Secondary Logo

Journal Logo

Metastatic Tumors of the Spine

An Analysis of Decision Making and Treatment in Thoracolumbar Metastases

Polly, David W. Jr, MD*; Chou, Dean, MD; Sembrano, Jonathan N., MD*; Ledonio, Charles G. T., MD*; Tomita, Katsuro, MD

Author Information
doi: 10.1097/BRS.0b013e3181ba6d02
  • Free

The most common site of symptomatic spinal metastasis is in the thoracic and lumbar segments in 80% to 90% of spine metastasis cases. In 85% to 96%, axial pain is the most common symptom.1–4 There is a significant debate about the best surgical approach for the treatment of thoracolumbar metastasis. The objective of this study is to determine whether the surgical approach and technique to be used to surgically treat thoracolumbar metastases is influenced by anatomic region of the spine involved. To accomplish this, a systematic review of the literature was undertaken.

Materials and Methods

Two clinically relavant questions below were determined through a consensus among a panel of spine oncology experts (The Spine Oncology Study Group), and a systematic review of literature was conducted using Pubmed and a review of bibliographies of reviewed articles.

Specific Clinical Questions

  1. In patients with metastatic disease in the upper thoracic spine (T2–T5) spine region, and undergoing decompressive and stabilization surgery, what is the impact of different surgical approaches (anterior, posterior, and combined AP) on local tumor recurrence, perioperative morbidity and mortality, pain alleviation, and neurologic recovery?
  2. Within the thoracolumbar spine mid thoracic [T6–T10], thoracolumbar [T11–L2], and lumbar [L3–L5]), what is the impact of different surgical approaches (anterior, posterior, and combined AP) on local recurrence, perioperative morbidity and mortality, pain alleviation, and neurologic recovery?


Four outcomes were determined to be important in the comparison of different surgical approaches. Tumor recurrence and perioperative morbidity were deemed the most critical because these can be impacted by surgical approach. Pain alleviation was considered the next most important, primarily because of its importance in quality of life. Neurologic recovery was likely to be affected more by preoperative neurologic status and by disease stage than by surgical approach.

Search Criteria

The search terms included “spin*” and “metasta*.” This was to include the terms, “metastatic,” “metastasis,” “metastases,” “spinal,” and “spine.”

Criteria for possible inclusion are as follows:

  • Articles published between 1990 and 2008.
  • All articles in English or with an English translation.
  • Articles with 30 or more subjects.
  • Adult age group (18 years and older).
  • Articles describing surgical treatment of thoracolumbar metastatic disease.

Exclusion criteria include the following:

  • Primary tumors.
  • Intradural tumors.
  • Pediatric age group.
  • Reports of nonsurgical treatment of thoracolumbar metastatic disease.
  • Reports of cervical-only metastatic disease.
  • Articles with fewer than 30 subjects.
  • Articles with nonhomogeneous pathology (e.g., trauma, primary tumors, etc. in the same series).

Abstracts of articles that matched the search terms and criteria were reviewed by 4 independent reviewers and full text versions of acceptable articles were gathered. These articles were then each analyzed in detail, and study results pertaining to the research questions and outcome measures were tabulated.

The quality of evidence was evaluated as high, moderate, low or very low. The results of the systematic review and rating of the evidence were presented to a multidisciplinary, international group of spine oncology surgeons, oncologists, and methodologists (Spine Oncology Study Group). The group then went through a consensus based decision making process using a modified Delphi technique to arrive at treatment recommendations related to the key clinical questions. This process and the strength of the recommendation were based on the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) method.5,6


From the search terms used a total of 5176 abstracts were found. Each of these abstracts was personally reviewed, and based on these abstracts, 161 abstracts were deemed acceptable. Based on this, 161 abstracts were further strictly reviewed according to inclusion and exclusion criteria mentioned above, and this further trimmed the number of articles to 60. These 60 articles were pulled and reviewed in detail. Articles that were found to not meet inclusion criteria were again excluded, leaving a final number of 32 articles included in this review (mentioned in Table 1).

