Efficacy of Steroids in TFESI

One of the most frequently cited studies comes from Riew and colleagues in 2000. A total of 55 patients with unilateral lumbar radicular pain and radiographically confirmed herniated nucleus pulposus (HNP) or spinal stenosis were enrolled. Every patient was considered a surgical candidate by the patient and the surgeon. These patients were randomized into a group with 1 to 4 TFIs with bupivacaine (n = 27) versus a group with 1 to 4 TFESIs with bupivacaine and betamethasone (n = 28). Both the subjects and the treating surgeon were blinded to the intervention. Follow-up occurred at 13 to 28 months postinjection, and 67% of patients in the control group went on to have surgery, whereas only 29% of the treatment group pursued surgical intervention. This difference was statistically significant ( P <.004). These results present strong evidence that in surgical candidates with radicular pain secondary to HNP or spinal stenosis, a TFESI is a surgery-sparring intervention compared with TFI of bupivacaine alone.

In 2006, Riew and colleagues followed up on all the patients who avoided surgery in the initial study. There were 29 patients who avoided surgery in the first 28 months post-injection, of which 9 were in the control group and 20 were in the treatment group. Eight patients in the treatment group were lost to follow-up. Of those who initially avoided surgery in 2000, 8 of 9 in the control group and 9 of 12 in the treatment group still had not had surgery at the 5-year follow-up. The difference between the groups was not statistically significant, which suggests that TFESIs do not make a significant difference in the long-term outcome when it comes to pursuing surgery, although all patients lost to follow-up were in the treatment group.

In 2001, Karppinen and colleagues used a single saline TFI as the placebo (n = 80) against a single TFESI with methylprednisolone and bupivacaine (n = 80). There were 160 patients who reported unilateral leg pain greater than back pain for 3 to 28 weeks. The patients were randomized and the intervention was double blinded. Outcome measures included back visual analog scale (VAS), leg VAS, Oswestry Low Back Disability Questionnaire, Nottingham Health Profile, cost, and physical examination maneuvers (straight leg raise, modified Schober measure, muscle strength testing, sensation testing, and reflexes). A single injection was given to each patient. Immediately after the injection, leg pain decreased by 61% in the treatment group and by 44% in the control group. At 2 weeks post-injection, there were statistically significant improvements in all measures (except the Schober measure) for both groups. At 3 months, the control group had a greater reduction in back pain than the treatment group. At 6 months, the control group had a greater reduction in both back and leg pain than the treatment group. At 1-year follow-up, both groups demonstrated statistically significant improvements compared with their status before injections, but there were no between-group differences. In addition, 23% of the treatment group and 19% of the control group went on to pursue a surgical intervention. The study concluded that the combination of methylprednisolone and bupivacaine in a TFESI seemed to have a short-term effect in both subjective and objective improvement of pain, but the effect did not last longer than 3 months post-injection.

Using the same 160 patients, Karppinen and colleagues published an in-depth subgroup analysis of the outcomes from the original study. They looked at contained HNP versus disc extrusions. In contained herniations, the treatment group (methylprednisolone and bupivacaine TFESI) had significantly improved leg pain at 2 and 4 weeks postinjection compared with the control group (saline TFI). The same trend was not seen in the disc extrusion subgroup. This study also concluded that patients receiving steroid treatment of a contained herniation were less likely to undergo back surgery and they also had significantly fewer days on sick leave at 3 to 6 months postinjection. There were also significant medical cost savings as early as 4 weeks postinjection and for as long as 1 year postinjection. In contrast, patients with disc extrusions who received the saline injection tended to have less cumulative medical costs after 3 and 12 months than those given the steroid injection, suggesting that a steroid injection seemed to be more harmful for disc extrusions, based on the hypothesis that macrophages are more prominent in disc extrusions and play a role in resorption that is disrupted by the introduction of steroids. Within the disc extrusion group, there were significantly more subjects who required surgical intervention after 10 weeks postinjection. The administration of steroids seems to be countereffective in the disc extrusion group.

Ng and colleagues studied 86 subjects with chronic unilateral radicular pain from HNP or foraminal stenosis that failed conservative treatment. The control group received a single TFI with bupivacaine only (n = 43), and the treatment group received a single TFESI with bupivacaine and methylprednisolone (n = 43). The outcomes were assessed at 6 and 12 weeks postinjection. The study found improvement in both groups for leg pain, Oswestry score, and walking distances, with no significant differences between the 2 groups. Only the duration of symptoms demonstrated a statistically significant predictive response. The shorter the duration of symptoms, the better the outcomes of the injection in both the control and treatment groups.

In 2010, Ghahreman and colleagues conducted a study on 150 patients who had pain radiating in the lower limb with a positive straight leg raise (SLR) and a disc herniation confirmed by computed tomography (CT) or magnetic resonance imaging (MRI). The subjects where then randomized into receiving a fluoroscopically-guided lumbar TFI of a steroid (triamcinolone), a local anesthetic (bupivacaine) or saline versus a placebo intramuscular injection of steroid or saline. If patients did not respond to the initial injection, they were allowed additional injections up to 3 total injections. The primary outcome measure was the proportion of patients who obtained complete relief or at least 50% relief of pain for at least 1 month after treatment. Results demonstrated that there was a significantly greater proportion of subjects treated with a TFESI (54%) who achieved relief of pain when compared with patients treated with TFI of local anesthetic (7%) or saline (19%) or intramuscular injections of steroids (21%) or saline (13%). Relief of pain was accompanied by significant improvements in function and disability. Over time, the relief of pain diminished in all groups equally.

