Metastatic spine tumors.

Abstract: Cancer is the second leading cause of death in the United States, and vertebral body metastases often occur in systemic malignancy. Metastatic spinal tumors may present with pain or neurologic deficit, or may be detected during screening examinations in patients with known malignancy. Management of spinal metastases remains controversial. The role of surgery, especially decompressive laminectomy without stabilization, has been questioned. Recent series attest to the beneficial role of surgery, emphasizing anterior and combined decompression and stabilization procedures. We review the relevant literature on metastatic spinal tumors, assessing imaging strategies, adjuvant treatment, patient selection, and results and complications. Operative decompression and stabilization is an important tool in the management of spinal metastatic disease. Patient selection and appropriate use of anterior and/or posterior decompression and stabilization are necessary to optimize surgical results.

Key Words: cancer, metastatic disease, vertebral body metastasis 

**********

The management of metastatic neoplasms involving the spinal column constitutes an area of continued controversy in the orthopedic and neurosurgical spine literature. Cancer is the second leading cause of death in the United States, and roughly two-thirds of cancer patients develop metastases. (1,2) Common sites of metastasis are lung and liver, with the skeletal system the third most common. Within the skeletal system, metastases favor the spine. (3) Vertebral body metastases are found in over one third of cancer patients in autopsy series. (4) Clinical evidence of spinal cord, cauda equina, or nerve root compromise is found in 5% of all cancer cases. (5,6)

Spread to the vertebral column may occur through hematogenous seeding or direct extension of a paravertebral tumor. The vertebral bodies are involved in 80% of cases, presumably due to their highly vascular red marrow and their size relative to posterior elements. (7) The thoracic region of the spine is most frequently involved in metastatic disease, followed by the cervical and lumbar regions. (8,9) Intramedullary metastases occur, but are much less common. (10) Colorectal carcinoma, an unusual source of spinal column metastasis, favors the lumbar spine (55% of lesions). (11)

Any tumor may spread to the spine. The most common cancers involving the spine are lung, breast, prostate, and renal cell. (3,6,8,12) A recent review found metastatic spinal involvement in 90% of prostate, 75% of breast, 55% of melanoma, 45% of lung, and 30% of renal cell carcinoma patients. (13) However, symptomatic spinal cord compression is less common and is found in 22% of breast, 15% of lung, and 10% of prostate cancer patients. (6) Either lytic or blastic lesions may occur with spinal cord metastasis. Breast and prostate carcinomas often generate blastic lesions, whereas lung, renal, and gastrointestinal tumors produce lytic lesions. (14) Spinal metastases may occur at any time during the course of a patient's primary disease. Vertebral body metastasis in colorectal cancer and melanoma are generally late findings. (11,15) The goal of therapy in most cases is palliation; treatment of spinal column metastases is seldom curative.

A significant source of controversy in the management of spinal column metastasis arises from previous reports comparing results from radiotherapy alone with results from radiation combined with laminectomy without stabilization. The lack of benefit from addition of laminectomy without stabilization produced skepticism regarding the role of surgical therapy in spinal metastases, with both interventions yielding improvement in roughly 44% of patients. (16,17) Because metastatic lesions usually involve the vertebral bodies, compression of neural elements arises ventrally. Laminectomy alone uses a dorsal decompressive procedure to alleviate ventral compression. Loss of vertebral body structural integrity due to metastatic destruction may produce instability, which can be exacerbated by resection of the spinal column's posterior elements. This may actually increase neurologic deficit or pain. At present, laminectomy alone is infrequently used in management of spinal column metastasis. (8)

However, laminectomy combined with posterior stabilization provides improved outcomes when compared with laminectomy alone. Recent reviews record neurologic improvement in approximately 72% of patients. (8,18) Newer approaches emphasize anterior decompression and vertebral body reconstruction, combined with posterior stabilization when deemed necessary. These series document neurologic improvement in approximately 75% of patients. (8,19-23)

Because modern surgical decompression and stabilization techniques result in improved outcomes, surgery should be considered as an initial management option in a greater percentage of patients presenting with spinal metastatic lesions. The majority of patients are nevertheless often referred initially for adjuvant therapy, usually radiotherapy, followed by surgical consultation. Preoperative radiation has been correlated with increased perioperative complications. (18,24) Early involvement of spine surgeons early in the evaluation and management of patients with metastatic spinal column malignancy is necessary.