Table 1
Table 1:
References Included in the Review
Table 1
Table 1:
Table 1
Table 1:
Table 1
Table 1:
Table 1
Table 1:

Of the 32 articles reviewed, there was not a single level I study. There was 1 level II study,7 5 level III studies,3,4,8–10 and 26 level IV studies.1,2,7,12–34 These articles were then evaluated independently by the authors according to the GRADE criteria.35

There were 5 studies that reported exclusively on the results of anterior approach surgery, with a total of 264 patients.1,2,13,17 Each of these articles presented very low quality evidence supporting use of an anterior approach over a posterior approach for thoracic and lumbar spine metastases. One study routinely used an anterolateral thoracotomy for the thoracic spine and a retroperitoneal flank approach for the lumbar spine.17 Another study recommended region-specific anterior approach within the thoracic spine as follows: sternotomy for T1–T2; “trap door” exposure for T3–T4; posterolateral thoracotomy for T5–T10; and thoracoabdominal for T11–T12.1

Seven studies reported almost exclusively on the results of posterior approach surgery, with a total of 545 patients.9,11,20,25,27,34 Each of these studies presented very low quality evidence supporting use of a posterior approach over an anterior approach for thoracic and lumbar spine metastases. One study presented very low quality evidence to support laminectomy without stabilization.9 All other studies used some form of posterior stabilization either with hooks or screws. Four studies each presented very low quality evidence to support posterior decompressive laminectomy with stabilization.11,20,25,27 Each of the 2 studies presented very low quality evidence in support of a posterolateral approach. One presented evidence to support single-stage posterolateral vertebrectomy (SPLV) with posterior stabilization34; the other presented evidence to support a posterolateral transpedicular approach to decompression and reconstruction, with posterior stabilization.14 In the latter study, 3 cases of combined AP approach cases were included in the series; however, these were vastly outnumbered by the 28 posterior approach cases. None of these studies presented region-specific differences within the thoracic and lumbar spine in terms of type of posterior approach.

No study reported exclusively on the results of a combined AP approach for thoracic and lumbar metastases.

Twenty studies included various combinations of anterior, posterior, and combined AP approaches in their series.3,4,8,12,15,16,18,19,21–24,26,28–33 The aggregate total number of cases for each approach are: 495 anterior, 1066 posterior, and 279 combined AP approaches. In 4 studies, no clear distinction was made between approaches. These include 57 “anterior or posterior”32; 18 “anterior or AP”4; 22 “posterior after previous anterior”15; and 24 “anterior, posterior, or AP.”10 In 16 of these 20 studies, the results were reported as an aggregate, with no attempt to differentiate the results of each approach to determine superiority of one approach over another. In most of these studies, the choice of approach was selected largely based on tumor topography in relation to the anatomic parts of the vertebra involved. With the exception of 1 study, each of these 16 studies presented very low quality evidence supporting the use of anterior, posterior, and AP surgical approaches for thoracic and lumbar metastases. The lone exception is a study exclusively on adenocarcinoma of unknown primary, which concluded results are invariably poor for this particular diagnosis, regardless of whether surgery is performed or not, and regardless of approach.28

Four studies made some attempt at comparing results by approach. In 1 study, multiple regression analysis was performed to look for factors associated with longer survival, and a posterior approach was found to be one such factor.8 However, this study included a disproportionately high number of posterior approach cases (79) compared to anterior (13) and AP (9), and the evidence is considered very low quality. Another study retrospectively compared posterolateral costotransversectomy approach against anterior or AP approach, and found a similar incidence and severity of perioperative complications.4 A third study retrospectively compared results of anterior, posterior, and AP approaches for lumbar spine metastases.3 They found that outcomes for either of these approaches are similar when the approach chosen mirrors the anatomic distribution of the disease. However, when lumbar vertebrectomy is necessary, there is very low quality evidence that an anterior approach is preferable in terms of minimizing blood loss and wound-related complications. The fourth article was a prospective cohort study on 42 cases of SPLV, and compared this against a group that underwent a variety of other approaches (anterior, posterior, AP). The authors concluded that the results were superior for the SPLV group; however, because detailed reporting of results was only done for this group, and lack of detail on the surgical approach and technique used in the heterogeneous group it was compared against, this is considered low quality evidence.10

Most of the 20 mixed-surgical approach studies did not break down their surgical approach according to specific regions in the thoracic and lumbar spine. The authors of 1 study preferentially treated upper thoracic (T2–T5) lesions via bilateral transpedicular approach; however, there was no attempt at randomization.14

The single level II study in this review was a prospective, international, multicenter study that compared the results of en bloc excision, debulking, and palliative surgery.7 Results showed that those who underwent excision survived significantly longer than those in the palliative group. However, since this was not a randomized study, there is expectedly a strong selection bias in that patients with a shorter life expectancy would be getting palliative rather than excisional surgery. As such, this evidence is considered low quality in favor of en bloc excisional surgery. Unfortunately, the surgical approach used was not included in the reporting of the results, and thus no evidence in that regard is gleaned from this study.

