| PARAPHRASED and ANNOTATED ISIS GUIDELINES The ISIS Guidelines (available from www.spinalinjection.com) represent the state of the art in defining the most scientific approach to spinal injections. The guidelines should be purchased and viewed personally, primarily because of the expansive discussions, reproducible post-injection patient assessment forms, and the spectacularly clear fluoroscopic images and diagrams representing years of work from the best minds in the field. This annotated, abridged, paraphrased version presented here is not a substitute for the book ISIS Practice Guidelines (hereafter called the "Guidelines") and does not at all replace the Guidelines. The book of the Guidelines give detailed fluoroscopic images with artist drawings of the neural structures, the approaches, and the targets. Viewing these images are essential for understanding the execution of the Guidelines. The Guidelines however are not "standards of care", and deviation from the Guidelines does not necessarily constitute a breach of patient care or substandard care. The Guidelines serve as a benchmark for delivery of quality spinal care but are not exclusionary to other approaches. If a practitioner follows the guidelines, they will have an excellent chance of achieving optimal outcomes. The comments included are mine alone and do not reflect any consultation with or prior debate with the learned authors of the Guidelines. Such comments are designed not to detract from the masterpiece of the guidelines themselves, but serve as an expansion, amplification, and at times alternative opinion regarding the guidelines implementation. |
I. Sedation "Sedation is not indicated for any of the procedures described in these
Guidelines. Notwithstanding practices and instructions to which practitioners in the USA may
have been accustomed, elsewhere in the world these procedures have been conducted, and
continue to be conducted, without sedation. There are no features of any of the procedures
covered by these Guidelines that warrant preemptive or routine sedation." (Guidelines page xx)
The preferences and routine patient care in the US differs substantially from these guidelines and not without good
reason. Many patients consider the use of radio frequency, discography, or protracted spinal procedures to be
barbaric and tantamount to torture without sedation. Those with chronic pain states often require sedation. A
recent poll of ISIS instructors found the vast majority use sedation and many use heavy sedation with propofol.
Therefore the US practices will continue to differ from the practices of those outside the US.
Guidelines. Notwithstanding practices and instructions to which practitioners in the USA may
have been accustomed, elsewhere in the world these procedures have been conducted, and
continue to be conducted, without sedation. There are no features of any of the procedures
covered by these Guidelines that warrant preemptive or routine sedation." (Guidelines page xx)
The preferences and routine patient care in the US differs substantially from these guidelines and not without good
reason. Many patients consider the use of radio frequency, discography, or protracted spinal procedures to be
barbaric and tantamount to torture without sedation. Those with chronic pain states often require sedation. A
recent poll of ISIS instructors found the vast majority use sedation and many use heavy sedation with propofol.
Therefore the US practices will continue to differ from the practices of those outside the US.
II. Intravenous Lines- a. IV access is necessary for the pre-injection administration of
medications, if there is a significant co-morbidity of another illness that may require
intervention during the procedure, or in all procedures near the epidural space where there
may be a risk of local anesthetic reaching the CSF or a blood vessel l
III. General contra-indications for procedures:
Absolute: Unwilling or physically/mentally incapable of consent
Contrast medium cannot be used (and is absolutely needed for the procedure) because of known
anaphylactic reactions
There is an active or untreated localized infection around the injection site or in the field.
The patient has a known bleeding disorder
Inability to cooperate during the procedure
Relative: Allergy to any planned administered drugs
Pregnancy
Anticoagulants are being used or were not stopped in time
Anatomical aberrations that may affect the safety or success of the procedure
Systemic infection
Respiratory or cardiovascular disease that would affect the patient safety
Immune suppression.
IV Required Fluoroscopy: Fluoroscopy is required for all Guidelines procedures. Hard copy
capability should be available. Live fluoroscopic imaging is recommended. Pulse ox and ECG are
recommended.
V. Lumbar Spinal Nerve Blocks: Fluoro setup: AP view with cephalocaudad beam
angulation to a point that the superior end plate of the vertebral body is non-overlapping
(horizontalize the end plate). If a SAP in overlapping the target, then tilt the beam more
lateral oblique until the space of interest inferior to the pedicle is clear. The beam will
ultimately be tilted oblique approximately 25-40 degrees. There are two possible target
points, both in the mid-pedicular line. The dorsal approach places the needle in the posterior
1/4 of the neuroforamen approximately 2-3 mm inferior to the pedicle approximately half
way across the neuroforamen on lateral fluoro view. Comment: Figure 6 on page 11 is incorrectly
labeled...should be rostral approach instead of dorsal approach. Real time fluoroscopic injection of
contrast is performed until contrast is seen to enter the lateral recess but not medially.
There should be no vascular uptake. Repositioning of the needle is used if there is vascular
uptake or if there is seen only egress out the neuroforamen without contrast ingress. Local
anesthetic is subsequently injected at the same volume used to outline the spinal nerve.
Comment: the pathology present with disc prolapse creating a radiculopathy is due to TNF-alpha leaking onto the
nerve at the level of the disc, not at the level of the neuroforamen. Therefore, injection of contrast distal to this level
may not produce any pain relief unless the pathology is due to tracking of TNF alpha to the DRG or edema of the
DRG due to vascular supply compromise. . This technique therefore relies on continued diffusion after injection to
and above the level of the disc TNF alpha effect in order to achieve relief or assumes the pathology produced by the
disc lies primarily at the level of the DRG.
Post procedure care involves monitoring the patient's pain relief by an observer who is not
the physician. The Guidelines call for medical personnel reassessment every 15-30min for 2
hours after the procedure. A positive block is 100% relief of the radicular pain. Anything
less does not qualify as a positive block. Comment: This aspect of monitoring the patients every 15
minutes prior to discharge is clearly impractical in a busy pain practice when there may be 10 such blocks
performed within 2 hours.
VI. S1 Transforaminal Injection The fluoro beam is positioned cephalo-caudad and
slightly ipsilateral oblique. The needle is walked off the sacrum into the lateral aspect of the
S1 neuroforamen with frequent AP and lateral checks. The target is the dorsal aspect of the
anterior sacral plate immediately opposite the S1 lateral neuroforaminal border. The needle
is advanced slightly medially and superiorly within the canal. The remainder of the block is
the same as for the lumbar block.
VII Lumbar Disc Stimulation (Provocative Discography).
Precepts: A. Provocation of the target disc reproduces the patient's pain, but provided that
provocation of adjacent disc does not reproduce pain. B. Internal disruption criteria are: 1.
The patient has discogenic pain and 2. the affected disc exhibits a grade 3 or great fissure,
upon CT discography. C. Concordant pain is the only type of pain that is diagnostic.
D. Manometry is used to determine the type of disc from a pressurization pain reproduction
standpoint. Patients with concordant pain at<15psi over opening pressure have a positive
disc. Those with concordant pain reproduced at 15-50psi are indeterminate, and those
without pain until above 50psi are negative. E. The degree of pain produced during the
manometric injection on a scale of 0 to 10 must be at least 7/10.
Given the above information, ISIS has developed interpretive criteria that categorize positive
discography into one of 3 categories:
Unequivocal Discogenic Pain- stimulation of the target disc reproduces concordant pain at
a level of 7/10 or greater at a pressure of less than 15psi above opening pressures and
stimulation of two adjacent discs produces no pain at all
Definite Discogenic Pain- stimulation of the target disc reproduces concordant pain at a
level of 7/10 or greater and
a. pain is reproduced <15psi above opening pressure + stimulation of one adjacent disc
produces no pain or
b. pain is reproduced at a pressure of less than 50psi above opening pressure and
stimulation of two adjacent discs does not reproduce pain at all
Probable Discogenic Pain- stimulation of target disc reproduces concordant pain at least
7/10 at a pressure of <50psi above opening pressure and stimulation of one adjacent disc
reproduces no pain at all while another adjacent disc produces pain at greater than 50psi but
not concordant
Indeterminate Disc- Does not meet all the above criteria yet still produces pain
There is an elaborate point system (page 30 of the Guidelines) which has been developed
empirically by ISIS (ISIS Discography Score) that gives further guidance on categorization of
the discs. This score may ultimately become widely used in discography.
ISIS guidelines also include evaluation of patients for other possible sources of pain prior to
discography. ISIS Guidelines repeatedly discuss the false positive rate of 75% in patients
with somatisation disorders yet does not make any recommendation regarding
pre-discography psychological testing. Comment: When there is any question at all about the patients
psychiatric status with respect to somatisation, this author suggests either an MMPI-2 or a BHI-2 evaluation be
performed preoperatively. If somatization is seen in these tests, then it is prudent to consider abandoning the
discography as part of the diagnostic algorithm.
ISIS Discography Technique Summary: Prophylaxis against disciitis with cephazolin 1g,
clindamycin 900mg, or ciprofloxacin 400mg IV 15-60min before commencing the procedure
Comment: New standards for surgical device implantation are 30-60 min in advance- given the extremely slow
circulation into the disc from the peripheral blood, it may be wise to administer the antibiotic 45-60min in advance.
If patient has a known allergy to contrast medium, they should be pre-treated with H1 and
H2 blockers and corticosteroids [prednisone 50mg PO 24 and 12 hours before the procedure]
prior to the procedure. Comment: Since the contrast may be found in the disc 6-24 hours later and
continually leach out through the end plates, the prophylaxis above may be insufficient to prevent late contrast
reactions. Instead, my preference is to use undiluted gadolinium penta-acetate in cases where prior severe contrast
reactions (eg. anaphylaxis, hypotension or airway compromise during non-ionic contrast administration, etc). Should
the prophylaxis above be given, it is important to warn diabetics and hypertensive patients of the alterations in blood
sugar and blood pressure produced by steroid administration. Positioning is prone or oblique. AP fluoro
view with cephalo-caudad angulation of the beam to pass parallel to the inferior end plate.
Beam is then rotated laterally until the SAP lies opposite the axial midline of the target disc
except at L5-S1 where the rotation must be less due to the overlying iliac crest. The target
(using the down the beam technique) is the midpoint of the disc just lateral to the lateral
border of the SAP. Comment: The safe zone is much wider at the inferior portion of the disc where there will
be less chance of nerve penetration. This is especially important at L5S1 where the obligatory needle trajectory may
be less oblique than desired due to the overlying iliac crest. Therefore, in cases where safety of the ventral ramus
nerve may be in question, using a target in the inferior aspect of the disc is appropriate. It also appears the
Guidelines are designed for primarily straight needle trajectories. The set up angles used are not optimal for curved
needle techniques where a less oblique beam rotation is needed. Local anesthetic infiltration into the skin
and muscles (up to 5ml) is used to provide adequate anesthesia for deep needle placement.
Sedation may be administered IV but the patient must be responsive during disc
pressurization, therefore careful titration of sedative and opioid medication is essential.
Comment: Withholding the patients usual oral opiate or NSAID medications may cause a significant exacerbation of
pain during the needle placement and during pressurization. On the other hand, excessive IV opiates can diminish
the patient's pain response. Excessive benzodiazepine can cause the patient to become over sedated, sluggishly
responsive, or incoherent. My preferred technique is to have the patient take their usual medications as scheduled
prior to the procedure and titrate propofol (has no analgesic properties on the nerves or disc) for needle placement
only. Then with the propofol redistributed (requires approximately 5 min after the last dose), the patient is
repeatedly questioned about orientation to date, place, and person until the patient is conversing coherently. Only at
that time is pressurization performed. This technique has the advantages of avoiding over sedation and opiates
during the pressurization but permits the patient to be comfortable and amnestic during needle placement. Propofol
is administered in boluses or as a continuous infusion. A single or two needle technique may be used.