Presentation

Back pain occurs in at least 90% of patients who present with metastatic vertebral body lesions. (6,11,18,19,21,22) Comey and Haid (7) stated that back pain in a patient with known systemic malignancy must be regarded as spinal metastasis until proved otherwise. The pain may increase gradually as the vertebral body is progressively destroyed. Movement may exacerbate the patient's pain complaints. Conversely, an acute onset of moderate or severe back pain may correlate with a pathologic fracture of a compromised vertebral body.

Pain secondary to vertebral body metastasis may take three general forms. Local pain arises from either tumor mass effect or stretching and distortion of the periosteum from tumor destruction. This pain is constant, and is not exacerbated by movement. Radicular pain develops from nerve root compression by epidural tumor extension. Axial pain is mechanical in nature, with worsening with motion and alleviated with rest. Axial pain arises from a structural abnormality of the spinal column, and may indicate instability. (20)

A variety of neurologic deficits, from mild radicular weakness to paraparesis, may be present in patients with spinal malignancy. Weakness may be initially mild with gradual progression, perhaps due to slowly worsening epidural compression. Acute deterioration may develop in pathologic fractures or vertebral body fracture/dislocation. A patient with systemic malignancy and deficit referable to spinal cord or nerve root compromise requires urgent evaluation. Patients who are ambulatory at the time of initial treatment for a spinal metastasis are much more likely to remain ambulatory after intervention. Conversely, patients with profound deficit are less likely to regain function. (6) A high percentage of patients reviewed in recent series were nonambulatory at time of surgical evaluation and operative intervention. (8,18-22) Poor preoperative neurologic function correlates with poor postoperative results.

Initial history and physical examination provides a baseline to assess progression of disease and the effect of therapy. Pain intensity may be quantified through a simple visual analogue scale. Motor function should be assessed and recorded. Sensory examination may reveal a sensory level or a radicular pattern of sensory deficit. Hyperreflexia may correlate with spinal cord compression. Deep tendon reflexes, presence or absence of pathologic reflexes, and rectal tone are noted. In patients with suspected sphincter disturbance, a postvoid residual is obtained.

Imaging

Evaluation of patients with suspected spinal column metastasis focuses on identification and localization of the lesion(s), assessment of neural element compromise, extent of bony destruction, and presence or absence of instability. Other bone metastasis, and evidence of involvement of other organ systems, may temper therapy. Patient evaluation will usually incorporate magnetic resonance imaging (MRI), computed tomographic (CT) scanning, and plain films. An example is presented in Figure 1.

Plain Films

Plain films are often used as a screening tool when vertebral body metastatic involvement is suspected. However, normal plain films often are found even in patients with clinical evidence of spinal cord compression from metastatic disease. (25) Plain films cannot rule out vertebral body metastases. One of the more common findings in the thoracic and lumbar spine is pedicle erosion. In the cervical spine, vertebral body destruction is more easily visualized. Compression fractures may be seen. Neoplastic vertebral body lesions spare the disc spaces, enabling differentiation from pyogenic osteomyelitis. Vertebral body alignment and stability may be assessed via plain films supplemented with dynamic views.

The sensitivity of plain radiographs is low, especially early in metastasis development. Significant vertebral body destruction must occur before abnormalities are visualized on plain films. (26)

Bone Scan 

Technetium-99 bone scintigraphy is more sensitive than plain films in detecting bony metastases. The entire body may be imaged in a single study. Evidence of bone disease will be detected earlier, although specificity is low, mandating further evaluation. False-positives may occur in patients with degenerative disease, trauma, or metabolic bone disease. In cases of plasmacytoma, false-negatives may occur. Bone scan may also appear normal in melanoma metastases. (15) Renal cell and thyroid metastases, including cases with significant bone destruction, may also be missed on isotope studies. This is possibly due to their high vascularity. (6) Systemic dissemination may be revealed through visualization of other bony metastases, and may impact on anticipated patient survival. (27,28) MRI may be more sensitive in detecting vertebral body and epidural metastases. (29)

[FIGURE 1 OMITTED]

Computed Tomography/CT Myelography 

Computed tomography is sensitive to bone destruction and will identify the extent of bony compromise and lesion distribution. Computed tomography best visualizes neural element compromise from retropulsed bony elements in pathologic compression fractures. Sagittal reformatting visualizes vertebral body integrity and can show spinal alignment at the cervicothoracic and thoracolumbar junction, where plain films may be insufficient. Computed tomography remains the best imaging modality for visualization of the bony anatomy of a vertebral body metastasis, and is extremely useful in surgical planning.

