Transforaminal epidural steroid injection for lumbosacral radiculopathy: pregang

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

dc2md

Full Member
15+ Year Member
Joined
Aug 21, 2006
Messages
1,137
Reaction score
101
Here's another article I came across tonight:

OBJECTIVE: The present study was undertaken to evaluate the effectiveness of transforaminal epidural steroid injection (TFESI) with using a preganglionic approach for treating lumbar radiculopathy when the nerve root compression was located at the level of the supra-adjacent intervertebral disc. MATERIALS AND METHODS: The medical records of the patients who received conventional TFESI at our department from June 2003 to May 2004 were retrospectively reviewed. TFESI was performed in a total of 13 cases at the level of the exiting nerve root, in which the nerve root compression was at the level of the supra-adjacent intervertebral disc (the conventional TFESI group). Since June 2004, we have performed TFESI with using a preganglionic approach at the level of the supra-adjacent intervertebral disc (for example, at the neural foramen of L4-5 for the L5 nerve root) if the nerve root compression was at the level of the supra-adjacent intervertebral disc. Using the inclusion criteria described above, 20 of these patients were also consecutively enrolled in our study (the preganglionic TFESI group). The treatment outcome was assessed using a 5-point patient satisfaction scale and by using a VAS (visual assessment scale). A successful outcome required a patient satisfaction scale score of 3 (very good) or 4 (excellent), and a reduction on the VAS score of > 50% two weeks after performing TFESI. Logistic regression analysis was also performed. RESULTS: Of the 13 patients in the conventional TFESI group, nine showed satisfactory improvement two weeks after TFESI (69.2%). However, in the preganglionic TFESI group, 18 of the 20 patients (90%) showed satisfactory improvement. The difference between the two approaches in terms of TFESI effectiveness was of borderline significance (p = 0.056; odds ratio: 10.483). CONCLUSION: We conclude that preganglionic TFESI has the better therapeutic effect on radiculopathy caused by nerve root compression at the level of the supra-adjacent disc than does conventional TFESI, and the difference between the two treatments had borderline statistical significance.
---------------------------

Makes perfect sense to me. Don't know why I haven't thought of it earlier. If sure you all may have read this before (what good interventional pain doc doesn't get their monthly delivery of "Korean Journal of Radiology"). Oh, it's from 2006.

Anyway, does anyone else do it this way?? I mean, if you saw a presentation of L5 radic, but MRI findings of more central disc bulge at L4-5 level, would you do a TF at the L4-5 or L5-S1 level usually?? Probably best off hitting both levels eh? Of course spitting through the L5-S1 foramen will almost always get cephalad flow of steroid/volume to the adjacent level, but what if it isn't much (or what if most of the ceph flow is in the dorsal epidural space)?? Things that make you go hmmmmmm. Just thought I'd share.

Here's the link to the abstract: http://www.ncbi.nlm.nih.gov/pubmed/16799275?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA&linkpos=5&log$=relatedarticles&logdbfrom=pubmed

Looks like they continued the study or got more data and republished in "Radiology" Nov. 2007: http://www.ncbi.nlm.nih.gov/pubmed/17940309?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA&linkpos=1&log$=relatedarticles&logdbfrom=pubmed

Members don't see this ad.
 
Last edited:
It is poorly written from just what you posted.

The n is too low to get much info.
The conclusion is not supported by the data.
There are no objective outcomes.
Their reported outcomes were not statistically significant and there is no way of determining of that makes it clinically significant.

How did they determine where the pain was coming from? Imaging, exam, EMG? How did they limit the spread of the injectate from ascending or descending? The follow up study was worse. Do we assume the MRI's revealed a single disc HNP as far lateral recess to get the nerve root at that level? Because if not, then how do we know there is not an inflammatory leakage out of the L4-5 disc HNP irritating the L5, S1 roots?

RESULTS: Univariate analysis showed that the preganglionic group had a better treatment effect (99 of 112, 88.4%) than did the ganglionic group (90 of 127, 70.9%) at short-term follow-up (P = .001). Multiple logistic regression analysis showed that the only significant outcome predictor at short-term follow-up was injection level (odds ratio = 2.232, P = .037). No significant difference was identified regarding TFESI approach or cause of radiculopathy at midterm follow-up. CONCLUSION: TFESI for lumbosacral radiculopathy with a preganglionic approach is more effective than TFESI with a ganglionic approach at short-term follow-up.
I am not a statistician, but when I read this, it means- we screwed with the numbers long enough to come up with a p<.05

And the effects of picking the right level was no different after the 2 week mneasurement at follow up.

