Intended for healthcare professionals

Practice Guidelines

Diagnosis and management of patients at risk of or with metastatic spinal cord compression: summary of NICE guidance

BMJ 2008; 337 doi: (Published 27 November 2008) Cite this as: BMJ 2008;337:a2538
  1. B D White, consultant neurosurgeon1,
  2. A J Stirling, consultant orthopaedic spinal surgeon2,
  3. E Paterson, general medical practitioner3,
  4. K Asquith-Coe, project manager4,
  5. A Melder, senior researcher 4
  6. on behalf of the Guideline Development Group
  1. 1Nottingham University Hospitals NHS Trust, Queen’s Medical Centre, Nottingham NG7 2UH
  2. 2Royal Orthopaedic Hospital NHS Foundation Trust, Department of Spinal Surgery, Northfield, Birmingham B31 2AP
  3. 3Govan Health Centre, Glasgow G51 4BJ
  4. 4National Collaborating Centre for Cancer, Cardiff CF10 3AF
  1. Correspondence to: B White barrie.white{at}

    Why read this summary?

    Metastatic spinal cord compression is thought to affect more than 4000 people each year in the United Kingdom.1 2 Treatment before paralysis is clinically effective and cost effective. Despite the fact that spinal pain is often present for three months and neurological symptoms for two months before paraplegia, almost 50% of patients are unable to walk by the time of diagnosis.1 3

    Recognition is difficult as non-specific back pain is common in both the general population and patients with cancer.4 5 In addition, 23% of patients with spinal metastases have no prior cancer diagnosis.1 The added distress and disability caused by paralysis affecting someone already living with cancer cannot be overestimated.6

    This article summarises the most recent recommendations from the National Institute for Health and Clinical Excellence (NICE) on how to diagnose and manage patients at risk of or with metastatic spinal cord compression.7 The algorithm outlines the management of patients with suspected metastatic spinal cord compression (figure).


    Management of patients with suspected metastatic spinal cord compression


    NICE recommendations are based on systematic reviews of best available evidence. When minimal evidence is available, recommendations are based on the guideline development group’s opinion of what constitutes good practice. Evidence levels for the recommendations are given in italic in square brackets.

    Early recognition and diagnosis

    • For patients who have cancer and bone metastases or a high risk of developing bone metastases, and for patients with cancer who present with spinal pain, offer information on the symptoms of spinal metastases and metastatic spinal cord compression and what to do and whom to contact if symptoms develop or worsen while waiting for investigation or treatment. [Based on the experience of the Guideline Development Group]

    • In patients with cancer, the following suggest spinal metastases: pain in the thoracic or cervical spine; severe unremitting or progressive lumbar spinal pain; spinal pain aggravated by straining; nocturnal spinal pain preventing sleep; localised spinal tenderness. Contact urgently (within 24 hours) the metastatic spinal cord compression coordinator at the nearest centre treating this condition to discuss the care of such patients. [Based on the experience of the Guideline Development Group and moderate quality observational evidence]

    • In patients with cancer and pain suggestive of spinal metastases the following suggest metastatic spinal cord compression: neurological symptoms including radicular pain, limb weakness, difficulty in walking, sensory loss, or bladder or bowel dysfunction; neurological signs of spinal cord compression. This situation is an oncological emergency, so contact the metastatic spinal cord compression coordinator immediately. [Based on the experience of the Guideline Development Group and moderate quality observational evidence]

    • For patients with cancer and non-specific lumbar spinal pain and patients without cancer who have suspicious spinal pain, review frequently for development of progressive pain or other features suggestive of spinal metastases (refer within 24 hours) or development of neurological features of spinal cord compression (refer immediately).

    • Magnetic resonance imaging of the whole spine should be performed in time to plan definitive treatment: within 1 week if clinical features suggest spinal metastases; within 24 hours if clinical features suggest spinal cord compression; sooner (including out of hours) if emergency treatment is needed. [Based on the experience of the Guideline Development Group and moderate to high quality evidence of cost effectiveness]

    • Do not routinely do imaging of the spine of asymptomatic patients with malignancy.

