Young, Allan A. MD, MBBS, MSpMed, PhD, FRACS (Orth)*; Neyton, Lionel MD†; Molony, Diarmuid C. MD*; Boileau, Pascal MD‡; Walch, Gilles MD†
Glenoid loosening is the most frequent complication after total shoulder arthroplasty, with reported rates of up to 44%,1–7 and is also the major cause of revision with reported rates of up to 10% at 10 years.1,3,4,7–10 When revision surgery is necessary, management of glenoid bone loss can be challenging. A number of different options are available and have been described for glenoid reconstruction.9,11–14 The use of tricortical iliac crest bone graft has been proposed,15,16 main advantage being that it provides structural resistance to the prosthetic humeral head and, thereby, decreases the risk of progressive medial erosion. In cases with a large glenoid cavitary defect or a combined cavitary and segmental wall defect, even with implantation of a large structural tricortical graft and cancellous bone chips, we have observed that there are often persistent voids behind and around the bone graft. The use of synthetic bone substitutes is becoming increasingly popular in trauma and orthopedic surgery.17–21 Numerous products are currently available containing calcium phosphate, calcium sulfate, hydroxyapetite, dicalcium and tricalcium phosphate, or bioactive glass, or combinations of these materials.22 Our hypothesis was that the use of an injectable calcium phosphate bone cement [Graftys HBS (hardening bone substitute), Aix En Provence, France] would be ideal in filling the voids around and behind the iliac crest tricortical graft and also provide additional primary stability to the graft construct, thereby, assisting with bone healing.
MATERIALS AND METHODS
Between February and September 2009, 6 patients (5 male and 1 female) were operated on for aseptic glenoid loosening. The average patient age at the time of revision surgery was 66.3 years (range, 54 to 75 y). The initial total shoulder arthroplasty was performed in 4 cases for posttraumatic arthritis (3 of the patients had previously undergone 2, 3, and 4 surgeries, respectively, to treat either the fracture or fracture sequalae), and the mean time to revision surgery for glenoid loosening in these patients was 10 years. The initial TSA was performed in 2 cases for primary osteoarthritis (none of these patients had undergone prior surgery), and the mean time to revision surgery for glenoid loosening in these patients was 9 years. All of the initial TSA were performed using the third generation Aequalis shoulder arthroplasty system (Tornier, Montbonnot, France), with a polyethylene keeled glenoid component anchored with high viscosity cement.
Institutional Review Board approval was obtained for conducting this study.
Patients present with symptomatic aseptic glenoid loosening associated with significant glenoid bone deficiencies (cavitary, segmental, or combined) precluding immediate reimplantation of a glenoid component. In patients with associated rotator cuff deficiency, and insufficient native glenoid to provide adequate baseplate fixation with a reverse prosthesis, this technique is also useful to reconstitute the glenoid bone stock before second-stage implantation of a reverse shoulder arthroplasty.
The patients were initially positioned supine for harvesting of the iliac crest graft. A folded towel was placed under the ipsilateral buttock to assist with positioning by elevating the iliac crest. An incision was made directly over the iliac crest in the region of the iliac tubercle, and the dissection continued to the iliac crest. The periosteum and muscular attachments to the inner and outer tables of the iliac crest were elevated with the use of an osteotome, thereby, keeping some bone flakes attached with the soft tissues to assist with repair and bony reconstitution of the iliac crest defect. An oscillating saw was used to harvest a tricortical graft, approximately 4 cm in length, from the widest part of the iliac crest, that is, tubercle. A curette was used to harvest additional cancellous bone chips. A drain was placed, and the wound was closed in layers.
The patient was then repositioned in a beach chair position. A deltopectoral approach was performed in all cases using the previous skin incision. The axillary nerve was identified and preserved in all cases. A subscapularis tenotomy was performed and repaired at the end of the procedure with nonabsorbable sutures, except in 1 case in which there was a deficient subscapularis noted at the time of surgery. Multiple deep tissue samples were taken for routine histologic and microbiological examination. The humeral head was removed to assist with glenoid exposure and a new humeral head, typically of the same size, was implanted at the conclusion of the procedure. The glenoid was exposed with careful soft tissue dissection. The loose glenoid component was removed and cement was then carefully cleared from the glenoid with the use of curettes. The glenoid vault was cleaned of any soft tissues, to completely expose the bony margins and reveal the extent of the cavitary and wall defects. A 2-mm drill was used to create multiple holes in the base of the glenoid vault to assist with union of the bone graft. The cancellous graft was impacted into the glenoid defect followed by insertion of the tricortical iliac crest graft, always aiming to place the graft in the superior most aspect of the glenoid vault. The tricortical graft was then impacted firmly with the use of a bone tamp and the stability of the graft checked manually. In 3 cases there were large combined cavitary and wall defects, and therefore, the tricortical iliac crest craft was additionally stabilized with 2 resorbable screws (Phusis, Saint-Ismier, France). Graftys HBS was then injected into the remaining voids around the bone graft (Fig. 1).
