Prolonged structural compromise of spinal nerve roots can lead to chronic changes that surgical decompression might not be able to reverse. In this study, it was hypothesized that if there were a reversible structural pain component, a steroid injected into the patient's symptomatic nerve root should provide temporary pain relief and that these patients should have a favorable surgical outcome. It also was hypothesized that duration of radicular symptoms would correlate inversely with surgical outcome. For postoperative relief of radicular pain, the results showed that patients with pain lasting less than 1 year had a positive surgical result (89%), regardless of response to steroid. Patients with pain lasting more than 1 year and who have had a positive response to steroid injected into the symptomatic nerve root (roots) had a positive surgical outcome of 85%. Patients who did not respond to the steroid and had pain for more than 1 year (95%) generally had a poor surgical outcome. Although the poor outcome in the last group might be explained in some cases by an inadequate structural correction, inadequate stabilization, or functional reasons, the majority of these failures represented irreversible changes in the neural structures.