loss on either the glenoid side or humeral head side significant enough to warrant bone grafting procedures. Warner and colleagues (4) reported on their experience in a tertiary referral shoulder practice and noted that only 4% of patients undergoing reconstructive procedures for traumatic anterior instability had defects large enough to necessitate bone graft reconstruction of the glenoid rim. Humeral-sided grafting is even more rare. A few papers outline treatment strategies for large reverse Hill-Sachs lesions acquired from locked posterior dislocations. (5,6) However, there are only small case reports and series of bone grafting procedures for large classic Hill-Sachs lesions. Most patients in either group who require grafting have a history of multiple dislocations or failed repairs, or both.

Burkhart and De Beer (7) were the first to coin the terms "inverted pear glenoid" and "engaging Hill-Sachs lesion." Their work drew attention to the essential role of bone loss in instability and fueled needed and further clinical research into this topic. (8) Recent biomechanical and observational (9) work has better defined the "critical" size defects on both the glenoid and humeral head. (10-13) Advanced imaging is now extremely helpful in preoperative planning and, in most cases, can accurately predict the need for grafting procedures when bone loss is severe. In general, glenoid defects in the 33% range and Hill-Sachs defects that either engage in a physiological range of motion or are in the 30% to 40% range should be considered for grafting. (5,14)

Historically, bone grafting for instability reconstruction has been performed most commonly via the Bristow-Latarjet procedure; there is excellent long-term data supporting this procedure. (15-17) However, there are some notable shortcomings to both the procedure and the reported data. The procedure is technically challenging and produces a non-anatomical repair and a complicated revision surgery if failure occurs. Furthermore, the data is from a nonhomogenous population, in that surgeons who report on this procedure often indicate it for cases of mild or moderate bone loss, not just severe.

Direct bone grafting of the anterior glenoid margin with autograft or allograft iliac crest has recently been advocated by some as a more anatomical approach. (4) It shares some of the same limitations as the Bristow-Latarjet procedure, such as potential hardware complications and technical difficulty. Clinical results have been reported sparingly, as indications for the procedure remain limited; however, early results seem promising.


In addition, direct bone grafting for large Hill-Sachs lesions with allograft humeral head or femoral head has recently been advocated for "engaging Hill-Sachs lesions" as well. There are few published clinical reports of this strategy. (18,19) Multiple surgical approaches have been employed: arthroscopic, anterior deltopectoral only, combined arthroscopic, and posterior.

We report on a small series of cases from 2002-2008 at our institution of anterior glenohumeral instability, with severe bone loss, treated using either open femoral head and humeral head allograft reconstruction of the humeral head for large Hill-Sachs lesions, or open allograft and autograft iliac crest reconstruction of the glenoid rim for severe glenoid bone loss.

Materials and Methods Patient Selection

The institution's medical records from 2002-2008 were searched and all instability patients (humeral head defect and glenoid defect patients) were identified. Six patients were identified who were operated on by two surgeons and underwent iliac crest bone grafting of the glenoid and humeral head for severe bone loss due to instability. All patients had multiple dislocations (Table 1). All patients also demonstrated severe apprehension and a positive relocation test on physical examination. Bone loss was documented preoperatively in all cases by means of 3D CT reconstruction or MRI, or both, using the method of Gerber and Nyfeller, as documented previously by Warner and coworkers (4) (Fig. 1). Bone loss was confirmed in the operating room in some of the cases by means of arthroscopy, using the method of Burkhart to measure the glenoid width. (7)


For humeral head defects, patients were identified by MRI or CT as having "large Hill-Sachs" lesions (33% to 40% of articular arc) (Fig. 2) and "engaging Hill-Sachs" lesions at the time of the preoperative examination, under anesthesia. Two different surgical techniques were employed. Two surgeons used an open bone grafting technique with allograft femoral head, and one surgeon used an arthroscopic technique with osteochondral plugs.

