Back to Physician Literature

Contemporary Concepts in Spine Care
Magnetic Resonance Imaging
Use in Patients With Low Back or Radicular Pain
Edward D. Simmons, MD, Richard J. Herzog, MD, Richard D. Guyer, MD, Arnold Graham-Smith, MD

With the current emphasis on cost containment, it is important to order the single best diagnostic test when clinical uncertainties must be resolved. Magnetic resonance imaging is currently the optimal imaging modality to provide the maximum amount of information when evaluating patients with suspected spinal disorders. A comprehensive magnetic resonance imaging study is needed along with a subspecialty interpretation to provide the greatest amount of useful clinical information. [Key words: lumbar, magnetic resonace imaging, MRI, spinal imaging] Spine 1995;20:1834-1839

Patients presenting with low back pain or pain radiating into the lower extremities challenge a physician who desires a precise diagnosis for the etiology of the pain. Back pain and neural dysfunction are a frequent symptom complex for a myriad of processes afflicting the lumbar spine and paraspinal tissues, therefore a clinician must carefully consider a large number of disorders in the work-up of a patient with low back pain or radicular pain. Fortunately, most acute episodes of low back pain are self-limited, and imaging tests are not needed. If pain persists or worsens, diagnostic imaging may provide the additional clinical information needed to rationally choose an appropriate treatment. The more precise the information provided by a diagnostic test, the more focused a therapeutic regimen can be. In this statement, we provide guidelines for the use of magnetic resonance imaging (MRI) for the evaluation of low back pain and pain radiating from the lumbar spine.

Historical Review

Before 1970, plain film radiography, with or without contrast, was the primary radiologic examination used to evaluate patients with lumbar spine dysfunction. Now, advanced imaging techniques, such as highresolution computed tomography (CT) and MRI, make it possible to evaluate noninvasively soft tissue and bone structures of the spine in any desired plane. ¹²

For CT examination of the lumbar spine, x-rays are used to generate cross-sectional images. Computed tomographic images are representations of differential x-ray attenuation by tissue. Spatial and contrast resolution are dependent on the energy of the x-ray source, slice thickness, field of view, and scanning matrix.⁸ High resolution CT provides excellent delineation of osseous structures.'o Its main limitations are radiation exposure, slightly restricted field of view, and poor delineation of intrathecal anatomy and pathology.

Whereas a CT image created with an x-ray source is determined by the electron density of the tissue, MRIs are based on totally different physical properties. An MRI is created when pulsed radio waves of a specific frequency induce the transition of a fraction of the spinning protons in the body into a higher energy state. With the termination of the radio-frequency pulse, the excited nuclei release energy and return to their lower energy state. Construction of images from this pattern of absorption-release of energy is called MRI.⁶ Magnetic resonance imaging produces superb delineation of soft tissue structures, excellent characterization of medullary bone, direct multiplanar imaging, and no radiation exposure.

Rationale
Back pain and pain that radiates from the back can be caused by disorders that threaten life, well-being, or major neurologic function. The need for early and accurate identification of such problems as tumor, infection, or major compression of spinal cord or cauda equina is evident. Even when the source of such symptoms is not as threatening, precise information about the anatomic source of pain may become necessary to good medical care. Information about the spine and its adjacent structures, such as the quality of tissue, the forms of the various components, and the sizes of spaces within and between the bones, can be derived from MRI. Magnetic resonance imaging is easily obtained (at least in the United States), noninvasive, and risk- and pain-free for most people and requires no special preparation of the patient. Although there are limitations to its usefulness, some relative and absolute contraindications, and some pitfalls in the applications of results, MRI gives more specific and complete information about the anatomy and pathology of the lumbar spine than any other available diagnostic tool.