    • Do not use plain radiographs to investigate spinal metastases or metastatic spinal cord compression.

    • Do not unnecessarily investigate or transfer patients who are too frail or unfit for specialist treatment.

    Centres treating metastatic spinal cord compression

    • Identify or appoint individuals responsible for the roles of metastatic spinal cord compression coordinator and senior medical advisers (clinical oncologists, radiologists, and spinal surgeons with expertise in treating this condition) and ensure their availability at all times.

    • The metastatic spinal cord compression coordinator coordinator should:

      • -Provide the first point of contact for clinicians

      • -Advise on the immediate care of the spinal cord and spine

      • -Agree the time and place for investigations and treatment if required

      • -Liaise with the acute receiving team about admission and mode of transport.

    • Senior medical advisers should determine the optimal care of patients with metastatic spinal cord compression, with advice from primary tumour site clinicians where possible and with the full involvement of the patient.

    Initial care of the patient with spinal cord compression

    • For patients with severe mechanical pain suggestive of spinal instability or neurological features suggestive of metastatic spinal cord compression, nurse the patient flat with the spine in neutral alignment (using, for example, “log rolling” or turning beds) until spinal and neurological stability are ensured and cautious remobilisation may begin. [Based on the experience of the Guideline Development Group]

    • Unless dexamethasone is contraindicated (for example, when substantial suspicion of lymphoma exists), give a loading dose of at least 16 mg dexamethasone, followed by a short course of 16 mg dexamethasone daily until definitive treatment is delivered or decided against, then wean and stop.

    • Manage postural hypotension with positioning and devices to improve venous return; avoid overhydration. Manage bladder dysfunction with a urinary catheter on free drainage. Clear airway secretions with breathing exercises, assisted coughing, and suctioning.

    • Follow NICE guidance for prophylaxis of venous thromboembolism,8 the prevention and treatment of pressure ulcers,9 10 and the management of bowel dysfunction.11

    • Offer specialist psychological and spiritual support at diagnosis, during treatment, and on discharge.

    Treatments primarily for spinal pain

    • Offer conventional analgesia, escalating from simple proprietary preparations to opioids (see the World Health Organization’s “pain ladder” at

    • Offer bisphosphonates to all patients with vertebral involvement from myeloma and breast cancer and to patients with prostate cancer in whom conventional analgesia is inadequate.

    • For patients with non-mechanical spinal pain, 8 Gy single fraction palliative radiotherapy may be appropriate even in complete paralysis.

    • For patients with mechanical spinal pain, consider spinal orthoses, vertebroplasty or kyphoplasty, or spine stabilisation surgery; these treatments may be appropriate even in complete paralysis.

    • For patients with intractable pain, consider specialist pain care procedures (including epidural and intrathecal analgesia) and neurosurgical interventions.

    Treatments primarily to prevent paralysis

    • If definitive treatment is appropriate, it should always be started before patients lose the ability to walk or other neurological deterioration occurs, and ideally within 24 hours. [Based on the experience of the Guideline Development Group and moderate quality observational studies]

    • When planning definitive treatment:

      • -Whenever practical, establish the primary histology of spinal metastases including tumour biopsy if necessary

      • -Determine the sites and extent of spinal and visceral metastases

      • -Take into account: patient preferences; neurological and functional status; general health and fitness; previous treatments; magnitude of surgery and likelihood of complications; fitness for general anaesthesia; overall prognosis.

    • Do not deny surgery or radiotherapy on the basis of age alone.


    • Consider major surgical treatments only if patients are expected to survive longer than three months.

    • Laminectomy alone should be performed only for isolated epidural tumour or neural arch metastases without bony instability.

    • Posterior decompression should be accompanied by internal fixation (with or without graft) if spinal metastases involve the vertebral body or threaten spinal stability.

    • For patients fit to undergo more extensive surgery, consider vertebral body reinforcement with cement for those expected to survive for up to one year and reconstruction with anterior bone graft for patients expected to survive one year or longer

    • Do not offer surgery as a definitive treatment to patients with metastatic spinal cord compression who have been completely paraplegic or tetraplegic for more than 24 hours, unless spinal stabilisation is required for pain relief.