Surgical Procedure: Glenoid Bone Grafting

All patients underwent open reconstruction of the glenoid bone defect with either autogenous iliac crest bone graft or fresh frozen allograft iliac crest graft, and procedures were reviewed. In two of the cases, open reconstruction was preceded by an arthroscopic evaluation and measurement of the glenoid defect. Patients underwent regional anesthesia in all cases, followed by sedation with laryngeal mask airway (LMA) support. Patients were placed in the beach chair position in a "Captain's Chair" table attachment, positioned with the head inclined 30[degrees] (referenced from the floor), and all bony prominences padded appropriately. The contralateral iliac crest was then prepped. A similar method described by Warner and associates (4) was used for iliac crest harvest and preparation.

For the open shoulder reconstruction, a standard deltopectoral approach was used. An incision was made from the palpable landmark of the coracoid process to the skin overlying the deltoid insertion. The cephalic vein was preserved and retracted laterally, and the subdeltoid and subconjoint tendon spaces were mobilized. A Kobel retractor was placed in the subdeltoid and subconjoint tendon space, exposing the subscapularis and rotator interval. The circumflex vessels were cauterized. A tenotomy of the subscapularis was made approximately 1 cm medial to its origin, superiorly, from the rotator interval to the inferior one-third of the subscapularis. A Cobb elevator was inserted between the muscular subscapularis and the glenohumeral capsule, thus, protecting the axillary nerve. The subscapularis tendon was gently dissected from the capsule and controlled with nonabsorbable No. 2 Merseline[TM] retention sutures. A T-shaped capsulotomy was performed, with the lateral margin, approximately, 0.5 cm medial to the subscapularis tenotomy. The capsulotomy was taken to the 6 o'clock position on the humeral neck.

The glenohumeral joint was exposed, and the humeral head was retracted posterolaterally with a Fukuda retractor. The anterior glenoid margin was identified, and the capsule was bluntly dissected with an elevator from the scapular neck. A curved anterior single-spike retractor was placed along the anterior scapular neck. The glenoid width was measured and, in all cases, the defect measured. The glenoid defect then was roughened with a high speed burr.

The iliac crest graft was prepared using an oscillating saw and high speed burr. The graft was contoured to match the glenoid curvature. The graft was preliminarily fixed in place with three K-wires, placed parallel and medial to the glenoid surface. Care was taken to place the graft in an anatomical position. Three 4.0 mm partially threaded cannulated screws (Synthes[R], Paoli, Pennsylvania) were placed over the K-wires. Prior to fully seating the screws, a No. 2 FiberWire[R] (Arthrex[R], Naples, Florida) suture was tied under each screw head, to be used later to repair the capsule. The graft was fully seated and the articular surface was assessed for a smooth transition. A burr was used to contour the graft further, if needed. The Fukuda retractor was removed, and the glenohumeral articulation was assessed. The No. 2 FiberWire[R] sutures were placed in a horizontal mattress fashion through the capsule, and the capsule was secured to the edge of the iliac crest bone graft. The capsulotomy was repaired with interrupted non-absorbable No. 2 Merseline[TM] sutures. During the repair, the arm was held in 30[degrees] of flexion and 30[degrees] of external rotation. The subscapularis tenotomy and rotator interval were repaired in a similar fashion. The remainder of the closure was performed in a layered manner, and a sterile dressing was placed.

Humeral Head Bone Graft Technique: Open Technique

For the open shoulder reconstruction, a standard deltopectoral approach was used (see previous section for details). Following completion of the deltopectoral approach and joint exposure, the humeral head was externally rotated, a retractor was placed in the joint to protect the glenoid surface, and the Hill-Sachs lesion was exposed. The lesion was roughened with a high speed burr back to a bleeding bone bed. The graft was obtained from the back table, shaped accordingly and placed with two headless, buried Acutrak[R] (Acumed,[R] Hillsboro, Oregon) screws. The graft is placed about 2 mm proud to allow for expected subsidence. Stability was checked during a full range of motion evaluation, and a standard capsular shift with labral repair was performed, if warranted. A layered closure was performed.