Addition of myelography, previously the preferred imaging modality in patients with deficits from presumed compressive lesions, will clearly show the level and extent of spinal cord compression. Due to its invasive nature, however, MRI has replaced CT myelography in most centers. CT myelography is still used in patients with pacemakers, nontitanium aneurysm clips, and other contraindications to MRI.

MRI

MRI is the imaging modality of choice for evaluation of spinal metastases. Neural compression from epidural tumor or bony destruction may be easily visualized. The entire spine can be imaged with multiple planes of view. Paravertebral tumor extension and delineation of tumor extent may be assessed. Multiple sites of bony involvement may be revealed. (30,31) Involvement of contiguous vertebral bodies may complicate instrumentation placement in operative stabilization, and is hence important in operative planning. (6,32)

Most lesions appear hypointense on T1-weighted images and hyperintense on T2-weighted images. Gadolinium enhancement may obscure metastatic bone lesions, due to normal vertebral body enhancement, and is generally contraindicated. MRI has high sensitivity for detecting changes in bone marrow and may therefore provide earlier evidence of vertebral body involvement in systemic malignancy. (29)

MRI offers the best visualization of spinal cord and nerve root compression in both sagittal and axial planes. This information is critical in planning operative decompression. MRI may be difficult to obtain in claustrophobic patients and may be contraindicated in patients with certain implants, including pacemakers. Image degradation due to patient motion or spinal instrumentation may hamper visualization.

Angiography/Embolization

Angiography with embolization may be used in highly vascular tumors (melanoma, hypernephroma, thyroid carcinoma) and is an important adjunct in minimizing intraoperative blood loss during resection. (8,12,18,19,33) Embolization is performed between 18 and 24 hours before surgery to prevent revascularization of the metastasis.

Even with embolization, blood loss in renal cell and thyroid metastases can be great. Average blood loss in hypernephromas treated with posterior decompression and fusion averaged 3,000 ml in one series. (18) Other series attest to the large blood loss that may be anticipated in surgical treatment of renal metastases. (8) Complications may occur with embolization. Transient or permanent worsening of neural function, sometimes severe, has been described. (8,22) Use of somatosensory evoked potential monitoring during embolization has been described; this may limit periprocedural neurologic dysfunction. (34)

Adjuvant Treatment

Most patients referred for surgical evaluation of vertebral body metastases will have already received some form of adjuvant therapy. Often, patients will have received chemotherapy, radiotherapy, or both. These measures may complicate surgery, through increasing the rate of perioperative complications, immunosuppression, or impaired local wound healing.

Chemotherapy

The primary tumor type determines chemotherapeutic intervention. Possible complications of chemotherapy depend on the agent(s) used, but may result in immunosuppression or delayed wound healing, increasing the risk of wound dehiscence or perioperative wound infections. (6) Soft tissue tumors such as lymphoma may undergo rapid shrinkage after treatment with high-dose corticosteroids. In general, however, chemotherapy has little effect in reversing spinal cord compression and improving neurologic deficit.

Radiation Therapy

Radiotherapy is the initial therapeutic modality in the majority of patients with vertebral body metastases. (8,12,18-20,22,23) In patients with poor prognosis due to impaired systemic control, radiation may be the only line of therapy used. Radiation is appropriate therapy for patients with minimal or no neurologic findings, tumors known to be radioresponsive, no evidence of instability, and no neural compression by bone. (19)

The usual dose of radiation depends on tumor type and length of treatment. Three thousand cGy delivered in 300-cGy fractions over 10 days is typical. The dose is usually limited to a maximum of 4,500 cGy to prevent damage to the spinal cord. The treatment field should incorporate at least one vertebral body above and below the lesion. Serial neurologic examinations must be performed. Evidence of neural deterioration during treatment is an indication of radiation failure; these patients should be promptly referred for operative evaluation.

