Long-term survival after surgical intervention for bone disease in multiple myeloma

F. Zeifang1,*, A. Zahlten-Hinguranage1, H. Goldschmidt2, F. Cremer2, L. Bernd1 and D. Sabo1

Departments of 1 Orthopedic Surgery and 2 Internal Medicine V, University of Heidelberg, Heidelberg, Germany

* Correspondence to: Dr F. Zeifang, Department of Orthopedic Surgery, University of Heidelberg, Schlierbacher Landstrasse 200, D-69 181 Heidelberg, Germany. Tel: +49-6221-969254; Fax: +49-6221-969288; Email: felix.zeifang{at}ok.uni-heidelberg.de


    Abstract
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
Background: We describe the surgical treatment, outcome and long-term survival of patients with multiple myeloma (MM) in response to conventional (CC) or high-dose (HDT) chemotherapy.

Patients and methods: Eighty-four patients diagnosed with MM were recruited for the study (51 male, 33 female; median age 62 years) and consecutively surgically treated in a single institution during a 12-year period. The main end point of the study was overall survival after surgery. Cox regression analysis was used to estimate the effect of factors that may predict survival.

Results: Spinal surgery was performed in 54 cases, and 30 patients were surgically treated at the extremities. The post-surgical complication rate was low (17%; 14/84 patients). The median overall survival time was 47 months. Patients receiving HDT had a longer 5-year overall survival rate than patients receiving CC (51% versus 33%). Univariate predictors of mortality included age >65 years [risk ratio (RR) 1.62; P=0.023], osteolyses in long weight-bearing bones (RR 2.23; P=0.007) and an elevated C-reactive protein level >5 mg/l (RR 1.82; P=0.016); the latter remained significant as a predictor in multivariate analysis (RR 2.66; P=0.0209).

Conclusions: Given the high number of patients reaching 5-year overall survival and the low post-surgery complication rate, surgery should pursue a long-term stable reconstruction of the affected bone.

Key words: bone, mortality, multiple myeloma, oncological surgery


    Introduction
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
The major clinical manifestation of multiple myeloma (MM) results from bone destruction. Sixty to eighty per cent of newly diagnosed patients present with clinical symptoms related to bone disease such as bone pain, hypercalcaemia, fractures and spinal cord compression [1Go]. The surgical outcome of lytic bone lesions in MM is often compared with that of bone metastases. The overall survival time in metastatic bone disease ranges between 6 and 22 months [2Go, 3Go] depending on the type of primary tumour. The prognosis for patients with MM has improved significantly in recent years. Owing to advances in systemic therapy, e.g. high-dose chemotherapy (HDT) and peripheral blood stem cell transplantation (PBSCT), overall survival rates of >5 years can be achieved [4Go, 5Go].

To date, only a few studies have been published reporting the clinical course of patients after surgical treatment of MM [6Go–8Go]. These studies have dealt with small numbers of patients and did not include an assessment of preoperative biochemical findings that are of prognostic significance for MM patients.

In this explorative study we analysed surgically treated patients with MM. Our primary aim was to describe the surgical treatment, management and outcomes in response to conventional chemotherapy (CC) or HDT, as well as to assess long-term survival. Multivariate analysis was used to estimate the effect of factors, e.g. serum lactate dehydrogenase (LDH), serum ß2-microglobulin level (B2M), C-reactive protein (CRP) and performance status, measured at the time of surgery that may function as predictors of survival.


    Material and methods
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
Patients
Eighty-four patients with MM who were consecutively surgically treated in the authors' institution between January 1990 and March 2002 were retrospectively recruited. The median age at surgery was 62 years (range 29–90). Fifty patients (60%) were <65 years old. Patient survival was followed up for a median time of 46 months. The baseline characteristics of patients are summarised in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Baseline characteristics of the 84 patients surgically treated for multiple myeloma of bone

 
Additional systemic therapy was given as either single CC dose in 38 patients with a median of six cycles, or as HDT with PBSCT in 30 patients. No chemotherapy was administered to 16 patients. All patients received supportive care measures with bisphosphonates, which is currently the therapy of choice in MM patients with hypercalcaemia.