Humeral Head Bone Graft Technique: Arthroscopic Technique

Two patients underwent arthroscopic osteochondral allograft transplant (OATS). The technique was performed in a fashion similar to that used by Chapovsky. (19) In each case, the Bankart bony bed was debrided with a rasp and the Arthrex[R] OATS instrument set was employed. Three osteochondral plugs EBI[R] were fit into the defect and fit flush to the surface. The labral reattachment was performed via arthroscopic techniques with 3 suture anchors in the anterior inferior quadrant of the anterior lateral glenoid rim.

Postoperative Care

Patients were maintained in an abduction sling for 2 weeks postoperatively and were encouraged to wear the sling at night and while out of the house until week 4. They began a monitored therapy program at 2 weeks, starting with gentle passive range of motion and working-up to active assisted and full active range of motion. At 3 months, strengthening was begun.


Patients were seen in the office postoperatively at 2 weeks, 6 weeks, 3 months, 6 months, and 1 year. Functional data using the ASES and UCLA validated instruments were collected during the office visit or via telephone interviews. All patients were asked to return to the office for a physical examination. Both the normal and affected shoulders of patients were examined for range of motion, strength, apprehension, relocation testing, and translation. Range of motion testing was done with a goniometer, and strength testing was performed manually. At a follow-up visit, radiographs of the affected extremity were obtained for anterior-posterior (AP) internal rotation, AP external rotation, axillary, and scapular Y-planes.



Six patients were identified who underwent bone graft reconstruction of the glenoid; four were available for full follow-up. The average age of these patients at the time of surgery was 26 years; all patients were male. Two of the patients had previously failed surgery, both of which were arthroscopic Bankart repairs. The dominant arm was affected in 3/6 cases. The average follow-up was 30.6 months. Five patients were identified who underwent allograft reconstruction of large Hill-Sachs lesions (1 female and 4 males). The patients were operated on by three surgeons. Four patients were available for full follow-up at an average of 27.4 months. One patient had a failed previous Putti-Platt procedure. The dominant arm was effected in 3/5 cases. Average age of the patients was 33 years (Table 1).

Physical Examination and Operative Findings

Operative findings correlated well with preoperative clinical history, physical examination, and imaging studies. In all cases, a severe anterior glenoid defect was noted, comprising greater than one-third of the width of the glenoid. All patients were found to have 3+ humeral head laxity, with frank dislocation under anesthesia through a physiological range of motion. In all cases, an engaging Hill-Sachs lesion was noted on preoperative exam under anesthesia; preoperative and intraoperative direct observation demonstrated that the Hill-Sachs lesions were large (approximately 40% of the articular surface).

Clinical Follow-Up

All 11 patients were available for limited phone follow-up, and there were no reported recurrences of subluxation or dislocation. Results for full follow-ups are listed below.

Glenoid Defect Group

Average postoperative ASES score was 91.3 (range, 90.6 to 96.1) and average postoperative UCLA score was 33 (range, 31 to 35), with a follow-up, on average, of 39.2 months (range, 7 to 63). Average loss of forward flexion, when compared to the normal side, was 3[degrees] (range, 0 to 10), average loss of external rotation was 10[degrees] (range, 0 to 20). All patients demonstrated full strength. No patients demonstrated any recurrent instability. Subscapularis testing was intact for all patients. No complications or reoperations were noted. All grafts remained in place, with no evidence of prominent hardware. Radiographs did not demonstrate any appreciable resorption. Measures were made using digital radiography with standard calibrating software.

Hill-Sachs Group

Average postoperative ASES score was 85.3 (range, 61 to 94.5) and average postoperative UCLA was 28.4 (range, 23 to 31). Follow-up, on average, was 27.4 months (range, 11.2 to 40.1). Average loss of forward flexion, when compared to the normal side, was 23[degrees] (range, 0[degrees] to 70[degrees]), average loss of external rotation was 8[degrees] (range, 5[degrees] to 10[degrees]), and average loss of internal rotation was two levels (0 to 3). All patients demonstrated full strength. No patients demonstrated any recurrent instability. Subscapularis testing was intact for all patients. One patient developed reflex sympathetic dystrophy (RSD), and one patient developed prominent hardware that required removal. This was the only case of reoperation.


