Spinal Cord Compression

Spinal cord compression develops in 1% to 5% of patients with systemic cancer. It should be considered an emergency, because treatment delays may result in irreversible paralysis and loss of bowel and bladder function.

Etiology

Compression of the spinal cord is predominantly due to extradural metastases (95%) and usually results from tumor involvement of the vertebral column. A tumor may occasionally metastasize to the epidural space without bony involvement.

Site of involvement

The segment most often involved is the thoracic spine (70%), followed by the lumbosacral spine (20%) and cervical spine (10%).

Most common malignancies

Spinal cord compression occurs in a variety of malignancies; the most common are lung, breast, unknown primary, prostate, and renal cancers.

Signs and Symptoms

Early signs

More than 90% of patients present with pain localized to the spine or radicular in nature (ie, not due to bony involvement but rather to neural compression). Pain, which is usually secondary to bony involvement, is often exacerbated with movement, recumbency, coughing, sneezing, or straining. The majority of patients experience pain for weeks to months before neurologic symptoms appear.

Intermediate signs

If cord compression goes untreated, weakness often develops next. It may be preceded or accompanied by sensory loss.

Late signs

Symptoms of autonomic dysfunction, urinary retention, and constipation are late findings. Once autonomic, motor, or sensory findings appear, spinal cord compression usually progresses rapidly and may result in irreversible paralysis in hours to days if untreated.

Physical findings

These may include tenderness to palpation or percussion over the involved spine, pain in the distribution of the involved nerve root, muscle weakness, spasticity, abnormal muscle stretch reflexes and extensor plantar responses, and sensory loss. Sensory loss occurs below the involved cord segment and indicates the site of compression. In patients with autonomic dysfunction, physical findings include a palpable bladder or diminished rectal tone.

Diagnosis

The first step in the diagnosis of spinal cord compression is an accurate neurologic history and examination.

Radiography

More than 66% of patients with spinal cord compression have bony abnormalities on plain radiographs of the spine. Findings include erosion and loss of pedicles, partial or complete collapse of vertebral bodies, and paraspinous soft tissue masses. Normal spine films are not helpful for excluding epidural metastases.

MRI

The standard for diagnosing and localizing epidural cord compression is MRI. Gadolinium-enhanced MRI has been especially helpful in assessing cord compression secondary to spinal epidural abscesses, because gadolinium enhances actively inflamed tissues and defines anatomic boundaries. An abnormal signal within the disk space suggests the possibility of infection.

Primary or secondary neoplasms involving the vertebral bodies generally demonstrate a long T1, resulting in decreased signal intensity on a T1-weighted image, and a long T2, with increased signal intensity on the T2-weighted image.

CT and myelography

If MRI is unavailable, a CT scan and/or myelogram may be used to diagnose and localize epidural cord compression.

Prognosis

Treatment outcome correlates with the degree and duration of neurologic impairment before therapy. In a prospective analysis of 209 patients treated for spinal cord compression with radiotherapy and corticosteroids, Maranzano and Latini reported that of patients who were ambulatory, nonambulatory, or paraplegic before treatment, 98%, 60%, and 11%, respectively, were able to ambulate following therapy. Treatment outcome in the most radiosensitive malignancies (eg, lymphoma) was superior to that in the less sensitive cancers (eg, renal cell carcinoma). Almost all ambulatory patients treated with either irradiation alone or laminectomy followed by postoperative irradiation remained ambulatory after treatment, whereas approximately 10% of patients whose lower extremities were paralyzed could walk after treatment.

Treatment

The goals of treatment of spinal cord compression are recovery and maintenance of normal neurologic function, local tumor control, stabilization of the spine, and pain control. Choice of treatment depends on clinical presentation, availability of histologic diagnosis, rapidity of the clinical course, type of malignancy, site of spinal involvement, stability of the spine, and previous treatment.

In general, radiation therapy has been the treatment of choice for these patients. This is based on the belief that radiotherapy is as effective as surgery in terms of pain relief and maintaining neurologic function. In other words, the potential complications and convalescence associated with surgery can be avoided in this group of patients with a limited life expectancy.

However, this approach has been further investigated in a randomized clinical trial reported by Patchell et al. A total of 101 patients with spinal cord compression caused by metastatic cancer were randomized to undergo either surgery followed by adjuvant radiation therapy (n = 50) or radiation therapy alone (n = 51). Radiotherapy for both groups consisted of 10 fractions of 300 cGy each. The primary endpoint was the ability to walk.

The study was stopped after an interim analysis of the 101 patients revealed that 42 of 50 patients (84%) in the surgery group were able to walk after treatment, compared with 29 of 51 patients (57%) in the radiotherapy group (P = .001). In addition, patients treated with surgery retained the ability to walk significantly longer than patients treated with radiation therapy alone (median, 122 days vs 13 days; P = .003). Of the 32 patients who entered the trial unable to walk, 10 of 16 (62%) in the surgery arm regained the ability to walk, compared with 3 of 16 (19%) in the radiotherapy arm (P = .01). On the basis of the results of this trial, decompressive surgery followed by adjuvant radiation therapy should be considered in the treatment of patients with spinal cord compression.

The ability to regain ambulatory function after surgery had been recognized before this study. This finding represented the rationale for strong consideration of surgery in this group of patients. The authors advocated the wider use of surgery in most patients with spinal cord compression. Still, there are reasons to consider radiotherapy alone as appropriate initial treatment. They include the disappointing results in the radiotherapy-alone arm in this study compared with the experiences of previous studies and the possible reluctance to consider spinal surgery by patients and/or physicians (based on limited life expectancy). These issues should, of course, be thoroughly reviewed during the process of informed consent.

