Cord Compression

Updated by Jason D. Kehrer and Bronwyn R. Stall

BACKGROUND

What % of cancer pts develop cord compression?

5%–10% of cancer pts develop cord compression.

What is the median survival in pts with cord compression?

Median survival in pts with cord compression is ∼3 mos (dependent on Hx of primary, extent of visceral/osseous met, degree of motor dysfunction, and performance status)

What is the most important prognostic factor?

Pre-Tx ambulatory status directly influences functional outcome and survival.

What are 3 routes of metastatic spread to the spine?

Routes of metastatic spread to the spine: hematogenous, direct extension, and CSF. (Abeloff MD et al., Abeloff’s Clin Oncol. 4th ed. 2008)

What malignancies commonly cause cord compression?

Lung, breast, and prostate each account for 15%–20% of cases, whereas renal cell, lymphoma, and multiple myeloma each account for 5%–10% of cases of cord compression.

What is the most common site of origin?

Cord compression origin: epidural (extradural) in 95% > leptomeningeal (intradural extramedullary) in 4%–5% > intramedullary in 0.5%–1%

How do pts with cord compression present?

Presenting Sx of cord compression: back pain, radicular pain, weakness, sensory deficits, bowel/bladder dysfunction, and paralysis

What is the most common presenting Sx of cord compression?

The most common Sx of cord compression is back pain present in 90%–95% of pts (often precedes neurologic Sx > 2 mos).

Describe the presentation of cauda equina syndrome?

Presenting Sx of cauda equina syndrome: pain, lower extremity weakness, sensory disturbance (“saddle anesthesia”), and autonomic dysfunction including urinary retention and fecal incontinence

What is the pathophysiology of cord compression?

Gradual progression results in epidural venous congestion, vasogenic edema, and demyelination (potentially reversible) while rapid progression results in disruption of arterial blood resulting in ischemia and cord infarction (irreversible)

What part of the vertebra is most commonly involved by metastatic Dz?

Metastatic Dz typically involves the vertebral body rather than the posterior elements. Compression originates directly from the vertebrae (85%–90%) or via neural foramina extension (5%–10%)

What part of the spine is most often involved in cord compression?

The thoracic spine (70%) is most commonly affected by cord compression > lumbosacral (20%) > cervical (10%).

WORKUP/STAGING

Describe the workup of cord compression.

Cord compression workup: H&P with careful attention to complete neurologic exam to include DRE, evaluation of sensation to determine level of the lesion, assessment of pain, prior cancer management to include prior RT, assessment of bowel/bladder function, and screening MRI spine

Why is a screening MRI of the spine ordered to evaluate cord compression?

Pts with suspected cord compression should be evaluated with a screening MRI of the spine b/c multilevel involvement is not uncommon.

Is a gadolinium-enhanced MRI necessary to evaluate cord compression?

No, it is not required to diagnose epidural compression. Contrast improves identification of leptomeningeal and intramedullary mets.

Why is CT useful in evaluating cord compression?

CT evaluation of spinal cord compression helps to delineate osseous structures, including retropulsed fragments, and aids in surgical planning.

TREATMENT/PROGNOSIS

What modalities are used to treat spinal cord compression?

Modalities used to treat spinal cord compression: steroids, surgery, and RT

What is the initial management of cord compression?

For initial management of cord compression, start dexamethasone (include GI prophylaxis with proton-pump inhibitor or H2 blocker) and consult neurosurgery or orthopedics, depending on the institution, to assess spine stability.

What initial bolus dose of dexamethasone should be used in cord compression?

For newly diagnosed cord compression, a loading dose of 10 mg IV is generally given → 4 mg orally q6hrs. Vecht CJ et al. randomized 37 pts to 10 mg IV vs. 100 mg IV, both → 16 mg daily in divided oral doses. There was no difference in pain control, rate of ambulation, or bladder function. (Neurology 1989)

Historically, what type of surgery was used to treat spinal cord compression?

Historically, laminectomy was used to treat spinal cord compression. However, this was abandoned b/c it can lead to instability, and improved surgical stabilization techniques have allowed for anterior decompressive approaches.

