Reconstruction and limb salvage after
resection for malignant bone tumour of the
proximal humerus
A SLING PROCEDURE USING A FREE VASCULARISED FIBULAR
GRAFT
T. Wada, M. Usui, K. Isu, S. Yamawakii, S. Ishii
From Sapporo Medical University and Sapporo National Hospital, Japan
e assessed the intermediate functional results of
W eight patients after wide resection of the
proximal humerus for malignant bone tumour. We
used a free vascularised fibular graft as a functional
spacer and a sling procedure to preserve passive
scapulohumeral movement. Scapulohumeral
arthrodesis was not carried out. Five patients had
osteosarcoma, two achondrosarcoma and one a
malignant fibrous histiocytoma of the bone. The mean
duration of follow-up was 70 months (median, 76) for
the seven patients who were still alive at the time of
the latest follow-up. One patient died from the disease
12 months after surgery. There were no local
recurrences.
The functional results were described and graded
quantitatively according to the rating system of the
Musculoskeletal Tumour Society. Our results were
satisfactory with regard to pain, emotional acceptance
and manual dexterity. Function and lifting ability
were unsatisfactory in two patients. One patient had
delayed union between host and graft, but this united
after six months without further surgery. Radiographs
of the shoulder showed absorption or collapse of the
head of the fibula in four of the eight patients and a
fracture in another. No functional problems related to
absorption or fracture of the head of the fibula were
noted. There was no infection or subluxation of the
head. We conclude that this is a reasonably effective
technique of limb salvage after resection of the
proximal humerus.
J Bone Joint Surg [Br] 1999;81-B:808-13.
Received 25 August 1998; Accepted after revision 18 November 1999
T. Wada, MD, Assistant Professor
M. Usui, MD, Associate Professor
S. Ishii, MD, Professor and Chairman
Department of Orthopaedic Surgery, Sapporo Medical University School
of Medicine, S-1, W-16, Sapporo 060-8543, Japan.
K. Isu, MD, Chief of Orthopaedic Clinic
S. Yamawaki, MD, President
Sapporo National Hospital, 4-2 Kikusui, Sapporo 030, Shiroishiku,
Japan.
Correspondence should be sent to Dr T. Wada.
©1999 British Editorial Society of Bone and Joint Surgery
0301-620X/99/59430 $2.00
808
In patients who have a malignant bone tumour of the
proximal humerus, it is often possible to carry out curative
resection which spares the limb. Reconstruction, however,
remains a problem because none of the prostheses which
are currently available adequately compensates for the
functional loss after amputation.
There are two main groups of reconstructive procedures
for the proximal humerus. One involves arthrodesis and
1,2 3
includes autogenous grafts, allografts and composite
1,4
allografts. The other aims to preserve glenohumeral
4,5
movement using functional pacers, prostheses for
4,6,7
replacement of the proximal humerus or osteoarticular
8
allografts. The advantages and disadvantages have been
widely discussed.
After resection of a malignant bone tumour of the prox-
imal humerus, we used a free vascularised fibular graft as a
functional spacer, and a sling procedure to preserve scap-
2,9
ulohumeral passive movement. We reviewed a consec-
utive series of eight patients who had this reconstructive
procedure, assessing function at a mean of 63 months after
surgery.
Patients and Methods
Between 1988 and 1995 five women and three men of
mean age 27 years (10 to 47) underwent this procedure
(Table I). All had a primary malignant tumour of the
proximal humerus. Excluded from the study were patients
with soft-tissue sarcomas or tumours of the clavicle, scap-
ula or proximal part of the humeral diaphysis which did not
involve the humeral head. Patients were selected only if
preoperative imaging had shown that a satisfactory surgical
margin could be achieved and, if metastatic disease was
present, that this was also amenable to resection. Pre-
operative studies included plane radiography of the shoul-
der, MRI or CT of the region around the tumour, full body
technetium bone scanning and CT of the chest. None of the
patients had distant metastasis at the time of operation.
There were five osteosarcomas, two chondrosarcomas and
one malignant fibrous histiocytoma (MFH) of the bone.
The surgical stage was Ib for one patient with chon-
drosarcoma and IIb for the remaining seven. All patients
presented with pain and a palpable mass, and two had
pathological fractures.
THE JOURNAL OF BONE AND JOINT SURGERY
 RECONSTRUCTION AND LIMB SALVAGE AFTER RESECTION FOR MALIGNANT BONE TUMOUR OF THE PROXIMAL HUMERUS 809