Further specific analysis of each of the articles did not yield compelling evidence for one approach over another, but it did shed some light onto different parameters. With regards to wound complications related to approach, it did appear that the posterior approach had a higher rate of wound problems. From a posterior approach, the range of wound healing or infection problems were: 19%,11 12%,10 5%,13 7%,3 12%,19 12%,20 2.4%,25 3%,2911.4%,34 and 6.7%.4 The range of wound healing or infection problems from an anterior approach were: 5%,13 3%,1 1%,2 and 1.7%.4 For anterior-posterior approaches, the range of wound infection or healing problems were: 11%,15 11%,16 13.7%,24 and 25%.32

We further subanalyzed the wound healing issue with regards to previous radiation. Holman et al did not find an increased rate of wound healing problems in patients who had previous radiation. However, Pascal-Moussellard et al found that they had a 12% rate of wound healing problems in patients with previous radiation compared with patients who did not have previous radiation (1.1%). They also stated that 55% of their complications were wound-related. In the study by Rompe et al (1999), they cited a 2.4% infection rate, and 2 of 3 of the infected patients had been previously radiated. Shehadi et al did not find an association between radiation and wound healing problems. Street et al also found that preoperative radiation did not affect wound healing.


Overall quality of evidence is very low to support the decision-making for surgical approach in spinal metastases. However, using the current paradigm for evidence-based medicine, the evidence is reviewed and placed into the context of the clinical experience. The results of the systematic review were presented and discussed by a group of experts in the Spinal Oncology Study Group and a series of recommendations has emerged. For the T2 through T5 range, there is very low quality evidence but a strong recommendation for a posterolateral approach. Anterior access at this level is problematic. While it can be done, if a conventional high thoracotomy is used, there is the morbidity of the detachment of the periscapular musculature. A direct midline approach—via sternotomy or manubriectomy—can be done, but the heart, great vessels limit direct access to the spinal column.36,1 Spinal fixation techniques have improved and allow rigid stabilization from a posterior approach.37,38 This is primarily due to the increasing facility of surgeons using pedicle screw fixation in this region.39–41 In addition, there has been greater dissemination of the understanding of posterolateral approaches to the vertebral bodies in a “posterolateral vertebrectomy” technique. This can be achieved via a transpedicular, costotransversectomy, or lateral extracavitary approach. Sacrificing one or more ipsilateral nerve roots greatly facilitates access in this fashion (as mentioned in Figure 1 example). Therefore, there is a high degree of agreement (unanimity) among the SOSG surgeons that a posterior approach in the upper thoracic spine (T2–T5) is preferred because of the difficult access via an anterior approach.

Figure 1
Figure 1:
An adult male with metastatic prostate cancer to the upper thoracic spine. The AP and lateral radiographs reveals a pathologic fracture at T5 (A, B) and epidural spinal cord compression seen on MRI (B). He underwent a posterior only decompression, corpectomy and stabilization shown in the postoperative radiographs and intraoperative photo (C, D). This was well tolerated, gave excellent pain relief, and allowed discharge to home care. He subsequently died of disease approximately 11 months after the surgery.

For metastatic lesions from T6 to L5, the very low quality evidence demonstrates viability of anterior, posterior, and circumferential surgery (as mentioned in Figure 2 example). There is a strong recommendation for use of any of the 3 procedures (anterior, posterior or combined anterior and posterior) depending on the clinical presentation, surgeon and patient preference. There are factors that influence surgeon preference. In the lumbar spine (especially L3–L5), sacrifice of nerve roots carries greater functional morbidity. Sometimes, the patient’s cardiopulmonary physiology may preclude their ability to tolerate an anterior approach. However, for patients who have had prior conventional radiation therapy, there may be a higher wound complication rate associated with a posterior approach. These factors can influence the approach.

Figure 2
Figure 2:
An adult female with metastaic breast cancer and pathologic fractures and epidural spinal cord compression at different levels shown in the MRI and radiographs of the thoracic and lumbar spine (A, B). She underwent an anterior approach, 2 level corpectomy and reconstruction and subsequent posterior stabilization as shown in the postoperative radiographs (C). The hospital course was somewhat prolonged. This markedly improved her pain and function. At greater than 3 year follow-up she remains active with minimal pain from her spine.

There are studies in the literature supporting the effectiveness of surgical treatment for thoracic and lumbar spinal metastases, using either an anterior, posterior, or combined AP approach. There is little evidence to support the superiority of one approach over another. While there are many reports including cases performed using different approaches, most of these did not report their results separately by approach, and thus no comparison can be made. Most of these studies selected their approach by tumor topography (i.e., which part or parts of the vertebra is involved). There are no studies in which cases were randomized by surgical approach.

There likewise is very little evidence on region-specific approaches within the thoracic and lumbar spines, although there appears to be some preference for a posterior (posterolateral) approach for the upper thoracic spine (T2–T5).