Most of the patients were mobile and their overall functional status was equivalent to a Eastern Cooperative Oncology Group (ECOG) performance status of 1 or 2 [9Go]. Four of 84 patients (5%) were capable of only limited self-care and were confined to a bed or chair for >50% of waking hours. One patient was completely disabled.

Methods
The patients were retrospectively assessed for clinical stage according to Durie and Salmon [10Go] and for established biochemical markers [11Go], i.e. B2M, CRP and LDH, at surgery. Cytogenetic abnormalities were not addressed.

Particular details of surgical and oncological outcome were obtained during physical examination or retrieved from the Orthopaedic Oncology Cancer Registry that is maintained at the authors' institution.

The indication for surgical treatment at the extremities included lesions with elevated fracture risk according to Mirels' Scoring System [12Go] and known fractures. Indication for surgical intervention at the vertebral column included progressive neurological impairment (complete or incomplete neurological deficits, corresponding to Frankel grade A–D [13Go] assessing sensory and motor function) or severe vertebral body lesions. Irrespective of the anatomical site, surgical intervention was indicated for intractable pain. The neurological outcome was assessed prior to and 1 month after surgery.

Based on Ruggieri et al.'s classification [14Go], post-treatment complications were categorised as ‘minor’, not requiring surgery (grade I or II), or ‘major’, requiring surgery (grade III–V). Complications were categorised into four groups. Type A (local wound complications): wound infection, fistula, seroma, haematoma, prolonged wound healing and skin necrosis. Type B (implant failure): septic or aseptic endoprosthetic or internal fixation device loosening, dislocation, luxation, implant fracture, pseudarthrosis. Type C (non-orthopaedic complications): pulmonary embolism, haemodynamic failure, deep venous thrombosis. Type D (others): perioperative nerve injuries, differences in leg length, lymphoedema, decubitus ulcer.

Statistical methods
Descriptive statistics were performed for all variables. Categorical data are expressed as percentages with associated 95% confidence intervals (CIs). Continuous data are expressed as medians with 5th and 95th percentiles. Observed overall survival estimates were based on Kaplan–Meier survival curves [15Go]. Univariate and multivariate Cox regression analyses were performed to identify effects of factors that might predict survival [16Go]. Risk ratios were estimated and given with 95% CIs. On account of the observational character of this study, which included patients treated from 1990 to 2002, a stratified Cox proportional hazard model was calculated standardising for differences in adjuvant treatment such as conventional or HDT and PBSCT in the subgroups. Complete sets of data were available for 72 patients (86%) and were included in the analysis. All calculations for significance were two-sided and were performed at the 5% level. The statistical software used for the calculation was SAS, version 8.2 (SAS Inc., Cary, NC, USA).

For power calculations, a rule of thumb for regression equation was applied, proposing that when using six or more predictors, an absolute minimum of 10 participants per independent predictor is appropriate [17Go].


    Results
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
Patient characteristics
At first presentation in the department of orthopaedic surgery, the patients' most common complaints were bone pain. In 35 patients, compression fractures of the thoracic or lumbar vertebral body and, in 10 patients, pathological fractures of the long bones led to the diagnosis of MM.

Anatomical sites of bone lesions were non-extremities (29 thoracic vertebrae, 14 lumbar vertebrae, six cervical vertebrae, three costa, one sternum, one os ilium) or extremities (17 femur, 11 humerus, one ulna, one tibia).

Primary indications for surgery of long bones were pathological fractures (15 femur, 11 humerus, one ulna and one tibia) or an impending fracture risk with a Mirels score >8 (two femur). Fifteen of the patients underwent local irradiation prior to surgery. Surgery of the vertebral column was indicated for progressive neurological deficiency (18 thoracic vertebrae, five lumbar vertebrae) or impending instability (six cervical vertebrae, 11 thoracic vertebrae and nine lumbar vertebrae). Indications for surgery of the thoracic skeleton or the pelvic bone were pathological fractures (two costa), pain (one sternum, one costa) and an impending fracture (one os ilium). Neurological impairments were classified according to Frankel A (one), Frankel B (three), Frankel C (six) and Frankel D (12).

