Manual of RF Techniques

A PRACTICAL MANUAL OF
RADIOFREQUENCY PROCEDURES
IN CHRONIC PAIN MANAGEMENT
3rd Edition

by

Dr. Charles A Gauci MD FRCA FIPP FFPMRCA

Consultant in Pain Management,

Whipps Cross University Hospital, London

U.K.

Graphics by
Basia Jankowiak

Co MEDICAL
Radiofrequency Pain Management
© 2011 CoMedical , Ridderkerk, the Netherlands

No part of this publication may be reproduced or transmitted in any form

or by any means without permission from the copyright owner.

Graphics: Basia Jankowiak, Nottingham, United Kingdom

Print: Koninklijke Van Gorcum BV, The Netherlands

ISBN 978 90 81 7928 0 6

 DEDICATION

To the dear memory of my late

parents without whom none of this

would have been possible

----- - - - ---- - - - - ~- . -- ---------_.

-?".....-:-~
CONTENTS

INTRODUCTION

A THE PHYSICS OF RADIOFREQUENCY & PULSED

RADIOFREQUENCY 16

Section 1: Dr. Eric R. Cosman Jr., Dr. Charles A. Gauci &

Prof. Eric R. Cosman Sr. 16
Section 2: Prof. Eric R. Cosman Sr. & Dr. Eric R. Cosman Jr. 35
Section 3: Prof. Alex Cahana, Prof. Phillippe Richebe' & Dr. Cyril Rivat 39

B RADIOFREQUENCY FACET JOINT DENERVATION

1 . Cervical facets, including the technique of an alternative posterior

approach to the cervical medial branch nerves, by Prof. Miles Day 54
2. Thoracic facets, including the use of Monopolar Cool RF 67
3 . Lumbar facets, including a parallel approach to the lumbar
medial branch nerves by Prof. Phillip Finch 78
4. Sacroiliac joint, including use of the Bipolar RF Palisade
technique by Dr. Eric R. Cosman jr and the use of the
Neurotherm Simplicity Ill™ RF probe 91

C DORSAL ROOT GANGLION RADIOFREQUENCY /PULSED

RADIOFREQUENCY

1 . Cervical dorsal root ganglia 110

2. Thoracic dorsal root ganglia, including use of an alternative
technique 125
3. Lumbar dorsal root ganglia & S 1 nerve 132
D. SYMPATHETIC NERVOUS SYSTEM
RADIOFREQUENCY /PULSED RADIOFREQUENCY
1. Sphenopalatine ganglion 146
2. Superior cervical ganglion 152
3. Stellate ganglion 158
4. Thoracic sympathetic chain 163
5. Splanchnic nerves 167
6. Lumbar sympathetic chain 173
7. Ganglion lmpar 182

MISCELLANEOUS PROCEDURES
1. Trigeminal ganglion radiofreguency 188
2. Intervertebral Disc needling & disc biaculoplasty 197
3. Pulsed radiofreguency of peripheral targets 205
• Occipital nerve 205
• Suprascapular nerve 207
• Other targets 209
};> Upper limb
Gleno-humeral joint 211
Rotator Cuff 211
Humeral epicondyles 211
Elbow joint 212
Wrist joint 212
};> Lower Limb
Hip Joint 212
Trochanteric bursa 212
Kneejoint 213
Ankle Joint 213
Cluneal nerve 213
Lateral femoral cutaneous nerve 214
Coccygeal nerve 215
4 . Percutaneous cervical RF Cordotomy 216
5 . Atlanta-Occipital & Lateral Atlantoaxial joints,
including the modified technique described by Dr. 0. Rohof 223
6. Glossopharyngeal nerve 229
7. Phrenic nerve 231
8 . Pudendal nerve 232
INTRODUCTION

I said in my introduction to the 2nd . Edition of this book that the first edition of
Manual of RF Techniques, which was printed in 2004, was a worldwide
success and was also produced in Korean and Spanish editions.
I was then asked to write a 2nd. Edition, which I did in 2008 .
Both editions have sold out, which is why I have been asked to produce this
3rd. Edition .

Radiofrequency and Pulsed radiofrequency are modalities which are used all
over the world by interventional pain physicians; the techniques in use are
constantly being refined and modified as more targets are located; in my
writing, I always ensure that the Manual keeps up to date with recent
developments.

I have added quite a lot of material to this edition, viz . alternative ways of
targeting the lumbar facet joints, as well as the sacroiliac joint (the latter
includes use of the Cosman TM Bipolar Palisade technique and of the
Neurotherm TM Simplicity probe) .
I have added alternative ways of targeting the thoracic facet joints (using cool
RF) and the thoracic dorsal root ganglia, alternative ways of targeting the
cervical facet joints and the lateral atlanto-axial joint. I have also included
pulsed radiofrequency of the pudendal , phrenic and coccygeal nerves.
In addition the physics section has been rewritten and expanded .

I am very grateful for the contributions of Prof. Alex Cahana & his team , Prof.
Eric R. Cosman Sr. , Dr. Eric R. Cosman Jr. , Prof. Phil Finch, Dr. Olav Rohof
and Prof. Miles Day; these are all big names in interventional pain as well as
being good drinking colleagues at WIP meetings!

I am also very grateful to Enrico Cohen, the Managing Director of CoMedical

who is publishing this edition . He has been extremely helpful in getting this
project off the ground after the previous publisher Flivopress was acquired by
the US-registered Neurotherm Ltd and was consequently precluded from the
field of medical publishing under US law and surrendered copyright.