The histologic nature of the tumor is important in predicting response to radiation. Breast, prostate, and hematopoietic tumors are considered radiation sensitive. Lung and thyroid are considered intermediately responsive. Melanoma, renal cell, and gastrointestinal tumors are radioresistant. (6,33) Surgical therapy may be the initial therapy in radioresistant lesions and in patients where vertebral body metastasis is the first manifestation of an occult primary tumor. (19)

Understanding of the source of neural element compression, and the cause of neoplastic pain, is necessary. In pain control, radiation may alleviate local pain derived from metastasis and radicular pain generated by epidural compression in radiosensitive tumors. Axial or mechanical pain will not respond to radiotherapy and requires operative stabilization. (20)

The best predictor of neurologic outcome after radiotherapy is the pretreatment neurologic function. Patients with severe neurologic deficit before radiation are unlikely to improve. (17,33) Patients receiving radiotherapy for symptomatic neural compression who deteriorate during therapy may also have poor outcomes, with a recent report noting 77% of patients not recovering lost function. (24)

Radiation may increase the risk of perioperative complications. In a review by Ghogawala et al, (24) preoperative radiation increased the rate of wound infection nearly threefold (12-32%). Preoperative radiotherapy correlates with increased perioperative complications in other studies. (8,18) Radiation is routinely used postoperatively for local control after operative decompression and stabilization. Radiation should be delayed at least 2 weeks to allow for wound healing. (35)

Vertebroplasty

Percutaneous vertebroplasty using polymethylmethacrylate (PMMA) cement has been used in the treatment of vertebral body metastases. Some authors report use of PMMA vertebroplasty in treating local and axial pain secondary to metastasis. (36) Biopsy of vertebral body lesions may also be obtained. Vertebral body biopsy in treatment of presumed osteoporotic collapse may reveal a previously unsuspected metastasis. (37) Although vertebroplasty often achieves pain control in osteoporotic compression fractures, consistent pain relief in malignancies is often not achieved. (38) Epidural extrusion of PMMA is more common in vertebral body metastases and may produce significant neurologic deficits. (36,39)

Patient Selection for Surgery

The appropriate selection of patients for surgery is an important determinant of operative outcome. Eradication of metastatic tumor is seldom possible. Other systemic metastases make cure unlikely or impossible. Hence, the goal of surgical treatment is to relieve pain and prevent or reverse neurologic compromise. Table 1 lists the relative indications for surgery and for radiation, as summarized by Cooper et al. (19) Each patient and spinal metastasis is different, however; the relative indications and contraindications must be individualized.

Indications for Surgery

Clear indications for surgery include unknown primary where computed tomography-guided biopsy is not possible or has not been successful, progressive neurologic dysfunction despite radiotherapy, steroids, or chemotherapy, radioresistant tumors or tumors with poor radioresponsivity, tumors that have progressed or recurred after radiation, instability due to bony destruction, and neural element compression by displaced bone fragments. Axial pain due to metastatic vertebral body destruction also requires operative stabilization. (19,20)

Contraindications to Surgery

Relative contraindications to surgery include poor life expectancy (<3 months), severe immunocompromise from primary disease or previous chemotherapy, and abnormal coagulation that cannot be corrected with replacement therapy. Radioresponsive tumors (lymphoma, myeloma) in patients with minimal symptoms are treated with radiation initially. Patients with complete absence of motor and sensory function of greater than 12 hours' duration are unlikely to regain function with operative decompression.

Operative Therapy

Preoperative Evaluation

Thorough preoperative evaluation will reduce the incidence of postoperative complications. Systemic malignancy, chemotherapy, and prior radiotherapy result in increased operative morbidity. Other medical conditions, such as diabetes mellitus, hypertension, coronary artery disease, and lung disease may also be present. Each patient's medical condition must be optimized to limit perioperative complications. Pulmonary testing may be required if a transthoracic approach is contemplated. Nutritional status must be assessed and optimized. (7) Incidence of deep venous thrombosis (DVT) is relatively high in cancer patients, especially in cases of decreased mobility due to lower extremity weakness. (6) Patients should be preoperatively evaluated with lower extremity Doppler ultrasound. Presence of lower extremity DVTs may require vena cava filter placement. Prophylactic treatment in patients without evidence of DVT includes lower extremity compression boots and low-dose subcutaneous heparin.