Table I. Details of the eight patients who had a sling procedure for sarcoma of the proximal humerus
Graft Latissimus
Age Pathological Surgical Type of Time of length dorsi
Case (yr) Gender diagnosis staging resection* reconstruction (cm) rotation Complications
1 43 M Chondrosarcoma Ib IA Primary 22 - Delayed union
2 20 M Osteosarcoma IIb VB Secondary† 26 - None
3 10 F Osteosarcoma IIb VB Primary 15 - Absorption of the fibular head
4 17 F Osteosarcoma IIb VB Primary 20 - Absorption of the fibular head
5 30 M Osteosarcoma IIb VB Secondary‡ 25 - None
6 34 F MFH IIb IA Primary 17 - Absorption of the fibular head
7 14 F Osteosarcoma IIb VB Primary 15 Primary Absorption of the fibular head
8 47 F Chondrosarcoma IIb VB Primary 23 Secondary Fracture, absorption of the fibular head
* see text
† wide resection only was carried out at initial operation
‡ reconstruction after failed prosthetic implant
11
Table II. Functional rating system of the Musculoskeletal Tumour Society for the upper limb
Emotional Positioning of Manual Lifting
Rating Pain Function acceptance the hand dexterity ability
5 None No restrictions Enthusiastic Unlimited No limitation Normal load
4 Intermediate Intermediate Intermediate Intermediate Intermediate Intermediate
3 Modest/non-disabling Recreational restrictions Satisfied Not above shoulder Loss of fine Limited
or no pronation movements
or supination
2 Intermediate Intermediate Intermediate Intermediate Intermediate Intermediate
1 Moderate/intermediately Partial occupational Accepted Not above waist Cannot pinch Helping only
disabling restriction
0 Severe/continuously Total occupational Dislikes None Cannot grasp Cannot help
disabling restriction

We classified the extent of each resection according to was fixed to the proximal stump of the humerus with a
10
the scheme proposed by Malawer, Meller and Dunham. 3.5 mm dynamic compression plate. The tendons of biceps
The functional results were described and graded quantita- femoris and palmaris longus together with a transient
tively according to the most recent rating system of the
11
Musculoskeletal Tumour Society (Table II).
The mean length of follow-up was 63 months. One
patient died from lung metastases 12 months after surgery.
The mean length of follow-up for the seven surviving
patients was 70 months (median, 76; range 32 to 91).
All five patients with osteosarcoma and the one with
MFH of the bone had systemic chemotherapy, both pre-
operatively and postoperatively. Neither of the two patients
with chondrosarcoma had chemotherapy.
Oncological and reconstructive procedure. A wide surgi-
cal margin was achieved in all eight patients. We carried
out either an intra-articular resection of the proximal
humerus with the abductor mechanism intact (type IA) or
an extra-articular humeral and glenoid resection with the
abductor mechanism disrupted (type VB).
A free vascularised fibular graft was inserted as a func-
tional spacer, using a sling procedure in six patients who
had primary resection, in one in whom a replacement
proximal humeral prosthesis had failed, and in one who had
resection without reconstruction. The mean duration of the
operation was 7.4 hours (5.5 to 11) and the mean blood loss
was 1131 ml (635 to 1634). The mean length of the graft
was 20.4 cm (15 to 26).
Sling procedure. An appropriate length of free vascu- Fig. 1
larised fibula, including the head, was harvested from the Diagram of the operative procedure, reproduced with permission of
ipsilateral or contralateral leg. The distal end of the fibula Medical View Inc.

VOL. 81-B, NO. 5, SEPTEMBER 1999

810 T. WADA, M. USUI, K. ISU, S. YAMAWAKII, S. ISHII

Fig. 2a Fig. 2b Fig. 2c

Fig. 2d Fig. 2e Fig. 2f

23,24
Case 3. Radiographs a) at presentation (chemotherapy had been given preoperatively) b) after extra-articular resection type VB and
reconstruction with the sling procedure and c) 91 months after operation, showing absorption of the head of the fibula. Photographs (d
to f) show postoperative active movement.