Great strides have been made in biomechanically and clinically defining the size of glenoid and humeral head defects likely to progress to recurrent instability. Controversy still exists, however, regarding the best treatment plan for a given defect.

In an early biomechanical study, Lippitt and Matsen (9) elegantly demonstrated that the labrum significantly deepens the glenoid and aids in preventing dislocation. They showed a decreased translational force to dislocation when the labrum was removed from the edge of the glenoid. Itoi and colleagues (12) created standardized bone cuts to simulate graduated bone loss in a similar cadaver model. They found that when the missing bone was equal to 21% of the length of the glenoid, significantly less translational force was needed to dislocate the shoulder. Restoration of bone and the Bankart lesion restored stability. Of note, Itoi and coworkers created a model in which the bone loss was created at an oblique 45[degrees] angle to the superior inferior axis of the glenoid. Saito and associates (20) later work with CT reconstruction would show that our perception of this bone loss pattern was slightly off. The orientation of bone loss in the general population of anterior shoulder dislocators tends to be parallel with the superior-inferior axis of the glenoid. Sugaya showed via CT with 3D reconstruction that large defects were rare. (1)

Due to the 2000 study findings of Burkhart and colleagues, (7) in which they described the inverted pear glenoid, Burkhart modified his practice and now performs a Bristow-Latarjet procedure for such patients. In 2007, Burkhart and coworkers (21) followed-up on 102 patients and found good and excellent results for a majority of patients, with a recurrence rate of 4.9%. Other long-term studies have touted the Bristow-Latarjet as an effective procedure for the treatment of instability. Hovelius and his colleagues (15-17) have produced some of the best follow-up for this procedure. However, these series included all patients with instability, not just those with significant bone loss. This makes it difficult to draw conclusions when comparing the Bristow-Latarjet procedure to the anatomical approach. Proponents of the Bristow-Latarjet cite the long track record and similar types of complications as the anatomical approach. A recent biomechanical study suggested that the Bristow-Latarjet procedure may provide greater resistance to anterior translation than the anatomical glenoid reconstruction with iliac crest graft provides, although it is not clear whether this is clinically significant. (22) It remains unclear which approach is superior for this difficult group of patients. Warner and associates (4) series of 11 patients, who underwent autograft iliac crest grafting of the anterior glenoid, showed a high success rate, excellent graft incorporation, and a low complication rate.

We reviewed six of our patients who underwent anatomical bone grafting with iliac crest for large glenoid defects. Our results compare favorably to Warner's 2006 case series from a clinical outcomes standpoint. (4) We report no complications, no recurrences, and favorable clinical outcomes. Although we did not obtain CT scans on all postoperative patients, radiographic studies did not demonstrate any evidence of hardware prominence or significant graft dissolution. In distinction to Warner and colleagues, (4) we have often employed an iliac crest allograft instead of an autograft. It is unknown whether and to what extent allograft incorporates in this application. The spine literature shows us, however, that autograft iliac crest harvest leads to chronic donor site pain for a subset of these patients. Therefore, this technique is certainly worth considering. While there is no long-term data on this approach, we believe it is a viable and promising technique.

The Hill-Sachs defect group, overall, performed well with regard to instability: no cases of recurrence were noted. However, the clinical outcomes did not compare as favorably as the glenoid group. Part of this can be explained by the one patient who developed RSD and significant motion loss. Since the sample size was small, one patient could have a relatively large effect on the general group outcome. Regardless, it seems that the motion loss for these patients is somewhat greater, overall, than for patients with glenoid grafting. It is unclear why this may be so. There did not appear to be an appreciable difference in the number of previous surgeries or dislocations between groups. Of the patients who underwent bone grafting of the humeral head, those who underwent arthroscopic grafting had slightly higher outcome scores. However, our numbers are too small too draw reasonable conclusions about this finding.