Again, this is not supported in their own paper. YOu breech the trust of the reader once in the paper and all credibility disappears.
I would not take this study with a grain of salt, I'd disregard it entirely.

You can go to the level where you suspect the pathology to be causing pain is residing, or you can go below that level because flow is more likely to go rostral than caudal.
 
you have some great points steve, but...

n=239 total patients (not too many other spine studies i've seen with more patients)

the conclusion is that the preganglionic approach is more effective than the ganglionic group in the short-term (73.2 vs 46.5% excellent pain relief, and 88.4 vs 70.9% excellent/good pain relief). and the p-value is 0.001.

BUT, at midterm followup (mean ~1yr), the pain relief was equal between the two.

what i can't figure out is did they only include patients with radiculopathy (per exam) and a supraadjacent paracentral herniation (and excluded pts with posterolateral herniation AT the level of nerve exit).

anyway, anatomically it makes sense to me to treat the level above with central/paracentral herniation. that's how we learned an L5-S1 paracentral disc bulge irritates the S1 (instead of L5) nerve.

tried to post the PDF file of the full article but it's too large. any way to decrease the file size??
 
Members don't see this ad :)
This seems like a similar technique the lateral recess block described by Algos.
It seems to me that the key is to see where your spread is and if you aren't satisfied, then use another approach. If the area is covered, it should work equally by either approach.
From an entirely unscientific point of view, my experience is that I tend to get less flow distally along the nerve with the lateral recess approach and I wonder if that might translate into a better outcome.

Why would they need 12 inch needles for this? Even on discos in the notoriously Rubenesque New Orleans population, I can't recall ever needing more than 10.
 
Why would they need 12 inch needles for this? Even on discos in the notoriously Rubenesque New Orleans population, I can't recall ever needing more than 10.

i know. i have no idea. maybe 12 inch needles so there's no need to attach tubing to inject the contrast and roids. kidding...but could be i guess. ???
 
Either adjacent or supra-adjacent should work.

The great majority of HNPs do not encroach on the root at the level of the lesion. They usually catch the next nerve passing through. So an L4/5 HNP usually catches L5, not L4.

If you inject at L4/5 for an L4/5 HNP you'll still get medication above the site where the HNP compresses L5. The contrast and medication typically track up the uncompressed L4 sleeve into the epidural space. If you go to the supra-adjacent level you are relying on the meds to run caudally. If you inject 3 ccs or more you'll see it run down the lateral recess.

It could be argued that if you have a far lateral foraminal HNP a supra-adjacent approach would be more likely to get medication above the compression. Or maybe not.

This assumes you don't subscribe to the newly-described "distal effects" theory of local anesthesia.
 
Either adjacent or supra-adjacent should work.

The great majority of HNPs do not encroach on the root at the level of the lesion. They usually catch the next nerve passing through. So an L4/5 HNP usually catches L5, not L4.

If you inject at L4/5 for an L4/5 HNP you'll still get medication above the site where the HNP compresses L5. The contrast and medication typically track up the uncompressed L4 sleeve into the epidural space. If you go to the supra-adjacent level you are relying on the meds to run caudally. If you inject 3 ccs or more you'll see it run down the lateral recess.

It could be argued that if you have a far lateral foraminal HNP a supra-adjacent approach would be more likely to get medication above the compression. Or maybe not.

This assumes you don't subscribe to the newly-described "distal effects" theory of local anesthesia.

the L4/5 HNP would impinge the L5 nerve usually right? we all agree on that. so then wouldn't you want to inject at that level of impingement (L4/5) instead of the level of L5 neuroforaminal exit (L5/S1). all those fun prostaglandins and other chemicals would be at the level of impingement (mostly) wouldn't they??

are you saying your thinking is that why not just inject at the level of nerve IVF exit and know that it'll get that level for sure, and the level of the actual HNP above. instead of injecting just at the level of the impingement and likely NOT getting caudal flow to the level of IVF exit?? sounds good to me. that's of course if you're just doing one level. why not just do the L5/S1 AND L4/L5 neuroforamen for a likely L5 radic??
 
You need to know it. There were several questions about it on the boards last year.

The thread "SNRB vs TFESI" has a very detailed discussion of the theory including comments from its original proponent.
 
You need to know it. There were several questions about it on the boards last year.