stainless-steel wire were used to suspend the head of the Results

fibula from the remaining part of the scapula, usually the
acromion. The peroneal artery was anastomosed to the
circumflex humeral or deep brachial artery (Figs 1 and 2).
A latissimus dorsi flap was rotated in two patients (cases 7
and 8) to create adequate muscle coverage around the
shoulder (Fig. 3).
Graft union and hypertrophy. The graft united in under
four months in seven of the eight patients. In the other (case
1), union took six months. None of the patients needed a
second operation to achieve union.
Hypertrophy did not develop in any of the eight grafts.
THE JOURNAL OF BONE AND JOINT SURGERY
 RECONSTRUCTION AND LIMB SALVAGE AFTER RESECTION FOR MALIGNANT BONE TUMOUR OF THE PROXIMAL HUMERUS 811

Case 7. A 15-year-old girl with osteosarcoma of the

proximal humerus. Photographs taken 49 months after
operation show the acceptable appearance of the left
shoulder after latissimus dorsi rotation.

Fig. 3a Fig. 3b

Table III. Functional and oncological results for the eight patients after the sling procedure for sarcoma of the proximal
humerus
Functional evaluation
Length of
follow-up Emotional Positioning Manual Lifting Overall
Case (mth) Pain Function acceptance of hand dexterity ability (%) Prognosis*
1 90 5 3 5 3 5 4 83 CDF
2 12 (30)† 5 2 3 3 5 3 70 DOD
3 91 5 3 4 4 5 4 83 CDF
4 90 5 3 4 4 5 4 83 CDF
5 76 (191)† 5 3 3 4 5 3 77 CDF
6 64 5 3 3 4 5 3 77 CDF
7 49 5 3 4 4 5 4 83 CDF
8 32 4 2 4 3 5 2 67 NED
* CDF, continuous disease-free; NED, no evidence of disease; DOD, died of disease
† indicates follow-up period after wide resection of the tumour

