Interventions for the treatment of metastatic extradural spinal cord compression

Cancer that spreads to the spine can compress the spinal cord and nearby spinal structures. If this is not treated, it can lead to pain, disability (including paraplegia) and incontinence. Radiotherapy, steroids and different surgical techniques have been used to reduce the pressure on the spinal cord. It is important to select patients carefully for the different types of treatments as the prognosis and outcome vary greatly depending on the type of cancer and the stage of the illness. This review included trials of radiotherapy, surgery, and steroids to assess if these treatments helped improve walking ability. One study showed that in some common cancers, or where survival was expected to be short, two radiotherapy treatments a week apart were as effective as longer courses of radiation treatment. Another trial showed that with decompressive surgery before radiotherapy compared to radiotherapy alone, more patients maintained the ability to walk and more patients regained their ability to walk. Survival in the radiotherapy alone group was 100 days versus 126 days in those who had surgery first. An older trial reported no additional benefit with the surgical procedure of laminectomy before radiotherapy. It is difficult to give definite recommendations based on these few trials which included a relatively small number of patients, and had different selection criteria. Radiotherapy will be required in the majority of patients, with better results seen in those who have not lost walking ability. Carefully selected patients with a single site of cord compression, who are fit for surgery and have not been paraplegic for more than 48 hours may be considered for decompressive surgery before radiotherapy. High doses of steroids (96 to 100 mg of dexamethasone) significantly increased the risk of serious side effects as compared to moderate doses of 10 to 16 mg dexamethasone or placebo.

Authors' conclusions: 

Patients with stable spines retaining the ability to walk may be treated with radiotherapy. One trial indicates that short course radiotherapy suffices in patients with unfavourable histologies or predicted survival of less than six months. There is some evidence of benefit from decompressive surgery in ambulant patients with poor prognostic factors for radiotherapy; and in non-ambulant patients with a single area of compression, paraplegia < 48 hours, non-radiosensitive tumours and a predicted survival of more than three months. High dose corticosteroids carry a significant risk of serious adverse effects.

Read the full abstract...

Metastatic epidural spinal cord compression (MESCC) is often treated with radiotherapy and corticosteroids. Recent reports suggest benefit from decompressive surgery.


To determine effectiveness and adverse effects of radiotherapy, surgery and corticosteroids in MESCC.

Search strategy: 

CENTRAL, MEDLINE, EMBASE, CINAHL, LILACS and CANCERLIT were searched; last search ran July 2008

Selection criteria: 

We selected randomized controlled trials (RCTs) of radiotherapy, surgery and corticosteroids in adults with MESCC.

Data collection and analysis: 

Three review authors independently assessed quality of included studies and extracted data. We calculated risk ratios (RR) and numbers needed to treat to benefit (NNT) with 95% confidence intervals (CI) and assessed heterogeneity.

Main results: 

We identified six trials (n = 544). One trial (n = 276) compared radiotherapy 30 Gray in eight fractions with 16 Gray in two fractions and showed no difference. Overall ambulatory rates were 71% versus 68%, (RR 1.02, CI 0.90 to 1.15); 91% versus 89% of ambulant patients maintained ambulation (RR 1.02, CI 0.93 to 1.12); 28% versus 29% of non-ambulant patients regained ambulation (RR 0.98, CI 0.51 to 1.88). In one trial (n = 101) decompressive surgery had significantly better outcomes than radiotherapy in selected patients. Overall ambulatory rates were 84% versus 57% (RR 0.67, CI 0.53 to 0.86, NNT 3.70 CI 2.38 to 7.69); 94% versus 74% maintained ambulation (RR 0.79, CI 0.64 to 0.98, NNT 5.00 CI 2.78 to 33.33); 63% versus 19% regained ambulation (RR 0.30, CI 0.10 to 0.89; NNT 2.27 CI 1.35 to 7.69). Median survival was 126 days versus 100 days. Laminectomy offered no advantage (n = 29, 1 trial). Three trials provided insufficient evidence about the role of corticosteroids (n = 105, Overall ambulation RR 0.91, CI 0.68 to 1.23). Serious adverse effects were significantly higher in high dose corticosteroid arms (n = 77, two RCTs, RR 0.12, CI 0.02 to 0.97).