Efficacy of TFESI Versus ILESI Versus Caudal ESI

Not all level 1 studies used a placebo as a control. In 2007, Ackerman and Ahmad compared the 3 different approaches to the epidural space (TFESI, ILESI, and caudal ESI). There were 90 subjects with L5-S1 disc herniations on imaging and severe S1 radicular pain confirmed as S1 radiculopathy by electromyography. The subjects were evenly randomized into each treatment group, and they all received the same dose of triamcinolone and saline. The same injection was given every 2 weeks for a total of 3 injections if adequate pain relief was not achieved. On average, subjects in the TFESI group received 1.5 injections, the ILESI group received 2.2 injections, and the caudal group received 2.5 injections. At 2 weeks postinjection, pain scores were reduced in all 3 groups; however, the TFESI group had a significantly greater reduction. Oswestry and Beck depression scores improved in all 3 groups without any between-group differences. There were also significantly more cases of complete pain relief in patients who had a ventral dispersion pattern of contrast flow, a flow pattern that occurred more consistently with the TFESI group. This study suggests that a transforaminal approach offers the benefit of increased analgesic efficacy, possibly because of an increased ventral spread of the steroid solution in the case of a herniated or extruded disc.

Efficacy of ILESI Versus Discectomy

In 2004, Buttermann looked at surgical outcomes versus conservative care in a patient population with a large HNP. A total of 100 patients who had no improvement after a minimum of 6 weeks of noninvasive care were followed for a 3-year period. They were randomly assigned to either receive an ILESI or have a discectomy. Those in the ILESI group were given repeated injections administered 1 week apart up to a maximum of 3 injections if needed. About 76% of the patients in the ILESI group received fluoroscopically-guided injections. Follow-up was performed at 1–3, 4–6, and 7–12 months and at 1–2 and 2–3 years after treatment. There was a crossover group of 27 patients who considered ILESI treatment a failure and subsequently went on to discectomy. The results demonstrated that the discectomy group had earlier motor recovery than the ESI group. Fewer patients in the discectomy group had a motor deficit at 1–3 months posttreatment ( P = .001). However, at the 2–3 year follow-up, there were no significant differences between the 2 groups in regard to motor weakness. Both groups demonstrated a significant decrease in both back and lower extremity pain VAS scores compared with pretreatment levels; however, the decrease in lower extremity pain scores in the discectomy group was significantly greater than in the ILESI group at the 1–3 month and 4–6 month follow-up intervals. Functional outcome improved significantly in both groups at all follow-up intervals. This study supported the use of ILESI in patients with continued severe symptoms after 6 weeks of noninvasive treatment because nearly half of these patients who received the ESI had a fairly rapid decrease in their symptoms without requiring surgery. The degree of improvement was similar to that for those who underwent a discectomy. Those who failed the ESI and went on to have the discectomy also demonstrated the same decline in pain scores and improvement in functional outcome scores despite the delay of surgery by at least 1 month. The delay to surgery did not seem to adversely affect the overall outcome for the patient.

Efficacy of Steroids in Non-guided ILESI

In 1997, Carette and colleagues examined nonguided ILESI as a means to treat sciatica caused by HNP. In this randomized, double-blind trial, there were 156 patients with intermittent pain in one or both legs, with a minimum duration of 4 weeks and a maximum duration of 1 year. They also had signs of nerve root irritation (positive SLR) or nerve root compression (motor, sensory, or reflex deficits). All patients had evidence of HNP by CT scan. Patients were randomized to receive either 2 mL of methylprednisolone mixed with 8 mL of saline or 1 mL of saline into the epidural space. All injections were non-guided, and repeat injections were done after 3 and 6 weeks if the patient did not report a marked improvement in their symptoms. At 3 weeks post-injection, the ILESI group demonstrated improvements in spine mobility and sensory deficits compared with the placebo group. At 6 weeks, there was more improvement in leg pain in the ILESI group. After 3 months, there were no significant differences between the groups. After 1 year, both groups had similar surgical rates (ILESI group, 25.8% and placebo group, 24.8%). Overall, non-guided ILESI demonstrated short-term improvements in radicular pain and spinal mobility compared with an interlaminar epidural injection of saline. There were no long-term differences seen between both the groups.

In 2003, Valat and colleagues also examined the efficacy of steroids in non-guided ILESI. Eighty-eight patients with unilateral sciatica lasting 15 to 180 days were randomized to receive 3 non-guided interlaminar epidural injections at 2-day intervals with either 2 mL of prednisolone (n = 39) or 2 mL of isotonic saline (n = 35). Self-assessment at day 20 was the outcome measure. On analysis, 12 of 35 (34%) in the control group and 22 of 39 (56%) in the steroid group had successful pain relief, but this result was not statistically significant between groups. The investigators concluded that ESI provided no additional improvement compared with isotonic saline when administered using a non-guided interlaminar approach. This outcome differs from that observed by Carette and colleagues ; however, a different volume of injectate (2 mL) was used in the 2 studies (2 mL Valat ; 10 mL Carette ).

Efficacy of Steroids in Caudal ESI

Although caudal ESIs have been around the longest, there are not many randomized controlled trials published recently on this topic. The most recent study was done by Sayegh and colleagues in 2009. They looked at 183 patients with severe chronic low back pain and sciatica. The patients were randomized into 2 groups. Both the groups were given a non-guided caudal epidural injection; group A was given Xylocaine and betamethasone (n = 93), and group B was given Xylocaine and water (n = 90). The investigators found that the steroid group experienced faster relief at the 1-week post-injection follow-up and had significant functional improvements on the Oswestry scores at all follow-up visits. SLR improved in both groups after the injection but improved earlier for the group given the betamethasone. This study concluded that a caudal epidural injection containing local anesthetic and steroids or water seems to be effective in treating patient with low back pain and sciatica; however, those receiving the steroid preparation demonstrated better and faster efficacy from the injection.