No longitudinal growth of the fibular graft occurred in
either of the two patients (cases 3 and 7) with open
12
epiphyses at the time of surgery.
Oncological evaluation. Six patients remained free from
disease at a mean of 74 months (median, 73) after surgery.
One patient (case 2) with osteosarcoma died from the
disease 12 months after operation. A patient with chon-
drosarcoma (case 8) developed a lung metastasis and was
successfully treated by thoracotomy. There were no local
recurrences.
Reconstruction. Functional data were available for all
eight patients (Table III). The mean overall functional
rating was 79% (67 to 83). With regard to pain, emotional
acceptance and manual dexterity, the results were rated as
satisfactory with a score of 3.0 points or more in all
patients. In two (cases 2 and 8), they were unsatisfactory
with a score of less than 3.0 points as regards function and
VOL. 81-B, NO. 5, SEPTEMBER 1999
lifting ability. In one patient (case 2) the sling procedure
was carried out 18 months after wide resection of the
tumour. Marked preoperative elbow contracture could not
be relieved. In another (case 8) the short head of biceps
could not be preserved because of involvement of the
tumour leading to poor function and lifting ability.
Complications. Radiographs of the shoulder showed
absorption or collapse of the head of the fibula in five of the
eight patients (Fig. 2c) and one had a fracture of the head.
Neither of these complications caused functional problems.
One patient (case 8) with a disrupted deltoid muscle had
pain and irritation of the skin over the acromion. This was
successfully treated by a latissimus dorsi flap, which pro-
vided adequate muscle coverage for the acromion and head
of the fibula. There was no superficial or deep infection.
Subluxation or dislocation of the head was not seen. No
problems were encountered at the donor site.
812 T. WADA, M. USUI, K. ISU, S. YAMAWAKII, S. ISHII
Discussion
Resection of the proximal humerus, the glenoid fossa and
abductor mechanism (type VB) presents a challenge.
Reconstruction with a spacer or proximal humeral pros-
thesis offers immediate distal fixation and makes it possible
to administer chemotherapy and/or radiotherapy in the
early postoperative period. It is generally the least time-
consuming of the available options. Although it restores
minimal function to the shoulder, the implant provides a
stable fulcrum for elbow and hand function and prevents
pain arising from traction on the neurovascular bundle.
Complications may arise, however, due to subluxation,
4
bone loss arising from stress-shielding, and deep infection.
13
Using an intramedullary nail as a spacer, Courpied et al
reported functional results which were fair in two patients
and poor in three. In three patients it was necessary later to
remove the nail because of loosening and pain in the
6
surrounding soft tissue. Ross et al noted that 19 of 25
patients with proximal humeral implants developed sub-
luxation or dislocation of the head.
8
Gebhardt et al, reviewing 20 patients who received
osteoarticular allografts to replace the proximal end of
humerus after wide resection of bone tumours, found the
procedure to be unsuitable for those with high-grade extra-
compartmental neoplasm requiring resection of the abductor
mechanism. The high rate of infection caused concern.
4
O’Connor et al reported that after type-VB resection
arthrodesis gave better results than reconstruction with a
proximal humeral prosthesis or a spacer. The functional
results of glenohumeral arthrodesis, undertaken for condi-
14
tions not related to tumours, do not deteriorate with time.
It may, however, take a long time to achieve bony union,
1
even when a vascularised bone graft is used since it is
difficult to fix the reconstructed humerus to the vestigial
scapula. The necessary prolonged immobilisation in a
shoulder spica causes difficulties for patients who need
postoperative chemotherapy. The concomitant use of a con-
1,4
ventional fibular graft can contribute to stable fixation.
4
Deep infection after arthrodesis remains a problem.
There is a high risk of fracture with an arthrodesis.
4 1. Wood MB. Free vascularized bone transfers for nonunions, segmental
O’Connor et al reported this in four of their ten patients.
14
Cofield and Briggs described fracture of the humerus in
eight of 71 patients who had glenohumeral arthrodesis for
treatment of conditions other than tumours. Similarly,
15
Richards and Kostuik noted fractures in two of their 53
patients.
We used a free vascularised fibular graft as a functional
spacer, preserving passive scapulohumeral movement. Pos-
itioning of the hand and lifting ability were not as good as
those obtained with arthrodesis, but were better than those
4 grafts in the reconstruction of segmental skeletal defects. J Bone Joint
reported by previous authors who used proximal humeral
prostheses or spacers. An autogenous bone graft facilitates
the healing of soft tissue and results in better function of
the preserved biceps and triceps. We also stabilised the
reconstructed humerus by attaching tendon to the bone.
This may be why none of our patients developed subluxa-
tion or dislocation of the head of the fibula.
The sling procedure is easier than an arthrodesis, there is
no need for postoperative immobilisation and the rate of
fracture is low. One complication of the use of a free
vascularised fibular graft is bony protrusion of the acro-
mion under the skin, which may cause pain and be unsight-
ly. This can be prevented by a primary rotation of the
latissimus dorsi. Absorption or collapse of the head of the
fibula may occur which may have caused the fracture of the
head in one of our patients. This complication is probably
due to a deficient blood supply to the head through the