What pts with cord compression are appropriate for decompressive surgery?

Pts with MRI evidence of cord compression in a single area and a life expectancy >3 mos who do not have radiosensitive tumors (lymphomas, leukemias, germ cell tumors, multiple myeloma) may be good candidates for decompressive surgery → RT. (Patchell R et al., Lancet 2005)

What are further indications for surgery?

Spinal instability and/or bony retropulsion, previous RT, Dz progression despite RT, unknown primary tumor (therapeutic and diagnostic), paraplegia <48 hrs

What was the trial design and outcome of the Patchell study of decompressive surgery for cord compression?

The Patchell cord compression trial was a multi-institutional RCT of 101 pts with MRI-confirmed spinal cord compression restricted to a single area with >3-mo life expectancy. Exclusion criteria included being paraplegic >48 hrs, radiosensitive tumors, Hx of prior cord compression, and other pre-existing neurologic conditions. Pts were randomized to decompressive surgery + RT vs. RT alone. RT was 30 Gy/10 delivered to the lesion + 1 vertebral body above and below. Surgery was tailored to the individual lesion to provide circumferential decompression and stabilization as needed (anterior corpectomy for 60% of cases involving only the vertebral body). The study was stopped at interim analysis. Surgery significantly improved the ambulatory rate (84% vs. 57%), duration of ambulatory status (122 days vs. 13 days), and survival (122 days vs. 100 days). Pts nonambulatory prior to Tx were more likely to walk after surgery (62% vs. 19%). (Patchell R et al., Lancet 2005)

What are some criticisms of the Patchell data?

RT alone results were worse than historical prospective controls, small sample size, 18 pts in the RT alone group had an “unstable” spine, surgery provides immediate decompression in pts with rapid onset of Sx (delayed response with RT).

Can surgery be delayed following RT?

Due to a decline in neurologic function (nonambulatory), 10 pts in the radiation group (20%) underwent surgery; 3 regained the ability to walk (30%) with results inferior to surgery upfront

Is SBRT utilized for primary Tx of cord compression?

No, direct tumor contact with the spinal cord (epidural compression within 3 mm) and/or spinal instability are contraindications.

What pts with cord compression should be treated with RT alone?

Cord compression pts treated with RT alone: life expectancy <3 mos, no spinal instability or bony compression, multilevel involvement and radiosensitive tumor

Does the interval between development of motor deficits and RT predict response?

Yes, a longer time interval results in improved functional outcome. A retrospective review of 96 pts demonstrated improved function in 86% of pts when motor Sx were present >14 days. In contrast, only 10% improved when Sx were present <7 days. (Rades D et al., IJROBP 1999)

How are conventional RT fields arranged to treat the cervical, thoracic, and lumbar spine?

Field arrangement for cord compression:

Cervical: opposed lats

Thoracic: AP/PA or PA alone, respecting cord tolerance

Lumbar: AP/PA

Encompass the lesion + 1–2 vertebral levels above and below.

Are there data to support the use of hypofractionation for cord compression?

Marranzano et al. enrolled 300 pts with metastatic spinal cord compression and short life expectancy (≤6 mos) randomized to short-course (8 Gy × 2 fx, 1-wk apart) or split-course (5 Gy × 3 fx, 4-day rest, and then 3 Gy × 5 fx). No significant difference was observed between the 2 schedules with median follow-up of 33 mos (response, duration of response, survival or toxicity). (JCO 2005) A follow-up trial randomized 327 pts to 8 Gy × 2 fx (1 wk apart) vs. 8 Gy × 1 fx without a difference in outcome. No myelopathy was registered with a median follow-up of 31 mos. (Radiother Oncol 2009)

Is reirradiation possible with recurrent cord compression?

Rades et al. reviewed the outcome of 124 pts with in-field recurrence (69% ambulatory) with improvement in 36% and stable motor function in 50% with reirradiation. No radiation myelopathy was observed at a median follow-up of 11 mos with cumulative biological effective dose ≤120 Gy₂ in 92%. (Cancer 2008)