The most common source of lumbar pain is related to some form of spinal degeneration. Degenerative processes may affect the disc or other support structures and may occur as acute, subacute, or chronic problems. Imaging studies may be needed to assess the entire lumbar motion segment, including the intervertebral disc, facet joints, and vertebral body endplates. Magnetic resonance imaging is the best examination to provide such information noninvasively. ^2,11,18

Disc degeneration may affect the chemical composition, ligament integrity, and the morphology of the disc. Magnetic resonance imaging may delineate pathoanatomic and chemical changes of a degenerating disc before disc herniation. ^9,13,21 Magnetic resonance imaging is capable of delineating small tears in the anular fibers. ^25,29,31 When nuclear material is displaced, it is possible with MRI to differentiate between a contained, noncontained, or sequestered herniation. ^11,19,30

Facet arthrosis may result from the altered functional status of a degenerated disc because increased load is transferred to the facet joints. Magnetic resonance imaging may reveal such changes as facet hypertrophy, osseous proliferation, cartilage narrowing, joint effusion, and capsular hypertrophy.

Spinal stenosis is defined as a local, segmental, or generalized narrowing of the central spinal canal or neural foramina by bone or soft tissue elements-an abnormality of passageways, which demands imaging that portrays spaces. Often associated with stenosis are hyperostotic ridges of the vertebral endplates, osseous proliferation of the facets, and hypertrophy of the ligamenta flava. Magnetic resonance imaging may show the stenosis and define which structures cause the stenotic process. ²³ Magnetic resonance imaging permits the evaluation of the true sagittal dimensions and crosssectional area of the thecal sac.

Because of exquisite delineation of cancellous bone and soft tissues, MRI is the optimal modality to detect the presence of any inflammatory process involving disc, vertebral body, or paravertebral soft tissue. ²² Magnetic resonance imaging is also the optimal examination to detect primary and metastatic neoplasms in the spinal column, spinal cord, or paravertebral soft tissues. ^{1,16, 29} Because it can noninvasively determine the exact position of the comis medullaris, MRI may be important when assessing the significance of posttraumatic spinal deformities or planning surgery for fracture of the spine. ³

When back pain occurs secondary to retroperitoneal, pelvic, or other visceral source, the role of a negative MRI in ruling out musculoskeletal etiologies may facilitate the diagnostic work-up.

Results
Disc Disease
With the superb soft tissue resolution obtainable from MRI, it is possible to determine if a disc herniation is contained by the outer anular-posterior longitudinal ligament complex or is extruded through this complex to become a noncontained disc herniation, as shown by Grenier's excellent correlation between MRI and anatomic abnormalities.¹¹ In a study reported by Masaryk,¹⁹ the accuracy of MRI in differentiating sequestered disc fragments from other forms of lumbar disc herniation was 85% compared with a 65% accuracy for CT and myelography. Early degeneration of the vertebral body endplates cannot be imaged well by CT, but as Modic ²⁴ reported, MRI can detect these degenerative changes. Forristall' compared MRI and contrast CT to surgical findings and found that overall MRI provided the most information about the state of disc degeneration or herniation. Thornbury ³⁰ compared MRI to CT in the evaluation of patients with nerve compression and acute low back pain and found no significant difference between the studies and concluded that factors of cost, radiation dose, and invasiveness must be considered when selecting a diagnostic examination. There have been several reports on the spontaneous resolution of disc herniation in which MRI was the optimal, noninvasive imaging study to document these changes. ⁵

Spinal Stenosis
Modic ²³ reported in a prospective study evaluating patients with lumbar disc disease and central stenosis using MRI, CT, and myelography that MRI was comparable to CT in the evaluation of spinal stenosis and eliminated the need for contrast injection and radiation exposure. Bolender ⁴ reported that the cross-sectional area of the thecal sac correlated best to patients' stenotic symptoms and found MRI ideally suited to provide this information noninvasively. Such unusual causes of spinal stenosis as intra- or extradural. tumors, synovial cysts, and osseous abnormalities may be defined accurately by MRI. ^15,20,29

Spinal Infection
Modic ²³ studied the use of MRI to evaluate spinal inflammation and found the sensitivity of MRI was 96%, specificity 93%, and accuracy 94% in the diagnosis of osteomyelitis. Tlie accuracy of these results was similar to a combined bone scan and gallium study. The main advantages of MRI over radionuclide studies include the acquisition of additional information about the spinal cord, thecal sac, and paravertebral soft tissues. Smith ²⁸ reported the value of MRI in detecting and characterizing tuberculous spondylitis and myelitis.