16
peroneal artery. Theoretically, concomitant anastomosis
of a lateral inferior genicular artery or a branch of the
anterior tibial artery could improve the blood supply, but
we did not attempt this. Alternatively, the use of the
anterior tibial artery as a pedicle may provide better blood
17
supply to the graft. None of our patients with absorption
of the head of the fibula developed symptoms.
Conventional or non-vascularised fibular autografts have
18-21
been used as passive spacers for many years. Non-
vascularised fibular autografts longer than 12 cm tend to
6
fracture, and the fractures do not always unite. Although
no graft hypertrophy was noted, we believe that the use of
vascularised fibula reduces the risk of fracture.
The extent of resection and the needs of the patient
should dictate the choice of the procedure used for recon-
struction. Despite the minor complications, we believe that
the functional results of the sling procedure are comparable
to, or better than, those obtained by other methods. Accord-
22
ing to Springfield, a flail shoulder is often the best
alternative for the patient who does not need to use the
hand in space. The procedure which we describe is a useful
limb-salvaging technique after resection of the proximal
humerus, particularly for patients who do light work.
This work was supported in part by grants from the Ministry of Education,
Science and Culture and the Ministry of Health and Welfare.
No benefits in any form have been received or will be received from a
commercial party related directly or indirectly to the subject of this
article.
References
gaps and following tumor resection. Orthopedics 1986;9:810-6.
2. Usui M, Naito T, Yamawaki S, Ishii S. Free vascularized fibular graft
for treatment of malignant tumor of proximal humerus. In: Vastamaki
M, Jalovaara P, eds. Surgery of the shoulder. Amsterdam: Elsevier
Science, 1995:471-4.
3. Alman BA, De Bari A, Krajbich JI. Massive allografts in the
treatment of osteosarcoma and Ewing sarcoma in children and adoles-
cents. J Bone Joint Surg [Am] 1995;77-A:54-64.
4. O’Connor MI, Sim FH, Chao EYS. Limb salvage for neoplasms of
the shoulder girdle: intermediate reconstructive and function results. J
Bone Joint Surg [Am] 1996;78-A:1872-88.
5. Enneking WF, Eady JL, Burchardt H. Autogenous cortical bone
Surg [Am] 1980;62-A:1039-58.
6. Ross AC, Wilson JN, Scales JT. Endoprosthetic replacement of the
proximal humerus. J Bone Joint Surg [Br] 1987;69-B:656-61.
7. Malawer MM, Chou LB. Prosthetic survival and clinical results with
use of large-segment replacements in the treatment of high-grade bone
sarcomas. J Bone Joint Surg [Am] 1995;77-A:1154-65.
THE JOURNAL OF BONE AND JOINT SURGERY
 RECONSTRUCTION AND LIMB SALVAGE AFTER RESECTION FOR MALIGNANT BONE TUMOUR OF THE PROXIMAL HUMERUS
8. Gebhardt MC, Roth YE, Mankin HJ. Osteochondral allografts for
reconstruction in the proximal part of the humerus after excision of a
musculoskeletal tumor. J Bone Joint Surg [Am] 1990;72-A:334-45.
9. Usui M, Ishii S, Naito T, et al. Limb-saving surgery in osteosarcoma
by vascularized fibular graft. J Orthop Sci 1996;1:4-10.
10. Malawer MM, Meller I, Dunham WK. A new surgical classification
system for shoulder-girdle resections: analysis of 38 patients. Clin
Orthop 1991;267:33-44.
11. Enneking WF, Dunham W, Gebhardt MC, Malawar M, Pritchard
DJ. A system for the functional evaluation of reconstructive proce-
dures after surgical treatment of tumours of the musculoskeletal
system. Clin Orthop 1993;286:241-6.
12. De Boer HD, Wood MB. Bone changes in the vascularised fibular
graft. J Bone Joint Surg [Br] 1989;71-B:374-8.
13. Courpied JP, Tomeno B, Langlais F, et al. Functional results
following resection of tumors in the proximal humerus: excluding
arthrodesis and massive allografts. In: Yamamuro T, ed. New develop-
ments for limb salvage in musculoskeletal tumors. Tokyo, etc: Spring-
er-Verlag, 1989:531-6.
14. Cofield RH, Briggs BL. Glenohumeral arthrodesis: operative and
long-term functional results. J Bone Joint Surg [Am] 1979;61-A:
668-77.
15. Richards RR, Kostuik JP. Shoulder arthrodesis: indications and
techniques. In: Watson MS, ed. Surgical disorders of the shoulder.
Edinburgh, etc: Churchill Livingstone, 1991:443-57.
VOL. 81-B, NO. 5, SEPTEMBER 1999
813
16. Taylor GI, Wilson KR, Rees MD, Corlett RJ, Cole WG. The
anterior tibial vessels and their role in epiphyseal and diaphyseal
transfer of the fibula: experimental study and clinical applications. Br
J Plastic Surg 1988;41:451-69.
17. Innocenti M, Ceruso M, Manfrini M, et al. Free vascularized growth
plate transfer after bone tumor resection in children. J Reconstr
Microsurg 1998;14:137-43.
18. Albee FH. Restoration of shoulder function in cases of loss of head
and upper portion of humerus. Surg Gynecol Obstet 1921;32:1-19.
19. Schauffler RMcE. Transplant of the upper extremity of the fibula to
replace the upper extremity of the humerus. J Bone Joint Surg 1926;
8:723-6.
20. Clark K. A case of replacement of the upper end of the humerus by a
fibular graft reviewed after twenty-nine years. J Bone Joint Surg [Br]
1959;41-B:365-8.
21. Wilson PD, Lance EM. Surgical reconstruction of the skeleton
following segmental resection for bone tumors. J Bone Joint Surg
[Am] 1965;47-A:1629-56.
22. Springfield DS. Orthopaedic oncology. In: Sledge CB, Poss RP, eds.
The year book of orthopedics. St Louis: Mosby, 1997:98-100.
23. Rosen G. Preoperative (neoadjuvant) chemotherapy for osteogenic
sarcoma: a ten year experience. Orthopedics 1985;8:659-64.
24. Wada T, Isu K, Takeda N, et al. A preliminary report of neoadjuvant
chemotherapy NSH-7 study in osteosarcoma: preoperative salvage
chemotherapy based on clinical tumor response and the use of granu-
locyte colony-stimulating factor. Oncology 1996;53:221-7.