Issues in Emerging Health Technologies

Radiofrequency Neurotomy for

Lumbar Pain
Issue 83 • May 2006

from two levels, medial branch RFN must typically be

Summary
performed at several spinal levels, either unilaterally
 Chronic lumbar (lower back) pain, which affects or bilaterally.
many Canadians, imposes a large economic
burden.
 Symptoms may occur in the vertebral facet joints
of 15% to 40% of patients with lower back pain.
 Medial branch radiofrequency neurotomy is a
minimally invasive outpatient procedure that
reduces pain by interrupting the nerve supply to
painful facet joints.
 Four systematic reviews of this procedure offer
disparate conclusions.
 One small well designed observational study has
shown positive results, but no equally rigorous
randomized controlled trial has been conducted.
Source: US National Institute of Arthritis and
Musculoskeletal and Skin Diseases
The Technology
Regulatory Status
Radiofrequency neurotomy (RFN) is known by many
names, including percutaneous RF facet denervation, Medial branch RFN is a medical procedure, and is not
percutaneous facet coagulation, percutaneous RFN, RF subject to Health Canada regulatory approval.
facet rhizotomy, and RF articular rhizolysis.1 It is a min-
imally invasive, interventional procedure used to treat
chronic spinal pain of facet joint origin. The facet or Patient Group
zygapophyseal joints2 are bilateral structures that link
The economic burden associated with chronic back
each vertebra to its neighbours. Lumbar facet joints
pain is high, and may be comparable to that due to
receive their nerve supply from the medial branches
depression, diabetes, and other common disorders.4 A
of the dorsal rami of the spinal nerves, each facet joint
survey of Canadians aged >12 years (n=118,533) esti-
being innervated by the branch specific to its own
mated that the prevalence of chronic back pain (pain
vertebral level and the branch from the level above.3
for >6 months) during the previous 12 months was
9%, with 19% of respondents reporting pain that was
In medial branch RFN, an insulated electrode with an
severe, 55% moderate, and 26% mild.5 A second
exposed tip is percutaneously introduced into the
Canadian study (n=13,756) identified persistent back
spinal area, and under X-ray fluoroscopic guidance, it
problems as the most common chronic problem
is positioned parallel to the nerve supplying a painful
among those <60 years old, and the third most com-
facet joint. Once positioning has been confirmed, a
mon in those >60 years old; prevalence was estimated
current is passed through the electrode. The resultant
at 15% to 18%.6 Canadian data are unavailable, but
heat destroys adjacent tissue, including the target
studies from the US and Australia indicate that in 15%
nerve, thereby interrupting the transmission of pain
to 40% of patients, chronic back pain may be
signals. Because lumbar facet joints are innervated
attributable to the lumbar facet joints.7,8

