Background: A previous report demonstrated predictive power of short-term leg pain relief after lumbar transforaminal epidural steroid injections for 1-month treatment response. The question whether the long-term response could be similarly predicted remained unanswered.

Methods: A prospective cohort of 57 patients who underwent a transforaminal epidural steroid injection for sciatica secondary to a lumbar disc herniation was followed for 24 months. Leg and back pain on the visual analog scale, health-related quality of life using the 12-Item Short Form Survey, and functional outcome using the Oswestry Disability Index were assessed. Responders were defined as not receiving any additional invasive treatment after a single injection. Patients who underwent a second injection or surgery were defined as treatment failures (nonresponders).

Results: At 24 months, 31 (54.4%) patients were responders, and 26 (45.6%) were nonresponders. Nonresponders left follow-up at 1 month (n = 9), 3 months (n = 9), 6 months (n = 6) and 12 months (n = 2). No patients were injected again or operated on between the 12- and 24-month follow-up. Responders at 24 months had significantly lower visual analog scale leg pain (P < 0.05) than nonresponders starting from the second week after TFESI and better 12-Item Short Form Survey scores and less disability on the Oswestry Disability Index.

Conclusions: Most patients with a symptomatic lumbar disc herniation who opt for a second injection or surgery do so within the first 6 months. Reliable prediction of the long-term treatment response based on short-term pain relief is not possible.