The needle is advanced to the SAP then slowly across the intervertebral foramen, towards the
surface of the target disc. Care is needed to avoid impaling the ventral ramus and
reintroduction is needed if the ventral ramus is encountered. Comment: Furcal nerves, which are
preganglionic nerves traversing the neuroforamina from superior to inferior, are present in 15% of patients. These
nerves can be impaled by the needle tip without being anywhere near the ventral ramus. When present, they are
found 93% of the time at the L4-5 disc level. Also, if the patient has an inflammatory mass on the surface of the
annulus, contact with the needle may cause severe pain. Neither of these are addressed by the Guidelines, but
discographers should be made aware of their presence. The needle is advanced to the center of the disc.
Comment: more correctly, the needle should be advanced to the center of the nucleus pulposis. The center of the
nucleus pulposis does approximate the center of the disc except at L5S1 where the center of the nucleus pulposis lies
at the junction of the posterior third and anterior 2/3 of the disc on lateral projection. The L5S1 disc has an
especially thick anterior annulus fibrosis. Once the needle is correctly positioned in the center of the
disc, its stylette is removed and the needle is connected to a system that will inject non-ionic
contrast with antibiotic. Comment: If there is any chance the dura was penetrated during the needle placement,
injection of an cefazolin containing solution could be hazardous. Contrast is often seen tracking along the course of
the discography needle and if cephazolin tracks subarachnoid, it can cause seizures (delayed) or death. In such
cases where there is a question regarding any dural penetration, it is wise to inject the antibiotic at the end of the
procedure rather than mixed with the contrast. A line-pressure transducer with readout and/or graph
may be used. If this is not available, a 3ml syringe with Luer-lock may be used. Comment: Here
the Guidelines inexplicably abandon the use of pressure manometric analysis which is used as part of their final
determination of a positive disc. It is patently unclear how this incongruity can be explained in the context of the
incorporation of manometric data into disc analysis. However, due to the cost of the manometric system ($70-120 per
use in the USA), the Guidelines may be waffeling. It should be noted that most discographers can generate a
sustained pressure of over 160psi with a 3ml syringe as opposed to less than 80 psi with a 20ml syringe.
As the disc is pressurized with contrast aliquots of 0.5-1ml, the pressure is noted along with
the patient's response. The appearance of the target disc must be monitored continuously,
using a lateral view. Injection should continue until a. pain is produced, b. contrast medium
escapes from the disc, c. the volume of injection reaches 3ml, or d. firm resistance to
injection is encountered. If pain is produced, the pressure and volume of injection should be
documented, and the patient should report the concordancy with the accustomed pain and
the current intensity of pain (scale 1-10). The injection may be repeated a short period of
time later in order to reaffirm the response. Upon injection, contrast medium should outline
the nucleus pulposis of the disc and any features of its internal structure. Once a disc has
been tested for pain reproduction, a local anesthetic agent can be injected into the disc,
either or both as a measure to relieve any pain produced by the previous injection, or as an
attempt to relieve the patient's accustomed pain temporarily. Comment: it is uncommon in my
experience for the patient's pain to be relieved by a small amount of local anesthetic in an already pressured disc.
The One should consider a separate anesthetic injection into the disc at a later date if anesthetic discography is
being used as a confirmatory test, however this may only relieve pain in cases of significant annular fissure
presence. The Guidelines contain excellent assessment sheets for disc stimulation and contain a
scoring system sheet (pp 43 and 44).
VIII. Lumbar Medial Branch Blocks
Lumbar MBB have diagnostic utility in that once positive, they identify the source of pain and
serve as a predictor of a good response to percutaneous radio frequency neurotomy. Because
of a high rate of false positive responses at least two separate blocks should be performed on
different occasions with the first block containing an anesthetic to demonstrate prima facie
the joint is symptomatic. Three blocks are preferred by ISIS. As to what constitutes the
criteria for a positive anesthetic block, the Guidelines do not specifically say except to discuss
the consequences of accepting less than 100% relief from the diagnostic anesthetic block.
ISIS does state however that a true positive response to comparative local anesthetic blocks
"...is one in which the patient reports complete relief of pain for a shorter duration when a
short-acting agent is used. Such a response is referred to as "concordant" in that it is
concordant with the expected action of the agents used. If a patient obtains complete relief
of pain but does not report an appropriate differential response in terms of duration of relief,
the joint is not excluded as the source of pain. Such paradoxical responses may be consistent
with the hypothesis being tested." Comment: Whereas it has been demonstrated medial branch block false
positive rates with a single anesthetic injection range from 29-40% and dual injections (one with saline) reduce this
to 12%, there are financial considerations in the US that cannot be ignored. If a patient must be charged a hospital
facility fee or ASC facility fee plus physician fees for each injection, the diagnosis of zygapophyseal syndrome via
performance of these injections could well exceed $6,000. Therefore performance of these injections in a physician's
office can significantly reduce the patient charges and out of pocket expenses. Also, some insurers will not pay for
the second or third injection used for diagnostic purposes thereby shifting the entire out of pocket cost to the
patients. This is unacceptable to the patients and should be unacceptable to the physicians performing these
injections, and therefore an alternative billing model or movement of the procedure to the fluoroscopically equipped
physician's office is appropriate. It is also controversial whether zygapophyseal syndrome exists in a large
percentage of the population as an isolated pain generator, given that discogenic disease with disc space narrowing
progressively transfers more weight to the z-joints thereby potentially damaging the joints. The adoption of a
standard of 100% pain relief from a medial branch block as evidence of a positive block, effectively eliminates those
with concomitant pain conditions from receiving the benefits from radio frequency neurotomy.
Patient selection criteria are documented in the Guidelines with an excellent discussion.
Technical Performance of the Block: Patient prone or oblique, fluoroscopic beam rotated
oblique to view the "scotty dog". Needle target L1-4 medial branch: superior portion of the
eye of the "scotty dog" at the lateral base of the SAP. The L5 dorsal ramus target is the
junction of the ala of the sacrum and the inferior-lateral aspect of the SAP. Once bone is
contacted, the bevel is angled inferiorly to avoid foraminal injection, then 0.1-0.3ml of
contrast may be injected in real time fluoroscopy to demonstrate vascular uptake. If there is
no vascular uptake, 0.5ml local anesthetic is injected. One must then inject the more
cephalad adjacent medial branch in order to anesthetize the descending branch. Comment: it is
unclear why a much smaller volume of contrast is injected (which usually outlines the medial branch), then is
followed by up to five times the volume of local anesthetic. It is possible this amount of local anesthetic could
anesthetize the neural contents of the intervertebral foramen, therefore it may be prudent to limit the volume of local
anesthetic used to the same amount of contrast injected since at least this small amount is demonstrated in real time
to remain outside the intervertebral foramen.
The Guidelines make the point the most important part of the block is in the assessment of
pain reduction, not in the technical performance. The Guidelines use an independent
assessor to evaluate pain for several hours after a block, preferably with a VAS scale of which
the patient has not seen their previous answers during the repeated assessment. Although
the guidelines do not select any specific thresh-hold for pain reduction, 50% of the usual pain
is mentioned in the assessment section. Some physicians require 100% pain reduction.
IX. Sacroiliac Joint Blocks- Premises: 1. The innervation of the joint is diffuse and
controversial therefore it is unlikely individual nerve blocks could be used to diagnose the SI
joint as a pain generator 2. The volume of the joint is approximately 1.5ml 3. Fluoroscopy
is absolutely necessary for this block 4. Physical exam is insufficient to make a diagnosis 5.
Control blocks are necessary-either a placebo control block or a sequence of blocks ruling out
other targets (eg. facet joints, discography, etc).
SI Injection Technique: AP view will reveal a medial distal SI joint shadow (the posterior part
of the joint) and a separate lateral SI shadow (the anterior joint line). Contra lateral oblique
fluoroscopic beam rotation 5-20 degrees is used to image the inferior aspect of the posterior
joint line. The stopping point of beam rotation is when the inferior aspect of the posterior
joint line is maximally "crisp" rather than attempting to overlap the anterior and posterior
joint lines. A cephalad tilt of 20-25 degrees of the fluoroscopy beam may then be used to
help identify the posterior from the anterior joint lines....the posterior-inferior aspect of the
joint is displaced in a caudal direction compared with the anterior joint. The target for
injection is the inferior joint just above the inferior border of the joint.
Alternative approaches in case of failure: a. direct the needle to the inferior location of
overlap of the two joint lines b. slide off the inferior aspect of the joint into the inferior
capsular recess c. cannulate the mid portion of the joint using 20-30 degrees oblique
rotation of the beam d. superior capsular recess using a cephalocaudad approach
After needle placement, 0.3-0.5ml contrast is injected. If there is a venous drainage,
impossible injection due to placement in ligaments, or aberrant pattern seen, re-attempts at
needle placement up to 3 times may be used. Joint arthrography does not correlate well with
painful joints. Injection of local anesthetic is performed until a firm resistive endpoint is
appreciated, extra-capsular contrast is seen, or a total volume of 2.5ml is injected. Steroid
use in the joint is not part of the Guidelines. The Guidelines comment several times on how it
may be impossible to enter the joint no matter the approach. Comments: The concept the synovial
part of the joint (the only part addressed by the Guidelines) is responsible for pain production in the joint is not well
established in the literature. There may be just as much validity to the ligamentous dorsal portion of the joint
producing pain through neural ingrowth as happens with an injured ACL joint. However, the use of these Guidelines
will effectively rule out the synovial joint as a pain generator. Conditions in which the synovial aspect of the joint
are seen to be abnormal on MRI include rheumatoid arthritis, ankylosing spondyliitis, pyogenic arthritis, Reiters
syndrome, and psoriatic arthritis. Following trauma, it is rare to see positive synovial joint MRI abnormalities,
therefore one must consider ligamentous sources of pain after trauma.
X. Cervical Discography-The false positive rate is very high and the procedure is
technically demanding. In order to have a positive discogram, 5 criteria must be met: 1.
technically correct needle placement is a requirement, 2. Z-joint pain generators must have
been previously eliminated (without this, there is a 67% false positive rate), 3. Concordant
pain is produced during pressurization of the target disc, 4. Pain level produced is 7/10, 5.
Adjacent disc pressurization does not reproduce the patient's pain. The protocol states 4
levels should be tested. The C2-3 level is not necessary (as a 5th level) unless there is a
significant headache symptomatology. The everyday pain patterns produced by cervical discs
are diffuse but roughly correlate to the pain maps constructed for cervical z-joints. Mercifully
the authors of the guidelines recognize sedation may be necessary for this extremely painful
procedure but make the pain the patient must be responsive during pressurization.