The Canadian Agency for Drugs and Technologies in Health (CADTH) is funded by Canadian federal, provincial and territorial governments. (www.cadth.ca)
The only valid and reliable diagnostic test for facet
joint pain is fluoroscopically guided, controlled
diagnostic medial branch blocks, or intra-articular
facet joint blocks.2,7,9-12 Patients who are unresponsive
to conservative therapy and who have positive pain
relief from controlled diagnostic blocks may be
suitable for medial branch RFN.
Current Practice
Conservative treatments for chronic lumbar pain
include exercise, oral medications, physical therapy,
spinal manipulation, and behavioural therapy. Most
have, at best, modest efficacy.13,14 Interventional thera-
pies for facet joint pain include intra-articular facet
joint injections, medial branch nerve blocks, and medi-
al branch RFN.11,12 No accurate utilization data are avail-
able for these procedures. However, the use of medial
branch RFN is relatively uncommon in Canada (Dr. D.
Vincent, Victoria, BC: personal communication, 2006
Mar 02).
The Evidence
Four systematic reviews (SRs) address the efficacy of
medial branch RFN for lumbar facet joint pain.15-18
Two15,17 include only randomized controlled trials
(RCTs), while the others16,18 include RCTs and observa-
tional studies. The SRs reach divergent conclusions:
• Geurts et al.15 found moderate evidence that the
procedure is more effective than placebo.
• Manchikanti et al.16 found strong evidence that
the procedure offers short- and long-term pain
relief.
• Niemesto et al.17 found conflicting evidence on the
short-term effect.
• Boswell et al.18 found moderate to strong evidence
in favour of efficacy.
The SRs include four relevant RCTs19-22 and six observa-
tional studies.8,13,23-26 Most of the included studies relied
on single diagnostic blocks [which have a false posi-
tive rate of 27% (95% CI: 22% to 32%) in the lumbar
region],7 rather than controlled diagnostic blocks in
the selected patients. As a result, many enrolled
patients may not have had facet joint pain.
Consequently, experts such as Hooten et al. have
argued that the results of several RCTs19,20,22 are
invalid.27 This criticism may be applied to all the
studies cited in this bulletin, with the exception of
the Dreyfuss et al. study.8
The Canadian Agency for Drugs and Technologies in Health (CADTH) is funded by Canadian federal, provincial and territorial governments. (www.cadth.ca)
In a small, well designed observational study (n=15),
Dreyfuss et al. used meticulous techniques, including
controlled diagnostic blocks, and they achieved signifi-
cant, sustained reductions in lumbar facet pain.8 In
contrast, a RCT designed to reflect common clinical
practice, including reliance on single diagnostic blocks,
found that medial branch RFN offered no benefit28
(Table 1).
Adverse Effects
Possible adverse effects (AEs) of medial branch RFN
include painful cutaneous dysesthesias, neuritis or
neurogenic inflammation pain, anesthesia dolorosa,
cutaneous hyperesthesia, pneumothorax, and deaf-
ferentation pain.11,12 AEs reported in two studies23,28
included treatment-related pain, transient neuro-
pathic pain, transient leg pain, dysesthesia, and
subjective leg weakness; the other studies reported
no AEs. One centre performing lumbar medial branch
RFN estimated the incidence of minor AEs at 1% per
lesion site.29
Administration and Cost
Historically, medial branch RFN was a neurosurgical
procedure, but currently, anesthetists and other physi-
cians specializing in pain medicine perform it most
often.9 The time required depends on the number of
levels to be treated, but estimates range from 20 min-
utes (most likely) to five hours.30 Local anesthesia is
generally used, although general anesthesia may be
preferred in some centres.23 The procedure can be
repeated at three-month intervals if >50% pain relief
is obtained for 10 to 12 weeks post-RFN.11,12
Payment may come from provincial health plans or
third-party payers such as workers’ compensation
boards. No cost-effectiveness evaluations were locat-
ed, and little costing data are available.11,12 A 2001
Canadian review addressing medial branch RFN for
cervical facet joint pain estimated the procedural cost
at C$401, excluding physician fees, the cost of diag-
nostic blocks, and overhead costs (e.g., RF neurotomy
needles, RF generator, fluoroscopy equipment).31 Van
Wijk et al. reported total actual treatment-related
costs for medial branch RFN at US$285, but details
about the cost calculations are lacking.28
 e 1:: Dreyfuss et al. and Van Wijk et al. studies
Table

Authorss and
d Studyy
Dreyfuss et al.8 Texas and
Australia; prospective audit
with 12 months follow-up;
n=15; funder was
International Spinal
Injection Society
van Wijk et al.28 the
Netherlands; multicentre,
randomized, double-blind,
sham treatment controlled
trial; n=81; funder was
Dutch Health Insurance
Council & Pain Expertise
Center Nijmegen
Treatmentt and
d Outcome
Comparator
RFN coagulation of
nerve for 8 to 10 mm
(90 seconds @85C);
no control group
treatment group
(n=40): RFN of medial
branch of dorsal
ramus (60 seconds
@80C); control group
(n=41): identical
procedure but no RFN
e Measuress
at 1.5, 3, 6, and 12 months: VAS
for pain; McGill Pain
Questionnaire; Roland-Morris
Inventory; NASS Treatment
Expectations; isometric push
and pull, lift tasks, dynamic floor
to waist lift, isometric above
shoulder lift; electromyography
of L2 to L5 multifidus bands
primary: number of successes at
3 months; secondary: GPE and
SF-36
Results
all but lifting tasks and
push-pull tasks improved
significantly; 60% had 90%
pain relief; 87% had 60%
pain relief; scores not
significantly different
throughout 12-month
follow-up
primary: number of
successes in treatment
group 11 (27.5%) versus 12
(29.3%) in control group
(p=0.86); secondary: only
GPE showed significant
difference in favour of
treatment group

