Locked intramedullary nailing of symptomatic metastases in the humerus
S. J. Spencer, G. Holt, J. V. Clarke, A. Mohammed, W. J. Leach, J. L. B. Roberts From the Western Infirmary, Glasgow, Scotland
S. J. Spencer, MRCS(G), Specialist Registrar G. Holt, MRCS, Specialist Registrar J. V. Clarke, MRCS(G), Specialist Registrar W. J. Leach, FRCS(Ed)(Orth), Consultant Orthopaedic Surgeon J. L. B. Roberts, FRCS(Orth), Consultant Orthopaedic Surgeon Department of Orthopaedic Surgery Western Infirmary, Dumbarton Road, Glasgow G11 6NT, UK. A. Mohammed, MD, FRCS(Orth), Consultant Orthopaedic Surgeon Department of Orthopaedic Surgery Southern General Hospital, 1345 Govan Road, Glasgow G11 6NT, UK. Correspondence should be sent to Mr S. J. Spencer; e-mail:
[email protected] ©2010 British Editorial Society of Bone and Joint Surgery doi:10.1302/0301-620X.92B1. 22399 $2.00 J Bone Joint Surg [Br] 2010;92-B:142-5. Received 4 February 2009; Accepted after revision 27 July 2009
142
The humerus is a common site for skeletal metastases in the adult. Surgical stabilisation of such lesions is often necessary to relieve pain and restore function. These procedures are essentially palliative and should therefore provide effective relief from pain for the remainder of the patient’s life without the need for further surgical intervention. We report a retrospective analysis of 35 patients (37 nails) with symptomatic metastases in the shaft of the humerus which were treated by locked, antegrade nailing. There were 27 true fractures (73.0%) and ten painful deposits (27.0%). Relief from pain was excellent in four (11.4%), good in 29 (82.9%) and fair in two (5.7%) on discharge. Function was improved in all but one patient. One case of palsy of the radial nerve was noted. The mean postoperative survival was 7.1 months (0.2 to 45.5) which emphasises the poor prognosis in this group of patients. There were no failures of fixation and no case in which further surgery was required. Antegrade intramedullary nailing is an effective means of stabilising the humerus for the palliative treatment of metastases. It relieves pain and restores function to the upper limb with low attendant morbidity.
The humerus is the second most common long bone to be affected by metastatic disease. This is often associated with pain and loss of function, particularly if a pathological fracture occurs.1-3 While the management of potentially resectable skeletal malignancy is the province of specialist centres, the general orthopaedic surgeon may be called upon to undertake the surgical stabilisation of symptomatic humeral metastases. The principal aim of surgery in such cases is to relieve pain and to restore the function of the upper limb by stabilising the humerus with fixation which should last for the lifetime of the patient.4 The surgical options described1 for the management of symptomatic metastases of the shaft of the humerus include intramedullary (IM) nailing and plate fixation. Such techniques may be combined with curettage of the tumour and/or augmentation with polymethylmethacrylate (PMMA) cement.5 The use of a locked antegrade IM nail has the advantage that it is a procedure familiar to most orthopaedic surgeons for the stabilisation of traumatic nonpathological fractures of the humerus. While some information is available about survival after the development of bony metastases,3 there is little specific data on survival after the development of symptomatic metastases
in a long bone, particularly in the upper limb. Such information would be invaluable in the preoperative planning, counselling and consenting of patients who require such surgery. Our aim was to assess whether locked antegrade IM nailing is effective as a single, durable surgical intervention for the stabilisation of symptomatic diaphyseal metastases in the humerus. In addition, we wished to provide survivorship data for specific types of tumour which commonly metastasise to the humerus.