The serum B2M level ranged between 1.3 and 11.2 mg/l (median 5.5) at surgery. In 40 patients (55%), the B2M level exceeded 2.5 mg/l; three (8%) of these patients had values >3 mg/l and 25 (62%) had >3.5 mg/l. The median CRP level was 3.6 mg/l (0.2–303) and was above the standardised upper normal limit of 5 mg/l in 14 patients (16%). An elevated LDH level >240 U/l was found in seven patients (9.7%).

Age-related chronic health problems (11 heart disease, three pulmonary disease, five diabetes, 11 hypertension) reflect the general health status of the study population.

Fifteen patients with single thoracic or lumbar vertebral body lesions were treated by a combined approach with anterior resection and posterior instrumentation [18Go, 19Go]. When contiguous vertebral bodies were involved or the patients' general health status was reduced, a one-stage posterior decompression–stabilisation procedure was performed (n=18) [8Go, 20Go]. Tumour surgery in the cervical vertebrae was performed only by ventral decompression and stabilisation (n=6). Owing to the risk of vertebral instability, decompressive laminectomy alone was not indicated [21Go, 22Go].

Femoral or humeral articular or metaphyseal fractures were replaced with cemented endoprostheses. In the subtrochanteric region long-stem femoral endoprostheses were used (n=8), in one case following a failed proximal femoral nail. In 16 patients diaphyseal osteolytic lesions were reconstructed with compound osteosynthesis.

The overall complication rate following surgery was 17% (14/84). Two patients had more than one complication. Local wound complication (type A) occurred in six cases, predominantly following plate osteosynthesis. Three required re-operation. Implant failure (type B) was observed in five cases. One patient died during the first month after surgery due to heart failure. Other non-orthopaedic complications (type C) were pulmonary embolism (two) and venous thromboembolism (one) (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Type of complications and therapies

 
Local recurrences occurred in five out of 84 patients (6%), four times following surgery of the vertebral column and once following internal hemipelvectomy. Forty-eight of 84 patients developed additional skeletal lesions during the course of the disease. The majority of these were locally irradiated; 14 needed surgical intervention.

Overall survival
Observed overall survival since diagnosis for surgically treated patients is shown in Figure 1. Overall survival estimates at 1, 3, 5 and 10 years were 86.8%, 68%, 50% and 30.1%, respectively.



View larger version (11K):
[in this window]
[in a new window]
 
Figure 1. Overall survival estimates after diagnosis for 84 surgically treated patients with bone disease in multiple myeloma (median 47 ± 6.43 months).

 
The median overall survival times for patients since surgery were 47 (±17) months. They were followed up for a mean of 3.5 years. The probability of surviving 1 year after surgery was 76%; 3- and 5-year survival probabilities were 59% and 40%, respectively.

Predictors of survival
Univariate regression analysis revealed significantly poorer predictive values for age >65 years [risk ratio (RR) 1.62; P=0.023], osteolyses in long weight-bearing bones (RR 2.23; P=0.007) and an elevated CRP >5 mg/l (RR 1.82; P=0.016). HDT and PBSCT significantly improved survival (RR 0.49; P=0.0225) (Table 3).


View this table:
[in this window]
[in a new window]
 
Table 3. Univariate Cox regression analysis

 
The median survival time for age >65 years was 33 versus 59 months for patients <65 years old. A tumour in long weight-bearing bones was less favourable to survival than a thoracic or vertebral tumour location (21 versus 66 months). Patients who had an elevated CRP level of >5 mg/l had a poorer life expectancy (median 25 versus 59 months). Improved 5-year overall survival rates were shown for patients treated with HDT and PBSCT compared with patients receiving CC or without chemotherapy (51% versus 33% and 25%, respectively) (Figures 24).