GPE=global perceived effect; NASS=North American Spine Society; RFN=radiofrequency neurotomy; SF-36=Short Form 36;
VAS=visual analogue scale.

research inconclusive. The most rigorous assessment

Concurrent Development
While there are some practitioners who advocate the
use of pulsed rather than continuous RF current, this
is a modification of the existing technique.32 One
study that was identified recommends intra-articular
(in the joint) RFN rather than medial branch RFN.21
Rate of Technology Diffusion
The diffusion of medial branch RFN in Canada is
unknown. Demand may increase because of its per-
ceived success compared with many conservative
therapies, and because of the social and personal
costs of unresolved chronic back pain. The limited
number of physicians practising interventional
pain management techniques is likely to restrict
availability.
Implementation Issues
Almost 30 years have passed since the first reports
about medial branch RFN were published, yet the pro-
cedure has not gained uptake, and remains an emerg-
ing technology.17 Some studies suggest that medial
branch RFN is efficacious, but procedural and other
methodological shortcomings render much of this
to date suggests that meticulous attention to diagno-
sis and treatment may generate positive results,8 but
this was extracted from a small observational study,
and equally rigorous RCTs have yet to be conducted.
Medial branch RFN is a specialized procedure.
Physicians must place an electrode, depending on its
size, within a millimetre of the target nerve to suc-
cessfully interrupt the nerve supply to the facet joint.
The contrast between the study results of Dreyfuss et
al.8 and Van Wijk et al.28 may highlight the importance
of training, and the need for practice guidelines
related to medial branch RFN.
References
1. Cigna healthcare coverage position: radiofrequency
ablation for chronic spinal pain. Atlanta: CIGNA; 2005.
Available:
http://www.cigna.com/health/provider/medical/pro-
cedural/coverage_positions/medical/mm_0144_cover-
agepositioncriteria_radiofrequency_ablation_for_chro
nic_spinal_pain.pdf.
2. Manchikanti L. Curr Rev Pain 1999;3(5):348-58
3. Lumbar radiofrequency neurotomy for chronic
zygapophysialJoint pain: a pilot study using dual medial
branch blocks. I S I S Sci Newsl 1999;3(2). Available:
http://www.spinalinjection.com/a/newsltrs/Feb1999.pdf.