Patients and Methods Our study was undertaken at the Western Infirmary, Glasgow, a teaching hospital which provides orthopaedic services to the Beatson Oncology Centre, a specialist regional oncology unit servicing 16 hospitals in the West of Scotland with a catchment population of approximately 2.6 million. We retrospectively identified all patients who had undergone antegrade, locked IM nailing of a symptomatic diaphyseal metastasis in the humerus, between January 1996 and October 2006 for a true fracture, or an impending fracture, associated with unremitting pain despite regular analgesia with opiates. Their medical notes and radiographs were retrieved and studied. We identified 35 patients (37 nails). Two, both with primary THE JOURNAL OF BONE AND JOINT SURGERY
LOCKED INTRAMEDULLARY NAILING OF SYMPTOMATIC METASTASES IN THE HUMERUS
Table I. Details of humeral metastases according to the type of primary tumour in the 35 patients Type of tumour
Number of patients (%)
Breast Bronchial Renal Unknown Gastrointestinal track Malignant melanoma Prostate
16 5 4 4 3 2 1
Fig. 1a
(45.7) (14.3) (11.4) (11.4) (8.6) (5.7) (2.9)
Fig. 1b
Radiographs showing pathological fracture of the humerus from metastasis a) before and b) after intramedullary fixation.
breast cancer, had bilateral nails inserted under the same anaesthetic. There were 24 women and 11 men with a mean age at the time of presentation of 61.5 years (31 to 80). In all, there were 27 cases of true fracture (73.0%) and ten of painful metastatic deposits (27.0%). The primary types of tumour are listed in Table I. Of the 35 patients, 25 received adjuvant local radiotherapy. Metastases had been diagnosed by plain radiography, Tc99m scanning or MRI. All the patients had full-length plain radiography of the humerus for surgical planning. Each had advanced local disease with metastatic deposits at the time of surgery for which palliative stabilisation was felt to be the most appropriate treatment. Of the 35 patients, 30 had other bony metastases and 21 had visceral metastases at the time of presentation. Only three had a solitary metastasis. All patients were thought to be fit enough to undergo surgery which was only undertaken after discussing all the treatment options, including conservative management. The operative technique was essentially the same in each case and used a deltoid-splitting approach to the rotator cuff to insert the humeral nail. Proximal locking was VOL. 92-B, No. 1, JANUARY 2010
143
Table II. The scoring system of Perez et al6 for pain Score
Definition
Excellent Good
Complete resolution of pain with no use of analgesia Nearly complete resolution and occasional use of analgesia Some reduction in pain with regular use of analgesia No difference in pain when compared with pre-operative levels
Fair Poor
performed using the appropriate guides and distal locking using a freehand technique. We used the AO humeral nail (Synthes, Solothurn, Switzerland) until 2004 (29 nails), and the Stryker T2 humeral nail (Stryker, Kalamazoo, Michigan) thereafter (eight nails). Radiographs of the humerus were taken post-operatively before discharge and in the fracture clinic, with a mean radiological time of follow-up of 5.2 weeks (median 2; Fig. 1). Adjuvant radiotherapy was performed at the discretion of the oncologists. Post-operative relief from pain was assessed using the scoring system of Perez, Bradfield and Morgan6 which assesses pain on the basis of the patient’s use of analgesia (Table II). The score is based on the individual’s requirements for analgesia at the time of discharge. This was usually on the third post-operative day. The outcome was determined from the patients’ medical records or death certificates. All the patients were discharged to the care of the oncologists after assessment of pain and function by the operating orthopaedic consultant or specialist registrar. In all, 18 patients (51.4%) returned to the orthopaedic clinic after a mean of 9.5 weeks (2 to 110). While formal function scores were not recorded, patients were assessed with respect to their subjective ability to perform activities of daily living (ADLs). Because of the nature of the underlying disease process with limited predicted survival, in many cases it was not possible (or appropriate) to review the patient routinely in the orthopaedic clinic. Five had died before follow-up (14.3%), nine were too unwell (25.7%), two had been placed in a hospice (5.7%) and one had a large distance to travel to hospital (2.9%). If the patient was not able to attend for any of the reasons, postoperative follow-up was undertaken by the general practitioner, oncologist or palliative care team. All were given open appointments to return should any problems be identified by the clinician responsible for care in the home/ hospice setting. Statistical analysis. This was performed using SPSS version 15.0 software (SPSS Inc., Chicago, Illinois). A p-value < 0.05 was considered statistically significant.