For patients with symptomatic metastatic disease in the upper thoracic region (T2–T5), a strong recommendation is made for a posterior-based (posterolateral) approach because of technical challenges with anterior or combined approaches. The posterior-lateral approach specifically when anterior decompression and/or stabilization is necessary. The quality of evidence supporting this is very low.

In the thoracolumbar (T6–L5) spine region afflicted with symptomatic metastatic disease there is a strong recommendation for use of either anterior, posterior or combined anterior or posterior surgery depending on the clinical situation, surgeon and patient preference. This recommendation was based primarily on clinical experience as the on evaluation of the best available evidence, as to which surgical approach (anterior, posterior, or combined AP) is superior in terms of local tumor recurrence, perioperative morbidity and mortality, pain alleviation, and neurologic recovery was very low quality.

Key Points

  • Spinal thoracolumbar metastasis is common and often requires intervention.
  • There is a significant debate about the best surgical approach for treatment of symptomatic thoracolumbar metastasis.
  • The quality of existing literature is limited.


1. Gokaslan ZL, York JE, Walsh GL, et al. Transthoracic vertebrectomy for metastatic spinal tumors. J Neurosurg 1998;89:599–609.
2. Harrington K. Metastatic tumors of the spine: diagnosis and treatment. J Am Acad Orthop Surg 1993;1:76–86.
3. Holman PJ, Suki D, McCutcheon I, et al. Surgical management of metastatic disease of the lumbar spine: experience with 139 patients. J Neurosurg Spine 2005;2:550–63.
4. Wiggins GC, Mirza S, Bellabarba C, et al. Perioperative complications with costotransversectomy and anterior approaches to thoracic and thoracolumbar tumors. Neurosurg Focus 2001;11:e4.
5. Brozek JL, Akl EA, Alonso-Coello P, et al. Grading quality of evidence and strength of recommendations in clinical practice guidelines. Allergy 2009;64:669–77.
6. Schunemann HJ, Jaeschke R, Cook DJ, et al. An official ATS statement: grading the quality of evidence and strength of recommendations in ATS guidelines and recommendations. Am J Respir Crit Care Med 2006;174:605–14.
7. Ibrahim A, Crockard A, Antonietti P, et al. Does spinal surgery improve the quality of life for those with extradural (spinal) osseous metastases? An international multicenter prospective observational study of 223 patients. Invited submission from the Joint Section Meeting on Disorders of the Spine and Peripheral Nerves, March 2007. J Neurosurg Spine 2008;8:271–8.
8. Klekamp J, Samii H. Surgical results for spinal metastases. Acta Neurochir (Wien) 1998;140:957–67.
9. Landmann C, Hunig R, Gratzl O. The role of laminectomy in the combined treatment of metastatic spinal cord compression. Int J Radiat Oncol Biol Phys 1992;24:627–31.
10. Street J, Fisher C, Sparkes J, et al. Single-stage posterolateral vertebrectomy for the management of metastatic disease of the thoracic and lumbar spine: a prospective study of an evolving surgical technique. J Spinal Disord Tech 2007;20:509–20.
11. Bauer HC. Posterior decompression and stabilization for spinal metastases: analysis of sixty-seven consecutive patients. J Bone Joint Surg Am 1997;79:514–22.
12. Chataigner H, Onimus M. Surgery in spinal metastasis without spinal cord compression: indications and strategy related to the risk of recurrence. Eur Spine J 2000;9:523–7.
13. Chen LH, Chen WJ, Niu CC, et al. Anterior reconstructive spinal surgery with Zielke instrumentation for metastatic malignancies of the spine. Arch Orthop Trauma Surg 2000;120:27–31.
14. Chen YJ, Chang GC, Chen HT, et al. Surgical results of metastatic spinal cord compression secondary to non-small cell lung cancer. Spine 2007;32:E413–8.
15. Fourney DR, Abi-Said D, Rhines LD, et al. Simultaneous anterior-posterior approach to the thoracic and lumbar spine for the radical resection of tumors followed by reconstruction and stabilization. J Neurosurg 2001;94:232–44.
16. Hammerberg KW. Surgical treatment of metastatic spine disease. Spine 1992;17:1148–53.
17. Hosono N, Yonenobu K, Fuji T, et al. Vertebral body replacement with a ceramic prosthesis for metastatic spinal tumors. Spine 1995;20:2454–62.