The Canadian Agency for Drugs and Technologies in Health (CADTH) is funded by Canadian federal, provincial and territorial governments. (www.cadth.ca)
4. Maetzel A, et al. Best Pract Res Clin Rheumatol
2002;16(1):23-30
5. Currie SR, et al. Pain 2004;107(1-2):54-60
6. Rapoport J, et al. Chronic Dis Can 2004;25(1):13-21.
Available: http://www.phac-aspc.gc.ca/publicat/cdic-
mcc/25-1/c_e.html.
7. Manchikanti L, et al. BMC Musculoskelet Disord
2004;5:15. Available:
http://www.biomedcentral.com/1471-2474/5/15.
8. Dreyfuss P, et al. Spine 2000;25(10):1270-7
9. Lord SM, et al. Best Pract Res Clin Anaesthesiol
2002;16(4):597-617
10. Slipman CW, et al. Spine J 2003;3(4):310-6
11. Manchikanti L, et al. Pain Physician 2003;6(1):3-81.
Available:
http://www.asipp.org/documents/Guidelines%20200
3.pdf.
12. Boswell MV, et al. Pain Physician 2005;8(1):1-47
13. Vad VB, et al. Pain Physician 2003;6(3):307-12
14. Bogduk N. Med J Aust 2004;180(2):79-83. Available:
http://www.mja.com.au/public/issues/180_02_190104
/bog10461_fm.pdf.
15. Geurts JW, et al. Reg Anesth Pain Med 2001;26(5):394-
400
16. Manchikanti L, et al. Pain Physician 2002;5(4):405-18.
Available:
http://www.painphysicianjournal.com/2002/octo-
ber/2002;5;405-418.pdf.
17. Niemisto L, et al. Cochrane Database Syst Rev
2005;(4):CD004058
18. Boswell MV, et al. Pain Physician 2005;8(1):101-14.
Available:
http://www.painphysicianjournal.com/2005/jan-
uary/2005;8;101-114.pdf.
19. Gallagher J, et al. Pain Clinic 1994;7(3):193-8
20. van Kleef M, et al. Spine 1999;24(18):1937-42
21. Sanders M, et al. Pain Clinic 1999;11(4):329-35
22. Leclaire R, e al. Spine 2001;26(13):1411-6
23. Tzaan WC, et al. Can J Neurol Sci 2000;27(2):125-30.
Available: http://cjns.metapress.com/media/84vpwn-
qqyh79l4kmexej/contributions/a/a/q/j/aaqjwkhph8jg
6pyg.pdf.
24. North RB, et al. Pain 1994;57(1):77-83
25. Schaerer JP. Int Surg 1978;63(6):53-9
26. Schofferman J, et al. Spine 2004;29(21):2471-3
27. Hooten WM, et al. Pain Med 2005;6(2):129-38
The Canadian Agency for Drugs and Technologies in Health (CADTH) is funded by Canadian federal, provincial and territorial governments. (www.cadth.ca)
28. van Wijk RM, et al. Clin J Pain 2005;21(4):335-44
29. Kornick C, et al. Spine 2004;29(12):1352-4
30. Curatolo M, et al. Spine 2005;30(2):263-5
31. Bassett K, et al. Percutaneous radio-frequency neuro-
tomy treatment of chronic cervical pain following
whiplash injury:reviewing evidence and needs.
Vancouver: BC Office of Health Technology
Assessment; 2001. Available:
http://www.chspr.ubc.ca/bcohta/pdf/bco01-
05T_PRFN.pdf.
32. Mikeladze G, et al. Spine J 2003;3(5):360-2
Cite as: Murtagh J, Foerster V. Radiofrequency neurotomy
for lumber pain [Issues in emerging health technologies
issue 83]. Ottawa: Canadian Agency for Drugs and
Technologies in Health; 2006.
***********************
CADTH appreciates comments from its reviewers.
Reviewers: W. Mark Erwin, DC PhD, Toronto Western
Hospital, Toronto ON; Daniel Denis Vincent, MD FRCPC
ABDA, Vancouver Island Health Authority, Vancouver BC.
This report and the French version entitled La neurotomie
par radiofréquence dans le traitement des lombalgies
are available on CADTH’s web site.
Production of this report is made possible by financial con-
tributions from Health Canada and the governments of
Alberta, British Columbia, Manitoba, New Brunswick,
Newfoundland and Labrador, Northwest Territories, Nova
Scotia, Nunavut, Ontario, Prince Edward Island,
Saskatchewan, and Yukon. The Canadian Agency
for Drugs and Technologies in Health takes sole
responsibility for the final form and content of this report.
The views expressed herein do not necessarily
represent the views of Health Canada or any provincial or
territorial government.
ISSN 1488-6324 (online)
ISSN 1488-6316 (print)
PUBLICATIONS MAIL AGREEMENT NO. 40026386
RETURN UNDELIVERABLE CANADIAN ADDRESSES TO
CANADIAN AGENCY FOR DRUGS AND TECHNOLOGIES IN HEALTH
600-865 CARLING AVENUE
OTTAWA ON K1S 5S8