This study has some limitations. It is retrospective in nature, and outcome scores were not kept preoperatively. Due to the rare nature of the procedures involved, the overall sample size for each group was relatively small; therefore, statistical analysis was not possible. Additionally there was more than one surgeon involved in performing the surgical procedures. While we believe that technique was fairly uniform across surgeons for the various approaches, this cannot be proven. A conventional goniometer was used for measurements; however, no standardization or intra-observer ratings were done.


Overall, bone grafting of large defects of the humeral head or glenoid can prevent recurrent instability in the short- to medium-term, with relatively low complication rates. Glenoid graft patients seem to have better clinical functioning than humeral head graft patients. Further study is needed to determine long-term efficacy of these procedures and to enable better conclusions to be drawn regarding their effectiveness and long-term outcome.

Disclosure Statement

None of the authors have a financial or proprietary interest in the subject matter or materials discussed, including, but not limited to, employment, consultancies, stock ownership, honoraria, and paid expert testimony.


(1.) Sugaya H, Moriishi J, Dohi M, et al. Glenoid rim morphology in recurrent anterior glenohumeral instability. J Bone Joint Surg Am. 2003 May;85(5):878-84.

(2.) Taylor DC, Arciero RA. Pathologic changes associated with shoulder dislocations. Arthroscopic and physical examination findings in first-time, traumatic anterior dislocations. Am J Sports Med. 1997 May-Jun;25(3):306-11.

(3.) Rowe CR, Zarins B. Recurrent transient subluxation of the shoulder. J Bone Joint Surg Am. 1981 Jul;63(6):863-72.

(4.) Warner JJ, Gill TJ, O'Hollerhan J D, et al. Anatomical glenoid reconstruction for recurrent anterior glenohumeral instability with glenoid deficiency using an autogenous tricortical iliac crest bone graft. Am J Sports Med. 2006 Feb;34(2):205-12.

(5.) Gerber C, Lambert SM. Allograft reconstruction of segmental defects of the humeral head for the treatment of chronic locked posterior dislocation of the shoulder. J Bone Joint Surg Am. 1996 Mar;78(3):376-82.

(6.) McLaughlin HL. Posterior dislocation of the shoulder. J Bone Joint Surg Am. 1952 Jul;243:584-90.

(7.) 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.

(8.) Mologne TS, Provencher MT, Menzel KA, et al. Arthroscopic stabilization in patients with an inverted pear glenoid: results in patients with bone loss of the anterior glenoid. Am J Sports Med. Aug 2007;35(8):1276-83.

(9.) Lippitt S, Matsen F. Mechanisms of glenohumeral joint stability. Clin Orthop Relat Res. 1993 Jun;291:20-8.

(10.) Montgomery WH Jr, Wahl M, Hettrich C, et al. Anteroinferior bone-grafting can restore stability in osseous glenoid defects. J Bone Joint Surg Am. 2005 Sep;87(9):1972-7.

(11.) Itoi E, Lee SB, Amrami KK, et al. Quantitative assessment of classic anteroinferior bony Bankart lesions by radiography and computed tomography. Am J Sports Med. 2003 Jan Feb;31(1):112-8.

(12.) Itoi E, Lee SB, Berglund LJ, et al. The effect of a glenoid defect on anteroinferior stability of the shoulder after Bankart repair: a cadaveric study. J Bone Joint Surg Am. 2000 Jan;82(1):35-46.

(13.) Greis PE, Scuderi MG, Mohr A, et al. Glenohumeral articular contact areas and pressures following labral and osseous injury to the anteroinferior quadrant of the glenoid. J Shoulder Elbow Surg. 2002 Sep-Oct;11(5):442-51.

(14.) Chen AL, Hunt SA, Hawkins RJ, Zuckerman JD. Management of bone loss associated with recurrent anterior glenohumeral instability. Am J Sports Med. 2005 Jun;33(6):912-25.

(15.) Hovelius L, Sandstrom B, Sundgren K, Saebo M. One hundred eighteen Bristow-Latarjet repairs for recurrent anterior dislocation of the shoulder prospectively followed for fifteen years: study I-clinical results. J Shoulder Elbow Surg. 2004 Sep-Oct;13(5):509-16.