18. Jackson RJ, Loh SC, Gokaslan ZL. Metastatic renal cell carcinoma of the spine: surgical treatment and results. J Neurosurg 2001;94:18–24.
19. Jansson KA, Bauer HC. Survival, complications and outcome in 282 patients operated for neurological deficit due to thoracic or lumbar spinal metastases. Eur Spine J 2006;15:196–202.
20. Jonsson B, Sjostrom L, Olerud C, et al. Outcome after limited posterior surgery for thoracic and lumbar spine metastases. Eur Spine J 1996;5: 36–44.
21. King GJ, Kostuik JP, McBroom RJ, et al. Surgical management of metastatic renal carcinoma of the spine. Spine 1991;16:265–71.
22. Missenard G, Lapresle P, Cote D. Local control after surgical treatment of spinal metastatic disease. Eur Spine J 1996;5:45–50.
23. Onimus M, Papin P, Gangloff S. Results of surgical treatment of spinal thoracic and lumbar metastases. Eur Spine J 1996;5:407–11.
24. Pascal-Moussellard H, Broc G, Pointillart V, et al. Complications of vertebral metastasis surgery. Eur Spine J 1998;7:438–44.
25. Rompe JD, Eysel P, Hopf C, et al. Decompression/stabilization of the metastatic spine. Cotrel-Dubousset-Instrumentation in 50 patients. Acta Orthop Scand 1993;64:3–8.
26. Rompe JD, Eysel P, Hopf C, et al. Metastatic spinal cord compression–options for surgical treatment. Acta Neurochir (Wien) 1993;123:135–40.
27. Rompe JD, Hopf CG, Eysel P. Outcome after palliative posterior surgery for metastatic disease of the spine–evaluation of 106 consecutive patients after decompression and stabilisation with the Cotrel-Dubousset instrumentation. Arch Orthop Trauma Surg 1999;119:394–400.
28. Saengnipanthkul S, Jirarattanaphochai K, Rojviroj S, et al. Metastatic adenocarcinoma of the spine. Spine 1992;17:427–30.
29. Shehadi JA, Sciubba DM, Suk I, et al. Surgical treatment strategies and outcome in patients with breast cancer metastatic to the spine: a review of 87 patients. Eur Spine J 2007;16:1179–92.
30. Sioutos PJ, Arbit E, Meshulam CF, et al. Spinal metastases from solid tumors: analysis of factors affecting survival. Cancer 1995;76:1453–9.
31. Sundaresan N, Rothman A, Manhart K, et al. Surgery for solitary metastases of the spine: rationale and results of treatment. Spine 2002;27:1802–6.
32. Sundaresan N, Steinberger AA, Moore F, et al. Indications and results of combined anterior-posterior approaches for spine tumor surgery. J Neurosurg 1996;85:438–46.
33. Turgut M, Gul B, Girgin O, et al. Role of surgical treatment in 70 patients with vertebral metastasis causing cord or root compression. Arch Orthop Trauma Surg 1997;116:415–9.
34. Wang JC, Boland P, Mitra N, et al. Single-stage posterolateral transpedicular approach for resection of epidural metastatic spine tumors involving the vertebral body with circumferential reconstruction: results in 140 patients. Invited submission from the Joint Section Meeting on Disorders of the Spine and Peripheral Nerves, March 2004. J Neurosurg Spine 2004;1:287–98.
35. Guyatt GH, Oxman AD, Kunz R, et al. What is “quality of evidence” and why is it important to clinicians? BMJ 2008;336:995–8.
36. Anderson TM, Mansour KA, Miller JI Jr. Thoracic approaches to anterior spinal operations: anterior thoracic approaches. Ann Thorac Surg 1993;55:1447–51, discussion 51–2.
37. Ferguson RL, Tencer AF, Woodard P, et al. Biomechanical comparisons of spinal fracture models and the stabilizing effects of posterior instrumentations. Spine 1988;13:453–60.
38. Wood KB, Wentorf FA, Ogilvie JW, et al. Torsional rigidity of scoliosis constructs. Spine 2000;25:1893–8.
39. Daubs MD, Kim YJ, Lenke LG. Pedicle screw fixation (T1, T2, and T3). Instr Course Lect 2007;56:247–55.
40. Fisher C, Singh S, Boyd M, et al. Clinical and radiographic outcomes of pedicle screw fixation for upper thoracic spine (T1–T5) fractures: a retrospective cohort study of 27 cases. J Neurosurg Spine 2009;10:207–13.
41. Guzey FK, Emel E, Hakan Seyithanoglu M, et al. Accuracy of pedicle screw placement for upper and middle thoracic pathologies without coronal plane spinal deformity using conventional methods. J Spinal Disord Tech 2006;19:436–41.

thoracolumbar metastasis; surgical approach

© 2009 Lippincott Williams & Wilkins, Inc.