Patchell et al have recently reexamined the role of age in determining outcomes of surgery vs radiation therapy. Secondary data analysis of the randomized trial with age stratification demonstrated a strong interaction between age and treatment outcomes. Multivariate modeling and Kaplan-Meier curves revealed that for patients 65 years or older, there was no difference in the preservation of ambulation between the surgery and radiotherapy-alone arms. However, for patients younger than 65 years, surgery still resulted in prolonged ambulation (P = .002).

Corticosteroids

Dexamethasone should be administered if the patient’s history and neurologic examination suggest spinal cord compression. There is controversy as to whether an initial high dose of intravenous dexamethasone (100 mg) followed by 10 mg of dexamethasone every 6 hours is necessary. Some studies have suggested that lower doses are just as effective.

Radiation therapy

Radiation therapy alone is still usually the standard initial treatment for most patients with spinal cord compression due to a radiosensitive malignancy. Treatment outcome is contingent upon both the relative radiosensitivity of the malignancy and the neurologic status of the patient at the time radiotherapy is initiated.

Radiation portal. In general, the treatment volume should include the area of epidural compression (as determined by MRI or myelography) plus two vertebral bodies above and below. Consideration should be given to including adjacent areas of abnormalities if feasible. Careful matching techniques should be employed in patients treated to adjacent vertebral levels, a situation that is not uncommon.

Radiation dose and fractionation. The chosen regimen should take into account such factors as field size and normal tissue tolerance. Smaller fields are appropriately treated to 2,000 to 3,000 cGy over 1 or 2 weeks, respectively. Larger fields may occasionally necessitate longer courses, such as 4,000 cGy over 4 weeks, to minimize adverse effects.

Investigators from the University Hospital Hamburg reported their results on five fractionation schemes of radiation therapy for spinal cord compression. In this retrospective review, 1,304 patients were treated from January 1992 through December 2003. Radiation schedules included 1 × 8 Gy (n = 261), 5 × 4 Gy (n = 279), 10 × 3 Gy (n = 274), 15 × 2.5 Gy (n = 233), and 20 × 2 Gy (n = 257). Improvement in motor function was noted in 26% (1 × 8 Gy), 28% (5 × 4 Gy), 27% (10 × 3 Gy), 31% (15 × 2.5 Gy), and 28% (20 × 2 Gy). Motor function improvement and posttreatment ambulatory rates were not significantly different throughout all groups.

On multivariate analysis, age, performance status, pretreatment ambulatory status, and length of time that motor deficits were present before initiation of radiotherapy were all significantly associated with improved functional outcome, whereas the schedule of radiation therapy was not a significant indicator. Recurrence rates at 2 years were 24%, 26%, 14%, 9%, and 7%, respectively, in the five radiation-schedule groups described above. There was mild acute toxicity and no late toxicity. The authors concluded that shorter fractionation schemes should be considered for patients with poor predicted survival.

The ability to maintain local control in a patient with spinal cord compression has been recently reported. Rades et al conducted a prospective nonrandomized study evaluating recurrence rates in short-course versus long-course radiotherapy. A total of 231 patients received radiation therapy for spinal cord compression: 114 patients received short-course radiation therapy (1 × 8 Gy or 5 × 4 Gy), and 117 patients received long-course radiation therapy (10 × 3 Gy, 15 × 2.5 Gy, or 20 × 2 Gy). This study showed an improvement in progression-free survival at 12 months for long-course radiation therapy compared with short-course radiation therapy (72% vs 55%; P = .034). In addition, there was improvement in local control in favor of the long-course group (77% vs 61%; P = .032). However, there was no difference in functional outcome or overall survival.

In another study, the University Hospital Hamburg reported its prospective evaluation of 10 versus 20 fractions of radiation therapy for metastatic spinal cord compression. A total of 214 patients were irradiated with 30 Gy in 10 fractions (n = 110) or 40 Gy in 20 fractions (n = 104). Motor function improved in 43% of patients treated with 30 Gy and in 41% of patients treated with 40 Gy (P = .799). There was no significant difference in posttreatment ambulatory rates (60% and 64%, respectively; P = .708). As expected, being ambulatory before the initiation of treatment was associated with better functional outcome after irradiation (P = .035). Acute toxicity was mild, and no late toxicity was observed during the 12-month follow-up.

Re-treatment may be entertained, particularly when no effective alternative exists. Usually, doses of 2,000 cGy over 2 weeks can be used for re-treatment. It is important, however, to counsel the patient regarding the risk of radiation neuropathy. Furthermore, only patients who had a lasting response to the initial treatment should be re-irradiated, because tumors that were refractory to the first course of therapy or that recur within 3 months are unlikely to respond to subsequent courses.

Surgery

Vertebral body resection for a tumor anterior to the cord and posterior laminectomy for a tumor posterior to the cord may be appropriate treatment options for relieving spinal cord compression in patients who require spinal stability, have undergone previous radiotherapy in the area of the compression, require a tissue diagnosis of malignancy, or experience progression of the cord compression despite optimal treatment with corticosteroids and irradiation.

In general, surgical decompression should be strongly considered in patients whose cord compression is caused by a relatively radioresistant cancer and who have a severe neurologic deficit (such as bowel or bladder dysfunction). Unfortunately, many patients in this situation are not candidates for aggressive surgery. In these cases, radiotherapy is offered, albeit with limited expectations for neurologic recovery.

Chemotherapy

Chemotherapy may be an effective treatment of spinal cord compression in select patients with a chemosensitive metastatic tumor. It also may be considered in combination with other treatment modalities, such as radiotherapy, or as an alternative if those modalities are not suitable options for relieving spinal cord compression.