The Discography Controversy: Discussion

Risks of discography include discitis, neurologic and visceral injury, dye reactions, spinal headache and others. Spinal cord injury, vascular injury, pre-vertebral abscess, and subdural empyema have all been reported post-discography (9). An article in 1993 mentioned an overall complication rate of 13% with cervical discography, including the subsequent development of an acute epidural abscess that led to myelopathy and eventual quadriplegia (10). The same article stated that information gained from cervical discography was not worth its potential risks (10). Another publication in 1981 mentioned one patient in whom cervical spondylodiscitis developed 48 hours after discography, lasting 15 months, and a second patient developed quadriplegia within seconds after injection of the contrast material into the intervertebral space (11). In 1995 it was recognized that the incidence of complications with discography were 0.15% per patient and .08% per disc (6). Although the complications associated with discography may indeed be serious, the majority of serious and high complication rates were reported before 1970, and have decreased enormously since that time due to improvement in injection techniques, imaging, contrast materials, and physician training. The procedure performed in the new millennium is very different from that which was performed years ago (6). In addition, as with most procedures, there is an inverse relationship between complication rate and physician experience.

The possibility that discography may cause persistent back pain is another question that has been raised. This has been difficult to ascertain since discography is generally performed in subjects with pre-existing back pain. A correlation between ongoing back pain for greater than one year after discography in patients with significant emotional, psychological and chronic pain problems has been identified (15). Also of note is that subjects with disability claims account for more than 80% of those with persistent pain after experimental discography, and subjects with normal psychometric test results had no reports of significant long-term back pain after discography (15). Discographic pain reports are not only related to anatomic abnormalities, but are influenced by personality as assessed by the Minnesota Multiphasic Personality Inventory. It was found that patients with high scores on the hypochondriasis, hysteria and depression scales may tend to overreport pain during discographic injection (28).

Another ongoing debate involves a comparison between magnetic resonance imaging of the spine, and the information obtained by discogram. MRI is an excellent diagnostic tool that can differentiate between annulus fibrosis and nucleus pulposus, as can a discogram. MRI can also be used to define details of the normal aging process of the disk. The major advantage of discogram is that it can confirm that a patient's symptoms are discogenic, whereas MRI cannot (4). Discogram is far more than an imaging study, and the most valuable information from the test is gained from symptom reproduction. A correlation has been found between pain drawings and discographic pain responses (16). Another recent significant finding on discography was that painless, disrupted discs were found in elderly patients. Particularly In this patient population, discography was felt to be helpful in differentiating clinically significant abnormalities from those associated with the aging process (18).

Bernard looked at the value of lumbar discography followed by computed tomography in detail. He studied 250 patients with low-back pain and found that in 93% the combination of lumbar discography followed by computed tomography provided additional useful diagnostic information that affected patient management and the selection of treatment alternatives. Lumbar discography followed by CT proved valuable in determining the significance of equivocal or multiple level abnormalities, determining the type of disc herniation, defining surgical options, and evaluating the previously operated spine. He also noted that computed tomography-discography may be more sensitive than magnetic resonance imaging in the early stages of disc degeneration because 18 of 177 discs with a normal T2-weighted image were discographically abnormal and the CT-discogram revealed annular tears or fissuring (23).

With regards to the patient with an iodine contrast allergy, gadolinium may be used with magnetic resonance imaging subsequent to injection instead of computed tomography.

False positive findings on discography have also been investigated recently. The ability of a patient to separate spinal from non-spinal sources of pain on discography was questioned after a publication by Carragee found that patients with no history of back pain who had undergone posterior iliac bone graft harvesting for non-lumbar procedures often experienced a concordant painful sensation on lumbar discography with their usual gluteal area pain (20). Another study concluded that the side of an annular tear does not correlate with the side of a patient's back pain, and that the pain symptoms provoked during discography may not originate from a lumbar disc (25). One must be proficient and cautious when interpreting discography results.

Discogram has helped to prove that normal discs on MRI may still be responsible for producing a patient's pain (12,13). In addition, disks with abnormal morphology on MRI have been shown to be painless on discogram. Reproduction of pain of significance and familiarity to the patient, degenerative morphology, and accurate needle position constitute the criteria that define a positive discogram result. It is important that all three factors above are met in order for discography to maintain a high level of specificity as a diagnostic tool. It is felt at this time that MRI can neither replace, nor predict discography (12,13).

Cervical analgesic discography has also been studied (22), and may also be used in some cases to treat discogenic pain symptoms if other treatments fail.