View larger version (16K):
[in this window]
[in a new window]
 
Figure 2. Overall survival estimates since surgery by C-reactive protein (CRP) (81% and 46% versus 64% and 12%, P=0.016).

 


View larger version (18K):
[in this window]
[in a new window]
 
Figure 4. Five-year overall survival estimates since surgery by high-dose chemotherapy/peripheral blood stem cell transplantation and conventional chemotherapy or without chemotherapy (51% versus 33% and 25%, P=0.05).

 
Stratified multivariate analysis estimated an elevated CRP level >5 mg/l as the only significant independent risk factor for mortality (P=0.0209), with the risk of death increased to RR 2.6.


    Discussion
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
Owing to a change in methodology from CC followed by HDT with PBSCT, a considerable improvement in overall survival could be achieved for patients with newly diagnosed MM [5Go]. Their actuarial survival rate is far better than for patients with bone metastases. This accounts for the fact that in myeloma patients requiring orthopedic surgery, a treatment option should be chosen that consists of a fit and stable reconstruction of the bone defects. Only few studies, which were quite small, have been published reporting the clinical course of patients surgically treated for MM [6Go–8Go].

In this study population, presenting features, e.g. neurological impairments, prior to spinal surgery were more prevalent (22/49; 45%) than in other studies (6/27; 22%) [8Go]. However, 14 out of 49 patients (29%) showed improved neurological function after surgery, 10 of them following dorsal decompression and stabilisation. No patient developed progressive neurological impairment after surgery. Similar results were reported in patients following posterior decompression and stabilisation for metastatic spinal cord compression by Rompe et al. [23Go]. Other authors report neurological improvement in up to 81% of patients with spinal neoplasm following combined anterior–posterior approaches [19Go, 24Go]. Furthermore, from our own results we maintain that the prognosis for neurological recovery is adversely affected by both a greater degree and a longer duration of canal narrowing, indicating that patients may benefit from early decompression regardless of the chosen surgical procedure [25Go].

The high benefit of surgery for patients with MM is further demonstrated by the low post-surgical complication rate. With 8% (4/49) it was lower than that reported for patients with metastases following surgery. There was no evidence of re-fractures in operated limbs. Pascal-Moussellard et al. [26Go] report a complication rate of 19% (17/145) following vertebral metastasis surgery. Even with an anterior–posterior approach, a complete resection in the vertebral column is not possible and is contraindicated for patients with a poor health status. In any case, radiation therapy for residual tumour following incomplete resection is performed, as the vertebral column is the location mostly affected by local recurrences (four out of five).

To date, there is no evidence from controlled clinical trials with a long-term follow-up to show to what extent percutaneous vertebroplasty/kyphoplasty may be considered as an additional therapeutic option for patients with lytic lesions due to MM, but early results are promising [27Go]. Based on this study, further outcome studies should be undertaken to evaluate the effectiveness of kyphoplasty.

The overall complication rate after reconstructive surgery at the extremities was low, except for compound plate osteosynthesis (three local wound complications, one implant failure). Despite the generally assumed immunodeficiency for patients with MM and reduced partial thromboplastin time in 22% of the patients, local wound complications or haematomas occurred in only six patients.

While univariate analyses revealed patient age >65 years, osteolyses in long weight-bearing bones and an elevated CRP >5 mg/l at surgery negatively predict survival, the application of HDT with PBSCT decreased the relative risk of death by an RR of 0.5.

In contrast to the proven high prognostic value of B2M at diagnosis, this value was not found to be significant. This might be due to the fact that B2M, which reflects the tumour mass, was assessed in response to CC or HDT, i.e. at the time of maximum tumour reduction in the patient. The negative effect of an increased CRP level before surgery on survival might be attributed to a disease progression as CRP is regulated by interleukin-6, a growth factor for MM [28Go].