    • For patients who are judged best treated by definitive radiotherapy or are unsuitable for surgery, provide urgent radiotherapy (within 24 hours, seven days a week) unless:

      • -They have had complete tetraplegia or paraplegia for more than 24 hours and their pain is well controlled, or

      • -Their prognosis is too poor. [Based on poor to moderate quality case series studies and evidence of cost effectiveness]

    • Do not carry out preoperative radiotherapy if surgery is planned.

    • Offer postoperative fractionated radiotherapy to patients with a satisfactory outcome, once the wound has healed.

    Rehabilitation, support, and care at home

    • Start discharge planning on admission, led by a named healthcare professional, and involving the patient and carers, their primary oncology team, rehabilitation team, and community support—including primary care and specialist palliative care as required. Ensure community based rehabilitation and support are available when the patient returns home. [Based on the experience of the Guideline Development Group and on NICE guidance12 and the West of Scotland cancer network guidelines13 ]

    • Focus rehabilitation on the patient’s goals, including functional independence, participation in normal activities of daily life, and quality of life.

    • Offer families and carers any training and support required, including bereavement support.

    Overcoming barriers

    The biggest barrier to preventing paralysis from metastatic spinal cord compression is a lack of awareness of the condition and of the improvements in the treatments available, and a lack of urgency in its investigation and treatment. No economic barriers exist.

    This guideline is intended to raise awareness of the condition as a complication of cancer. It includes not only clinical guideline but also service guidance to improve access to treatment. Cancer networks are expected to incorporate this guideline into their overall strategy and promote local introduction of its recommendations. Commissioners are expected to ensure that facilities are in place, meetings of all interested parties occur twice a year, and audit is performed to assess the delivery of service and clinical outcomes of all patients affected.

    Implementation of the guideline and awareness of the issues it covers form the remit of NICE’s implementation support programme (see

    Further information on the guidance


    Improvements in the management of primary malignancy mean that many people now live for years with their disease and its complications. The true incidence of metastatic spinal cord compression in the United Kingdom is unknown because the diagnosis is not systematically recorded, but it is believed to be about 80 cases per million population. This represents about 4000 new cases each year in England and Wales, more than 100 cases per cancer network.

    Typically, symptoms of spinal metastases are present for three months before referral for a specialist opinion, and neurological symptoms often precede paralysis from metastatic spinal cord compression by two months. Despite this, almost half of patients cannot walk by the time of diagnosis, and of these, almost 70% remain immobile. Of those able to walk at treatment, about 80% remain ambulant.

    The added distress and disability caused by paralysis affecting someone already living with cancer cannot be overestimated. Patient survey discloses the personal tragedy of this potentially avoidable condition:

    • “I started experiencing discomfort when walking, a sort of dull ache down one side of my body. It slowed my walking down and I started walking with a slight limp. I also experienced this pain if I lifted items such as travel bags.”

    • “It still hadn’t occurred to me that all the pain I was experiencing could indicate that I had metastatic breast cancer in my back. I had no awareness of the risk of it developing there and still thought I had an unrelated back problem.”

    • “My general practitioner was treating me for backache. I had no toilet functions (and was prescribed laxatives) and had loss of walking ability and acute pain. He just gave me stronger painkillers.”

    • “A delay of nearly one week as doctors at local hospital diagnosed only back pain, cancelled MRI, and admitted me to a ward for observation.”

    • “The delays in diagnosis and the speed of onset of my symptoms in the last few days before treatment meant that there was no window of opportunity to consider whether I would benefit from surgical treatment, which would have entailed a transfer to another hospital some 30 or 40 miles away. This option was only mentioned to me after my radiotherapy treatment, when it was no longer an option.”

    • “When I asked why patients aren’t informed of the possibility of having metastatic spinal cord compression I was told that it’s a very difficult subject to broach without ‘freaking’ people out. My feeling on this, though, is that patients who might develop this do have an absolute right to know about this as a possibility because of the disastrous consequences that can happen if action isn’t taken quickly. Isn’t it better to be informed than the very real scenario of ending up being paralysed for the rest of your life?”