Technical conduct of the procedure: Supine position, right sided approach (to avoid the
esophagus), fluoro beam cephalocaudad angulation parallel to the end plates (usually the
beam source is about 10-20 degrees caudad angulation), then right oblique rotation 20-30
degrees. Great vessels are retracted laterally by hooking the fingers of one hand into the
groove between the trachea and the carotid sheath. Following local anesthetic infiltration, a
23-26 ga spinal needle is inserted through the retracted groove to the uncinate process using
the down the beam approach while maintaining retraction. Any sudden onset of coughing,
respiratory distress, sudden electrical shock sensation, or edema at the injection site require
the needle be withdrawn. The needle is then advanced to the vertebral end plate (superior)
and the retraction released and constant contact is maintained by the directing hand with the
needle. Subsequently the needle is walked off the end plate through the annulus and
immediately a lateral fluoro view is obtained. The needle is adjusted to reach the center of
the disc without crossing the mid line in either AP or lateral views. Frequent AP and lateral
views may be required during adjustment. The Guidelines state a small curve at the tip of
the needle will help with placement. During pressurization, the patient should be asked
about the onset of pain, severity of pain, concordancy of pain, and location of pain. Injection
of contrast is with a 3cc syringe and 0.1ml increments are injected under continuous
fluoroscopic visualization. The injection is terminated when the pain is 7/10 or greater,
contrast escapes from the disc, end plates are visibly displaced, there becomes firm resistance
to injection, the patient exhibits non-concordant pain, or there are any unexpected or
distressing sensations. The total volume of the disc is 0.5ml. Fluoro images pre and post
injection should be saved and CT post injection acquired. The Guidelines do state manometry
may be used. Discs are graded as concordant, asymptomatic, or dissimilar. Comments: It is very
difficult to inject 0.1ml increments with a 3ml syringe, therefore a luer lock 1cc syringe is preferred. Because the
opening pressures are very high (often well over 120-150 psi) manometry is not useful. On morphological analysis
of the discogram either in fluoroscopy or CT, one must remember the uncinate clefts are often mistaken for annular
tears by the unknowing. Uncinate clefts are an anatomical probability in discs, especially in aged discs. The disc
material in the cervical spine is much more firm than the lumbar making anatomical "fracture planes" much more
likely in the disc itself (the clefts). Antibiotics should be used in the injected contrast. If post injection CT scans are
obtained, the injectionist should view these personally as many radiologists have never seen a cervical discogram
and haven't a clue as to what constitutes normal vs abnormal findings. Also, 1mm slices should be taken directly
through the disc of interest and parallel to the end plates. Oblique slices through the disc are useless. There is no
need to CT the vertebral bodies nor discs not producing concordant pain. Such a specialized request may require
consultation with the radiologist to assure the CT is carried out in a technically correct manner.
XI. Cervical Medial Branch Blocks
Principles- If pain is not relieved in the area targeted, the target nerves are not mediating
the patient's pain. There may be other medial branches mediating the pain or the pain
generator may arise from a non-median branch mediated source. If pain is relieved, then
this constitutes evidence of median branch transmission of pain if there are no false positives.
Partial relief of pain may be possible if only part of the median branches transmitting the pain
are actually blocked and other sources are excluded. Cervical median branch blocks are
diagnostic and predict a good chance of obtaining complete pain relief from RF.
Which z-joints should be the targets? The following pain map created from volunteers during
stimulated pain in the joints should serve as a guide.
Statistically, the most likely joint to be involved in the cervical
spine is the C2-3 joint, followed by the C5-6 joint. The
guidelines recommend against medial branch blocks in cases
where there is widespread pain or hypersensitivity to touch
as it is unlikely medial branch blocks will be of any value.
Contraindications: In addition to the absolute and relative
contraindications listed at the beginning of the guidelines,
neurological signs are a relative contraindication for cervical
medial branch blocks. However if the neurological disorder
is be managed, then such blocks may be indicated. Radicular
pain is not a contraindication to medial branch blocks.
Agents: The most common agents used are bupivicaine 0.5% and lidocaine 2%, but other
concentrations of these agents may be used. The volume of the block should not exceed
0.3ml.
Technique: The most convenient and least demanding approach is the lateral approach with
the patient lying in the lateral position. The fluoro beam is aimed in a true lateral view with
complete superimposition of facet joint lines. The target for C3-6 is the centroid of the
articular pillar. For C7 the target lies high on the apex of the superior articular process. The
target for the third occipital nerve (necessary for blocking the C2-3 joint) is at three separate
points on a line bisecting the centroid of C3 and the center of the C2 lateral arch parallel to
the posterior vertebral line. The three points are at the level of the C2-3 disc, just above on
the arch of C2 and just below on the SAP of C3. The patient should be warned about the
probability of ataxia being quite high after the TON block and driving should be avoided.
The guidelines suggest errors in performance of the blocks and in interpretation can be
reduced by using double blind dual blocks, making certain the patient has sufficient pain
before the block to justify its performance (at least 50% of their pain at its worst and VAS at
least 4/10 prior to the block), and in performance of the least number of blocks necessary per
patient. An exhaustive discussion is undertaken in the guidelines to demonstrate the
complexities of both selection of nerves to block and in interpretation of the results of the
blocks.
XII. Lateral Atlanto-Axial Joint Blocks (LAAJB)
The guidelines sagely make the point inexperienced injectionists should absolutely not
attempt this block due to the extreme risk attached to even slight deviation of the needle tip.
There are few studies demonstrating efficacy of these blocks. Specifically Aprill and
separately Busch and Wilson reported relief of headache fowing performance of these blocks.
The guidelines suggest a triple block or a dual anesthetic block. The suggestion of using control
blocks is only meaningful if further interventions were to be entertained. The Guidelines do not contain a
description of a radiofrequency neurotomy of the nerve supplying the C1-2 joint, therefore it is this authors
opiniontriple blocks or dual anesthetic blocks of C1-2 are merely an academic exercise without any practical
application.
The primary indication for LAAJB is the need to know if the patient's pain stems from such
block. The literature suggests indications are headache and suboccipital pain, focal
tenderness in the suboccipital region, focal tenderness over the tip of the left or right
transverse process of C1, restricted rotation of C1 on C2 on manual exam, and aggravation of
the usual headache by passive rotation of the C1 vertebrae to the left or right.
Technique: 25 ga 90mm needle, pulse oximetry, prone patient position with the head and
face supported comfortably with a pad. Fluoroscopic evaluation is undertaken of the joint in
AP and the posterior arch of the atlas identified. If the arch overlies the joint, the C-arm
should be angled caudally or the patient should flex the head slightly. If the teeth obscure
the view when the mouth is opened, the patient may need to rotate their face slightly . This
rotation can be quite dangerous as the room for error is mere millimeters. This author suggests unless you are
extremely experienced, no rotation of the face or head laterally off the midline should be undertaken.
The target is the lateral third of the lower border of the atlas or the lateral quarter of the
axis. More medial placement (eg. the midline) may transect the spinal C2 nerve or the DRG.
Slow methodical advancement of the needle perpendicularly to the initial target point is
undertaken. If the tip strays lateral to the joint margin, it may pith the vertebral artery. If
the tip strays to the center of the joint, it may strike the C2 spinal nerve or ganglion. More
medial placement risks cord puncture. If the tip passes through the joint, it may impale the
carotid artery. Frequent fluoroscopic eval is required during advancement. After initial
osseous contact is confirmed, the needle is adjusted inferior from the atlas or superiorly from
the axis and advanced into the lateral 1/3 point of the joint. Once passage into the joint has
occurred, further advances of the needle are made in the lateral fluoroscopic view.
Advancement is then made 1/4 to 1/2 way across the joint on lateral fluoroscopic projection.
If blood is aspirated, the needle is advanced slightly anterior. Contrast is injected to reveal a
slit like filling of the joint on AP view and a partial filling of the joint on lateral view. The
joint slopes inferiorly on lateral fluoro view and therefore one may not outline the joint well.
There are anterior and posterior meniscoids that may fill and also filling may occur medially
through the joint into the spinal canal or even across the synovial connections to the
contralateral side. Once the volume of filling is determined the same volume of local
anesthetic is injected with a minimum of 0.5ml and a maximum of 1.0ml.
As the pathology of joints producing pain is often cartilage fragmentation or synovial inflammation, the use of
steroids is frequently used throughout the body as a therapeutic measure. The Guidelines do not mention steroid
injections into the C1-2 joints but this author routinely uses them in combination with 0.5% bupivicaine for a total
volume injection of approximately 3/4 ml. The overdistension of the joint with contrast followed by injection of a
steroid/local anesthetic combination can result in extreme pain after such injection, therefore this author uses
minimal contrast, only enough to outline the joint. In this author's mind, there is no point to perform a diagnostic
block if there is no potential therapy to be offered.
XIII. Lumbar Transforaminal Injections of Corticosteroids
The most compelling evidence for TFESI is a series of controlled trials outlined in the
expansive references of the Guidelines. The first by Riew demonstrated 29% of patients
treated with steroids vs. 67% of patients receiving bupivicaine alone ultimately required
surgery. Another study by Karppinen found that TFESI were more effective than saline at 2
and 4 weeks post injection. Finally, Vad compared TFESI with paraspinal injections of saline.
Of these 84% of the steroid group vs 48% of the saline group reported significant
improvement in pain after 12 months. Of course, all these studies deal with HNP with
radicular pain.
The main selection criteria for the injection is radicular pain who have not responded to
non-surgical interventions or for whom non-surgical interventions are not indicated, and
whose pain is likely to have an inflammatory basis.
The Guidelines suggest use of 25 or 23 ga needles. If a double needle technique is utilized,
both a larger diameter needle 17-20ga is employed in conjunction with a small diamter
central injection needle. An IV is necessary. Contrast is used as part of the Guidelines along
with a steroid, Betamethasone suspension 6-18mg, Triamcinolone 20-80mg, or
dexamethasone. It is curious methylprednisolone is not included in the list of steroids as it is one of the most
commonly employed steroids in use for TFESI. Consent should include dural puncture, infection,
allergic reaction, spinal cord injury, or arachnoiditis.
There are two basic approaches described by the Guidelines: the subpedicular approach with
placement of the needle tip on the anterior vertebral body superior to the exiting nerve root
(the most commonly taught approach) and the so called "retroneural approach" which
requires greater skills at placement.
The subpedicular approach uses the following fluoro setup: AP tangential to the inferior
endplates. The target is on the vertebral body at the 6 o'clock position of the pedicle and just
inferior to the inferior pedicular border. There is ostensibly a "safe triangle " that widens as
one goes laterally under the pedicle. If the foramen is not distorted such as in foraminal
stenosis or loss of disc height, then the triangle formed between the nerve root, the
infrapedicular bisecting line, and the lateral pedicular line constitutes the "safe triangle" that
contains only the sinuvertebral nerve. If the 6 o'clock position is obscured by the projection
of the superior articular process, then rotation of the beam slightly obliquely is employed in
concert with readjustment of the target to the new 6 o'clock position. Note hypertrophy of the
superior articular process may demand a significant adjustment of the needle tip more obliquely. Also note
foraminal height narrowing may substantially decrease the margin of error for proper needle placement. Also, if the
patient has superior nerve root displacement at the midpedicular line due to a prolapse of the disc at the foraminal
level, the root may be displaced directly into the path of the subpedicular needle approach.
For the single needle approach, the skin entry is slightly inferio-lateral to the target. The
needle is advanced with frequent fluoroscopic evaluations of needle trajectory, needle
withdrawal slightly, then re-advancement until ultimately bone is contacted. If the patient
suffers the onset of aggravated radicular pain, the operator should carefully consider the
wisdom of terminating the procedure. If the patient suffers only a temporary burst of
symptoms and the symptoms abate upon withdrawal of the needle, the procedure can be
resumed. Once the patient experiences a radicular paresthesia, ostensibly there has been a pithing of the nerve
with some obligatory (although may be subclinical or not result in weakness due to the more cephalad lying DRG)
neuronal damage. It is this authors opinion that due to the hazards of this approach with potential nerve injury that
no sedation at all be used for this procedure, or conversion to a procedure that is less known to cause paresthesia
may be prudent.
The two needle technique places the introducer needle to the lateral border of the superior
articular process and advanced to the dorsal half of the foramen. The slightly curved
injection needle is subsequently advanced through the introducer needle and the passed
under AP view fluoroscopy towards the target point.
