The spinal column is the most frequent site of bony metastasis.[1,2] Metastatic spinal disease is a significant problem for a large number of cancer patients. Between 5% and 10% of all cancer patients develop spinal metastases during the course of their disease. Approximately 40% of persons dying of cancer will have autopsy evidence of spinal metastases,[3-5] and 10% of these will develop spinal cord compression with subsequent neurologic deficits.[6,7] The annual incidence of spinal cord compression secondary to spinal metastases is estimated to be 20,000.
A recent autopsy study found metastatic involvement of the spine in 90% of patients with prostate carcinoma, 75% with breast carcinoma, 55% with melanoma, 45% with lung carcinoma, and 30% with renal carcinoma.[9-11] Clinically significant (symptomatic) spinal cord compression is observed in a higher percentage of patients with certain carcinomas22% of patients with breast cancer, 15% with lung cancer, and 10% with prostate carcinomas.
In almost 10% of cases of newly diagnosed spinal metastases, the patient is not known to have cancer, and spinal cord compression is the initial symptom of the malignancy. Half of these patients are subsequently found to have lung cancer. The time between the original diagnosis of cancer and the occurrence of spinal metastases varies widely.
There is a slight preponderance of metastatic spinal tumors in males (60%) compared to females. This gender difference may reflect the incidence of primary breast vs prostate tumors and the proclivity of prostate cancer to metastasize to bone in published series. Spinal metastases may occur in all age groups, but the period of highest prevalence coincides with the relatively high cancer risk period of 40 to 65 years of age.
Metastatic spinal lesions involve the vertebral column (85% of cases), paravertebral spaces (10% to 15%), or the epidural space (£ 5%).[15,16] The vertebral body is usually the portion of the vertebrae that is involved. Spinal metastases can also be noted intradurally and within the spinal cord itself (intramedullary).[16,17]
Metastatic spread to bone may be related to osteotropism of the malignant cells, direct spread, and/or the highly vascular supply of bone marrow. Batson and others have demonstrated that the vertebral venous plexus draining the thoracic, abdominal, and pelvic viscera (Batsons plexus) is a route of metastases to the spine.
Most metastatic spinal lesions (70%) are found at the thoracic level, 20% in the lumbar region, and 10% in the cervical region. The increased frequency of spinal metastases in the thoracic vertebrae may be related to the greater number of thoracic vertebrae, compared to vertebrae in the lumbar and cervical regions. Metastatic lesions occur at multiple noncontiguous levels in 10% to 38% of cases.[15,19,20]
The incidence of metastatic spinal disease is likely to increase in the future, due to several factors. Improvement in diagnostic capabilities, especially the widespread availability and sensitivity of magnetic resonance imaging (MRI), may increase the number of metastatic lesions detected and patients evaluated. Progress in the treatment of many cancers using multimodality regimens will undoubtedly lead to a prolongation of patient survival, with the resultant development of more spinal metastases. Finally, earlier detection of malignant primary lesions as part of screening programs and improved diagnostic tests will result in progressive increases in the number of spinal metastases detected.
History and Physical Examination
Pain is the most common presenting symptom of patients with metastatic spinal tumors.[15,21,22] The etiology of pain may be tumor spread, spinal instability, nerve root or dural irritation, or direct cauda equina or spinal cord compression. Pain may also be related to bony destruction or invasion of paraspinous tissues, such as muscles or ligaments.
Patients with spinal metastases frequently describe the pain as a dull, constant ache and often complain that the pain awakens them from sleep. Occasionally, the pain will occur in a radicular distribution, which may be caused by nerve root irritation.
Muscular weakness is frequently manifested by the progressive inability to ambulate or the loss of limb function. The patients ability to walk at the time of presentation is a sensitive predictor of the ability to walk after surgery.
Difficulty with ambulation is often slowly progressive, and the patient compensates for the difficulty by using a cane or other assistive device, as well as reducing activities. Alternatively, an ambulatory difficulty may occur rapidly and present as a neurologic emergency. Occasionally, patients will relate difficulty in walking to generalized weakness. Prognostically, the ability to walk, absence of myelopathy, and £ 75% myelographic block are all associated with preservation of the ability of a patient to ambulate after treatment.
Sensory dysfunction, while frequently noted on physical examination, is a rare presenting complaint, although paresthesias are often noted by patients who are questioned carefully. Similarly, patients tend to underestimate the loss of bladder and bowel control, or will relate these difficulties to other medical problems, such as prostatic hypertrophy.