Anesthesia and Monitoring

Induction of anesthesia and patient positioning are major concerns in patients with instability due to bony metastases. Cervical spine lesions require awake fiberoptic intubation with minimal neck manipulation. Dual-lumen endotracheal tubes are used in patients with thoracic metastases at T8 and above to permit deflation of the lung ipsilateral to the side of the thoracotomy. The surgeon should supervise patient positioning, especially when turn to a prone position is necessary for dorsal decompression and stabilization. Maintenance of neutral position in patients with cervical lesions is mandatory. In patients with cervical instability, preoperative traction with Gardner-Wells tongs is administered. We maintain inline traction in these cases during operation. Use of tongs, a halo ring, or a three-pin headholder may be helpful in patient positioning. Operative time may be lengthy; all extremities and pressure points must be well padded to prevent iatrogenic peripheral nerve compression. Radiographs will confirm adequate positioning before incision. Operative approach determines patient positioning. Corticosteroids are used preoperatively and continued in the perioperative and postoperative periods.

We use intraoperative somatosensory-evoked potential (SSEP) monitoring in patients with preserved neurologic function. In patients with moderate or severe neurologic deficits, responses may be absent or diminished, limiting the effectiveness of intraoperative SSEP monitoring. (7) Although disappearance of SSEPs intraoperatively may predict an adverse outcome, the use of this monitoring modality may not prevent untoward intraoperative neurologic deterioration. (40)

Postoperative Management

Patients are maintained in the intensive care unit after surgery. Postoperative bracing is used in patients where bony fusion is the goal of surgery. External bracing is generally not used in patients stabilized with PMMA constructs. Inpatient rehabilitation is used as necessary. Early walking reduces the risk of postoperative DVTs. Corticosteroids are continued postoperatively, with a rapid taper. H2-antagonists are used to minimize gastroduodenal ulceration. Plain films are taken postoperatively to confirm instrumentation placement and spinal alignment.

Postoperative wound management may be complicated by long-term steroid use, preoperative radiation or chemotherapy, poor nutritional status, and prior surgery. Pressure on wounds should be minimized. Superficial sutures should be left in place for an extended period. Postoperative radiation should be delayed for at least 2 weeks.

Close follow-up of patients with spinal metastatic disease is necessary to detect recurrence at the site of tumor resection. New lesions may develop in contiguous or isolated regions. Tumor recurrence may produce loss of stability or recurrent neural compression, perhaps requiring reoperation. Recurrence is particularly common in tumors resistant to radiation and chemotherapy (melanoma, renal cell carcinoma). We study patients at 3- to 6-month intervals with plain films, supplemented with MRI in cases of increased pain or return of neurologic deficit.

Results

Pain Control

Surgical stabilization in patients with axial pain, decompression of epidural neoplastic mass lesions, and correction of abnormal alignment may each contribute to successful pain management. Isolated pedicle screw stabilization achieved nearly 90% improvement in visual analogue scores for pain at I month after surgery. (18) Similar results are reported in more aggressive, anteroposterior combined approaches, with over 90% of patients achieving improvement in pain control after operative decompression and stabilization. (8,18,19) Multiple series note reduction in the use of pain medication. Regardless of the type of tumor or the level, control of pain may be achieved with operative stabilization.

Neurologic Deficit

Postoperative neurologic function depends on preoperative status. Patients with severe deficit of long-standing duration may not achieve improvement even with adequate decompression and fusion. Early intervention in patients with neurologic deficits is necessary to maximize chances of recovery. (7)

Assessment of operative outcomes varies in different series, making interseries comparisons difficult. However, in properly selected patients, approximately 70 to 80% of patients may achieve improvement in neurologic function. (8,19,20,41-43) Recovery of walking may be achieved in approximately 50% of patients nonambulatory before surgery. (8,12,18,19) Siegal and Siegal (41) emphasize the importance of operative stabilization, noting addition of decompressive laminectomy alone to radiation did not contribute to improvement, with 44% of patients in both groups improved. Addition of stabilization via anterior or posterior approaches achieved improvement in approximately 70% of patients, with pain control achieved in 85%. Patients who deteriorate during radiotherapy may manifest poor recovery, with 77% of patients unchanged after surgery in one series. (24)

Survival

Overall patient survival will depend on the primary malignancy, degree of systemic spread, and tumor biology. Median survivals of at least 1 year after surgery are reported by many series. (18,19) Considerably diminished survival presents in certain malignancies, such as metastatic colon carcinoma with spinal involvement, where median survival is less than 3 months. (11) Spinal metastasis in metastatic melanoma is also a late finding, with median survival 4 months and a 15% 1-year survival rate. (15)