Once the target point is reached, 0.2-0.5 ml contrast is injected to assure there is no
intrathecal or radicular artery injection. In either case, the procedure is abandoned and the
patient is brought back at another time for a repeat attempt. If on the other hand there is an
intravenous injection, the needle tip is repositioned slightly and a repeat injection of contrast
is performed. Optimally, contrast should spread into the foramen and into the lateral
epidural space. Once the safe triangle and nerve root sleeve is outlined by a given volume of
contrast, the same volume of therapeutic agent is subsequently injected. It is important to
note the area of the nerve considered to be inflamed must be covered with steroid depending
on the needed volumes determined via the neurogram. Specifically this means if the superior disc
adjacent to the level injected contains the inflammatory disc prolapse, then the neurogram must reach at least to this
level. This is commonly seen in paracentral disc herniations. On the other hand if the disc herniation is foraminal
at the level of the exiting nerve root, then only the lateral recess and infrapedicular area need to be injected. For far
lateral disc herniations, a different approach will be needed to adequately address the extraforaminal nature of this
inflammatory disc effect. It should be also noted there was no mention of "real time" neurography or even better,
digital subtraction neurography. One or the other of these approaches is critically important in order to avoid the
intra-artial injections thought to be associated with permanent paralysis from steroid injections into the artery.
The Guidelines state the steroid is the active agent in therapeutic TFESI but most operators
also use a local anesthetic, ostensibly in order to render the injection less painful. The
Guidelines correctly note the volume of local anesthetic is injected first, not to exceed half
the intended overall volume of injectate. This author is aware of several cases of unrecognized spinal
cord infarction, unrecognized nerve injury, or fractures from falls in the past due to the excessive injection of local
anesthesia volume or concentration before the publication of these Guidelines. If the Guidelines are followed, there
should be only minimal chance of these unrecognized or produced injuries.
The retroneural approach target is the dorsal neuroforamen directly over the nerve root. The
anatomic location is at the posterior 1/3 of the neuroforamen and the superior 1/3 of the
neuroforamen. Fluoroscopic setup is initially AP with the beam passing parallel to the
superior endplate. The AP target is the mid-pedicular line. If the lamina overlaps the plane,
then the beam is rotated slightly lateral oblique. If the superior articular process obscures
the target area, then rotate the fluoroscope beam cephalad until the SAP recedes caudad.
Initially he lateral laminar border is contacted as a depth finding measurement. The needle
is advanced off the lateral borer of the lamina with frequent fluoro guidance to the target
point above. No attempt should be made to elicit a paresthesia. The remainder of the
injection is as per the subpedicular description.
The transforaminal injection of S1 is undertaken for S1 radicular pain. The AP view of the
sacrum often fails to permit visualization of the S1 posterior neuroforamen, therefore a
slightly cephalad beam trajectory with a slightly lateral oblique approach makes the anterior
and posterior foraminal borders line up. The needle is placed on the lateral lip of the bone of
the sacrum around the neuroforamen and walked off into the neuroforamen approximately 2
millimeters. Then lateral fluoroscopy will reveal the depth of needle insertion into the
neuroforamen. The needle is advanced to the anterior sacral border, lateral to the anterior
sacral neuroforamen, passing 5mm from the floor of the sacral canal, but short of the
posterior border of the anterior border of the sacrum. Care is to be taken that the needle
does not pass through the anterior sacral foramen. The remainder of the injection Guidelines
are used as per the subpedicular injection. Note the pedicle of S1 always lies in a line inferior to the L5
pedicle.
XIV Radiofrequency Lumbar Medial Branch Neurotomy (RLMBN)
The only indication for RLMBN is complete relief from selective medial branch nerve
blocks.The medial branches of the posterior primary rami course from the dorsal ramus at the
neuroforamina running inferiorly to the SAP-transverse process of the vertebral body below
the neuroforamina, then posteriorly in the groove between the SAP-transverse process under
the mamillary-accessory ligament, then splits into a ipsilateral z-joint branch, multifidus
branch, and descending inferior branch. The latter courses inferiorly to the next z-joint.
Therefore the numbering of the appropriate innervating branch becomes important. The L4-5
Z-joint is innervated by the L3 and L4 medial branches which lie on the L4 and L5 transverse
processes. The Guidelines note the L5S1 Z-joint is innervated by the medial branches of L4
and L5. It should be noted the L5 medial branch does not arise until the inferior aspect of
the L5S1 Z-joint. Before this point, it is simply known as the dorsal ramus of L5. The
Guidelines go over the physics and the optimum time frame of lesioning (60-90 seconds) with
the needle tip being placed very close to and parallel to the nerve.
Technique: The target point is the junction of the SAP-TP with the tip at the middle to
anterior 1/3 line of the groove (on lateral view). The "pillar view" is approximately 15-20
degrees oblique and varying degrees of declination of the beam. The typical declination angle
is 30-40 degrees inferior to superior relative to the superior end plate angulation. This angle
maximizes the contact between the median branch nerve and the needle tip. AP, lateral, and
oblique fluoroscopic images are checked prior to lesioning (check the Guidelines extensive
fluoroscopic views to demonstrate appropriate placement). After checking for proper
placement with fluoroscopy, a small amount of local anesthetic (0.5ml) is applied through the
needle and subsequent lesioning is performed. The more cephalad level will also need
lesioning. The Temperature is measured during lesioning, but no pre-lesioning stimulation is
used. In the US, it is always prudent to perform at least motor stimulation for reasons of medicolegal
documentation.
XV. Intradiscal Electrothermal Therapy - Currently this therapy has no CPT code
in the US, is considered experimental by most insurance carriers, and lacks reimburement by
most insurers. Because of the above issues and only modest clinical improvements with
such therapy, it will not be discussed in this document at this time.
XVI. Cervical Transforaminal Injection of Corticosteroids
"Spinal cord infarction, due to intra-arterial injection of corticosteroids, is more than a
nominal risk of cervical transforaminal injection of corticosteroids. The severity of this
complication weighs heavily on the benefit-risk ratio of this therapy". Author's note: one should
burn the warning here into your mind before pursuing this therapy. The disasters are rare, but are very real, and
have happened to experienced physicians. This doesn't mean one shouldn't lean the technique or perform the
procedure, but rather that extreme caution should be exercised.
Indications: patients with cervical radicular pain or radiculopathy in which the presumptive
cause is an inflammatory process affecting the cervical spinal nerve or root.
Medications used for injection: water soluble non-ionic contrast suitable for iv and intrathecal
use (Isovue M or Omnipaque 180, 240, or 300), preservative free local anesthetics,
corticosteroid (betamethasone, triamcinolone, dexamethasone). Note this author believes there is
little to be gained and a significant degree of complication that may occur with the routine injection of local
anesthesia during this block. Certainly the presence of local anesthetic may delay the diagnosis of spinal cord
infarction and may result in a significant delay in diagnosis or a spinal artery injection.
Technique: Patient position is supine, lateral, or oblique depending on operator preference
and patient comfort and physician preference. This author strongly prefers the supine position with the
head oriented straight AP. Minute deviations from the angles and targets chosen can lead to significantly
malpositioned needles. Small movements of the patient after initial trajectory calculation must be prevented and the
supine position is easier to define the neutral or starting fluoroscopic views since any head rotation is obvious.
The fluoroscope is adjusted to view an oblique projection of the cervical spine to provide a full
view of the target foramen along its length. Often injectionists will initially position the fluoroscope
beam to 45 degrees since the neuroforamina are widest at this angulation from C3-4 to C6-7. This places the carotid
and vertebral artery directly in the line of fire at some vertebrae. Transvascular needle trajectories should be
avoided, especially in the elderly in whom plaques or calcific deposits may be dislodged into the brain. See the
fluoroscopic views on this website demonstrating cross-sectional anatomy of the cervical spine and various angles of
needle trajectory.
The needle puncture point overlies the posterior half of the selected foramen targeting the
posterior wall. The 25ga needle is brought through the skin and directed towards the
anterior half of the SAP advancing down the beam in small increments with frequent
adjustments of the needle, until the tip touches the SAP. The needle is advanced a few mm
off the posterior wall and directed to the middle third of the neuroforamen. Once the lateral
margin of the neuroforamen has been reached, an AP fluoroscopic view is used. The needle
is advanced to the mid-pedicular line but never beyond the lateral border of the uncinate
processes. If radicular pain occurs on needle advancement, no injection should be made
without repositioning of the needle and only if the radicular paresthesia resolves prior to any
injections. Once final positioning is achieved, spot films in the oblique and AP views are
made. The oblique view should demonstrate the needle position in the posterior aspect of the
foramen while the AP view documents the needle depth within the foramen. This author believes
passage of the needle into the middle 1/3 of the neuroforamen is seriously jeopardizing the spinal nerve and
therefore the tip should be maintained along the posterior foraminal wall periosteum during advancement.
A short length of low volume tubing is recommended in order to change syringes without
needle movement. Under direct , real-time fluoroscopy, a small volume of non-ionic contrast
material (1.0ml or less) is injected. The solution should disperse in the intervertebral
foramen outlining the spinal nerve and the DRG. It is essential to visually monitor the
fluoroscopic injection of contrast to look for arterial injection (manifested by a flash then
clearance) In either case, the needle should be withdrawn and no further injections
attempted. The procedure should be rescheduled until after a period of time to permit
healing of the vessel Injection of contrast should be continued until epidural spread is
achieved with outlining of the target nerve dural sleeve. Intrathecal injection will result in
rapid dilution of the contrast material. If during injection of contrast, the radicular pain is
tolerable, the injection may be continued. Local anesthetics must be used judiciously and
carefully due to potential for unintended subarachnoid injection. Use of a short acting
anesthetic is recommended. Lidocaine is injected in volumes of 0.5-1.5ml @1% strength or
1-3ml at 0.5% strength. Corticosteroid use is predicated on whether the procedure is
monoradicular or polyradicular. With monoradicular pathology, betamethasone 3-6mg,
tramcinolone 20-40mg, or dexamethasone 7.5-10mg are appropriate. For multiradicular
problems, the doses are doubled. This author believes the use of local anesthetics injected transforaminally
serve to detract the physician from diagnosis of a neurological event. They do not help facilitate the procedure nor
do they add any diagnostic utility to the procedure (unless they are used for a selective nerve root block, then must
be given after the steroid. This author also believes the volumes cited above are excessive given the size of the
spinal canal and the care being taken to use small volumes of contrast makes the local anesthesia volumes of up to
3cc inconsistent. The cord itself can be compressed by 3 volume aliquots. Finally, the use of multiradicular
treatment with a single shot transforaminal approach is inconsistent with the selectivity of the transforaminal
approach. This author recommends multiple TFESI needle placements or the use of an interlaminar ESI under
fluoroscopy.
XVII. Percutaneous Radiofrequency Cervical Media Branch Neurotomy
(CMBRF)
Anatomy: The structures innervated by the cervical medial branches include the cervical
zygapophysial joints, paramedia and deep cervical muscles, multifidus, semispinalis cervicis,
and interspinales. The C5 medial branch courses across the centroid of the articular pillar of
C5. The C4 and C6 medial branches are between the centroid and the base of the SAP while
the C3 and C7 branches lie on the SAP. Click the pics below to enlarge.
medications, if there is a significant co-morbidity of another illness that may require
intervention during the procedure, or in all procedures near the epidural space where there
may be a risk of local anesthetic reaching the CSF or a blood vessel l
III. General contra-indications for procedures:
Absolute: Unwilling or physically/mentally incapable of consent
Contrast medium cannot be used (and is absolutely needed for the procedure) because of known
anaphylactic reactions
There is an active or untreated localized infection around the injection site or in the field.
The patient has a known bleeding disorder
Inability to cooperate during the procedure
Relative: Allergy to any planned administered drugs
Pregnancy
Anticoagulants are being used or were not stopped in time
Anatomical aberrations that may affect the safety or success of the procedure
Systemic infection
Respiratory or cardiovascular disease that would affect the patient safety
Immune suppression.
IV Required Fluoroscopy: Fluoroscopy is required for all Guidelines procedures. Hard copy
capability should be available. Live fluoroscopic imaging is recommended. Pulse ox and ECG are
recommended.
