Epidural Steroids for Slipped Discs

Source: Carette S, et al. Epidural corticosteroid injections for sciatica due to herniated nucleus pulposus. N Engl J Med 1997;336:1634-1640.

Back pain affects up to 80% of americans at some time in their lives and up to 50% in any given year. It is the second leading cause of doctor’s visits, the third leading cause of surgical procedures, the fifth leading cause of hospitalizations, and burdens us with more than $24 billion a year in direct medical costs alone (Bodenstein D, et al. Low Back Pain, 2nd ed. Philadelphia: WB Saunders; 1995). Yet controversy still exists as to the ideal method for treating low back pain, including pain due to disc herniation. Epidural corticosteroid injections, though costly, are commonly used despite the lack of rigorous evidence of efficacy.

In a double-blind, placebo-controlled, randomized trial, 158 patients with sciatica due to a herniated nucleus pulposus received injections of either epidural methylprednisolone acetate (Depo-Medrol, 80 mg in 8 mL isotonic saline; n = 78) or isotonic saline (1 mL; n = 80). Injections were administered up to three times, three weeks apart, if patients reported no marked or very marked improvement. Evaluations were performed at three weeks, six weeks, and three months and included a neurological examination and measurement of maximal forward flexion. Intercurrent use of other medications was recorded. Patients reported their own perceived degree of improvement on a descriptive seven-item scale as well as using two components of the McGill Pain Questionnaire (Melzack R. Pain 1975;1:277-299). Functional disability was measured using the Oswestry Low Back Pain Questionnaire (Fairbank JCT, et al. Physiotherapy 1980;66:271-273) and the Sickness Impact Profile (Bergner M, et al. Med Care 1981; 19:787-805).

At three weeks, no significant difference was appreciated between the groups except that the methylprednisolone group had greater improvement of forward flexion and a smaller proportion had sensory deficits. At six weeks, only the degree of improvement of leg pain was significant in the methylprednisolone group compared to placebo. At three months, no significant difference was evident between the groups in any of the outcome parameters. Throughout the trial, both groups experienced amelioration of symptoms, with 55% in each group reporting marked or very marked improvement.


Clearly, epidural steroid injections provide no long-term benefit in treating low back pain due to herniated nucleus pulposus, and their use in the short term affords neither functional benefit nor reduced need for surgery. This overused and costly treatment, if used at all, should be reserved for patients who cannot tolerate oral medication and are not surgical candidates. —mr