Complications

Surgical treatment of spinal malignancy is not without risk. Significant complications occur in many patients. Severe complications include perioperative neurologic deterioration, wound breakdown, significant blood loss, and mortality. Major and minor complications may occur in up to 30% of patients summarized in modern series. Pulmonary, cardiac, and gastrointestinal complications, along with cerebrospinal fluid leakage and wound breakdown, are all common. Instrumentation failures may occur. (12,18,20,22,43) The use of high-dose steroids, radiation, and chemotherapy all increase complication rates. Perioperative mortality should be expected in less than 5%. (3)

Neurologic Deterioration

Already compromised neural elements may be at greater risk of further damage through surgical manipulation. Rates of perioperative deterioration are fortunately low. In large recent series, less than 10% of patients experienced deterioration of neurologic function. (12,18,19,24,43) These neurologic complications varied from quadriparesis from cord ischemia to foot drop from nerve root retraction. In many cases, these neurologic deficits were transient. (18)

Wound Complications

Wound complications are unusually high in surgery for spinal metastases. The combination of previous radiation, steroid use, poor nutritional status, and immunosuppression all contribute to increased risk of postoperative wound infection or breakdown. Wound complications after radiotherapy in laminectomy patients have been reported in up to 28% of patients. (41,44) Interestingly, significant wound breakdown and even superficial wound infections are uncommon in anterior transthoracic approaches. (8)

Ghogawala et al (24) reviewed the role of radiation in predisposing operative candidates to perioperative wound breakdown. They found a nearly threefold increase in the incidence of wound complications, from 12 to 32%, in patients treated with radiation before operative decompression and stabilization. Surgery performed within 7 days of beginning radiotherapy, usually in patients who suffered neurologic deterioration, yielded a wound complication rate of 46%. Tumor type, patient age, and preoperative albumin were not found to correlate with the risk of wound complications.

Blood Loss

Blood loss may be significant in surgical decompression and stabilization of spinal malignancies. Highly vascular malignancies may entail tremendous intraoperative blood loss, even when preoperative embolization is used. Jackson et al, (12) in reviewing 79 patients undergoing 107 procedures for metastatic renal cell carcinoma, found average blood loss of 2,100 ml (range, 50-31,000 ml). Curiously, preoperative embolization did not correlate with decreased blood loss, possibly due to selection bias. The authors note that, empirically, embolization appeared to decrease intraoperative blood loss.

Conclusions

Surgical treatment of spinal metastatic disease remains controversial. Modern surgical approaches provide options for decompression and immediate stabilization. Laminectomy alone in cases of ventral compression is seldom used, due to poor outcomes reported in previous series. Anterior, posterior, or combined decompressions with immediate stabilization have been shown to provide improved patient outcomes when compared with historical reports of radiation, decompressive laminectomy without stabilization, or combined radiation and laminectomy. Surgery is often contemplated only after failure of radiation and chemotherapy to achieve tumor control. In many patients, surgery should be considered as initial therapy. In properly selected patients, surgery may provide excellent pain relief, a significant chance of recovery of neurologic function, acceptable perioperative morbidity and mortality, and prevention of late neurologic deterioration.

Table 1. Relative indications for surgery or radiation as primary
treatment Modality

Radiation as primary intervention        Surgery as primary intervention

Radioresponsive tumor                    Retropulsed bone producing
                                           neural compression
Moderately radioresponsive tumor in      Spinal deformity producing pain
  patient with minimal deficit/limited     or neural compression
  pain
Isolated epidural neural compression     Axial pain
Isolated local pain                      Spinal instability from
                                           metastatic bony or
                                           ligamentous destruction
Expected survival less than 3 months     Progressive neurological
                                           deficit
Poor operative candidate                 Failure of radiation
Complete neurological deficit              Progression of deficit/pain
                                             during radiation
                                           Previous radiation with
                                             recurrence/progression
                                             after treatment
                                         Unknown primary

Accepted September 26, 2002.