(16.) Hovelius L, Sandstrom B, Saebo M. One hundred eighteen Bristow-Latarjet repairs for recurrent anterior dislocation of the shoulder prospectively followed for fifteen years: study II-the evolution of dislocation arthropathy. J Shoulder Elbow Surg. 2006 May-Jun;15(3):279-89.

(17.) Hovelius LK, Sandstrom BC, Rosmark DL, et al. Long-term results with the Bankart and Bristow-Latarjet procedures: recurrent shoulder instability and arthropathy. J Shoulder Elbow Surg. 2001 Sep-Oct;10(5):445-52.

(18.) Kropf EJ, Sekiya JK. Osteoarticular allograft transplantation for large humeral head defects in glenohumeral instability. Arthroscopy. 2007 Mar;23(3):322.e1-5; Epub 2006 Nov 27.

(19.) Chapovsky F, Kelly JD 4th. Osteochondral allograft transplantation for treatment of glenohumeral instability. Arthroscopy. 2005 Aug;21(8):1007.

(20.) Saito H, Itoi E, Sugaya H, et al. Location of the glenoid defect in shoulders with recurrent anterior dislocation. Am J Sports Med. Jun 2005;33(6):889-93.

(21.) Burkhart SS, De Beer JF, Barth JR, et al. Results of modified Latarjet reconstruction in patients with anteroinferior instability and significant bone loss. Arthroscopy. 2007 Oct;23(10):1033-41.

(22.) Wellmann M, Petersen W, Zantop T, et al. Open shoulder repair of osseous glenoid defects: biomechanical effectiveness of the Latarjet procedure versus a contoured structural bone graft. Am J Sports Med. 2009 Jan;37(1):87-94.

Matthew DiPaola, M.D., is Assistant Professor, Wright State University, Dayton VA Medical Center, Dayton, Ohio. Laith Jazrawi, M.D., is Assistant Professor, Andrew Rokito, M.D., is Assistant Professor; Young Kwon, M.D., Ph.D., is Assistant Professor; and Joseph D. Zuckerman, M.D., is Walter A.L. Thompson Professor of Orthopaedic Surgery, New York University School of Medicine; and Lava Patel, and Brian Pahk, were Research Assistants, within the Department of Orthopaedic surgery, at NYU Hospital for Joint Diseases, NYU Langone Medical Center, New York, New York. Correspondence: Matthew DiPaola M.D., 6th floor, Department of Orthopedics, Dayton VA Medical Center, 4100 W. 3rd Street, Dayton, Ohio 45428;

DiPaola MJ, Jazrawi LM, Rokito AS, Kwon YW, Patel L, Pahk B, Zuckerman JD. Management of humeral and glenoid bone loss associated with glenohumeral instability: results with anatomical bone grafting. Bull NYU Hosp Jt Dis. 2010;68(4):245-50.
Table 1 Patient Demographics and Outcome Scores

Humeral Head     Age                              Number of
Patients       (Years)   Gender      Injury      Dislocations

1                46        F          Fall            6
2                36        M        Sailing           3
3                32        M      Martial arts        2
4                19        M        Seizure           3


1                16        M       Basketball       > 100
2                46        M          Fall            30

3                28        M          Fall            3
4                24        M       Wrestling         > 5

Humeral Head                   ASES    UCLA
Patients       Prior Surgery   Score   Score

1               Putti-Platt     61      23
2                  None        94.5     34
3                  None        78.1     31
4                  None         67      24


1                  None        96.1     23
2              Arthroscopic    95.1     34
3                  None        90.6     31
4                  None        93.5     24
COPYRIGHT 2010 J. Michael Ryan Publishing Co.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:DiPaola, Matthew J.; Jazrawi, Laith M.; Rokito, Andrew S.; Kwon, Young W.; Patel, Lava; Pahk, Brian;
Publication:Bulletin of the NYU Hospital for Joint Diseases
Article Type:Report
Date:Oct 1, 2010
Previous Article:Rheumatoid arthritis classification criteria: it's finally time to move on!
Next Article:Predictors of end stage renal disease in African Americans with lupus nephritis.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters