Advocates of scapular replacement suggest sparing the deltoid should be a prerequisite for using a scapular prosthesis. We attempted a different approach by using a scapular prosthesis as a means for reconstruction with and without a functioning deltoid. We evaluated the importance of a functioning deltoid in scapular replacement for malignant tumors.
In 1977, Marcove et al described early experience with the Tikhoff-Linberg procedure in 17 patients with cancer of the shoulder girdle.9 The tumors were widely excised, and the patients were left with a functioning hand and elbow.9 They either left the arm flail or reconstructed the shoulder by placing a Küntscher rod into the remaining humerus, which was secured proximally to a rib or clavicular remnant.9 This stabilization was effective temporarily, however bony erosion and breakout of the fixation lead to instability.9 The rods migrated into the axilla or through the overlying skin causing cosmetic and neurovascular problems.9
The purpose of reconstructing the shoulder girdle is to provide a stable post on which to position the hand and elbow. Active shoulder abduction is also a goal, but shoulder elevation is difficult to preserve.2,8,10,11 The cosmetic effects are typically a secondary concern to the patient and surgeon. The Tikhoff-Linberg procedure leaves a substantial defect in the shoulder contour. Some authors have described efforts to restore near normal shoulder contour after unreconstructed Tikhoff-Linberg resections with external prostheses or delayed transverse rectus abdominis musculocutaneous (TRAM) flaps.5,6 These measures were intended to improve the aesthetic appearance of the shoulder and improve clothing fit.5 Scapular prostheses were developed to improve function, restore better shoulder contour, and prevent pain and limb-length change because of gravitational stretch.6,12 Patterson et al (Patterson F, Cyran F, Munson D, Leeson M, Aboulafia A, Benevenia J. Total scapular replacement versus resection arthroplasty following radical excision for tumors of the scapula. Presented at the annual meeting of the Musculoskeletal Tumor Society; May 10-12, 2001; Baltimore, MD) and Wodajo et al (Wodajo F, Bickels J, Wittig JC, Kellar-Graney K, Kollendar Y, Meller I, Malawer MM. Reconstruction with scapular endoprosthesis provides superior results after total scapular resection: surgical technique and comparison to patients without endoprosthetic reconstruction. Presented at the annual meeting of the Musculoskeletal Tumor Society; April 27, 2002; Toronto, Ontario, Canada) reported superior functional and cosmetic results after scapula replacement compared with results of patients who did not have reconstruction with a scapular prosthesis. They stressed the importance of a functioning deltoid as a prerequisite to using a scapular prosthesis. However, the literature does not establish whether this is crucial.
Therefore, we asked whether the presence of a functioning deltoid influenced the results of scapular replacement.
MATERIALS AND METHODS
We retrospectively identified and reviewed 19 consecutive patients treated from 1989 to 2004 with a scapular replacement prosthesis after resection of a malignant bone or soft tissue tumor (Table 1). We included patients who had a total scapulectomy and excluded patients who had no reconstruction, who had reconstruction with an allograft scapula, or who had a subtotal scapulectomy. There were eight female and 11 male patients with an average age of 37 years (range, 10-73 years). We divided our patients into two groups based on whether they had a functioning deltoid after surgery. In addition to the status of the axillary nerve, we recorded the percentage of deltoid remaining after tumor resection (Table 2). Eight patients had their axillary nerve resected. There were three male and five female patients with an average age of 39 years. The second group did not have their axillary nerve resected. There were eight male and three female patients in this group with an average age of 38 years. The median followup was 18 months (range, 12-124 months). We excluded nine patients who died of disease. Prior Institutional Review Board approval was obtained.
The classification system for shoulder girdle resections is based on criteria proposed by Malawer et al and adopted by the Musculoskeletal Tumor Society (MSTS).7 The classification system has six categories based on the degree of bony resection. Each category is modified as an A or B resection depending on whether the abductor mechanism is left intact. The abductor mechanism is considered the deltoid and rotator cuff. A Type I resection is an intraarticular resection of the proximal humerus, Type II is an extraarticular partial scapulectomy, Type III is an intraarticular total scapulectomy, Type IV is an extraarticular total scapulectomy coupled with a humeral head resection, Type V is an extraarticular humeral resection with a glenoid resection, Type VI is an extraarticular humeral resection coupled with a total scapulectomy, and Type VI involves more than just the humeral head, as in a Type IV resection. In general, the magnitude of the surgical procedure increases based on the degree of bony resection.7
We most commonly perform total scapulectomy or the Tikhoff-Linberg procedure for bone and soft tissue tumors about the scapula and proximal humerus. We favor forequarter amputation when limb-sparing wide excision of the tumor is not possible because of contamination of the neurovascular structures. The surgical techniques used for scapulectomy and the Tikhoff- Linberg procedure have been described.2,8,9 Most often the defect left after the scapulectomy or the Tikhoff-Linberg procedure was not reconstructed with a prosthesis or allograft, which led to problems associated with a flail arm. We reconstructed the shoulder girdle to restore the normal force about the shoulder to maximize function and stability. An osteoarticular scapular allograft or scapular hemiarthroplasty are reconstructive options if the humeral head is preserved. We think a constrained scapulohumeral prosthesis is necessary if the proximal humerus is resected in addition to performing a scapulectomy.
All patients had reconstructions with a scapular prosthesis (Biomet, Warsaw, IN). The Tikhoff-Linberg procedure included an extraarticular excision of the shoulder. Patients had reconstructions with constrained scapulohumeral prostheses (Fig 1) that had a consistent design during the study period. The prosthesis was lightened by making the central body of the scapula a mesh surrounded by a solid frame. Suture holes in the frame allowed repair of the periscapular muscles. The prosthesis was contoured in the cephalocaudad and mediolateral directions to match the contour of the chest wall. The patients who had a total scapulectomy alone had reconstructions using a scapular hemiarthroplasty prosthesis (Fig 2). This prosthesis has a polyethylene-bearing surface that articulates with the humeral head. Reconstruction of the glenohumeral soft tissue envelope was necessary to ensure stability of the articulation. Synthetic graft material augmented deficient capsular reconstruction. The graft was formed into a cylinder to provide circumferential coverage. The graft then was secured to remaining soft tissues attached to the humerus after the resection. If no capsule or rotator cuff attachments were left on the humerus, then suture anchors were placed in the bone. The graft was secured to the prosthesis by running a purse string suture through the graft and tensioning it around the neck of the prosthetic glenoid.
Eleven patients had a Type IV resection with removal of their scapula, the distal clavicle, and humeral head. Four patients had a Type III intraarticular scapulectomy. Four patients had a Type VI resection including the scapula, distal clavicle, and proximal humerus in an extraarticular fashion. Fifteen patients had reconstruction with a scapular/humeral prosthesis, and four patients had only their scapula replaced.
We think partial or total resection of the rotator cuff musculature is important for all scapular cancers and extraarticular humeral resections. Loss of the rotator cuff substantially impedes function of the glenohumeral articulation. However, we usually retain the latissimus dorsi, and often retain the deltoid after scapulectomy and Tikhoff-Linberg resections. The retained muscles were identified by reviewing operative notes and during examinations at followup. Ten patients had a latissimus dorsi soft tissue flap at their index operation. In nine patients, the latissimus flap served primarily to cover the prosthesis. However, one patient had the latissimus flap inserted into the remaining rotator cuff musculature coaxial with the supraspinatus, which helped provide soft tissue coverage and provided a functional benefit to shoulder abduction. One patient had a secondary pectoralis flap rotated to cover an area of marginal wound necrosis.
We graded functional results using the International Society of Limb Salvage (ISOLS) scoring system. The ISOLS score is based on modifications of the initial MSTS rating scale using six criteria each graded from 0 to 5.3,4 The categories include: pain, emotional acceptance, hand positioning, function, manual dexterity, and lifting ability. The pain score is based on whether the patient uses pain medication and what type of medication (narcotic versus nonnarcotic). Emotional acceptance is a self-rated function of the patient's perception of the overall result. Hand positioning is a measure of the patient's ability to actively position the hand in space. Manual dexterity is a function of the patient's ability to perform increasingly complex functions with the hand (eg, buttoning, writing, eating). The lifting ability of the reconstructed limb is described using the contralateral limb as the standard control. Functional results are determined by the effect of restrictions (actual or prohibited) on the patient's lifestyle. Scores are presented as a percent of the maximum 30 points.
We performed statistical analyses using SPSS software version 11.0 (SPSS Inc, Chicago, IL). We used the Mann-Whitney test to compare ISOLS scores between patients with and without functioning deltoids.
Functional scores were superior in the group of patients who retained a functioning deltoid. The average ISOLS score was 25 points (range, 21-27 points), representing an 82% maximum functional score. Patients in the group that had resections (n = 8) had lower (p = 0.006) ISOLS scores (average, 24 points; range, 21-25 points; standard deviation [SD], 1.6) than patients in the spared group (n = 11) (average, 26 points; range, 23-27 points; SD, 1.6). Function (p = 0.004), hand positioning (p = 0.016), and pain scores (p = 0.032) were better in the spared group. There were no differences in emotional acceptance or lifting ability. Manual dexterity was not affected in either group.
The rate of major complications between the two groups did not reach statistical significance. There were six major complications resulting in additional surgery. Three occurred in each group. Wound necrosis developed in two patients. The first patient had marginal wound necrosis of the latissimus flap that was treated by soft tissue advancement and closure. Another patient did not receive a flap at the primary surgery and was treated with a pectoralis flap. The shoulder prostheses dislocated in two patients. One patient had open reduction with further stabilization of the soft tissue to the prosthesis. Another patient had revision surgery for the dislocation. Subsequently this patient had an infection develop that required débridement with retention of the prosthesis. Another infection occurred in a patient who did not have a flap at the index operation. She had a localized infection develop that was treated with removal of the scapular prosthesis. The humeral prosthesis was retained. In one patient, local recurrence necessitated a forequarter amputation.
Nine of the 19 patients were alive with no evidence of disease at the last followup. Nine patients died of disease and one patient is alive with disease. There was one local recurrence,
Advocates of scapular-replacing prostheses consider a functioning deltoid a prerequisite for their use. The purpose of our study was to determine whether a difference in function exists between patients with and without a functioning deltoid who have had scapular replacement.
Our study has several limitations. First, there are only 19 patients, and although it is the largest series of its kind, it is difficult to draw firm conclusions based on a study this size. A control group of patients who did not have scapular replacement would improve our study despite results showing superior function in patients who had scapular replacement (Patterson F, Cyran F, Munson D, Leeson M, Aboulafia A, Benevenia J. Total scapular replacement versus resection arthroplasty following radical excision for tumors of the scapula. Presented at the annual meeting of the Musculoskeletal Tumor Society; 2001 May 10-12; Baltimore, MD). Finally, our study is limited by its retrospective design.
Numerous authors have reported their results after scapulectomy and Tikhoff-Linberg resections.2,8-11 Some emphasized the importance of stabilizing the proximal humerus to the chest wall to avoid instability and neurapraxia because of gradual lengthening of the flail arm.9,12 A Küntscher nail secured to the second rib or remaining clavicle initially was used for this purpose.9 Unfortunately, the Küntscher nail often became detached and migrated superiorly leaving the arm flail and at risk for skin compromise.12 Surgeons have begun using a prosthesis for reconstruction.6,12
The first data regarding use of a scapular prosthesis for reconstruction after tumor excision showed that scapular replacement provided superior function when compared with a scapulectomy in patients who did not have reconstruction with a prosthesis (Patterson F, Cyran F, Munson D, Leeson M, Aboulafia A, Benevenia J. Total scapular replacement versus resection arthroplasty following radical excision for tumors of the scapula. Presented at the annual meeting of the Musculoskeletal Tumor Society; 2001 May 10-12; Baltimore, MD). They compared patients who had reconstruction with a scapular prosthesis with patients treated with a resection only. Preservation of the axillary nerve and deltoid function were prerequisites for inclusion. Patients who had reconstruction with a scapular prosthesis had better active shoulder flexion and abduction and better overall MSTS scores. Improved shoulder contour and cosmesis also was present in the scapular replacement group, although the measurement parameters were not clear.
Scapular replacement was reported to be superior to humeral suspension without scapular replacement for shoulder abduction and contour (Wodajo F, Bickels J, Wittig JC, Kellar-Graney K, Kollendar Y, Meller I, Malawer MM. Reconstruction with scapular endoprosthesis provides superior results after total scapular resection: surgical technique and comparison to patients without endoprosthetic reconstruction. Presented at the annual meeting of the Musculoskeletal Tumor Society; April 27, 2002; Toronto, Ontario, Canada). Twenty-three patients had scapulectomies for treatment of bone tumors. Seven patients had reconstruction with a scapular prosthesis. The remaining 16 patients had their proximal humerus suspended from their clavicle. Four patients had a prosthetic humeral component and 12 patients had their native humerus suspended. Patients who had reconstruction with a scapular prosthesis had better active shoulder abduction. Patients in the scapular prosthesis group also had a more natural contour of the shoulder girdle.
In a case series of patients who had reconstruction with a constrained scapula/humeral prosthesis, Wittig et al stressed the importance of a functioning deltoid.12 Their patients had ISOLS scores ranging from 24 to 27 points.12 The deltoid is an important shoulder abductor when the axillary nerve is preserved. However, the denervated deltoid continues to provide vascularized soft tissue coverage, allows myodesis with the functioning trapezius, and helps maintain better shoulder contour. Restoration of the normal shoulder contour is important for cosmetic reasons and improving clothing fit.
The latissimus dorsi has numerous reconstructive roles. It is an excellent source of soft tissue to rotate on its thoracodorsal pedicle and cover the prosthesis.1 It rarely contributes to motor power of the shoulder, yet it serves as an interposition graft between the trapezius and residual deltoid passively contributing to some potential abduction. The latissimus dorsi flap also helps to maintain the contour of the shoulder.
The status of the deltoid was an important determinant of function in patients who had scapular replacement. However, the patients' overall acceptance was not influenced by the status of the deltoid, and we do not think the lack of a functioning deltoid should be an absolute contraindication for scapular replacement. We agree with Patterson et al (Patterson F, Cyran F, Munson D, Leeson M, Aboulafia A, Benevenia J. Total scapular replacement versus resection arthroplasty following radical excision for tumors of the scapula. Presented at the Annual Meeting of the Musculoskeletal Tumor Society; 2001 May 10-12; Baltimore, MD) and Wodajo et al (Wodajo F, Bickels J, Wittig JC, Kellar-Graney K, Kollendar Y, Meller I, Malawer MM. Reconstruction with scapular endoprosthesis provides superior results after total scapular resection: surgical technique and comparison to patients without endoprosthetic reconstruction. Presented at the annual meeting of the Musculoskeletal Tumor Society; April 27, 2002; Toronto, Ontario, Canada) that the cosmetic result after scapular replacement is better than without replacement. However, we could not find a suitable way to measure this feature. The common use of a latissimus dorsi flap may have contributed to our lower wound infection and complication rate than reported in a previous study.1
We thank The Biomet Orthopaedic Oncology Fellowship.
1. Capanna R, Manfrini M, Briccoli A, Gherlinzoni F, Lauri G, Caldora P. Latissimus dorsi pedicled flap applications in shoulder and chest wall reconstructions after extracompartimental sarcoma resections. Tumori
2. Capanna R, van Horn JR, Biagini R, Ruggieri P, Ferruzzi A, Campanacci M. The Tikhoff-Linberg procedure for bone tumors of the proximal humerus: the classical “extensive” technique versus a modified “transglenoid” resection. Arch Orthop Trauma Surg
3. Enneking W. Modification of the System for Functional Evaluation in the Surgical Management of Musculoskeletal Tumors
. New York, NY: Churchill Livingstone; 1987:626-639.
4. Enneking WF, Dunham W, Gebhardt MC, Malawer M, Pritchard DJ. A system for the functional evaluation of reconstructive procedures after surgical treatment of tumors of the musculoskeletal system. Clin Orthop Relat Res
5. Hidalgo DA, Zenn MR, Marcove RC. Aesthetic reconstruction of Tikhoff-Linberg shoulder defects with a dual-pedicle TRAM free flap. Plast Reconstr Surg
6. Malawer M, Wittig J, Rupert C. Scapulectomy. In: Malawer M, Sugarbaker P, eds. Musculoskeletal Cancer Surgery: Treatments of Sarcomas and Allied Diseases
. Norwell, MA, Kluwer Academic Publishing; 2002:553-568.
7. Malawer MM, Meller I, Dunham W. A new surgical classification system for shoulder-girdle resections. Clin Orthop Relat Res
. 1991; 267:33-44.
8. Malawer MM, Sugarbaker PH, Lampert M, Baker AR, Gerber NL. The Tikhoff-Linberg procedure: report of ten patients and presentation of a modified technique for tumors of the proximal humerus. Surgery
9. Marcove RC, Lewis MM, Huvos AG. En bloc upper humeral interscapulo-thoracic resection: the Tikhoff-Linberg procedure. Clin Orthop Relat Res
10. O'Connor MI, Sim FH, Chao EY. Limb salvage for neoplasms of the shoulder girdle: intermediate reconstructive and functional results. J Bone Joint Surg Am
11. Voggenreiter G, Assenmacher S, Schmit-Neuerburg KP. Tikhoff-Linberg procedure for bone and soft tissue tumors of the shoulder girdle. Arch Surg
12. Wittig JC, Bickels J, Wodajo F, Kellar-Graney KL, Malawer MM. Constrained total scapula reconstruction after resection of a high-grade sarcoma. Clin Orthop Relat Res