In earlier reports, the anatomical site of lytic bone lesions, Salmon–Durie stage III and advanced age were indicated as negative predictors for survival [11Go]. The 5-year actuarial survival rate for non-extremity location was 52 months versus extremity location 17 months (RR 2.0; P=0.017). It can be assumed that plasma cells infiltrate first the axial skeleton, which leads to the compression of marrow. With increased cellular proliferation, extensive bone destruction, pathological fractures, hypercalcaemia and osteolyses in long weight-bearing bones becomes evident, indicating an advanced stage of disease [29Go]. The Salmon–Durie stage could not be identified as having a significant prognostic value; however, stage III increased the risk of death to RR 2.2 (P=0.126). Despite the fact that some authors consider age to be a prognostic factor for survival in patients with MM [30Go], we assume that the strong correlation between age >65 years and poor survival is partly accounted for by the fact that effective HDT and PBSCT were generally preserved in patients under the age of 65 years.

However, in multivariate analysis, stratified for HDT and PBSCT, the only independent predictor for survival that remained statistically significant was an elevated CRP >5 mg/l, with an increased risk for mortality of RR 2.6 (P=0.0209). With the given 72 complete datasets that were included in the analyses and eight independent variables, small effects probably could not be detected.


    Conclusions
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
This study confirmed improved long-term overall survival after diagnosis, especially for patients treated with HDT and PBSCT. The clinical implication of our findings, given the high percentage of patients reaching 5-year survival and the low post-surgery complication rate, is that the aim of surgery should be a fit and stable reconstruction of the affected bone.

Univariate predictors for mortality after surgery included age >65 years, osteolyses in long weight-bearing bones and an elevated CRP >5 mg/l before surgery. Although explorative in nature, based on these preliminary results, further investigation aiming to identify prognostic factors after surgical intervention for bone disease in MM, including kyphoplasty, should be undertaken.



View larger version (17K):
[in this window]
[in a new window]
 
Figure 3. One- and 5-year overall survival by anatomical site of tumour (85% and 52% versus 62% and 17%, P=0.0070).

 

    Acknowledgements
 
This work was supported in part by the University Research grant, Ministry of Science, Research and the Arts of Baden-Württemberg, Germany.

Received for publication July 12, 2004. Revision received September 16, 2004. Accepted for publication September 28, 2004.


    References
 Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
1. Lathinen R, Laakso M, Palva I et al. Randomised, placebo-controlled multicentre trial of clodronate in multiple myeloma. Lancet 1992; 340: 1049–1052.[ISI][Medline]

2. Tatsui H, Onomura T, Morishita S et al. Survival rates of patients with metastatic spinal cancer after scintigraphic detection of abnormal radioactive accumulation. Spine 1996; 21: 2143–2148.[CrossRef][ISI][Medline]

3. Wedin R, Bauer HC, Rutqvist LE. Surgical treatment for skeletal breast cancer metastases: a population-based study of 641 patients. Cancer 2001; 92: 257–262.[CrossRef][ISI][Medline]

4. Alexanian R, Dimopoulos M. The treatment of multiple myeloma. N Engl J Med 1994; 330: 484–489.[Free Full Text]

5. Child JA, Morgan GJ, Davies FE et al. High-dose chemotherapy with hematopoietic stem-cell rescue for multiple myeloma. N Engl J Med 2003; 348: 1875–1883.[Abstract/Free Full Text]

6. Kivioja AH, Karaharju EO, Elomaa I, Bowling TO. Surgical treatment of myeloma of bone. Eur J Cancer 1992; 28A: 1865–1869.[CrossRef][ISI][Medline]

7. Papagelopoulos PJ, Galanis EC, Greipp PR, Sim FH. Prosthetic hip replacement for pathologic or impending pathologic fractures in myeloma. Clin Orthop 1997; 341: 192–205.[Medline]

8. Dürr HR, Wegener B, Krodel A et al. Multiple myeloma: surgery of the spine: retrospective analysis of 27 patients. Spine 2002; 27: 320–324.[CrossRef][ISI][Medline]

