Predicting outcomes of transforaminal epidural injections for sciatica.

Abstract

Previous studies on epidural injections have focused on efficacy and have not evaluated factors predicting outcomes of epidural injections. To determine which patient factors are associated with transforaminal epidural injection outcomes for sciatica. Cross-sectional study design at a university spine center. Consecutive patients with sciatica who have had an electromyography study and a transforaminal epidural injection between January 1999 and July 2001. The outcome variable was a composite score of the patients' pain severity and medication use compared with their initial visit. Independent variables included demographics, previous back surgery, pain severity at initial presentation, pain medication use at initial presentation, litigation, having Social Security Disability Insurance (SSDI) or workers' compensation, electromyography results. Medical information was gathered on the variables listed above. Seventy-six patients, mean age 50.4 years and 49% female, participated. At a mean follow-up of 122 days (SD 146.3), 35 (47%) were improved, 21 (28% were unchanged) and 12 (16%) were worse. Initial correlation analysis showed some relationship between outcomes and having SSDI or workers' compensation (R=.43, p<.001), lifting requirements at work (R=.44, p=.002), working status (R=-.31, p=.02), and electromyogram evidence of radiculopathy (R=-.25, p=.04). Regression analysis showed that having SSDI or workers' compensation was the primary factor associated with worse outcomes. When controlled for SSDI/workers' compensation, lifting requirements at work, but not working status and radiculopathy, also were associated with outcomes but the association was not as strong. This paper brings into question the utility of offering epidural injections to patients who are on SSDI/workers' compensation and require heavy lifting at work.