Recurrent Disc Herniation
Hueftle et al¹⁴ described the use of gadolinium-enhanced MRI in the evaluation of patients after surgery. When gadolinium is given intravenously and MRI scanning performed within approximately 10 minutes of the injection, the gadolinium enhances scar tissue but not disc tissue, making it easier to differentiate between postoperative scar and recurrent disc herniation. It is generally beneficial to have MRI performed before and after gadolinium injection because it is easier to identify the areas of gadolinium uptake. Ross et al ²⁷ reported a 96% accuracy rate in distinguishing scar from recurrent disc herniation in back patients after surgery. They used MRI scanning before and after injecting gadopentetatedimeglumine.

Discussion
Radionuclide scintigraphy has one advantage over MRI during a search for tumor or infection when the location is not well-defined, in that a single MRI study is limited to one region, such as the lumbar spine, and could miss abnormal tissue in the pelvis or thoracic spine that might be discovered by a bone scan or leukocyte scan. Magnetic resonance imaging, as cited previously, may reveal infections or tumors that are not detected from radionuclide scans, and MRI evaluates soft tissues in ways that scans do not.

Because of sharp edge-enhancement and crisp internal bone detail, CT may reveal lesions such as stress fractures that MRI does not detect and may portray the bone pathology of fractures with clarity superior to MRI. With fractures, MRI has the advantage of evaluating adjacent cancellous bone and soft tissues, including nerve and cord, better than CT.

In some circumstances, particularly presurgical evaluation of stenosis, radiographic images obtained from motion studies after myelographic contrast injection can provide precise definition of the relationship of bone edge to contrast-defined dural sac and nerve sheath. This information may not be obtainable with the same precision by MRI.

Evaluation of pain that is provoked by disc injection during discography attempts to correlate the clinical manifestation of abnormalities with the observed radiographic abnormality in a way that MRI does not broach. Additionally, CT performed after discography may provide images of the interior of the disc, fissuring of the anulus, or definition of the extent of containment of injected contrasts that are uniquely different from assessments made from MRI.

One must consider that MRI, radiography, scintigraphy, myelography, CT, and discography are unique diagnostic tests and that no one test obviates the need for the others in all cases. In selecting which tests are appropriate for any one patient, the special diagnostic needs of the patient must be compared with the potential of the test along with consideration of such factors as availability, cost, pain, risk, and difficulties of interpretation and application.

When the test is done to rule out life-threatening disease, a high degree of sensitivity with ability to discern a small abnormality early in its course is desirable. When the diagnostic test is done to help with a decision about whether to treat a benign-but-painful condition invasively, the problems of clinical correlation become more important than the ability to discern small and early lesions. Magnetic resonance imaging provides a very high yield to detect abnormalities, but often such abnormalities are unrelated to the patient's pain.

Magnetic resonance imaging cannot discern what abnormalities produce symptoms and cannot, in most cases, define the duration an abnormality has been present or the cause of its presence. Magnetic resonance imaging, therefore, leaves considerable importance to such information as may be gained from other tests and, most especially, to the judgment of the treating physician. An MRI finding is not, in itself, an indication for surgical or other specific treatment. Criticisms that MRI yields "false-positives" and leads to unnecessary surgery are criticisms of the judgment used to apply the results of the test and are not valid criticisms of the test.

Technical advances related to MRI have been rapid over the past decade; the equipment needed to provide the best results from MRI has been expensive. Those who administer and interpret MRI must be highly skilled. This examination is relatively costly. In areas where health care is not generously funded, MRI may not be easily available.

An MRI examination demands little of the patient except motionlessness, but that may be intolerable for the duration of the test for claustrophobic, disoriented, or emotionally or mentally impaired patients and, in some cases, when pain is too severe. The quality of the images obtained may be severely degraded by the proximity of metal to the area of study, as in many patients who have had internal fixation surgery of the spine.

Future Studies
Multiple reports document the accuracy of MRI in the evaluation of abnormal lumbar changes, but there has been a paucity of well-controlled studies that determine its efficacy and impact on patient outcome.¹⁷ Because such studies would require control of many variables, such as how patients are selected, how results of imaging studies are applied clinically, how treatments are administered, and how outcomes are measured, they would require large, multidisciplinary teams with very generous funding.