9. Oken MM, Creech RH, Tormey DC et al. Toxicity response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol 1982; 5: 649–655.[ISI][Medline]

10. Durie BFM, Salmon SE. A clinical staging system for multiple myeloma. Cancer 1975; 36: 842–854.[ISI][Medline]

11. Bataille R, Boccadoro M, Klein B et al. C-reactive protein and serum ß2-Microglobulin produce a simple and powerful myeloma staging system. Blood 1992; 80: 733–737.[Abstract]

12. Mirels H. Metastatic disease in long bones. A proposed scoring system for diagnosing impending pathologic fractures. Clin Orthop 1989; 249: 256–264.[Medline]

13. Frankel HL, Hancock DO, Hyslop G et al. The value of postural reduction in the initial management of closed injuries to the spine with paraplegia and tetraplegia. Paraplegia 1969; 7: 182.

14. Ruggieri P, De Cristofaro R, Picci P et al. Complications and surgical indications in 144 cases of nonmetastatic osteosarcoma of the extremities treated with neoadjuvant chemotherapy. Clin Orthop 1993; 295: 226–238.[Medline]

15. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.[ISI]

16. Cox DR. Regression models and life tables. J R Stat Soc [B] 1972; 34: 187–220.[ISI]

17. Harris RJ. A Primer of Multivariate Statistics, 2nd edition. New York: Academic Press 1985.

18. Sundaresan N, Steinberger AA, Moore F et al. Indications and results of combined anterior-posterior approaches for spine tumour surgery. J Neurosurg 1996; 85: 438–446.[ISI][Medline]

19. Harrington KD. Anterior decompression and stabilisation of the spine as a treatment for vertebral collapse and spinal cord compression from metastatic malignancy. Clin Orthop 1988; 233: 177–197.[Medline]

20. Hosono N, Yonenobu K, Fuji T et al. Orthopaedic management of spinal metastases. Clin Orthop 1995; 312: 148–159.[Medline]

21. Nather A, Bose K. The results of decompression of cord or cauda equina compression from metastatic extradural tumors. Clin Orthop 1982; 169: 103–108.[Medline]

22. Lee CK, Rosa R, Fernand R. Surgical treatment of tumors of the spine. Spine 1986; 11: 201–208.[ISI][Medline]

23. Rompe JD, Eysel P, Hopf C, Heine J. Metastatic spinal cord compression—options for surgical treatment. Acta Neurochir 1993; 123: 135–140.[ISI]

24. Kluger P, Korge A, Scharf HP. Strategy for the treatment of patients with spinal neoplasms. Spinal Cord 1997; 35: 429–436.[CrossRef][ISI][Medline]

25. Helweg-Larsen S. Clinical outcome in metastatic spinal cord compression. A prospective study of 153 patients. Acta Neurol Scand 1996; 94: 269–275.[ISI][Medline]

26. Pascal-Moussellard H, Broc G, Pointillart V et al. Complications of vertebral metastasis surgery. Eur Spine J 1998; 7: 438–444.[CrossRef][Medline]

27. Levine SA, Perin LA, Hayes D, Hayer WS. An evidence-based evaluation of percutaneous vertebroplasty. Manag Care 2000; 9: 56–60.

28. Alexandrakis MG, Passam FH. Ki-67 Proliferation Index. Correlation with prognostic parameters and outcome in multiple myeloma. J Clin Oncol 2004; 27: 8–13.[CrossRef]

29. Hannisdal E, Kildahl-Andersen O, Grottum KA, Lamvik J. Prognostic factors in multiple myeloma in a population-based trial. Eur J Haematol 1990; 45: 198–202.[ISI][Medline]

30. Tsuchiya J, Murakami H, Kanoh T. Ten year survival and prognostic factors in multiple myeloma. Br J Haematol 1994; 87: 832–834.[ISI][Medline]





This Article
Abstract
Full Text (PDF)
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Zeifang, F.
Articles by Sabo, D.
PubMed
PubMed Citation
Articles by Zeifang, F.
Articles by Sabo, D.