One patient has subsequently undergone a second-stage implantation of a reverse total shoulder arthroplasty, performed at 7 months after the bone grafting surgery (Fig. 2). In this case, during the initial grafting procedure the humeral component was removed and none were reimplanted.
Patients were immobilized in a simple sling for 6 weeks. Immediate active wrist and elbow exercises were commenced, combined with pendulum shoulder exercises. At 6 weeks after surgery, active-assisted shoulder exercises were commenced and progressive use of the arm for gentle activities of daily living was allowed. Normal activities of daily living were permitted at 3 months after surgery.
Clinical and Radiographic Assessment
All patients were entered into a database. Preoperative and postoperative Constant scores, range-of-motion assessment, and subjective outcome measures were recorded. All patients were followed at 3, 6, and 12 months after surgery. The preoperative and postoperative radiographs, taken at 3, 6, and 12 months, were available and analyzed in terms of: radiographic evolution of the graft construct; healing of the structural and cancellous graft; erosion of the graft; and absence, partial, or complete resorption of the bone graft (Figs. 3, 4). We specifically observed for any potential osteolysis related to the use of the Graftys HBS, evidenced by either massive bone loss or limited defects, that is, radiolucent line around the structural bone graft. In addition, we were able to perform computed tomographic scans in 5 cases at 1-year follow-up.
Preoperative and postoperative Constant scores were compared using the Wilcoxon matched-pairs signed-ranks test. The level of statistical significance was set as a P value <0.05.
The mean follow-up in this series was 15.3 months (range, 12 to 18 mo) (Table 1). The average Constant score improved significantly (P<0.05) from 36.8 points (18 to 50) preoperatively to 59.8 points (55 to 67) postoperatively. The average gain in Constant score, therefore, after surgery was 23.0 points (8 to 49). The average active forward elevation after surgery was found to be 110 degrees (90 to 140). Subjectively, 5 patients were satisfied and 1 was disappointed.
In all cases, the stability and integrity of the construct was maintained in the early postoperative period with no evidence of change in position or collapse of the bone graft. Postoperative radiographs and computed tomographic scans showed that there was always a good integration of the Graftys bone cement with no lucent lines either between the glenoid bone and bone cement or between the graft (cancellous and structural) and bone cement. In 5 cases, the density of the graft appeared homogenous throughout the first year and was associated with complete healing of the graft. In 1 case, the graft did not heal and was observed to have undergone some degrees of resorption at 12 months postoperatively, in addition to severe erosion of the graft. This case (patient 1) still has ongoing shoulder pain, and although there is no objective evidence of infection, the clinical and radiologic failure could be related to a low-grade infection.
As mentioned earlier, 1 patient underwent second-stage reverse shoulder arthroplasty at 7 months after the glenoid reconstruction. A reverse prosthesis was implanted because of an associated rotator cuff deficiency observed at the time of the glenoid grafting procedure (supraspinatus and subscapularis). During the second-stage procedure, a biopsy of the neoglenoid was performed. Histologic analysis showed excellent osteointegration of Graftys HBS (Fig. 5). The iliac crest bone graft was seen to be well consolidated, solid, and dense. A very close intertwining was observed between Graftys HBS, in the process of degradation, and new bone formation. There was complete integration of Graftys HBS into the bone collagen matrix. This result confirmed the structural integrity of the graft construct and the beginning of resorption of Graftys HBS.
In the 3 cases, in which the tricortical graft was stabilized with bioresorbable screws, additional Graftys HBS was injected into the screw holes before insertion of screws. Postoperative radiographs showed extrusion of bone cement into the supraspinatus fossa. At 1-year follow-up, the bone cement was still present and appeared unchanged (Fig. 4).
When a glenoid component fails and becomes loose, cavitary defects are often created in an anatomically relatively small bone volume. When these defects become severe or are combined with defects of the thin cortical vault walls, immediate reimplantation of a new glenoid component may not be possible. The surgeon must decide to either simply remove the glenoid component3,9,10,14 or perform reconstruction of the glenoid with bone graft.11,12,15,16 On the basis of our earlier experience, our preferred technique for management of glenoid bone loss is tricortical iliac crest structural bone graft reconstruction.16 Our goal with glenoid reconstruction is to provide a solid articulating surface area and a fulcrum for the prosthetic humeral head. We believe that the ideal reconstruction technique should, therefore, result in a graft construct that offers sufficient primary resistance, heals to the native glenoid, avoids secondary subsidence, and resists medialization of the humeral head secondary to erosion. Moreover, the technique should allow either immediate or secondary glenoid component reimplantation once the bone stock has been reconstituted.
Our hypothesis was that injection of resorbable cement in conjunction with structural autograft would enhance the primary stability of the graft construct, while allowing healing of the structural graft to the native glenoid. In this series, we did not observe any change in position of the graft, and healing was observed in all but one case. In cases with intact glenoid vaults, that is, cavitary defects, the graft was impacted into the central defect, whereas in cases of combined central and peripheral defects, the graft needed to be secured to the remaining glenoid with absorbable screws. In both of these situations, it was impossible to contour the tricortical iliac crest graft exactly to match the glenoid defect, and the injection of the resorbable cement around the graft enabled us to fill the voids. We believe that the addition of injectable cement into these voids increased the primary stability of the graft construct and, potentially, enhanced the osteointegration and healing of the graft.
In our small series, it seemed that the injectable Graftys cement was very well tolerated. We did not observe any inflammatory reaction, either clinically or in the histologic analysis. Furthermore, sufficient numbers of bone marrow cells were seen within the graft construct and adjacent to the cement, confirming the biocompatibility of the biomaterial.23 No fibrous interface was observed between the cement and bone trabeculae. At 7 months, the histology showed partial resorption of the cement.
It has been shown that the use of tricortical structural iliac crest bone graft is better at resisting secondary subsidence or humeral head migration than allograft reconstruction9–13 or compaction of cancellous bone chips.15 In this series, we observed 1 case of secondary partial or incomplete resorption that could have been related to an unproven low-grade infection. Nonetheless, the goal of the resorbable cement was not to enhance the resistance of the bone graft against humeral head medialization, which is provided by the tricortical iliac crest structural graft, and we, therefore, do not consider this case as a failure of the cement.
We have found that this technique can be applied to the majority of glenoid deficiencies encountered at revision surgery that requiring grafting, either cavitary or combined. We specifically recommend that the graft be placed at the most superior aspect of the glenoid vault and impacted in a medial and superior direction. We recommend the use of additional screw fixation in those cases in which primary stability of the tricortical iliac crest graft cannot be obtained after the impaction of the graft into the defect. In those cases, in which there is a complete wall deficiency and, therefore, a very small, shallow glenoid vault, we would advise caution using this technique, given our concerns with regard to the immediate stability of the graft construct. In this scenario, we would suggest not re-implanting the humeral head initially, to reduce the shear forces on the graft construct; we would, instead, recommend performing a 2-stage procedure with either delayed reimplantation of the humeral head with or without glenoid component after graft healing, or our preferred option in the older patient would be to implant a reverse prosthesis during the second stage. We believe the reverse shoulder arthroplasty would be associated with a more predictable outcome in this setting, both in terms of pain relief and function.
There are a number of limitations to this study. We have reported on a very small series with a relatively short follow-up. However, the number of revision arthroplasties performed for glenoid loosening is generally small, and therefore large series in the literature are rare. The main purpose of this study was to report our surgical technique for glenoid reconstruction using Graftys HBS and the early results, particularly with respect to graft construct stability and healing potential. We will continue to follow our patients and report on the medium and long-term results.
In summary, we believe that the addition of injectable Graftys HBS resulted in improved mechanical stability of the iliac crest tricortical graft used for glenoid reconstruction. Five of 6 patients had a complete healing of the bone graft using the described technique, and we did not observe any radiolucent lines associated with resorption of cement. One patient underwent a successful second-stage implantation of a reverse shoulder arthroplasty after glenoid reconstruction.
1. Stewart MP, Kelly IG. Total shoulder replacement in rheumatoid disease: 7- to 13-year follow-up of 37 joints J Bone Joint Surg Br. 1997;79:68–72
2. Bohsali KI, Wirth MA, Rockwood CA Jr. Complications of total shoulder arthroplasty J Bone Joint Surg Am.. 2006;88:2279–2292
3. Dines JS, Fealy S, Strauss EJ, et al. Outcomes analysis of revision total shoulder replacement J Bone Joint Surg Am.. 2006;88:1494–1500
4. Sojbjerg JO, Frich LH, Johannsen HV, et al. Late results of total shoulder replacement in patients with rheumatoid arthritis Clin Orthop Relat Res.. 1999;366:39–45
5. Haines JF, Trail IA, Nuttall D, et al. The results of arthroplasty in osteoarthritis of the shoulder J Bone Joint Surg Br.. 2006;88:496–501
6. Raiss P, Aldinger PR, Kasten P, et al. Total shoulder replacement in young and middle-aged patients with glenohumeral osteoarthritis J Bone Joint Surg Br.. 2008;90:764–769
7. Torchia ME, Cofield RH, Settergren CR. Total shoulder arthroplasty with the Neer prosthesis: long-term results J Shoulder Elbow Surg.. 1997;6:495–505
8. Hawkins RJ, Greis PE, Bonutti PM. Treatment of symptomatic glenoid loosening following unconstrained shoulder arthroplasty Orthopedics.. 1999;22:229–234
9. Cheung EV, Sperling JW, Cofield RH. Revision shoulder arthroplasty for glenoid component loosening J Shoulder Elbow Surg.. 2008;17:371–375
10. Antuna SA, Sperling JW, Cofield RH, et al. Glenoid revision surgery after total shoulder arthroplasty J Shoulder Elbow Surg.. 2001;10:217–224
11. Scalise JJ, Iannotti JP. Bone grafting severe glenoid defects in revision shoulder arthroplasty Clin Orthop Relat Res.. 2008;466:139–145
12. Phipatanakul WP, Norris TR. Treatment of glenoid loosening and bone loss due to osteolysis with glenoid bone grafting J Shoulder Elbow Surg.. 2006;15:84–87
13. Norris TR, Kelly JD, Humphrey CS. Management of glenoid bone defects in revision shoulder arthroplasty: a new application of the reverse total shoulder prosthesis Tech Shoulder Elbow Surg.. 2007;8:37–46
14. Deutsch A, Abboud JA, Kelly J, et al. Clinical results of revision shoulder arthroplasty for glenoid component loosening J Shoulder Elbow Surg.. 2007;16:706–716
15. Neyton L, Sirveaux F, Roche O, et al. Results of revision surgery for glenoid loosening: a multicentric series of 37 shoulder prosthesis Rev Chir Orthop Reparatrice Appar Mot.. 2004;90:111–121
16. Neyton L, Walch G, Nove-Josserand L, et al. Glenoid corticocancellous bone grafting after glenoid component removal in the treatment of glenoid loosening J Shoulder Elbow Surg.. 2006;15:173–179
17. Russell TA, Leighton RK. Comparison of autogenous bone graft and endothermic calcium phosphate cement for defect augmentation in tibial plateau fractures: a multicenter, prospective, randomized study J Bone Joint Surg Am.. 2008;90:2057–2061
18. Giannoudis PV, Dinopoulos H, Tsiridis E. Bone substitutes: an update Injury.. 2005;36(suppl 3):S20–S27
19. Chau AM, Mobbs RJ. Bone graft substitutes in anterior cervical discectomy and fusion Eur Spine J.. 2009;18:449–464
20. Carson JS, Bostrom MP. Synthetic bone scaffolds and fracture repair Injury.. 2007;38(suppl 1):S33–S37
21. De Long WG Jr, Einhorn TA, Koval K, et al. Bone grafts and bone graft substitutes in orthopaedic trauma surgery: a critical analysis J Bone Joint Surg Am.. 2007;89:649–658
22. Van der Stok J, Van Lieshout EM, El-Massoudi Y, et al. Bone substitutes in the Netherlands: a systematic literature review Acta Biomater.. 2010;7:739–750
23. Wilke A, Orth J, Lomb M, et al. Biocompatibility analysis of different biomaterials in human bone marrow cell cultures J Biomed Mater Res.. 1998;40:301–306
© 2011 Lippincott Williams & Wilkins, Inc.