Selective Epidural Injection: Section 4

As is the case with provocative discography, there are a number of factors that can lead to both false positive and false negative results from selective epidural injections. These factors include changes in CNS nociceptive processing that occur with chronic pain, psychological factors, and placebo responses.

Siddle, P and Cousins, M. Spinal Pain Mechanisms. Spine 1997;22:98-104.

North, R, et al. Specificity of diagnostic nerve blocks: a prospective, randomized study of sciatica due to lumbosacral spine disease. Pain 1996;65:77-85.

However, in contrast to provocative discography, selective epidural injections are done primarily for radicular pain. Therefore, there is a well established gold standard, surgical exploration, which can be used to compare diagnostic tests to.

A number of studies have evaluated the clinical utility of selective epidural and selective nerve root injections. Individual studies have investigated the predictive value of pain provocation, immediate pain relief from local anesthetic infection, and prolonged pain relief from corticosteriod in evaluating patients with radiculopathy.

Derby, et al. Precision percutaneous blocking procedures for localizing spinal pain. Part 2: The lumbar neuraxial compartment. Pain Digest 1993;3:175-88. Ref’s 17, 23.

Dooley, et al. Nerve root infiltration in the diagnosis of radicular pain. Spine 1988;13(1): 79-83.

Stanley, et al. A prospective study of nerve root infiltration in the diagnosis of sciatica. A comparison with radiculography, computed tomography, and operative findings. Spine 1990;15(6):540-3.

Derby, et al. Response to steroid and duration of radicular pain as predictors of surgical outcome. Spine 1992;17(supplement):S176-8.

Synthesizing the results from the studies on clinical utility suggest that the following protocol should be used to interpret the response to selective nerve root injections:

These studies suggest that if a patient has concordant or exact provoked pain in response to injection of contrast, complete pain relief following injection of local anesthetic, and a prolonged steroid response (greater than one week) that the injected nerve root is mediating the patients symptoms, and that a good result can be expected from surgical decompression assuming a correctable lesion is demonstrated on imaging studies.

If a patient has discordant pain, incomplete immediate pain relief, and no prolonged steroid response the injected nerve is probably not mediating the patient’s symptoms, and an alternate pain generator should be searched for.

If an intermediate response occurs a number of different possibilities exist. The patient may still have symptoms arising from a single nerve root, they may have symptoms from multiple roots, or they may not have radicular pain. If the clinical situation is highly suggestive of radiculopathy, it may be reasonable to repeat the snr, potentially with a control injection at an adjacent nerve. If the control injection is negative the result from the active injection is more likely to be a true positive, particularly if there are multiple factors supporting a clinically significant response. If the control injection is positive or intermediate, and there is evidence of nerve root compression at both levels, it may be reasonable to perform a multilevel decompression, depending of course on the many clinical variables that may exist.

As with any diagnostic injection, if patients are psychologically distressed the criteria used to interpret the test may be invalid. Recent data from Caragee on lumbar discography reinforces the effect that psychosocial risk factors have on pain responses. While more study is needed in this area, at a minimum it is important to note these risk factors and understand how they might affect interpretation of the result.

The usefulness of a selective epidural injection is primarily related to the associated provocative and analgesic pain responses. However, the contrast enhanced images from the injection can reveal pathologic findings.

Dooley, et al. Nerve root infiltration in the diagnosis of radicular pain. Spine 1988;13(1): 79-83.

Derby, et al. Precision percutaneous blocking procedures for localizing spinal pain. Part 2: The lumbar neuraxial compartment. Pain Digest 1993;3:175-88.

Two examples of this are a perpendicular nerve root sign, resulting from up-down foraminal stenosis, and obstruction to proximal flow of contrast, which can occur with foraminal stenosis, disc herniations, and scar tissue. The primary utility of findings such as these is to confirm findings evident on advanced imaging studies.