    It is difficult to compare the costs of treatment to prevent paralysis with the cost of caring for someone with paralysis from cancer because evidence is sparse, quality of life improvements are uncertain, and relative longevities are artificially affected by case selection. Pseudorandomised modelling of different clinical scenarios confirmed that treatment to prevent paralysis is clinically effective and cost effective, provided that case selection follows the recommendations in this guideline and outcomes are satisfactory.

    The deficiencies identified by the Guideline Development Group during production of this guideline relate largely to unawareness of both the condition and the treatments to prevent it.

    In the distant past, laminectomy alone was often done in desperate circumstances, as an out of hours emergency, too late to make a difference, and the procedure occasionally worsened pain and instability without improving neurological function. Unreasonable requests and poor outcomes generated nihilistic opinions and a reluctance to refer, investigate, and treat.

    This poor reputation for the care of metastatic spinal cord compression comes partly from the difficulties encountered in making a clinical diagnosis early in the course of the condition, when treatment is most effective.

    Spinal pain is a very common self limiting symptom, sometimes accompanied by radicular pain. It affects a third of the population each year—a fifth of whom visit their general practitioner—and is managed conservatively and appropriately without investigation for the most part. Although less than 0.1% of people who visit their general practitioner with back pain have spinal metastases, 23% of patients with spinal metastases have no prior cancer diagnosis, and patients with cancer experience non-specific back pain as often as, if not more commonly than, the general population.

    This makes appropriate selection of patients for investigation of early metastatic spinal cord compression a challenge and will inevitably result in increased diagnostic demand. This added awareness or lowering of threshold should be offset by earlier and easier treatment of fitter patients with better outcomes using fewer healthcare resources during a shortened hospital stay and with less long term dependency.

    The services required to deliver this guideline are already available, and no new treatments are proposed. This NICE guideline is largely about early recognition of the condition and a coordinated response to ensure that timely referral and timely investigation lead to scheduled rather than emergency treatment and to better outcomes in terms of quality of life and longevity.

    Through the implementation team at NICE, national cancer networks, local commissioners, and cancer charities, the care of people with metastatic spinal cord compression is expected to improve markedly in the coming years.

    Primary care trusts should discuss the implementation of guideline recommendations, including those on service delivery, as soon as possible with the Cancer Action Team, Welsh Assembly government, and the cancer networks.


    The development of this guideline was based on methods outlined by the NICE guidelines manual.14 Four different versions of the guideline have been produced: a full version containing all the evidence and the recommendations; a quick reference guide; a version known as the “NICE guideline,” which lists the recommendations; and a version of the NICE guideline for patients and the public. All the versions are available from the NICE website (

    Future updates of the guideline will be produced as part of the NICE guideline development programme.14

    Future research and remaining uncertainties

    The Guideline Development Group also made the following research recommendations:

    • Further research should be undertaken on the reasons why patients with metastatic spinal cord compression present late

    • The use of radiotherapy to prevent the development of metastatic spinal cord compression in patients with identified spinal metastases but no pain should be investigated in prospective randomised controlled trials

    • The use of vertebroplasty and kyphoplasty in preventing metastatic spinal cord compression in patients with vertebral metastases should be investigated in prospective comparative studies

    • The use of surgery to prevent the development of metastatic spinal cord compression in patients with identified spinal metastases but no pain should be investigated in prospective randomised controlled trials

    • Further research should investigate (a) what are the most clinically and cost effective regimens of radiotherapy to treat patients with established metastatic spinal cord compression and (b) the use of new techniques, such as intensity modulated radiation therapy.


    Cite this as: BMJ 2008;337:a2538


    • This is one of a series of BMJ summaries of new guidelines, which are based on the best available evidence; they highlight important recommendations for clinical practice, especially where uncertainty or controversy exists.

    • Contributors: All authors contributed to the conception and drafting of this article and revising it critically. They have all approved this version.

    • Funding: The National Collaborating Centre for Cancer was commissioned and funded by the National Institute for Health and Clinical Excellence to write this summary.

    • Competing interests: None declared.

    • Provenance and peer review: Commissioned; not externally peer reviewed.


    View Abstract