Of considerable social concern is the cost of MRI. There would seem little other reason not to do a test that involves no pain and no risk. At what time, in which circumstances, and according to whose judgment an MRI should be done could be studied from the view of cost-effectiveness. Research that leads to lower costs of MRI machines and software, shorter examination times, and reduced demands on the time of highly skilled technicians and professionals would broaden the applicability of the test.

Current Recommendations
If there has been a major acute injury or there are symptoms of infection, neoplasia, or progressive neural dysfunction, MRI may be appropriate to the initial work-up. Because most patients with lumbar and radicular pain do not present those problems and most spontaneously improve, it is more appropriate to wait until symptoms have persisted for approximately 7 weeks without improvement, despite proper care, before performing MRI.

Of currently available tests, MRI provides the most accurate and complete information for the initial assessment of disorders of the lumbar disc, stenosis, infection, neoplasia, or trauma. In some situations, CT, scintigraphy, myclography, or discography may also be needed, particularly before surgical intervention.

It is critical when performing an MRI that a complete study be done. This includes sequential sagittal and axial high-resolution images to optimally evaluate all components of the spine. Spin echo sequences providing T1 and T2 information are standard. T2* imaging is frequently used but has some limitations, e.g., artifacts secondary to magnetic field or tissue inhomogeneity, decreased sensitivity to disc desiccation, and exaggeration of stenosis. High quality MRI can be obtained from mid- and high-field strength MRI systems if imaging protocols are optimized. To obtain maximum information about the spine, it is important that MRI be interpreted by a radiologist who subspecializes in musculoskeletal disease.

The absolute contraindications for performing MRI include the presence of the following materials in the patient's body: ferromagnetic cerebral aneurysm clips, cardiac pacemakers, certain cardiovascular stents or valves, metallic foreign bodies in the orbit or spinal canal, and ferromagnetic cochlear implants.²⁶ Relative contraindications include pregnancy, transcutaneous electrical nerve stimulators, and severe claustrophobia.

Acknowledgments
Each Contemporary Concepts review expresses a consensus representing general views of current practice and should not be used to dictate care of patients to the exclusion of innovation or tailoring to special circumstances.