A complete medical history, including a treatment history, must be obtained. Prior treatment may have included radiotherapy and chemotherapy. The radiotherapy fields and dosage must be well defined, as these factors may influence any decision to offer the patient further radiation therapy for either local disease or for treatment of metastatic spinal lesions. High skin dosages may also affect decisions regarding surgical approach and timing.
Significant cardiac dysfunction may influence the decision to offer surgery, the surgical procedure, or the use of preoperative surgical adjuncts, such as tumor embolization. Pulmonary dysfunction also increases perioperative morbidity and mortality. The presence of disease in these major organ systems may represent contraindications to surgery that would be expected to result in excessive blood loss or further lung dysfunction (eg, transthoracic procedures).
An understanding of the rate of disease progression, response of the tumor to prior treatments, and degree of tumor control is essential. The patients desires and the degree of importance that the patient places on the preservation of ambulatory function, bladder and bowel control, and other neurologic functions should also be considered. The patient and family must have realistic expectations of the risks and potential benefits of the surgery so that they can make an informed decision.
A detailed physical and neurologic examination is necessary to determine the patients eligibility for surgery and to establish a neurologic baseline against which responses to therapeutic interventions can be compared. The spine is examined for tenderness to palpation and paraspinous muscular spasm. A localized kyphotic deformity is an important positive finding. Patient resistance to active and passive movements is recorded. The neurologic examination includes an evaluation of mental status and long tract and cerebellar function to help determine whether intracranial lesions are present as well.
Plain spinal radiography is relatively sensitive and specific for metastatic disease. Plain radiographs may be obtained to provide baseline information and intraoperative comparison and can define spinal alignment, the presence of a fracture, and gross areas of bone involvement by tumor. The majority of spinal metastatic lesions are osteolytic, with only 5% being osteoblastic; however, 30% to 60% of the bone must be destroyed for a lytic lesion to be appreciated on plain radiographs.[9,28]
In one series comprising patients with clinical metastatic spinal compression, 60% of patients had normal plain radiographs.[27,29] Although radiographs are frequently employed in initial screening, a negative plain radiograph does not necessarily indicate the absence of disease.
Flexion and extension studies may be required for lesions of the cervical and lumbar spine where extensive bony destruction may be associated with spinal stability. If instability is not felt to be present, a nonoperative approach may be sufficient. However, movement on flexion and extension studies may alone determine the need for surgical spinal stabilization to prevent future neurologic injury.[30-32]
Magnetic resonance imaging is now the method of choice to detect the presence and extent of spinal metastasis.[16,21,33-35] It provides excellent visualization of soft-tissue involvement, bone marrow replacement, ligamentous involvement, spinal cord edema, and degree of canal compromise and cord compression. Magnetic resonance imaging has the further advantage of being able to image the entire spine; this may be especially helpful because multiple lesions that are not suspected clinically are found in up to 30% of cases.
Due to the high sensitivity of MRI for the detection of early changes in bone marrow, several reports have documented MRI detection of spinal metastases that were not appreciated on plain radiographs, radioisotope studies, or computed tomographic (CT) scans.[37-41] However, false-positive MRI studies have been reported in which signal characteristics suggestive of metastases were subsequently found to be nonpathologic reactive tissue within vertebrae.[41,42]
Computed tomography remains an important imaging tool in the evaluation of patients with spinal metastases. Axial CT images provide excellent bony detail of the spinal canal and vertebral bodies and can help determine whether canal compromise is due to bone or soft tissue.[43,44] The integrity of vertebral elements that may be required for internal fixation, such as pedicles and lamina, can also be determined. Computed tomographic scanning may be reserved for patients who cannot undergo MRI scanning for reasons such as the presence of cardiac defibrillators.
Myelography, especially when combined with postmyelographic CT studies, can provide an excellent analysis of both the spinal cord and the vertebral column. In addition, cerebrospinal fluid can be obtained for analysis during the procedure. A possible risk associated with myelography is the development of neurologic deficits in cases of high-grade block when the lumbar puncture is performed below the level of disease.[19,46-47] This complication may be avoided by gaining access to the subarachnoid space via a high cervical cisternal puncture.
Radioisotope studies, most commonly using technetium-99m pyrophosphate and diphosphonate compounds, are an efficient means of detecting bone metastases.[26,48] These studies identify areas of increased bone deposition, such as blastic-type metastases but are less useful in detecting purely lytic-type lesions unless sufficient bone repair is taking place.
Radioisotope studies are specific in only 60% to 80% of patients with metastatic bone disease. Both CT and MRI imaging have been shown to be more sensitive for the detection of cervical metastases than bone scintigraphy.