Steroids Versus Gabapentin

Randomized, Double-blind, Comparative-effectiveness Study Comparing Epidural Steroid Injections to Gabapentin in Patients With Lumbosacral Radiculopathy

Walter Reed National Military Medical Center, Brooke Army Medical Center, Landstuhl Regional Medical Center, Washington D.C. Veterans Affairs Medical Center, United States Naval Medical Center, San Diego, Case Western Reserve University, Milton S. Hershey Medical Center

The purpose of this study is to determine whether pharmacotherapy or epidural steroid injections are a better treatment for lumbosacral radicular pain. 142 patients referred to a participating pain clinic with lumbosacral radiculopathy will be randomized in a 1:1 ratio to receive one of two treatments. Half (n=71) of the patients will be allocated to receive an epidural steroid injection (ESI; group I), with an equal number allocated to receive gabapentin (group II). Patients & evaluating physicians will be blinded. Follow-up will be through 3-months after treatment.

142 patients referred to a participating pain clinic with lumbosacral radiculopathy will be randomized in a 1:1 ratio to receive one of two treatments. Half (n=71) of the patients will be allocated to receive an ESI (group I), with an equal number allocated to receive gabapentin (group II). Group I patients with unilateral symptoms will receive (unilateral) transforaminal ESI, while those with bilateral symptoms will receive (central) interlaminar ESI, as is common practice. In group II patients who receive gabapentin, the dose will be titrated to between 1800 mg/d and 2700 mg/d in TID dosing, but may be lowered or elevated (up to 3600 mg/d) depending on the clinical circumstances. To ensure blinding, these patients will also receive midline (for patients with bilateral symptoms who would receive interlaminar ESI) or unilateral paraspinal (for patients with unilateral symptoms who would receive transforaminal ESI) normal saline into the interspinal ligaments or paraspinal musculature, respectively. Injections and medication titration will commence on the same day. Rescue medications will consist of tramadol 50 mg 1 to 2 tablets every 6 hours PRN (up to 8/d) and/or ibuprofen 400-800 mg every 6 hours PRN (not-to-exceed 3000 mg/d). Patients already taking analgesics, including opioids, can continue on these medications "as needed". The first follow-up visit will be scheduled 1-month from the start of treatment. A positive outcome will be defined as a > 2-point decrease in leg pain coupled with a positive satisfaction rating. Subjects who obtain a positive outcome at their initial 1-month follow-up visit will remain in the study and return for the final 3-month follow-up visit. Those with a negative outcome will exit the study "per protocol" to receive standard care, which may consist of unblinded ESI, medical management with drugs such as gabapentin (for those who did not receive gabapentin) and antidepressants, and physical therapy. Subjects who obtain a positive outcome at 1-month but experience a recurrence before their 3-month follow-up visit will also exit the study per protocol, with their final outcome measures recorded before they receive standard care. At all follow-up visits, pill counts will be conducted to determine medication compliance.

This outcome measure compares the average leg pain at baseline to the average leg pain 1 month after the start of treatment. This is measured using the Numeric Pain Scale. The Numeric Pain Scale ranges from 0-10. 0 being no pain at all and 10 being the worst imaginable pain possible. The epidural steroid injection group is compared to the gabapentin group.

This outcome measure compares the average leg pain at baseline to the average leg pain 3 month after the start of treatment. This is measured using the Numeric Pain Scale. The Numeric Pain Scale ranges from 0-10. 0 being no pain at all and 10 being the worst imaginable pain possible. The epidural steroid injection group is compared to the gabapentin group.

This outcome measure compares the average leg pain at baseline to the average leg pain 1 month after the start of treatment. This is measured using the Numeric Pain Scale. The Numeric Pain Scale ranges from 0-10. 0 being no pain at all and 10 being the worst imaginable pain possible. The epidural steroid injection group is compared to the gabapentin group.

This outcome measure compares the worst leg pain at baseline to the worst leg pain at 3 months after the start of treatment. This is measured using the Numeric Pain Scale. The Numeric Pain Scale ranges from 0-10. 0 being no pain at all and 10 being the worst imaginable pain possible. The epidural steroid injection group is compared to the gabapentin group.

This outcome measure compares the average back pain at baseline to the average back pain 1 month after the start of treatment. This is measured using the Numeric Pain Scale. The Numeric Pain Scale ranges from 0-10. 0 being no pain at all and 10 being the worst imaginable pain possible. The epidural steroid injection group is compared to the gabapentin group.

This outcome measure compares the average back pain at baseline to the average back pain at 3 months after the start of treatment. This is measured using the Numeric Pain Scale. The Numeric Pain Scale ranges from 0-10. 0 being no pain at all and 10 being the worst imaginable pain possible. The epidural steroid injection group is compared to the gabapentin group.

Functional capacity measured using Oswestry disability index. The range of possible scores for Oswestry Disability Index are 0-100. 0 is equated with no disability and 100 is the maximum disability possible.

Functional capacity measured using Oswestry disability index. The range of possible scores for Oswestry Disability Index are 0-100. 0 is equated with no disability and 100 is the maximum disability possible.

The participant is asked "Are you satisfied with the treatment you have received so far? Participants could respond yes or no. This is measured 3 months after the start of treatment.

The participant is asked "Are you satisfied with the treatment you have received so far? Participants could respond yes or no. This is measured 1 month after the start of treatment.

This outcome measure compares the worst back pain at baseline to the worst back pain at 1 months after the start of treatment. This is measured using the Numeric Pain Scale. The Numeric Pain Scale ranges from 0-10. 0 being no pain at all and 10 being the worst imaginable pain possible. The epidural steroid injection group is compared to the gabapentin group.

This outcome measure compares the worst back pain at baseline to the worst back pain at 3 months after the start of treatment. This is measured using the Numeric Pain Scale. The Numeric Pain Scale ranges from 0-10. 0 being no pain at all and 10 being the worst imaginable pain possible. The epidural steroid injection group is compared to the gabapentin group.

Inclusion Criteria: Lumbosacral radicular pain based on history and physical exam (e.g. pain radiating into one or both lower extremities, sensory loss, muscle weakness, positive straight leg raising test etc.) Numerical Rating Scale leg pain score > 4 (or if 3/10, greater or equal to back pain) MRI evidence of spinal pathology consistent with symptoms Exclusion Criteria: Untreated coagulopathy Previous spine surgery No MRI study Leg pain > 4 years duration Epidural steroid injection within past 3 years Cauda equina syndrome Previous failed trials with gabapentin or pregabalin Allergic reactions to gabapentin or pregabalin Referrals from surgery for diagnostic injections for surgical evaluation Serious medical or psychiatric that condition that might preclude optimal outcome or interfere with participation, such as the need for uninterrupted anticoagulation. Pregnancy

Cohen SP, Hanling S, Bicket MC, White RL, Veizi E, Kurihara C, Zhao Z, Hayek S, Guthmiller KB, Griffith SR, Gordin V, White MA, Vorobeychik Y, Pasquina PF. Epidural steroid injections compared with gabapentin for lumbosacral radicular pain: multicenter randomized double blind comparative efficacy study. BMJ. 2015 Apr 16;350:h1748. doi: 10.1136/bmj.h1748.