References
1. Avrahami E, Tadmor R, Dally 0, Hadar H. Early MR demonstration of spinal metastases in patients with normal radiographs and CT and radionuclide bone scans. J Comp Assist Tomogr 1989;13:598-602. Lumbar MRI - Herzog et al 1837
2. Berns DB, Blaser SI, Modic MT. Magnetic resonance imaging of the spine. Chn Orthop 1989;244:78-100.
3. Blumenkopf B, Juneau PA 111. Magnetic resonance imaging (MRI) of thoracolumbar fractures. J Spinal Disord 1988; 1: 144-SO.
4. Bolender NF, Schonstrom NSR, Spengler DM. Role of computed tomography and myelography in the diagnosis of central spine stenosis. J Bone joint Surg (Ani] 1985;67:240-S.
S. Bozzao A, Gallucci M, Masciocchi C, et al. Lumbar disk herniation: MR imaging assessment of natural history in patients treated without surgery. Radiology 1992;18S:135-41.
6. Bushong SC. Magnetic resonance imaging: Physical and biological principles. St. Louis: CV Mosby, 1988:382.
7. Forristall RM, Marsh HO, Pay NT. Magnetic resonance imaging and contrast CT of the lumbar spine: Comparison of diagnostic methods and correlation with surgical findings. Spine 1988;13:1049-S4.
8. Genant H. Technical considerations. In Newton TH, Ports DG, eds. Computed Tomography of the Spine and Spinal Cord. San Anselmo: Clavadel Press, 1983:1-14.
9. Gill K, Blumenthal SL. Function results after anterior lumbar fusion at LS-S1 in patients with normal fusion at L.5-S1 in patients with normal and abnormal MRI scans. Spine 1992;17:940-2.
10. Glenn WV Jr, Rothman SLG, Rhodes ML, Kerber CW. An overview of lumbar computed tomography/multiplanar reformations: What are its elements and how do they fit together? In: Post NJD, ed. Computed Tomography of the Spine. Baltimore: Williams & Wilkins, 1984:135-S4.
11. Grenier N, Greselle JF, Vital JM, et al. Normal and disrupted lumbar longitudinal ligaments: Correlative MR and anatomic study. Radiology 1989;171:197-205.
12. Herzog RJ. Selection and utilization of imaging studies for disorders of the lumbar spine. Physical Medicine and Rehabilitation Clinics of North America 1991;2:7-S9.
13. Hickey DS, Aspden RM, Hukins DWL, et al. Analysis of magnetic resonance images from normal and degenerate lumbar intervertebral discs. Spine 1986;11:702-8.
14. Huefde MG, Modic MT, Ross JS, et al. Lumbar spine: Post-operati, c MRI imaging with Gd-DTPA. Radiology 198 8; 167:817-24.
1S. Jackson DE Jr, Atlas SW, Mani JR, Norman D. Intraspinal synovial cysts: MR imaging. Radiology 1989;170:527-30. 16. Kattapuram SV, Khurana JS, Scott JA, El-Khoury GY. Negative scintigraphy with positive magnetic resonance imaging in bone metastases. Skeletal Radiol 1990;19:113-6.
17. Kent DL, Haynor DR, Larson EB, Deyo RA. Diagnosis of lumbar spine stenosis in adults: A meta-analysis of the accuracy of CT, MR1 and myelography. AJR Am J Roentgenol 1992;lS8:1135-44.
18. Maravilla KR, Harding RP. Imaging decisions in degenerative spinal disease: A practical approach. MRI Decisions 1988;2:2-13.
19. Masaryk TJ, Ross JS, Modic MT, et al. High-resolution MR imaging of sequestered lumbar intervertebral disks. AJR Am J Roentgenol 198 8; 150: 115 5- 62.
20. Maupin WB, Naul LG, Kanter SL, Chang CS. Synovial cyst presenting as a neural foraminal lesion: MR and CT appearance. AJR Am J Roentgenol 1989;153:1231-2.
21. Miller JAA, Schmatz C, Schultz AB. Lumbar disc degeneration: Correlation with age, sex, and spine level in 600 autopsy specimens. Spine 1988;13:173-8.
22. Modic MT, Feiglin DH, Piraino DW, et al. Vertebral osteomyelitis: Assessment using MR. Radiology 1985;157: 157-66.
23. Modic MT, Masaryk T, Boumphrey F, et al. Lumbar hemiated disk disease and canal stenosis: Prospective evaluation by surface coil MR, CT and myelography. AJR Am J Roentgenol 1986;147:757-65.
24. Modic MT, Steinberg PM, Ross JS, et al. Degenerative disk disease: Assessment of changes in vertebral bone marrow with MR imaging. Radiology 1988;166:193-9.
25. Osti OL, Fraser RD. MRI and discography of annular tears and intervertebral disc degeneration. A prospective clinical comparison. j Bone joint Surg (Br] 1992;74:43 1-5.
26. Paveficek W, Geisinger M, Castle L, et al. The effects of nuclear magnetic resonance on patients with cardiac pacemakers. Radiology 1983;147:149-53.
27. Ross JS, Masaryk Tj, Schrader M, et al. MR imaging of the postoperative lumbar spine: Assessment with gadopentetate dimeglumine. AJR Am J Roentgenol 1990;155:867-72. 28. Smith AS, Weinstein MA, Mizushima A, et al. MR imaging characteristics of tuberculos spondylitis vs vertebral osteomyetitis. AJR Am J Roentgenol 1989;153:399-405.
29. Sze G, Krol G, Zimmerman RD, Deck M.DF. Malignant extradural spinal tumors: MR imaging with gadoliniumDT?A. Radiology 1988;167:217-23.
30. Thorbury JR, Fryback DG, Turski PA, et al. Disk-caused nerve compression in patients with acute low-back pain: Diagnosis with MRI, CT myelography, and plain CT. Radiology 1992;186:731-8.
31. Yu S, Haughton VM, Sether LA, et al. Criteria for classifying normal and degenerated lumbar intervertebral disks. Radiology 1989;170:523-6.

Address reprint requests to
Richard J. Herzog, INM Department of Radiology University of Pennsylvania Medical Center Hospital of the University of Pennsylvania 3400 Spruce Street Philadelpbia, PA 19104-4283