V. Lumbar Spinal Nerve Blocks: Fluoro setup: AP view with cephalocaudad beam
angulation to a point that the superior end plate of the vertebral body is non-overlapping
(horizontalize the end plate). If a SAP in overlapping the target, then tilt the beam more
lateral oblique until the space of interest inferior to the pedicle is clear. The beam will
ultimately be tilted oblique approximately 25-40 degrees. There are two possible target
points, both in the mid-pedicular line. The dorsal approach places the needle in the posterior
1/4 of the neuroforamen approximately 2-3 mm inferior to the pedicle approximately half
way across the neuroforamen on lateral fluoro view. Comment: Figure 6 on page 11 is incorrectly
labeled...should be rostral approach instead of dorsal approach. Real time fluoroscopic injection of
contrast is performed until contrast is seen to enter the lateral recess but not medially.
There should be no vascular uptake. Repositioning of the needle is used if there is vascular
uptake or if there is seen only egress out the neuroforamen without contrast ingress. Local
anesthetic is subsequently injected at the same volume used to outline the spinal nerve.
Comment: the pathology present with disc prolapse creating a radiculopathy is due to TNF-alpha leaking onto the
nerve at the level of the disc, not at the level of the neuroforamen. Therefore, injection of contrast distal to this level
may not produce any pain relief unless the pathology is due to tracking of TNF alpha to the DRG or edema of the
DRG due to vascular supply compromise. . This technique therefore relies on continued diffusion after injection to
and above the level of the disc TNF alpha effect in order to achieve relief or assumes the pathology produced by the
disc lies primarily at the level of the DRG.
Post procedure care involves monitoring the patient's pain relief by an observer who is not
the physician. The Guidelines call for medical personnel reassessment every 15-30min for 2
hours after the procedure. A positive block is 100% relief of the radicular pain. Anything
less does not qualify as a positive block. Comment: This aspect of monitoring the patients every 15
minutes prior to discharge is clearly impractical in a busy pain practice when there may be 10 such blocks
performed within 2 hours.
VI. S1 Transforaminal Injection The fluoro beam is positioned cephalo-caudad and
slightly ipsilateral oblique. The needle is walked off the sacrum into the lateral aspect of the
S1 neuroforamen with frequent AP and lateral checks. The target is the dorsal aspect of the
anterior sacral plate immediately opposite the S1 lateral neuroforaminal border. The needle
is advanced slightly medially and superiorly within the canal. The remainder of the block is
the same as for the lumbar block.
VII Lumbar Disc Stimulation (Provocative Discography).
Precepts: A. Provocation of the target disc reproduces the patient's pain, but provided that
provocation of adjacent disc does not reproduce pain. B. Internal disruption criteria are: 1.
The patient has discogenic pain and 2. the affected disc exhibits a grade 3 or great fissure,
upon CT discography. C. Concordant pain is the only type of pain that is diagnostic.
D. Manometry is used to determine the type of disc from a pressurization pain reproduction
standpoint. Patients with concordant pain at<15psi over opening pressure have a positive
disc. Those with concordant pain reproduced at 15-50psi are indeterminate, and those
without pain until above 50psi are negative. E. The degree of pain produced during the
manometric injection on a scale of 0 to 10 must be at least 7/10.
Given the above information, ISIS has developed interpretive criteria that categorize positive
discography into one of 3 categories:
Unequivocal Discogenic Pain- stimulation of the target disc reproduces concordant pain at
a level of 7/10 or greater at a pressure of less than 15psi above opening pressures and
stimulation of two adjacent discs produces no pain at all
Definite Discogenic Pain- stimulation of the target disc reproduces concordant pain at a
level of 7/10 or greater and
a. pain is reproduced <15psi above opening pressure + stimulation of one adjacent disc
produces no pain or
b. pain is reproduced at a pressure of less than 50psi above opening pressure and
stimulation of two adjacent discs does not reproduce pain at all
Probable Discogenic Pain- stimulation of target disc reproduces concordant pain at least
7/10 at a pressure of <50psi above opening pressure and stimulation of one adjacent disc
reproduces no pain at all while another adjacent disc produces pain at greater than 50psi but
not concordant
Indeterminate Disc- Does not meet all the above criteria yet still produces pain
There is an elaborate point system (page 30 of the Guidelines) which has been developed
empirically by ISIS (ISIS Discography Score) that gives further guidance on categorization of
the discs. This score may ultimately become widely used in discography.
ISIS guidelines also include evaluation of patients for other possible sources of pain prior to
discography. ISIS Guidelines repeatedly discuss the false positive rate of 75% in patients
with somatisation disorders yet does not make any recommendation regarding
pre-discography psychological testing. Comment: When there is any question at all about the patients
psychiatric status with respect to somatisation, this author suggests either an MMPI-2 or a BHI-2 evaluation be
performed preoperatively. If somatization is seen in these tests, then it is prudent to consider abandoning the
discography as part of the diagnostic algorithm.
ISIS Discography Technique Summary: Prophylaxis against disciitis with cephazolin 1g,
clindamycin 900mg, or ciprofloxacin 400mg IV 15-60min before commencing the procedure
Comment: New standards for surgical device implantation are 30-60 min in advance- given the extremely slow
circulation into the disc from the peripheral blood, it may be wise to administer the antibiotic 45-60min in advance.
If patient has a known allergy to contrast medium, they should be pre-treated with H1 and
H2 blockers and corticosteroids [prednisone 50mg PO 24 and 12 hours before the procedure]
prior to the procedure. Comment: Since the contrast may be found in the disc 6-24 hours later and
continually leach out through the end plates, the prophylaxis above may be insufficient to prevent late contrast
reactions. Instead, my preference is to use undiluted gadolinium penta-acetate in cases where prior severe contrast
reactions (eg. anaphylaxis, hypotension or airway compromise during non-ionic contrast administration, etc). Should
the prophylaxis above be given, it is important to warn diabetics and hypertensive patients of the alterations in blood
sugar and blood pressure produced by steroid administration. Positioning is prone or oblique. AP fluoro
view with cephalo-caudad angulation of the beam to pass parallel to the inferior end plate.
Beam is then rotated laterally until the SAP lies opposite the axial midline of the target disc
except at L5-S1 where the rotation must be less due to the overlying iliac crest. The target
(using the down the beam technique) is the midpoint of the disc just lateral to the lateral
border of the SAP. Comment: The safe zone is much wider at the inferior portion of the disc where there will
be less chance of nerve penetration. This is especially important at L5S1 where the obligatory needle trajectory may
be less oblique than desired due to the overlying iliac crest. Therefore, in cases where safety of the ventral ramus
nerve may be in question, using a target in the inferior aspect of the disc is appropriate. It also appears the
Guidelines are designed for primarily straight needle trajectories. The set up angles used are not optimal for curved
needle techniques where a less oblique beam rotation is needed. Local anesthetic infiltration into the skin
and muscles (up to 5ml) is used to provide adequate anesthesia for deep needle placement.
Sedation may be administered IV but the patient must be responsive during disc
pressurization, therefore careful titration of sedative and opioid medication is essential.
Comment: Withholding the patients usual oral opiate or NSAID medications may cause a significant exacerbation of
pain during the needle placement and during pressurization. On the other hand, excessive IV opiates can diminish
the patient's pain response. Excessive benzodiazepine can cause the patient to become over sedated, sluggishly
responsive, or incoherent. My preferred technique is to have the patient take their usual medications as scheduled
prior to the procedure and titrate propofol (has no analgesic properties on the nerves or disc) for needle placement
only. Then with the propofol redistributed (requires approximately 5 min after the last dose), the patient is
repeatedly questioned about orientation to date, place, and person until the patient is conversing coherently. Only at
that time is pressurization performed. This technique has the advantages of avoiding over sedation and opiates
during the pressurization but permits the patient to be comfortable and amnestic during needle placement. Propofol
is administered in boluses or as a continuous infusion. A single or two needle technique may be used.
The needle is advanced to the SAP then slowly across the intervertebral foramen, towards the
surface of the target disc. Care is needed to avoid impaling the ventral ramus and
reintroduction is needed if the ventral ramus is encountered. Comment: Furcal nerves, which are
preganglionic nerves traversing the neuroforamina from superior to inferior, are present in 15% of patients. These
nerves can be impaled by the needle tip without being anywhere near the ventral ramus. When present, they are
found 93% of the time at the L4-5 disc level. Also, if the patient has an inflammatory mass on the surface of the
annulus, contact with the needle may cause severe pain. Neither of these are addressed by the Guidelines, but
discographers should be made aware of their presence. The needle is advanced to the center of the disc.
Comment: more correctly, the needle should be advanced to the center of the nucleus pulposis. The center of the
nucleus pulposis does approximate the center of the disc except at L5S1 where the center of the nucleus pulposis lies
at the junction of the posterior third and anterior 2/3 of the disc on lateral projection. The L5S1 disc has an
especially thick anterior annulus fibrosis. Once the needle is correctly positioned in the center of the
disc, its stylette is removed and the needle is connected to a system that will inject non-ionic
contrast with antibiotic. Comment: If there is any chance the dura was penetrated during the needle placement,
injection of an cefazolin containing solution could be hazardous. Contrast is often seen tracking along the course of
the discography needle and if cephazolin tracks subarachnoid, it can cause seizures (delayed) or death. In such
cases where there is a question regarding any dural penetration, it is wise to inject the antibiotic at the end of the
procedure rather than mixed with the contrast. A line-pressure transducer with readout and/or graph
may be used. If this is not available, a 3ml syringe with Luer-lock may be used. Comment: Here
the Guidelines inexplicably abandon the use of pressure manometric analysis which is used as part of their final
determination of a positive disc. It is patently unclear how this incongruity can be explained in the context of the
incorporation of manometric data into disc analysis. However, due to the cost of the manometric system ($70-120 per
use in the USA), the Guidelines may be waffeling. It should be noted that most discographers can generate a
sustained pressure of over 160psi with a 3ml syringe as opposed to less than 80 psi with a 20ml syringe.
As the disc is pressurized with contrast aliquots of 0.5-1ml, the pressure is noted along with
the patient's response. The appearance of the target disc must be monitored continuously,
using a lateral view. Injection should continue until a. pain is produced, b. contrast medium
escapes from the disc, c. the volume of injection reaches 3ml, or d. firm resistance to
injection is encountered. If pain is produced, the pressure and volume of injection should be
documented, and the patient should report the concordancy with the accustomed pain and
the current intensity of pain (scale 1-10). The injection may be repeated a short period of
time later in order to reaffirm the response. Upon injection, contrast medium should outline
the nucleus pulposis of the disc and any features of its internal structure. Once a disc has
been tested for pain reproduction, a local anesthetic agent can be injected into the disc,
either or both as a measure to relieve any pain produced by the previous injection, or as an
attempt to relieve the patient's accustomed pain temporarily. Comment: it is uncommon in my
experience for the patient's pain to be relieved by a small amount of local anesthetic in an already pressured disc.
The One should consider a separate anesthetic injection into the disc at a later date if anesthetic discography is
being used as a confirmatory test, however this may only relieve pain in cases of significant annular fissure
presence. The Guidelines contain excellent assessment sheets for disc stimulation and contain a
scoring system sheet (pp 43 and 44).
VIII. Lumbar Medial Branch Blocks
Lumbar MBB have diagnostic utility in that once positive, they identify the source of pain and
serve as a predictor of a good response to percutaneous radio frequency neurotomy. Because
of a high rate of false positive responses at least two separate blocks should be performed on
different occasions with the first block containing an anesthetic to demonstrate prima facie
the joint is symptomatic. Three blocks are preferred by ISIS. As to what constitutes the
criteria for a positive anesthetic block, the Guidelines do not specifically say except to discuss
the consequences of accepting less than 100% relief from the diagnostic anesthetic block.
ISIS does state however that a true positive response to comparative local anesthetic blocks
"...is one in which the patient reports complete relief of pain for a shorter duration when a
short-acting agent is used. Such a response is referred to as "concordant" in that it is
concordant with the expected action of the agents used. If a patient obtains complete relief
of pain but does not report an appropriate differential response in terms of duration of relief,
the joint is not excluded as the source of pain. Such paradoxical responses may be consistent
with the hypothesis being tested." Comment: Whereas it has been demonstrated medial branch block false
positive rates with a single anesthetic injection range from 29-40% and dual injections (one with saline) reduce this
to 12%, there are financial considerations in the US that cannot be ignored. If a patient must be charged a hospital
facility fee or ASC facility fee plus physician fees for each injection, the diagnosis of zygapophyseal syndrome via
performance of these injections could well exceed $6,000. Therefore performance of these injections in a physician's
office can significantly reduce the patient charges and out of pocket expenses. Also, some insurers will not pay for
the second or third injection used for diagnostic purposes thereby shifting the entire out of pocket cost to the
patients. This is unacceptable to the patients and should be unacceptable to the physicians performing these
injections, and therefore an alternative billing model or movement of the procedure to the fluoroscopically equipped
physician's office is appropriate. It is also controversial whether zygapophyseal syndrome exists in a large
percentage of the population as an isolated pain generator, given that discogenic disease with disc space narrowing
progressively transfers more weight to the z-joints thereby potentially damaging the joints. The adoption of a
standard of 100% pain relief from a medial branch block as evidence of a positive block, effectively eliminates those
with concomitant pain conditions from receiving the benefits from radio frequency neurotomy.
Patient selection criteria are documented in the Guidelines with an excellent discussion.
Technical Performance of the Block: Patient prone or oblique, fluoroscopic beam rotated
oblique to view the "scotty dog". Needle target L1-4 medial branch: superior portion of the
eye of the "scotty dog" at the lateral base of the SAP. The L5 dorsal ramus target is the
junction of the ala of the sacrum and the inferior-lateral aspect of the SAP. Once bone is
contacted, the bevel is angled inferiorly to avoid foraminal injection, then 0.1-0.3ml of
contrast may be injected in real time fluoroscopy to demonstrate vascular uptake. If there is
no vascular uptake, 0.5ml local anesthetic is injected. One must then inject the more
cephalad adjacent medial branch in order to anesthetize the descending branch. Comment: it is
unclear why a much smaller volume of contrast is injected (which usually outlines the medial branch), then is
followed by up to five times the volume of local anesthetic. It is possible this amount of local anesthetic could
anesthetize the neural contents of the intervertebral foramen, therefore it may be prudent to limit the volume of local
anesthetic used to the same amount of contrast injected since at least this small amount is demonstrated in real time
to remain outside the intervertebral foramen.
The Guidelines make the point the most important part of the block is in the assessment of
pain reduction, not in the technical performance. The Guidelines use an independent
assessor to evaluate pain for several hours after a block, preferably with a VAS scale of which
the patient has not seen their previous answers during the repeated assessment. Although
the guidelines do not select any specific thresh-hold for pain reduction, 50% of the usual pain
is mentioned in the assessment section. Some physicians require 100% pain reduction.
IX. Sacroiliac Joint Blocks- Premises: 1. The innervation of the joint is diffuse and
controversial therefore it is unlikely individual nerve blocks could be used to diagnose the SI
joint as a pain generator 2. The volume of the joint is approximately 1.5ml 3. Fluoroscopy
is absolutely necessary for this block 4. Physical exam is insufficient to make a diagnosis 5.
Control blocks are necessary-either a placebo control block or a sequence of blocks ruling out
other targets (eg. facet joints, discography, etc).
SI Injection Technique: AP view will reveal a medial distal SI joint shadow (the posterior part
of the joint) and a separate lateral SI shadow (the anterior joint line). Contra lateral oblique
fluoroscopic beam rotation 5-20 degrees is used to image the inferior aspect of the posterior
joint line. The stopping point of beam rotation is when the inferior aspect of the posterior
joint line is maximally "crisp" rather than attempting to overlap the anterior and posterior
joint lines. A cephalad tilt of 20-25 degrees of the fluoroscopy beam may then be used to
help identify the posterior from the anterior joint lines....the posterior-inferior aspect of the
joint is displaced in a caudal direction compared with the anterior joint. The target for
injection is the inferior joint just above the inferior border of the joint.
Alternative approaches in case of failure: a. direct the needle to the inferior location of
overlap of the two joint lines b. slide off the inferior aspect of the joint into the inferior
capsular recess c. cannulate the mid portion of the joint using 20-30 degrees oblique
rotation of the beam d. superior capsular recess using a cephalocaudad approach
After needle placement, 0.3-0.5ml contrast is injected. If there is a venous drainage,
impossible injection due to placement in ligaments, or aberrant pattern seen, re-attempts at
needle placement up to 3 times may be used. Joint arthrography does not correlate well with
painful joints. Injection of local anesthetic is performed until a firm resistive endpoint is
appreciated, extra-capsular contrast is seen, or a total volume of 2.5ml is injected. Steroid
use in the joint is not part of the Guidelines. The Guidelines comment several times on how it
may be impossible to enter the joint no matter the approach. Comments: The concept the synovial
part of the joint (the only part addressed by the Guidelines) is responsible for pain production in the joint is not well
established in the literature. There may be just as much validity to the ligamentous dorsal portion of the joint
producing pain through neural ingrowth as happens with an injured ACL joint. However, the use of these Guidelines
will effectively rule out the synovial joint as a pain generator. Conditions in which the synovial aspect of the joint
are seen to be abnormal on MRI include rheumatoid arthritis, ankylosing spondyliitis, pyogenic arthritis, Reiters
syndrome, and psoriatic arthritis. Following trauma, it is rare to see positive synovial joint MRI abnormalities,
therefore one must consider ligamentous sources of pain after trauma.
X. Cervical Discography-The false positive rate is very high and the procedure is
technically demanding. In order to have a positive discogram, 5 criteria must be met: 1.
technically correct needle placement is a requirement, 2. Z-joint pain generators must have
been previously eliminated (without this, there is a 67% false positive rate), 3. Concordant
pain is produced during pressurization of the target disc, 4. Pain level produced is 7/10, 5.
Adjacent disc pressurization does not reproduce the patient's pain. The protocol states 4
levels should be tested. The C2-3 level is not necessary (as a 5th level) unless there is a
significant headache symptomatology. The everyday pain patterns produced by cervical discs
are diffuse but roughly correlate to the pain maps constructed for cervical z-joints. Mercifully
the authors of the guidelines recognize sedation may be necessary for this extremely painful
procedure but make the pain the patient must be responsive during pressurization.
Technical conduct of the procedure: Supine position, right sided approach (to avoid the
esophagus), fluoro beam cephalocaudad angulation parallel to the end plates (usually the
beam source is about 10-20 degrees caudad angulation), then right oblique rotation 20-30
degrees. Great vessels are retracted laterally by hooking the fingers of one hand into the
groove between the trachea and the carotid sheath. Following local anesthetic infiltration, a
23-26 ga spinal needle is inserted through the retracted groove to the uncinate process using
the down the beam approach while maintaining retraction. Any sudden onset of coughing,
respiratory distress, sudden electrical shock sensation, or edema at the injection site require
the needle be withdrawn. The needle is then advanced to the vertebral end plate (superior)
and the retraction released and constant contact is maintained by the directing hand with the
needle. Subsequently the needle is walked off the end plate through the annulus and
immediately a lateral fluoro view is obtained. The needle is adjusted to reach the center of
the disc without crossing the mid line in either AP or lateral views. Frequent AP and lateral
views may be required during adjustment. The Guidelines state a small curve at the tip of
the needle will help with placement. During pressurization, the patient should be asked
about the onset of pain, severity of pain, concordancy of pain, and location of pain. Injection
of contrast is with a 3cc syringe and 0.1ml increments are injected under continuous
fluoroscopic visualization. The injection is terminated when the pain is 7/10 or greater,
contrast escapes from the disc, end plates are visibly displaced, there becomes firm resistance
to injection, the patient exhibits non-concordant pain, or there are any unexpected or
distressing sensations. The total volume of the disc is 0.5ml. Fluoro images pre and post
injection should be saved and CT post injection acquired. The Guidelines do state manometry
may be used. Discs are graded as concordant, asymptomatic, or dissimilar. Comments: It is very
difficult to inject 0.1ml increments with a 3ml syringe, therefore a luer lock 1cc syringe is preferred. Because the
opening pressures are very high (often well over 120-150 psi) manometry is not useful. On morphological analysis
of the discogram either in fluoroscopy or CT, one must remember the uncinate clefts are often mistaken for annular
tears by the unknowing. Uncinate clefts are an anatomical probability in discs, especially in aged discs. The disc
material in the cervical spine is much more firm than the lumbar making anatomical "fracture planes" much more
likely in the disc itself (the clefts). Antibiotics should be used in the injected contrast. If post injection CT scans are
obtained, the injectionist should view these personally as many radiologists have never seen a cervical discogram
and haven't a clue as to what constitutes normal vs abnormal findings. Also, 1mm slices should be taken directly
through the disc of interest and parallel to the end plates. Oblique slices through the disc are useless. There is no
need to CT the vertebral bodies nor discs not producing concordant pain. Such a specialized request may require
consultation with the radiologist to assure the CT is carried out in a technically correct manner.
XI. Cervical Medial Branch Blocks
Principles- If pain is not relieved in the area targeted, the target nerves are not mediating
the patient's pain. There may be other medial branches mediating the pain or the pain
generator may arise from a non-median branch mediated source. If pain is relieved, then
this constitutes evidence of median branch transmission of pain if there are no false positives.
Partial relief of pain may be possible if only part of the median branches transmitting the pain
are actually blocked and other sources are excluded. Cervical median branch blocks are
diagnostic and predict a good chance of obtaining complete pain relief from RF.
Which z-joints should be the targets? The following pain map created from volunteers during
stimulated pain in the joints should serve as a guide.
Statistically, the most likely joint to be involved in the cervical
spine is the C2-3 joint, followed by the C5-6 joint. The
guidelines recommend against medial branch blocks in cases
where there is widespread pain or hypersensitivity to touch
as it is unlikely medial branch blocks will be of any value.
Contraindications: In addition to the absolute and relative
contraindications listed at the beginning of the guidelines,
neurological signs are a relative contraindication for cervical
medial branch blocks. However if the neurological disorder
is be managed, then such blocks may be indicated. Radicular
pain is not a contraindication to medial branch blocks.
Agents: The most common agents used are bupivicaine 0.5% and lidocaine 2%, but other
concentrations of these agents may be used. The volume of the block should not exceed
0.3ml.
Technique: The most convenient and least demanding approach is the lateral approach with
the patient lying in the lateral position. The fluoro beam is aimed in a true lateral view with
complete superimposition of facet joint lines. The target for C3-6 is the centroid of the
articular pillar. For C7 the target lies high on the apex of the superior articular process. The
target for the third occipital nerve (necessary for blocking the C2-3 joint) is at three separate
points on a line bisecting the centroid of C3 and the center of the C2 lateral arch parallel to
the posterior vertebral line. The three points are at the level of the C2-3 disc, just above on
the arch of C2 and just below on the SAP of C3. The patient should be warned about the
probability of ataxia being quite high after the TON block and driving should be avoided.
The guidelines suggest errors in performance of the blocks and in interpretation can be
reduced by using double blind dual blocks, making certain the patient has sufficient pain
before the block to justify its performance (at least 50% of their pain at its worst and VAS at
least 4/10 prior to the block), and in performance of the least number of blocks necessary per
patient. An exhaustive discussion is undertaken in the guidelines to demonstrate the
complexities of both selection of nerves to block and in interpretation of the results of the
blocks.