The thread "SNRB vs TFESI" has a very detailed discussion of the theory including comments from its original proponent.

yeah, i remember reading that now. sounds like complete bulls*** to me. OBVIOUSLY there's no mysterious efferent reflex loop distal to the lesion. and the effectivenss of the nerve block for surgery is b/c they (the surgeons) are working distal to the block. and for radic, it's obviously eventually leaking at least to the IVF at that level.

i hope you're joking about that crappy theory being a board question. :laugh:
 
It was right after the question about Sluijter's theory that pulsed RF works by an immunological mechanism.
 
This assumes you don't subscribe to the newly-described "distal effects" theory of local anesthesia.


whoever came up with that theory must be a huge jackass.........


but i remain stubbornly (and stupidly) unconvinced. hear me out here:

say you touch a spinal nerve with a needle during a TFESI. you get is shock down the leg. then, anethetize that nerve. THEN, touch the nerve proximal to where it is anesthetized. do you you still feel that shock down your leg?
 
whoever came up with that theory must be a huge jackass.........


but i remain stubbornly (and stupidly) unconvinced. hear me out here:

say you touch a spinal nerve with a needle during a TFESI. you get is shock down the leg. then, anethetize that nerve. THEN, touch the nerve proximal to where it is anesthetized. do you you still feel that shock down your leg?

not that i've done this, but yes! of course you would. now of course shooting some local during a TFESI would anesthetize a bunch of the nerve both proximally and distally to where you inject (as you can see with the spread of dye). the only thing the local is doing is blocking the Na-channels at the nodes of Ranvier (of A-delta fibers) so that the saltatory nerve transmissions are blocked at that point in the nerve. poking the nerve distal to that point will illicit a stimulation centrally along the afferent fibers, but this conduction up the nerve will be stopped at that point of temporary node blockage. BUT, if you stimulate/irritate the nerve more proximally (closer to the A-delta and C-fiber entrance into the tract of Lissauer), the pain transmission will continue contralateral up the lateralthalamic tract to the VPL nucleus of the thalamus and finally to the primary sensory cortex represented by the homonculus (sp?)...and yes, they'll then get the pain "down" the leg or whatever nerve distribution the irritated nerve came from.

according to the theory you're talking about, we could treat pronator teres syndrome by anesthetizing the median nerve in the carpal tunnel. that would be nice though.
 
not that i've done this, but yes! of course you would. now of course shooting some local during a TFESI would anesthetize a bunch of the nerve both proximally and distally to where you inject (as you can see with the spread of dye). the only thing the local is doing is blocking the Na-channels at the nodes of Ranvier (of A-delta fibers) so that the saltatory nerve transmissions are blocked at that point in the nerve. poking the nerve distal to that point will illicit a stimulation centrally along the afferent fibers, but this conduction up the nerve will be stopped at that point of temporary node blockage. BUT, if you stimulate/irritate the nerve more proximally (closer to the A-delta and C-fiber entrance into the tract of Lissauer), the pain transmission will continue contralateral up the lateralthalamic tract to the VPL nucleus of the thalamus and finally to the primary sensory cortex represented by the homonculus (sp?)...and yes, they'll then get the pain "down" the leg or whatever nerve distribution the irritated nerve came from.

according to the theory you're talking about, we could treat pronator teres syndrome by anesthetizing the median nerve in the carpal tunnel. that would be nice though.

hey Guyton, easy on the physiology lesson. despite what my previous posts might suggest, i do know my medicine. and gorback's examples of my foolishness were much more elegant.

again, my difficulty with this concept is the ISIS nomenclature and my previous instruction on the subject. if you are blocking a spinal nerve, and a spinal nerve alone, then there has got to be a reason why you'd feel pain relief. if you actually trying to get the LA to the foramen or disc-nerve interface, thats a whole different story.
 
sorry ssdoc. that basic sciences was more for my benefit than yours. i slightly refreshed my pathway knowledge today and typing it out just helps it stick more. plus, some (including myself at 10am today) have forgotten most of that by now.

i think (and no, i haven't tried to perform a SNB yet) what gorback and others are saying is that unless you're way the hell away from the foramen (probably a few inches or more), you're gonna get some flow of the local along the nerve and into the foramen. it just liking to travel along those nerves (at the contrast does).

what is it, btw, that makes the contrast travel along the nerve?? is there an adipose/nerve interface that has a potential space that we end up filling up with local/steroids/contrast?? it seems like it would just pool in one big blob...yet somehow, it doesn't. i'll have to pay more attention when we dissect the cadaver again this year. what does ISIS or Bogduk say about that??
 
Top