Results At the time of review, October 2006, 33 of 35 patients (94.3%) had died. Two had survived for 18.9 and 45.5 months, respectively (Fig. 2). The mean post-operative sur-
144
S. J. SPENCER, G. HOLT, J. V. CLARKE, A. MOHAMMED, W. J. LEACH, J. L. B. ROBERTS
Cumulative survival (%)
Table III. Mean (SD) survival by type of tumour
Survival function Censored
1
0.8
0.6
Tumour
Number of patients Mean survival (mths)
All types Breast Bronchial Renal Gastrointestinal track
35 16 5 4 3
7.1 9.7 3.2 3.7 1.3
(2.0) (3.4) (2.0) (2.4) (0.07)
0.4
0.2
0 0
30 40 10 20 Time from fracture to death (mths)
50
Fig. 2 A Kaplan-Meier survival curve showing the cumulative survival of the 35 patients.
vival for the entire series was 7.1 (SD 1.9, median 2.5). Death was the result of the primary malignancy in 31 patients (93.9%), and cerebrovascular accident secondary to cerebral metastases in two (6.1%). The mean survival was calculated for each type of tumour and is given in Table III. There was no significant difference in survival between cases of impending and true fractures (analysis of variance, p > 0.05). There was one case of pre-operative palsy of the radial nerve. Relief from pain was excellent in four patients (11%), good in 29 (83%) and fair in two (6%). Function was improved in 36 arms (97%) such that the patient could perform most activities of daily living. In the remaining patient function was poor because of paralysis of the deltoid by the tumour. The initial post-operative radiographs were available in each case. All the implants were found to be in a satisfactory position with appropriate proximal and distal locking. In the 18 patients (51.4%) who returned to the orthopaedic clinic after a mean of 9.5 weeks (median 8), no loss of fixation was observed. In the remaining patients who could not attend follow-up for the reasons stated previously, no further radiological examination was performed. It should be noted that no referrals to the orthopaedic clinic were made regarding these remaining patients by their supervising physician. No patient needed any further surgery on the affected humerus.
Discussion The mean survival after presenting with bony metastases at any site is said to be 24 months for breast cancer and 20 months for prostate cancer.3 The patients in our study were referred for palliative stabilisation and were known to have additional bony and/or visceral metastases. Consequently, the mean survival in our series of 7.1 months
(median 2.5) reflects the advanced nature of the disease. Other studies of palliative IM stabilisation of the humerus have shown similarly poor survival rates of between four and 12 months.2,7,8 They have also shown a marked difference in survival rates for types of tumour. Patients with metastatic breast disease had a mean survival of 9.7 months (SD 3.4) but much shorter survival times were observed for those with bronchial (3.2 months), renal (3.7 months) and gastrointestinal carcinomas (1.3 months). Painful humeral metastases without fracture and a low risk of progression to fracture may be treated effectively by radiotherapy and immobilisation in a sling.1,5,9-11 In appropriate patients, once a pathological fracture has occurred or there is a high risk of impending fracture1,5,9,10 we prefer surgical stabilisation to conservative treatment. Patients with established fractures are poorer candidates for conservative treatment than those without.5 Unlike its nonpathological counterpart, the potential for healing of a pathological humeral fracture is low12,13 and if healing occurs, it tends to be at a slower rate.5 Braces and casts are not effective for controlling pain after fracture.5,13 If the decision to operate has been made, internal fixation followed by radiotherapy is appropriate for the patient with a limited life expectancy.14 Most of the literature on antegrade locked humeral nailing reports good early relief from pain and restoration of function without significant morbidity.2,7,15 Possible options for supplementing fixation of the nail are augmentation with PMMA cement15 or open curettage and allografting of the tumour.7 At present there is no good evidence to suggest that either of these methods is any better than locked IM nailing. As far as we are aware, our study is the largest published series in which unaugmented locked IM nailing was used for stabilisation of metastasic deposits in the diaphysis of the humerus. Most of our patients had already sustained a pathological fracture by the time of referral. We acknowledge that the identification of patients who will benefit from fixation for an impending humeral fracture remains controversial. In the upper limb, when loss of cortical bone approaches 75%, the risk of fracture is high.1,5 When only one cortex is intact the humerus may fracture after only minor activity.1 Purely lytic lesions are more prone to fracture, as are lesions in the diaphysis of the humerus which are more sensitive to cortical bone loss than those in the metaphysis or the epiphysis.1 THE JOURNAL OF BONE AND JOINT SURGERY