XII. Lateral Atlanto-Axial Joint Blocks (LAAJB)
The guidelines sagely make the point inexperienced injectionists should absolutely not
attempt this block due to the extreme risk attached to even slight deviation of the needle tip.
There are few studies demonstrating efficacy of these blocks. Specifically Aprill and
separately Busch and Wilson reported relief of headache fowing performance of these blocks.
The guidelines suggest a triple block or a dual anesthetic block. The suggestion of using control
blocks is only meaningful if further interventions were to be entertained. The Guidelines do not contain a
description of a radiofrequency neurotomy of the nerve supplying the C1-2 joint, therefore it is this authors
opiniontriple blocks or dual anesthetic blocks of C1-2 are merely an academic exercise without any practical
application.
The primary indication for LAAJB is the need to know if the patient's pain stems from such
block. The literature suggests indications are headache and suboccipital pain, focal
tenderness in the suboccipital region, focal tenderness over the tip of the left or right
transverse process of C1, restricted rotation of C1 on C2 on manual exam, and aggravation of
the usual headache by passive rotation of the C1 vertebrae to the left or right.
Technique: 25 ga 90mm needle, pulse oximetry, prone patient position with the head and
face supported comfortably with a pad. Fluoroscopic evaluation is undertaken of the joint in
AP and the posterior arch of the atlas identified. If the arch overlies the joint, the C-arm
should be angled caudally or the patient should flex the head slightly. If the teeth obscure
the view when the mouth is opened, the patient may need to rotate their face slightly . This
rotation can be quite dangerous as the room for error is mere millimeters. This author suggests unless you are
extremely experienced, no rotation of the face or head laterally off the midline should be undertaken.
The target is the lateral third of the lower border of the atlas or the lateral quarter of the
axis. More medial placement (eg. the midline) may transect the spinal C2 nerve or the DRG.
Slow methodical advancement of the needle perpendicularly to the initial target point is
undertaken. If the tip strays lateral to the joint margin, it may pith the vertebral artery. If
the tip strays to the center of the joint, it may strike the C2 spinal nerve or ganglion. More
medial placement risks cord puncture. If the tip passes through the joint, it may impale the
carotid artery. Frequent fluoroscopic eval is required during advancement. After initial
osseous contact is confirmed, the needle is adjusted inferior from the atlas or superiorly from
the axis and advanced into the lateral 1/3 point of the joint. Once passage into the joint has
occurred, further advances of the needle are made in the lateral fluoroscopic view.
Advancement is then made 1/4 to 1/2 way across the joint on lateral fluoroscopic projection.
If blood is aspirated, the needle is advanced slightly anterior. Contrast is injected to reveal a
slit like filling of the joint on AP view and a partial filling of the joint on lateral view. The
joint slopes inferiorly on lateral fluoro view and therefore one may not outline the joint well.
There are anterior and posterior meniscoids that may fill and also filling may occur medially
through the joint into the spinal canal or even across the synovial connections to the
contralateral side. Once the volume of filling is determined the same volume of local
anesthetic is injected with a minimum of 0.5ml and a maximum of 1.0ml.
As the pathology of joints producing pain is often cartilage fragmentation or synovial inflammation, the use of
steroids is frequently used throughout the body as a therapeutic measure. The Guidelines do not mention steroid
injections into the C1-2 joints but this author routinely uses them in combination with 0.5% bupivicaine for a total
volume injection of approximately 3/4 ml. The overdistension of the joint with contrast followed by injection of a
steroid/local anesthetic combination can result in extreme pain after such injection, therefore this author uses
minimal contrast, only enough to outline the joint. In this author's mind, there is no point to perform a diagnostic
block if there is no potential therapy to be offered.
XIII. Lumbar Transforaminal Injections of Corticosteroids
The most compelling evidence for TFESI is a series of controlled trials outlined in the
expansive references of the Guidelines. The first by Riew demonstrated 29% of patients
treated with steroids vs. 67% of patients receiving bupivicaine alone ultimately required
surgery. Another study by Karppinen found that TFESI were more effective than saline at 2
and 4 weeks post injection. Finally, Vad compared TFESI with paraspinal injections of saline.
Of these 84% of the steroid group vs 48% of the saline group reported significant
improvement in pain after 12 months. Of course, all these studies deal with HNP with
radicular pain.
The main selection criteria for the injection is radicular pain who have not responded to
non-surgical interventions or for whom non-surgical interventions are not indicated, and
whose pain is likely to have an inflammatory basis.
The Guidelines suggest use of 25 or 23 ga needles. If a double needle technique is utilized,
both a larger diameter needle 17-20ga is employed in conjunction with a small diamter
central injection needle. An IV is necessary. Contrast is used as part of the Guidelines along
with a steroid, Betamethasone suspension 6-18mg, Triamcinolone 20-80mg, or
dexamethasone. It is curious methylprednisolone is not included in the list of steroids as it is one of the most
commonly employed steroids in use for TFESI. Consent should include dural puncture, infection,
allergic reaction, spinal cord injury, or arachnoiditis.
There are two basic approaches described by the Guidelines: the subpedicular approach with
placement of the needle tip on the anterior vertebral body superior to the exiting nerve root
(the most commonly taught approach) and the so called "retroneural approach" which
requires greater skills at placement.
The subpedicular approach uses the following fluoro setup: AP tangential to the inferior
endplates. The target is on the vertebral body at the 6 o'clock position of the pedicle and just
inferior to the inferior pedicular border. There is ostensibly a "safe triangle " that widens as
one goes laterally under the pedicle. If the foramen is not distorted such as in foraminal
stenosis or loss of disc height, then the triangle formed between the nerve root, the
infrapedicular bisecting line, and the lateral pedicular line constitutes the "safe triangle" that
contains only the sinuvertebral nerve. If the 6 o'clock position is obscured by the projection
of the superior articular process, then rotation of the beam slightly obliquely is employed in
concert with readjustment of the target to the new 6 o'clock position. Note hypertrophy of the
superior articular process may demand a significant adjustment of the needle tip more obliquely. Also note
foraminal height narrowing may substantially decrease the margin of error for proper needle placement. Also, if the
patient has superior nerve root displacement at the midpedicular line due to a prolapse of the disc at the foraminal
level, the root may be displaced directly into the path of the subpedicular needle approach.
For the single needle approach, the skin entry is slightly inferio-lateral to the target. The
needle is advanced with frequent fluoroscopic evaluations of needle trajectory, needle
withdrawal slightly, then re-advancement until ultimately bone is contacted. If the patient
suffers the onset of aggravated radicular pain, the operator should carefully consider the
wisdom of terminating the procedure. If the patient suffers only a temporary burst of
symptoms and the symptoms abate upon withdrawal of the needle, the procedure can be
resumed. Once the patient experiences a radicular paresthesia, ostensibly there has been a pithing of the nerve
with some obligatory (although may be subclinical or not result in weakness due to the more cephalad lying DRG)
neuronal damage. It is this authors opinion that due to the hazards of this approach with potential nerve injury that
no sedation at all be used for this procedure, or conversion to a procedure that is less known to cause paresthesia
may be prudent.
The two needle technique places the introducer needle to the lateral border of the superior
articular process and advanced to the dorsal half of the foramen. The slightly curved
injection needle is subsequently advanced through the introducer needle and the passed
under AP view fluoroscopy towards the target point.
Once the target point is reached, 0.2-0.5 ml contrast is injected to assure there is no
intrathecal or radicular artery injection. In either case, the procedure is abandoned and the
patient is brought back at another time for a repeat attempt. If on the other hand there is an
intravenous injection, the needle tip is repositioned slightly and a repeat injection of contrast
is performed. Optimally, contrast should spread into the foramen and into the lateral
epidural space. Once the safe triangle and nerve root sleeve is outlined by a given volume of
contrast, the same volume of therapeutic agent is subsequently injected. It is important to
note the area of the nerve considered to be inflamed must be covered with steroid depending
on the needed volumes determined via the neurogram. Specifically this means if the superior disc
adjacent to the level injected contains the inflammatory disc prolapse, then the neurogram must reach at least to this
level. This is commonly seen in paracentral disc herniations. On the other hand if the disc herniation is foraminal
at the level of the exiting nerve root, then only the lateral recess and infrapedicular area need to be injected. For far
lateral disc herniations, a different approach will be needed to adequately address the extraforaminal nature of this
inflammatory disc effect. It should be also noted there was no mention of "real time" neurography or even better,
digital subtraction neurography. One or the other of these approaches is critically important in order to avoid the
intra-artial injections thought to be associated with permanent paralysis from steroid injections into the artery.
The Guidelines state the steroid is the active agent in therapeutic TFESI but most operators
also use a local anesthetic, ostensibly in order to render the injection less painful. The
Guidelines correctly note the volume of local anesthetic is injected first, not to exceed half
the intended overall volume of injectate. This author is aware of several cases of unrecognized spinal
cord infarction, unrecognized nerve injury, or fractures from falls in the past due to the excessive injection of local
anesthesia volume or concentration before the publication of these Guidelines. If the Guidelines are followed, there
should be only minimal chance of these unrecognized or produced injuries.
The retroneural approach target is the dorsal neuroforamen directly over the nerve root. The
anatomic location is at the posterior 1/3 of the neuroforamen and the superior 1/3 of the
neuroforamen. Fluoroscopic setup is initially AP with the beam passing parallel to the
superior endplate. The AP target is the mid-pedicular line. If the lamina overlaps the plane,
then the beam is rotated slightly lateral oblique. If the superior articular process obscures
the target area, then rotate the fluoroscope beam cephalad until the SAP recedes caudad.
Initially he lateral laminar border is contacted as a depth finding measurement. The needle
is advanced off the lateral borer of the lamina with frequent fluoro guidance to the target
point above. No attempt should be made to elicit a paresthesia. The remainder of the
injection is as per the subpedicular description.
The transforaminal injection of S1 is undertaken for S1 radicular pain. The AP view of the
sacrum often fails to permit visualization of the S1 posterior neuroforamen, therefore a
slightly cephalad beam trajectory with a slightly lateral oblique approach makes the anterior
and posterior foraminal borders line up. The needle is placed on the lateral lip of the bone of
the sacrum around the neuroforamen and walked off into the neuroforamen approximately 2
millimeters. Then lateral fluoroscopy will reveal the depth of needle insertion into the
neuroforamen. The needle is advanced to the anterior sacral border, lateral to the anterior
sacral neuroforamen, passing 5mm from the floor of the sacral canal, but short of the
posterior border of the anterior border of the sacrum. Care is to be taken that the needle
does not pass through the anterior sacral foramen. The remainder of the injection Guidelines
are used as per the subpedicular injection. Note the pedicle of S1 always lies in a line inferior to the L5
pedicle.
XIV Radiofrequency Lumbar Medial Branch Neurotomy (RLMBN)
The only indication for RLMBN is complete relief from selective medial branch nerve
blocks.The medial branches of the posterior primary rami course from the dorsal ramus at the
neuroforamina running inferiorly to the SAP-transverse process of the vertebral body below
the neuroforamina, then posteriorly in the groove between the SAP-transverse process under
the mamillary-accessory ligament, then splits into a ipsilateral z-joint branch, multifidus
branch, and descending inferior branch. The latter courses inferiorly to the next z-joint.
Therefore the numbering of the appropriate innervating branch becomes important. The L4-5
Z-joint is innervated by the L3 and L4 medial branches which lie on the L4 and L5 transverse
processes. The Guidelines note the L5S1 Z-joint is innervated by the medial branches of L4
and L5. It should be noted the L5 medial branch does not arise until the inferior aspect of
the L5S1 Z-joint. Before this point, it is simply known as the dorsal ramus of L5. The
Guidelines go over the physics and the optimum time frame of lesioning (60-90 seconds) with
the needle tip being placed very close to and parallel to the nerve.
