Are corticosteroids necessary in epidural injections?

[Epidural]

Ng L, Chaudhary N, Sell P. The Efficacy of Corticosteroids in Periradicular Infiltration for Chronic Radicular Pain. Spine 2005;30(8):857–862.

This was a randomized, double-blind, controlled study published recently. One group received bupivacaine and methylprednisolone while the other group received just bupivacaine. While the authors noted an improvement for both groups of patients, they concluded that the corticosteroids did not produce a treatment effect or benefit for patients’ with chronic radicular pain.

---

Members don't see this ad.

 

[f_w]

I wonder how well people would have done with saline (or just the needle).

In the back pain crowd an almost religious devotion to a chosen therapy seems to be a good part of the mechanism.

 

[drrinoo]

it is interesting....but it is one of the very few articles on percutaneous spine interventions that actually bothers to reference a pre-eminent neurobiologist....Dr. Marshall Devor....

The sad reality is that most injectionists are religiously devoted to cadaver anatomy...but largely ignore the literature published by pre-eminent researchers in the field of pain processing neurobiology and for that matter, most spinal interventionalists ignore the vast literature on the pharmacology of drugs commonly utilized in regional anesthesia...the latter comment is for another soap box on another day....

Two other excellent articles in Spine recently.....(long stretch on my part)...but elucidate some of the issues regarding the effect of HNPs on DRGs..both suggest that nuclear material can leak out free glutamate ( a known sensitizer of the DRG)...hence, you could conceivably have radicular pain in the absence of true inflammation or compression....but rather due to autonomous mechanisms, that persist after inflammation has subsided....

1. Score: 100 Article Type: Report Format
Herniated Lumbar Disc Material as a Source of Free Glutamate Available to Affect Pain Signals Through the Dorsal Root Ganglion.
Spine. 25(8):929-936, April 15, 2000.
Harrington, J. Frederick MD *; Messier, Arthur A. PhD +; Bereiter, David PhD *; Barnes, Bryan MD ++; Epstein, Mel H. MD * Abstract
HTML
PDF (2.47 M)

2. Score: 99 Article Type: Miscellaneous Article Format
Physiological and Behavioral Evidence for Focal Nociception Induced by Epidural Glutamate Infusion in Rats.
Spine. 30(6):606-612, March 15, 2005.
Harrington, J Frederick MD; Messier, Arthur A. PhD; Hoffman, Laurie BS; Yu, Elizabeth BS; Dykhuizen, Matthew BS; Barker, Kelly BS


in this case a steroid would only function as a membrane stabilizer that reduces C-fiber activity temporarily at best rather than as an anti-inflammatory agent....hence you may get as much effect....furthermore look at some of the conundrums with perineural injections of steroids

Reg Anesth. 1997 Jan-Feb;22(1):59-65. Related Articles, Links


Effect of local methylprednisolone on pain in a nerve injury model. A pilot study.

Johansson A, Bennett GJ.

Department of Anesthesiology, Lund University at Malmo General Hospital, Sweden.

BACKGROUND AND OBJECTIVES: Local injections of corticosteroids are frequently used in the treatment of regional pain. The rationale for this is not very clear, since an inflammatory cause of pain is rarely evident. There are few data on the effect of corticosteroids on nociception in experimental animals. However, corticosteroids have been found to suppress ectopic discharges from experimental neuromas and to have a short-lasting suppressive effect on transmission in normal C-fibers. In this study the influence of a locally applied depot form of a corticosteroid on neuropathic pain was investigated in a rat model. METHODS: Peripheral mononeuropathy was induced with a chronic constriction injury to the left sciatic nerve. This procedure has previously been shown to produce various signs of disturbed sensibility, including heat hyperalgesia, mechanical allodynia, and mechanical hyperalgesia, indicating that a neuropathic pain-like condition has developed. The occurrence of neuropathic pain in these animals was confirmed with behavioral testing after 9 days. The site of injury was then reexposed and treated locally with either a depot form of a corticosteroid (methylprednisolone) or saline. The animals were then tested for another 11 days. RESULTS: The heat hyperalgesia and mechano-allodynia but not the mechano-hyperalgesia were depressed in the animals receiving the corticosteroid but not in those treated with saline. The effect remained during the whole 11-day test period. CONCLUSIONS: It is hypothesized that the corticosteroid acts by suppression of ectopic neural discharges from the injured nerve fibers.