LOCKED INTRAMEDULLARY NAILING OF SYMPTOMATIC METASTASES IN THE HUMERUS
We do not suggest that all patients with humeral metastases should be treated surgically. Indeed, painful lesions without fracture or a low risk of fracture can be effectively treated by radiotherapy alone. Given the advanced nature of the disease in those considered for palliative stabilisation, some are not fit enough or simply have too short a predicted survival to make surgical intervention a sensible option. In these patients palliative external beam radiotherapy may help to reduce the pain from the lesion. This assessment of risk to benefit can be difficult, particularly in the presence of advanced disease or an intact bone. In our department, we make the final decision in conjunction with oncologists and anaesthetists. We acknowledge that other orthopaedic centres may use a different set of criteria and therefore choose to treat a higher proportion of cases conservatively. Our findings confirm that locked antegrade IM nailing provides good early relief from pain and return of function. The mean patient survival after surgery was only 7.1 months (median 2.5), reflecting the advanced nature of the disease in those undergoing palliative stabilisation. A marked difference was noted in the mean survival rate in primary types of tumour; breast (9.7 months), bronchial (3.2 months), renal (3.7 months) and gastrointestinal (1.3 months). Given the low rate of complications observed in our study, the poor prognosis associated with this group of patients and the fact that there were no cases in which the method of fixation failed, it is our opinion that locked IM humeral nailing remains a good treatment option for pathological fractures of the humeral shaft or impending fractures in patients who are fit enough for surgery.
VOL. 92-B, No. 1, JANUARY 2010
145
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 1. Frassica FJ, Frassica DA. Evaluation and treatment of metastases to the humerus. Clin Orthop 2003;415(Suppl):212-18. 2. Redmond B, Biermann J, Blasier R. Interlocking intra-medullary nailing of pathological fractures of the shaft of the humerus. J Bone Joint Surg [Am] 1996;78-A:891-6. 3. Rubens RD. Bone metastases: the clinical problem. Eur J Cancer 1998;34:210-13. 4. Tillman RM. The role of the orthopaedic surgeon in metastatic disease of the appendicular skeleton. J Bone Joint Surg [Br] 1999;81-B:1-2. 5. Frassica FJ, Frassica DA. Metastatic bone disease of the humerus. J Am Acad Orthop Surg 2003;11:282-8. 6. Perez CA, Bradfield JS, Morgan HC. Management of pathological fractures. Cancer 1972;29:684-93. 7. Tome J, Carsi B, Garcia-Fernandez C, Marco F, Lopez-Duran Stern L. Treatment of pathological fractures of the humerus with Sidel nailing. Clin Orthop 1998;350:51-5. 8. Bauze AJ, Clayer MT. Treatment of pathological fractures of the humerus with a locked intra-medullary nail. J Orthop Surg 2003;11:34-7. 9. Agarawal JP, Swangsilpa T, Van der Linden Y, et al. The role of external beam radiotherapy in the management of bone metastases. Clin Oncol (R Coll Radiol) 2006;18:747-60. 10. Van der Linden YM, Dijkstra PDS, Kroon HM, et al. Comparative analysis of risk factors for pathological fracture with femoral metastases: results based on a randomised trial of radiotherapy. J Bone Joint Surg [Br] 2004;86-B:566-73. 11. McQuay HJ, Carroll D, Moore RA. Radiotherapy for painful bone metastases: a systematic review. Clin Oncol (R Coll Radiol) 1997;9:150-4. 12. Ekholm R, Adami J, Tidermark J, et al. Fractures of the shaft of the humerus: an epidemiological study of 401 fractures. J Bone Joint Surg [Br] 2006;88-B:1469-73. 13. Flemming JE, Beals RK. Pathological fracture of the humerus. Clin Orthop 1986;203:258-60. 14. Katagiri H, Takahashi M, Wakai K, et al. Prognostic factors and a scoring system for patients with skeletal metastasis. J Bone Joint Surg [Br] 2005;87-B:698-703. 15. Atesok K, Liebergall M, Sucher E, et al. Treatment of pathological humeral shaft fractures with unreamed humeral nail. Ann Surg Oncol 2007;14:1493-8.