Copyright [c] 2004 by The Southern Medical Association

0038-4348/04/9703-0246

References

1. Silverberg E. Cancer statistics. J Clin Cancer 1984;34:7-23.

2. Jaffe W. Tumors and Tumorous Conditions of the Bones and Joints. Philadelphia, Lea & Febiger, 1958.

3. Gokaslan Z. Spine surgery for cancer. Curr Opin Oncol 1996;8:178-181.

4. Fornasier V, Horne JG. Metastases to the vertebral column. Cancer 1975;36:590-594.

5. Barron K, Hirano A, Araki S, et al. Experiences with metastatic neoplasms involving the spinal cord. Neurology 1959;9:91-106.

6. Nazzaro JM. Metastatic spinal lesions, in Benzel EC (ed); Spine Surgery: Techniques, Complication Avoidance, and Management. New York, Churchill Livingstone, 1999, vol 1, pp 679-695.

7. Comey C, Haid RW Jr. Chapter 159: Management of spinal metastases, in Kaye AH, Black PM (eds): Operative Neurosurgery. Philadelphia, Churchill Livingstone, 2000, vol 2, pp 1961-1971.

8. Gokaslan ZL, York JE, Walsh GL, et al. Transthoracic vertebrectomy for metastatic spinal tumors. J Neurosurg 1998;89:599-609.

9. Perrin R, McBroom RJ. Anterior versus posterior decompression for symptomatic spinal metastasis. Can J Neurol Sci 1987;14:75-80.

10. Potti A, Abdel-Raheem M, Levitt R, et al. Intramedullary spinal cord metastases (ISCM) and non-small cell lung carcinoma (NSCLC): Clinical patterns, diagnosis and therapeutic considerations. Lung Cancer 2001;31:319-323.

11. Brown P, Stafford SL, Schild SE, et al. Metastatic spinal cord compression in patients with colorectal cancer. J Neurooncol 1999;44:175-180.

12. Jackson RJ, Loh SC, Gokaslan ZL. Metastatic renal cell carcinoma of the spine: Surgical treatment and results. J Neurosurg 2001;94(1 Suppl): 18-24.

13. Wong D, Fornasier VL, MacNab I. Spinal metastases: The obvious, the occult, and the imposters. Spine 1990;15:1-4.

14. Constans J, DeDiviths E, Donzelli R, et al. Spinal metastases with neurological manifestations. J Neurosurg 1983;59:111-118.

15. Gokaslan Z, Aladag MA, Ellerhorst JA. Melanoma metastatic to the spine: A review of 133 cases. Melanoma Res 2000;10:78-80.

16. Young R, Post EM, King GA. Treatment of spinal epidural metastases: Randomized prospective comparison of laminectomy and radiotherapy. J Neurosurg 1980;53:741-748.

17. Gilbert R, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: Diagnosis and treatment. Ann Neurol 1978;3:40-51.

18. 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(2 Suppl):232-244.

19. Cooper P, Errico TJ, Martin R, et al. A systematic approach to spinal reconstruction after anterior decompression for neoplastic disease of the thoracic and lumbar spine. Neurosurgery 1993;32:1-8.

20. Jackson RJ, Gokaslan ZL. Spinal-pelvic fixation in patients with lumbosacral neoplasms. J Neurosurg 2000;92(1 Suppl):61-70.

21. Miller DJ, Lang FF, Walsh GL, et al. Coaxial double-lumen methylmethacrylate reconstruction in the anterior cervical and upper thoracic spine after tumor resection. J Neurosurg 2000;92(2 Suppl):181-190.

22. Walsh GL, Gokaslan ZL, McCutcheon IE, et al. Anterior approaches to the thoracic spine in patients with cancer: Indications and results. Ann Thorac Surg 1997;64:1611-1618.

23. York JE, Walsh GL, Lang FF, et al. Combined chest wall resection with vertebrectomy and spinal reconstruction for the treatment of Pancoast tumors. J Neurosurg 1999;91(1 Suppl):74-80.

24. Ghogawala Z, Mansfield FL, Borges LF. Spinal radiation before surgical decompression adversely affects outcomes of surgery for symptomatic metastatic spinal cord compression. Spine 2001;26:818-824.

25. Greenberg H, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: Results with a new treatment protocol. Ann Neurol 1980;8:361-366.

26. Edelstyn G, Gillespie PJ, Grebbell ES. The radiologic demonstration of osseous metastases: Experimental observations. Clin Radiol 1967;18:158-162.

27. McNeil B. Rationale for the use of bone scans in selected metastatic primary bone tumors. Semin Nucl Med 1978;8:336-345.

