ACR Appropriateness Criteria®

Clinical Condition: Myelopathy

Variant 1: Traumatic.

Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.

Variant 2: Painful.

Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.

Variant 3: Sudden onset.

Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.

Variant 4: Stepwise progressive.

Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.

Variant 5: Slowly progressive.

Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.

Variant 6: Infectious disease patient.

Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.

Variant 7: Oncology patient.

Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.

Summary of Literature Review

The term myelopathy is used to describe any neurological deficit related to the spinal cord itself. Most frequently, myelopathy is due to compression of the spinal cord by osteophyte or extruded disc material in the cervical spine. Osteophytic spurring and disc herniation may also produce myelopathy localized to the thoracic spine, though this is less common. The next most common sources of myelopathy are spinal cord compression due to extradural mass caused by carcinoma metastatic to bone, and blunt or penetrating trauma. Many primary neoplastic, infectious, inflammatory, neurodegenerative, vascular, nutritional, and idiopathic disorders may also result in myelopathy, though these are very much less common than discogenic disease, metastases, and trauma. A variety of cysts and benign neoplasms may also compress the cord; these tend to arise intradurally. The most common of these are meningiomas, nerve sheath tumors, epidermoid cysts, and arachnoid cysts.

In general, disorders of the spinal cord itself are uncommon and difficult to treat effectively. Therefore, most attention in the radiological evaluation of myelopathy is focused on extrinsic compression of the spinal cord. Classically, radiological evaluation of myelopathic patients consisted of positive contrast myelography. Later, this evaluation was supplemented by computed tomography (CT) and CT myelography, and then magnetic resonance imaging (MRI) became the mainstay in the evaluation of myelopathy. More recently, imaging of the spinal cord has improved to the point that reliable diagnosis of nonexpansile spinal cord lesions is routinely possible.

Despite the wide variety of causes of myelopathy, diagnosis and treatment rest on demonstration of mechanical stability of the spine, particularly in the cervical region and when tumor or trauma history is present. Depiction of direct neural involvement by a pathologic process is then required for more refined diagnosis and specific treatment decisions. Anatomical diagnosis of myelopathy rests principally in the distinction between extradural, intradural, and intramedullary lesions.

Clinically, the diagnosis of myelopathy depends on the neurological localization of the finding to the spinal cord, rather than the brain or peripheral nervous system, and then to a particular segment of the spinal cord. The antecedent clinical syndrome and other details of the patient's course help to refine diagnosis, but imaging plays a crucial role. In general, myelopathy is clinically divided into categories based on the presence or absence of significant trauma, the presence or absence of pain, and the mode of onset (slowly progressive or insidious onset vs. a stepwise progression vs. a sudden onset). Patients with known tumor history and those in whom infectious disease is likely may also be considered separately.

In the patient with traumatic myelopathy, the first priority for the spine is assessing its mechanical stability. Plain radiographs are useful for this purpose, but CT may be more useful when a high probability of bony injury or ligamentous injury is present. At some centers, routine multidetector CT with sagittal and coronal reconstructions is supplanting the role of plain radiographs, especially in the setting of multiple trauma.

MRI is widely considered the study of choice when paralysis is incomplete or under other circumstances where direct visualization of neural or ligamentous structures is clinically necessary. If surgery for herniated disc, hematoma, or other cause of incomplete paralysis is planned, MRI best depicts the relation of pathology to the cord, and can help predict which patients may benefit from surgery.

When local or radicular pain accompanies myelopathy, the most likely diagnoses are spondylosis, tumor, or infection. Plain radiographs may depict osteophytic narrowing of the spinal canal or bone destruction. CT improves the depiction of both bony encroachment on the spinal canal and compression of neural structures by herniated disc material that is occult to plain radiographic evaluation. Bone destruction and soft-tissue masses are also better seen. MRI has largely replaced CT scanning in the noninvasive evaluation of patients with painful myelopathy because of its superior soft-tissue resolution and multiplanar capability. Invasive evaluation by means of myelography and CT myelography may be supplemental when visualization of neural structures is required for surgical planning or other specific problem solving, though this is less frequent.