Technique: The target point is the junction of the SAP-TP with the tip at the middle to
anterior 1/3 line of the groove (on lateral view). The "pillar view" is approximately 15-20
degrees oblique and varying degrees of declination of the beam. The typical declination angle
is 30-40 degrees inferior to superior relative to the superior end plate angulation. This angle
maximizes the contact between the median branch nerve and the needle tip. AP, lateral, and
oblique fluoroscopic images are checked prior to lesioning (check the Guidelines extensive
fluoroscopic views to demonstrate appropriate placement). After checking for proper
placement with fluoroscopy, a small amount of local anesthetic (0.5ml) is applied through the
needle and subsequent lesioning is performed. The more cephalad level will also need
lesioning. The Temperature is measured during lesioning, but no pre-lesioning stimulation is
used. In the US, it is always prudent to perform at least motor stimulation for reasons of medicolegal
documentation.
XV. Intradiscal Electrothermal Therapy - Currently this therapy has no CPT code
in the US, is considered experimental by most insurance carriers, and lacks reimburement by
most insurers. Because of the above issues and only modest clinical improvements with
such therapy, it will not be discussed in this document at this time.
XVI. Cervical Transforaminal Injection of Corticosteroids
"Spinal cord infarction, due to intra-arterial injection of corticosteroids, is more than a
nominal risk of cervical transforaminal injection of corticosteroids. The severity of this
complication weighs heavily on the benefit-risk ratio of this therapy". Author's note: one should
burn the warning here into your mind before pursuing this therapy. The disasters are rare, but are very real, and
have happened to experienced physicians. This doesn't mean one shouldn't lean the technique or perform the
procedure, but rather that extreme caution should be exercised.
Indications: patients with cervical radicular pain or radiculopathy in which the presumptive
cause is an inflammatory process affecting the cervical spinal nerve or root.
Medications used for injection: water soluble non-ionic contrast suitable for iv and intrathecal
use (Isovue M or Omnipaque 180, 240, or 300), preservative free local anesthetics,
corticosteroid (betamethasone, triamcinolone, dexamethasone). Note this author believes there is
little to be gained and a significant degree of complication that may occur with the routine injection of local
anesthesia during this block. Certainly the presence of local anesthetic may delay the diagnosis of spinal cord
infarction and may result in a significant delay in diagnosis or a spinal artery injection.
Technique: Patient position is supine, lateral, or oblique depending on operator preference
and patient comfort and physician preference. This author strongly prefers the supine position with the
head oriented straight AP. Minute deviations from the angles and targets chosen can lead to significantly
malpositioned needles. Small movements of the patient after initial trajectory calculation must be prevented and the
supine position is easier to define the neutral or starting fluoroscopic views since any head rotation is obvious.
The fluoroscope is adjusted to view an oblique projection of the cervical spine to provide a full
view of the target foramen along its length. Often injectionists will initially position the fluoroscope
beam to 45 degrees since the neuroforamina are widest at this angulation from C3-4 to C6-7. This places the carotid
and vertebral artery directly in the line of fire at some vertebrae. Transvascular needle trajectories should be
avoided, especially in the elderly in whom plaques or calcific deposits may be dislodged into the brain. See the
fluoroscopic views on this website demonstrating cross-sectional anatomy of the cervical spine and various angles of
needle trajectory.
The needle puncture point overlies the posterior half of the selected foramen targeting the
posterior wall. The 25ga needle is brought through the skin and directed towards the
anterior half of the SAP advancing down the beam in small increments with frequent
adjustments of the needle, until the tip touches the SAP. The needle is advanced a few mm
off the posterior wall and directed to the middle third of the neuroforamen. Once the lateral
margin of the neuroforamen has been reached, an AP fluoroscopic view is used. The needle
is advanced to the mid-pedicular line but never beyond the lateral border of the uncinate
processes. If radicular pain occurs on needle advancement, no injection should be made
without repositioning of the needle and only if the radicular paresthesia resolves prior to any
injections. Once final positioning is achieved, spot films in the oblique and AP views are
made. The oblique view should demonstrate the needle position in the posterior aspect of the
foramen while the AP view documents the needle depth within the foramen. This author believes
passage of the needle into the middle 1/3 of the neuroforamen is seriously jeopardizing the spinal nerve and
therefore the tip should be maintained along the posterior foraminal wall periosteum during advancement.
A short length of low volume tubing is recommended in order to change syringes without
needle movement. Under direct , real-time fluoroscopy, a small volume of non-ionic contrast
material (1.0ml or less) is injected. The solution should disperse in the intervertebral
foramen outlining the spinal nerve and the DRG. It is essential to visually monitor the
fluoroscopic injection of contrast to look for arterial injection (manifested by a flash then
clearance) In either case, the needle should be withdrawn and no further injections
attempted. The procedure should be rescheduled until after a period of time to permit
healing of the vessel Injection of contrast should be continued until epidural spread is
achieved with outlining of the target nerve dural sleeve. Intrathecal injection will result in
rapid dilution of the contrast material. If during injection of contrast, the radicular pain is
tolerable, the injection may be continued. Local anesthetics must be used judiciously and
carefully due to potential for unintended subarachnoid injection. Use of a short acting
anesthetic is recommended. Lidocaine is injected in volumes of 0.5-1.5ml @1% strength or
1-3ml at 0.5% strength. Corticosteroid use is predicated on whether the procedure is
monoradicular or polyradicular. With monoradicular pathology, betamethasone 3-6mg,
tramcinolone 20-40mg, or dexamethasone 7.5-10mg are appropriate. For multiradicular
problems, the doses are doubled. This author believes the use of local anesthetics injected transforaminally
serve to detract the physician from diagnosis of a neurological event. They do not help facilitate the procedure nor
do they add any diagnostic utility to the procedure (unless they are used for a selective nerve root block, then must
be given after the steroid. This author also believes the volumes cited above are excessive given the size of the
spinal canal and the care being taken to use small volumes of contrast makes the local anesthesia volumes of up to
3cc inconsistent. The cord itself can be compressed by 3 volume aliquots. Finally, the use of multiradicular
treatment with a single shot transforaminal approach is inconsistent with the selectivity of the transforaminal
approach. This author recommends multiple TFESI needle placements or the use of an interlaminar ESI under
fluoroscopy.
XVII. Percutaneous Radiofrequency Cervical Media Branch Neurotomy
(CMBRF)
Anatomy: The structures innervated by the cervical medial branches include the cervical
zygapophysial joints, paramedia and deep cervical muscles, multifidus, semispinalis cervicis,
and interspinales. The C5 medial branch courses across the centroid of the articular pillar of
C5. The C4 and C6 medial branches are between the centroid and the base of the SAP while
the C3 and C7 branches lie on the SAP. Click the pics below to enlarge.
The medial branches are separated from the bone by an average of 1.1-1.7mm by loose
areolar tissue. The physics behind the process of RF involve placement of the RF needle
active tip parallel to the nerve to be lesioned and the more lesions, the more interrupted is
the nerve with subsequent longer results. In order to achieve this, the TON nerve being the
exception, the remainder of medial branches requires both a sagittal pass and an oblique
pass.
INDICATIONS: Complete relief (90% or greater) of pain following controlled diagnostic or
comparative blocks.
PATIENT SELECTION: Patients with at least 3 months pain with no signs of resolution, neck
pain has not responded to conservative treatment, or for which conservative treatment is
either not available or indicated, and in patients without an abuse history of analgesics for
pain management.
POSITION: Prone or lateral.
PROCEDURE: 3 needle approach to each level with a 2 stage insertion: Initially the medial
branch is anesthetized via a standard medial branch block and the muscles in the area are
also infiltrated with local anesthetic. This marker needle is left in place to facilitate the
placement of the oblique RF needle. Then the RF needles are inserted using oblique and
sagittal passes. Oblique is suggested to be performed first as it is the most difficult. Target
is anterior 1/3 of the SAP and on AP projection is on the convexity above the concavity of
the pillar. A 30 degree oblique approach is used and the needle is inserted without any
declination or cranio-caudad beam angulation. Several lesions are made serially as the
needle is linearly retracted after each lesion. Lesioning parameters are 80-85 degrees C for
90 seconds. The sagittal pass needle is placed without any obliquity in the beam but with a
declination of the fluoroscopy beam to the same angle as the zygapophysial joints on lateral
fluoro views. The lateral aspect of the lateral mass at the waist of the pillar is lesioned
using the same parameters. No stimulation is needed as the patient during the described
procedure is quite awake or only minimally sedated. See the guidelines for detailed
fluoroscopic views. The guidelines argue against stimulation and also against use of a single
pass curved needle vs. two straight needles due to their projection of 33% more coverage
with 2 straight needles than one curved needle.
In the medicolegal climate that exists in the US, it is very prudent to stimulate for motor capture of nerves. This
simple test adds to the safety profile of the procedure. The ISIS guidelines use the patient response during
thermocoagulation nerve injury as a point for immediate cessation of the procedure, however in the USA, with
stimulation available on all RF machines, it is wise to stimulate prior to coagulating.
XVIII. Emerging Procedures: ISIS is examining thoracic discography, thoracic
transforaminal injections, thoracic intra-articular blocks, and thoracic medial branch
blocks as presumptive procedures without any literature support. There are descriptions in
the Guidelines for each of these.
areolar tissue. The physics behind the process of RF involve placement of the RF needle
active tip parallel to the nerve to be lesioned and the more lesions, the more interrupted is
the nerve with subsequent longer results. In order to achieve this, the TON nerve being the
exception, the remainder of medial branches requires both a sagittal pass and an oblique
pass.
INDICATIONS: Complete relief (90% or greater) of pain following controlled diagnostic or
comparative blocks.
PATIENT SELECTION: Patients with at least 3 months pain with no signs of resolution, neck
pain has not responded to conservative treatment, or for which conservative treatment is
either not available or indicated, and in patients without an abuse history of analgesics for
pain management.
POSITION: Prone or lateral.
PROCEDURE: 3 needle approach to each level with a 2 stage insertion: Initially the medial
branch is anesthetized via a standard medial branch block and the muscles in the area are
also infiltrated with local anesthetic. This marker needle is left in place to facilitate the
placement of the oblique RF needle. Then the RF needles are inserted using oblique and
sagittal passes. Oblique is suggested to be performed first as it is the most difficult. Target
is anterior 1/3 of the SAP and on AP projection is on the convexity above the concavity of
the pillar. A 30 degree oblique approach is used and the needle is inserted without any
declination or cranio-caudad beam angulation. Several lesions are made serially as the
needle is linearly retracted after each lesion. Lesioning parameters are 80-85 degrees C for
90 seconds. The sagittal pass needle is placed without any obliquity in the beam but with a
declination of the fluoroscopy beam to the same angle as the zygapophysial joints on lateral
fluoro views. The lateral aspect of the lateral mass at the waist of the pillar is lesioned
using the same parameters. No stimulation is needed as the patient during the described
procedure is quite awake or only minimally sedated. See the guidelines for detailed
fluoroscopic views. The guidelines argue against stimulation and also against use of a single
pass curved needle vs. two straight needles due to their projection of 33% more coverage
with 2 straight needles than one curved needle.
In the medicolegal climate that exists in the US, it is very prudent to stimulate for motor capture of nerves. This
simple test adds to the safety profile of the procedure. The ISIS guidelines use the patient response during
thermocoagulation nerve injury as a point for immediate cessation of the procedure, however in the USA, with
stimulation available on all RF machines, it is wise to stimulate prior to coagulating.
XVIII. Emerging Procedures: ISIS is examining thoracic discography, thoracic
transforaminal injections, thoracic intra-articular blocks, and thoracic medial branch
blocks as presumptive procedures without any literature support. There are descriptions in
the Guidelines for each of these.