PMID: 9010948 [PubMed - indexed for MEDLINE]

19: Kingery WS, Castellote JM, Maze M. Related Articles, Links
Methylprednisolone prevents the development of autotomy and neuropathic edema in rats, but has no effect on nociceptive thresholds.
Pain. 1999 Apr;80(3):555-66.
PMID: 10342417 [PubMed - indexed for MEDLINE]


personally, in the near future, we should begin abandoning anatomic constructs of pain...and begin focusing on pain mechanisms (ala Woolf)....C Woolf's ideas regarding pain will minimize the role of anatomy in spinal pain (that we hold so dearly today).

so for those of you who love the mantra...that TFESIs are better than ILESIs because of a more 'targeted' delivery of steroids....buyer beware!

Rinoo Shah

 

[f_w]

so for those of you who love the mantra...that TFESIs are better than ILESIs because of a more 'targeted' delivery of steroids....buyer beware!

I strive to practice medicine in a scientific way, if like in the study the OP quoted there is indeed no evidence to support the steroid, I would be glad to drop it.

As for the 'targeted' delivery: If you mess up during a myelogram and end up with an epidural injection, guess where your contrast goes: all the way up to the clivus. I have no reason to believe that the drugs during an ESI behave much different.

 

[Epidural]

I think you meant to say that if you mess up an epidural injection and wind up with an intrathecal injection, there could be serious consequences. I think the risks of a dural puncture are less with TFESIs when compared to ILESIs. Additionally, the dosage of steroids typically used for TFESIs is the dosage that's frequently used for intrathecal injections in the treatment of postherpetic neuralgia. While you certainly do want to avoid this complication, I think intravascular injections are of greater concern.

I wouldn't say there's no evidence for the role of steroids. I simply posted the article to discuss it. This data is contrary to Riew's findings published in 2000. In this study, the experimental group was injected with betamethasone and bupivacaine while the control group was injected with just bupivacaine. The results showed that a significantly greater number of patients in the steroid + LA (local anesthetic) group were able to avoid a surgery. They concluded that selective nerve-root injections of corticosteroids are significantly more effective than those of bupivacaine alone in obviating the need for a decompression.

The issues I had with the study are:

1. I'm not sure that there was no difference between the demographics of each group. The study didn't state that there was no difference between the two groups. There were twice as many retired patients in the steroid + LA group versus just LA group. The LA group also seemed to have more patients that were students/housewives/others.

2. I think it would've been nice to see if there were differences past 12 weeks. While there may not have been a difference between the two groups at 12 weeks, there might've been at 6 months or 1 year.

3. This was only a single-injection study. Studies have shown that it can take more than one injection to get sustained relief. While there's no difference between the two groups after one injection, I wonder if there would've been a difference with more than one.

4. The study talks about how periradicular infiltration has been widely used as a diagnostic tool. What I wonder is if this was a SNRB (selective nerve root block) rather than a TFESI. In theory, a TFESI should get more spread to the anterior epidural space. I think it would've been nice if they included some images.

Nonetheless, I thought the article provided good references and gave me something to think about. I think further studies to elucidate the role of steroids in ESIs would be helpful.

Riew et al. Can nerve root injections obviate the need for operative treatment of lumbar radicular pain? A prospective, randomized, controlled, double-blinded study. Journal Bone Joint Surgery 2000; 82A(11):1589-1593.

 

[f_w]

I think it will be hard to get a truly randomized controlled trial of ESI vs surgery through the IRB at any US institution. As a result, your groups will allways be biased into patients who are willing to have sx vs the ones who will rather have glowing coals shoved down their throat than going for surgery. The other issue confounding many studies on ESI is the fact that often the groups are heterogeneous containing patients with LBP only who wouldn't be surgical candidates (at least for ethical neurosurgeons).

Thanks for posting that study. My only criticism is that there was no true 'control' group (a sham procedure, IM injection or the like). There have been other studies before that compared IM injection and ESI, and the results weren't that convincing either.


I think you meant to say that if you mess up an epidural injection and wind up with an intrathecal injection, there could be serious consequences.