Although painful myelopathy is most commonly due to spondylosis and disc herniation, a significant proportion is caused by tumor or infection. Demyelinating disease may present with pain symptoms as well. Occasionally, syringomyelia may present with the anesthetica dolorosa syndrome. The ability of MRI to depict the spinal cord directly, and to assess its contour and internal signal characteristics reliably and noninvasively, has resulted in general acceptance of MRI as the study of choice in evaluating cervical myelopathy when spondylosis or disc herniation is the most likely cause. When MRI is not available, or to answer specific questions before surgical intervention, myelography and CT myelography may be useful.

In slowly progressive myelopathy, the ability of MRI to depict the spinal cord noninvasively is most valuable. Some specifically treatable disorders may be localized and depicted quite well by means of myelography followed by CT myelography. However, the occasional complication of myelography in cases of spinal block, difficulty in visualizing the upper extent of lesions, and relative "blind spots" at the cervical thoracic and craniocervical junctions limit the utility of myelography. CT myelographic techniques may help avoid these pitfalls and may be useful to answer specific preoperative questions about bony anatomy.

Enlargement of the spinal cord by intramedullary mass is well depicted by myelography only when large masses are present. CT myelography can be extremely useful in supplementing the plain radiographic examination. These techniques, however, are less useful than MRI because the distinction between solid and cystic masses is usually not possible, even when delayed examination is performed. The distinction of syrinx from tumor, location of tumor nodule, extent of cyst, and distinction of nodule and cyst from edema are crucial in treatment planning for intramedullary disease and virtually necessitate MRI.

When myelopathy progresses stepwise or is of sudden onset, vascular processes become significant diagnostic possibilities. Vascular malformations, spinal cord infarct, and epidural hematoma account for most of the vascular lesions of the cord. In practice, they are difficult to distinguish clinically from other nontraumatic causes of myelopathy because the classic history is frequently absent or difficult to elicit from a seriously ill patient.

If arteriovenous malformation (AVM) is considered clinically likely, gadolinium-enhanced MRI, MR angiography, and myelography to demonstrate abnormal vasculature may be useful adjuncts to guide spinal arteriography. More recently, progress in CT angiography has led to its use in preangiographic evaluation of patients with suspected spinal vascular abnormalities.

If myelopathy is painless and slowly progressive, the differential diagnosis is quite broad. Neoplastic disease of the spinal cord and extrinsic compression by epidural or intradural tumor may present in this manner. Demyelinating disease, degenerative diseases, and metabolic or deficiency diseases may also present in this fashion. Spondylosis may present painlessly as well, particularly in the elderly. In these cases, visualization of the spine as well as the spinal cord is useful, and this is best accomplished noninvasively by MRI.

In oncology and infectious disease patients, multiple sites of involvement are possible. In these patients it is often necessary to study the entire spine or even the entire skeleton despite a specifically localized myelopathic level. MRI is considered more sensitive at an individual site, but the convenience of radionuclide bone scanning makes it useful in this setting as well. Acquired immune deficiency syndrome (AIDS) patients may present with myelopathy due to primary cord disease caused by human immunodeficiency virus (HIV) infection. No high-quality evidence supports the use of discography, thermography, epidural venography, ultrasound, or central spinal fluid (CSF) flow studies in the evaluation of myelopathy. Radionuclide bone scan may play an adjunctive role, for example, to locate a safer biopsy site in patients with suspected metastatic cord compression.

An important limitation of MRI in the diagnosis of myelopathy is its high sensitivity. The ease with which the study depicts expansion and compression of the spinal cord in the myelopathic patient may lead to false positive examinations and inappropriately aggressive therapy if findings are interpreted incorrectly. For example, transverse myelitis due to demyelinating disease may demonstrate cord enlargement and be mistaken for tumor. Spondylosis, which occurs with normal aging, may be mistaken for clinically significant osteophytic compression of the spinal cord in a patient who is myelopathic for other reasons. These problems are minimized by experienced observers and meticulous clinical correlation with radiologic findings. Similar problems are present in the interpretation of any anatomical study of the spinal cord and are not unique to MRI. Careful patient selection and clinical correlation are essential in interpretation of imaging findings everywhere.

Anticipated Exceptions

The following are available:

The National Guideline Clearinghouse™ (NGC) does not develop, produce, approve, or endorse the guidelines represented on this site.

All guidelines summarized by NGC and hosted on our site are produced under the auspices of medical specialty societies, relevant professional associations, public or private organizations, other government agencies, health care organizations or plans, and similar entities.