28. Belliveau R, Spencer RP. Incidence and sites of bone lesions detected by [.sup.99m]Tc-polyphosphate scans in patients with tumors. Cancer 1975;36:359-363.

29. Sarpel S, Sarpel G, Yu E, et al. Early diagnosis of spinal-epidural metastases by magnetic resonance imaging. Cancer 1987;59:1112-1116.

30. Jones KM, Schwartz RB, Mantello MT, et al. Fast spin-echo MR in the detection of vertebral metastases: Comparison of three sequences. AJNR Am J Neuroradiol 1994;15:401-407.

31. Aichner F, Poewe W, Rogalsky W, et al. Magnetic resonance imaging in the diagnosis of spinal cord diseases. J Neurol Neurosurg Psychiatry 1985;48:1220-1229.

32. Smoker W, Godersky JC, Knutzon RK, et al. The role of MR imaging in evaluating metastatic spinal disease. AJR Am J Roentgenol 1987;149:1241-1248.

33. Sundaresan N, Scher H, Digiacinto GV, et al. Surgical treatment of spinal cord compression in kidney cancer. J Clin Oncol 1986;4:1851-1856.

34. Dimar J, Mehta S, Glassman SD, et al. The use of somatosensory evoked potentials to prevent ischemic neural damage during preoperative embolization of a vascular renal metastasis. Orthop Rev 1994;23:950-956.

35. Lawrence WT. Wound healing in the cancer patient: Effects of the disease and its treatment. in Moosa AR, Schimpf SC, Robson MC (eds): Comprehensive Textbook of Oncology. Baltimore, Williams & Wilkins, 1991, ed 2, pp 1697-1705.

36. Deramond H, Depriester C, Toussaint P, et al. Percutaneous vertebroplasty. Semin Musculoskelet Radiol 1997;1:285-296.

37. Minart D, Vallee JN, Cormier E, et al. Percutaneous coaxial transpedicular biopsy of vertebral body lesions during vertebroplasty. Neuroradiology 2001;43:409-412.

38. Barr J, Barr MS, Lemley TJ, et al. Percutaneous vertebroplasty for pain relief and spinal stabilization. Spine 2000;25:923-928.

39. Ratliff J, Nguyen T, Heiss J. Root and spinal cord compression from methylmethacrylate vertebroplasty. Spine 2001;26:E300-E302.

40. Ginsburg H, Shetter AG, Raudzens PA. Postoperative paraplegia with preserved intraoperative somatosensory potentials. J Neurosurg 1985;63:296-300.

41. Siegal T, Siegal T. Surgical decompression of anterior and posterior malignant epidural tumors compressing the spinal cord: A prospective study. Neurosurgery 1985;17:424-432.

42. Hellar M, McBroom RJ, MacNab T, et al. Treatment of metastatic disease of the spine with postero-lateral decompression and Luque instrumentation. Neuroorthopedics 1986;2:70-74.

43. Jackson RJ, Gokaslan ZL. Occipitocervicothoracic fixation for spinal instability in patients with neoplastic processes. J Neurosurg 1999;91(1 Suppl):81-89.

44. McPhee I, Williams RP, Swanson CE. Factors influencing wound healing after surgery for metastatic disease of the spine. Spine 1998;23:726-733.

RELATED ARTICLE: Key Points

* Vertebral body metastases often occur in systemic malignancy.

* Metastatic spinal tumors may present with pain or neurologic deficit, or may be detected during screening examinations in patients with known malignancy.

* Operative decompression and stabilization is an important tool in the management of spinal metastatic disease.

* Patient selection and appropriate use of anterior and/or posterior decompression and stabilization are necessary to optimize surgical results.

John K. Ratliff, MD, and Paul R. Cooper, MD

From the Department of Neurosurgery, New York University School of Medicine, New York, NY, and the Department of Neurosurgery, Rush-Presbyterian-St. Luke's Medical Center, Chicago Institute of Neurosurgery and Neuroresearch, Chicago, IL.

Reprint requests to John K. Ratliff, MD, Department of Neurosurgery, Rush-Presbyterian-St. Luke's Medical Center, Chicago Institute of Neurosurgery and Neuroresearch, 1725 W. Harrison Street, Suite 970, Chicago, IL 60612. Email: [e-mail]