No, I was talking about the fact that the epidural space is one large bottle, a bit larger at the bottom but extending all the way to the clivus anteriorly. In a diagnostic myelogram, you WANT to put your contrast intrathecal, but s%& happens and once in a while somebody will dump 10ccs of iodinated contrast into the epidural space. And there are CT's done hours later demonstrating the contrast all the way up to the neck. In the living human, the epidural space is large and interconnected, unless there are post-surgical changes, whatever you inject anywhere will diffuse throughout, albeit with a fairly steep gradient. That is a good thing, not a bad one.

 

[greywater]

AMEN F_W! I posted a similar opinion to yours in the ISIS forum recently.

I have to laugh at the so-called "spinal diagnosticians" in my area who do "selective nerve blocks" thinking they can keep the medication limited to only one nerve. In order to be effective it has to reach the epidural space, and from there it can spread. Hell, you can see the contrast tracking up and down before you put in the steroid.

I wonder how many TFESIs would be done if ILs paid the same or better?

 

[f_w]

Well, I believe that there will actually be a gradient, with more steroid at the injection site, and less at other locations. So far, I just haven't found any serious research on the distribution dynamics of steroids vs bupivacaine vs omnipaque (there's got to someone who injected a couple of pigs and slaughtered them 2,12,24,48,196 hours post injection). Maybe I just haven't looked hard enough.

 

[drrinoo]

f_w...

10 cc of contrast reaching the clivus....is by definition a subdural injection.....although theoretically the subdural compartment connects with the cranium...the epidural space in the spine is not contiguous with the epidural space in the cranium and it in the spine it ends at the foramen magnum (technically the base of the clivus)

40 cc can be routinely injected in the epidural over the period of an hour safely during epiduroscopy...but I would not go over that number


typically in the lumbar spine....1cc of contrast will rise 1.8 vertebral body levels

if anyone gets such a dye spread....in an office setting.....do not inject local anesthetic under any circumstance...and if you do...do not let the patient leave for at least 1/2 to 1 hour...even in the absence of sedation

 

[drrinoo]

grey water...it is good to see that you, like me, have been resurrected from the POL grave

lets make George Romero proud!

 

[f_w]

the epidural space in the spine is not contiguous with the epidural space in the cranium and it in the spine it ends at the foramen magnum (technically the base of the clivus)

I was talking about the base of the clivus at the foramen magnum, this is typically the last image you obtain on a head-CT.


typically in the lumbar spine....1cc of contrast will rise 1.8 vertebral body levels


The prized question is: how many vertebral body levels will Depo or Bupivacaine raise over the course of 24-72 hours ? (I can see how far the contrast agent goes while I am injecting, but frankly I don't care)

 

[drrinoo]

The prized question is: how many vertebral body levels will Depo or Bupivacaine raise over the course of 24-72 hours ? (I can see how far the contrast agent goes while I am injecting, but frankly I don't care)[/QUOTE]

This question is a complex one.....the degree of epidural spread following a single shot epidural injection is not really relevant 24-72 hours following the injection....the octanol/partition coefficient and the PKas of a solution are more relevant with a single shot epidural...i.e., pharmacokinetics are more relevant 24-72 hours out. the other factors are the differential spread and blockade of symapthetic vs. somatic sensory vs. motor fibers

Bolus doses during patient controlled epidural analgesia...typically have localizaed segmental spread, but they do not linger for 24-72 hours...rather they linger for 15-20 minutes


only in the setting of continuous infusions is the steady state degree of epidural spread relevant...even then the volume infused over an hour correlates with the degree of segmental spread...but it reaches a steady state, i.e. 6cc/hour in the thoracic area may cover 6-8 dermatomes; slow intermittent 5cc boluses...up to 40 cc injected in the lumbar epidural space can reach the low cervical high thoracic levels and may be good for hypotensive epidural anesthesia and providing analgesia for the shoulders---only done by a few anesthesiologists, e.g. for hip replacements


there is a lot of literature on this in the regional anesthesia literature with respect to rib fractures/epidurals for general surgery/ob anesthesia.

as I said, 10 cc of contrast reaching the 'clivus', by definition is a subdural injection (assuming you did not get a subarachnoid injection)

 

[greywater]

Returning to the original post: A few years ago there was a Japanese paper that said you got just as good relief with ESI with marcaine as with steroid for spinal stenosis. I tried it, but ended up having to re-do almost all of the marcaine pts with steroid.