Basia Jankowiak, my graphic designer has, yet again , provided her great skills
to make this book what it is .
Once again , I hope that interventional pain physicians across the world will
continue to find this Manual of RF Techniques, useful to them in their daily
practise of their chosen field.

Charles A. Gauci

London, UK
August 2011
LIST OF CONTRIBUTORS TO THIS BOOK
(in alphabetical order)

Prof. Alex Cahana MD DAAPM FIPP

Hughes M. & Katherine Blake Endowed Professor
Professor & Chief, Division of Pain Medicine
Professor Radiology (adj .)
Professor Bioethics & Humanities (adj.)
Department of Anesthesiology and Pain Medicine
University of Washington Medical Center, WA, USA

Prof. Eric R. Cosman Sr. PhD

Professor Emeritus, M .I.T. Physics Department.
Boston, USA

Dr. Eric R. Cosman Jr. MEng PhD

Scientific Director, Cosman Medical

Prof. Phillip Finch MBBS DRCOG FFARCS FFPMANZCA FIPP

Specialist in Pain Management & Adjunct Professor
in the School of Psychology, Division of Health Sciences,
Murdoch University , Perth, Australia

Prof. Miles Day MD DABA FIPP DABIPP

International Pain Center
Department of Anesthesiology
Texas Tech University Health Sciences Center
Lubbock, Texas, USA

Prof. Philippe Richebe MD PhD

Associate Professor,
Department of Anesthesiology and Pain Medicine,
University of Washington, Seattle, WA, USA

Dr. Cyril Rivat PhD

Acting Instructor,
Department of Anesthesiology and Pain Medicine,
University of Washington, Seattle, WA, USA.

Dr. 0. Rohof PhD MD FIPP

Sittard, The Netherlands
INTRODUCTION TO 1sr. EDITION (2004)

It is now ten years since publication of my first book on pain relief therapy
'A handbook of clinical techniques in the management of chronic pain' which
I co-authored with my colleague and friend Dr. John Wedley. This book was
a great success and is still being used by many pain relief specialists. Since its
publication, more techniques have been added to our armamentarium and old
techniques have been modified in the light of further detailed anatomical
studies coupled with clinical observations .

Over the past eight years, I have acquired considerable experience in teaching
radiofrequency lesioning techniques at cadaver workshops, worldwide . I have
always found this form of teaching very useful as nothing beats seeing the
procedures in question being demonstrated by an experienced practitioner.
Indeed, I am indebted to many a colleague for instructing me in techniques
I had never carried out before when attending these workshops as a faculty
member. Not only do I teach , I learn!

Following each teaching session, it has always been my practise to jot down
points, which were raised by the participants . I have used these points to
develop my own mode of instruction, which I freely admit is strongly didactic
- probably the result of the fact that I served for so many years in the British
Army! Many participants at my cadaver workshop sessions have repeatedly
requested me to come up with a simple and straight forward 'how to do it'
radiofrequency textbook which they could take with them into the operating
theatre when carrying out procedures; a book which th ey could use to help
them get through the FIPP examination , since this is now fast becoming the
gold standard for interventional pain practice .

This book has been produced in response to these requests.

I have written an easy to follow instruction manual, which I trust, will be of use
to relatively inexperienced colleagues until such a time as they feel confident
enough to do without it. It is not a detailed, exhaustive textbook covering all
the aspects of radiofrequency. Dr. Menno Sluijter's two volumes on the subject
remain the bible as far as I am concerned .
The format used is one of spring bound laminated pages which allows the
reader to add his/her own notes and comments enabling the manual to
become a trusted friend in times of need! In this book, I have attempted to get
straight down to the object of the exercise i.e. 'how to do it'. I have not gone
into the indications for each technique and I have not carried out a detailed
review of evidence.

For evidence on the use of radiofrequency (RF) in spinal pain, I strongly

recommend 'Chronic Spinal Pain' (FiivoPress, 2003) by my colleague
Pauline Newnham. Pain Practice, the journal of the World Institute of Pain also
reviews evidence on RF techniques as it appears from time to time.

Repetition is kept to a bare minimum. Thus, for example, I assume that the
reader is aware that intravenous access should be established before starting
the procedure, that meticulous attention should be paid to sterility and that the
patient should not be on anticoagulant or on antiplatelet medication. The
information on after-care covers the most important and reasonably common
possible complications.

The emphasis on the book is pictorial with text reduced to a bare (heavily
didactic) minimum; this should make it very easy to use .

When using pulsed radiofrequency on peripheral nerves, one is limited only

by one's imagination! I have provided two 'specimen procedures' i.e. Occipital
and Suprascapular nerve pulsed radiofrequency, in the full knowledge that
colleagues have targeted many other peripheral nerves with this very useful
technique . One need only consult a standard textbook on regional blocks in
order to work out the best way to access the nerve in question. Inclusion here
is therefore totally unnecessary. Balogh's work on the use of pulsed
radiofrequency on neuromas and trigger points is a particularly exciting new
development in this field, but I did not consider it necessary to include such
technically simple (albeit effective) procedures in this book.
My sincere thanks to Radionics Inc. and to RDG Ltd. for their financial help
in defraying the costs of the graphics .

I am very grateful to my graphic designer Basia Jankowiak who patiently

worked on the X-rays and amateur illustrations, which I provided in order to
produce graphics of superb clarity. These pictures should greatly assist the
reader as he/ she attempts to carry out the relevant procedures . It was Basi a
who gave me the idea for the 'cookery book' format, which I have adopted,
and which I think admirably serves the desired purpose. In truth, this book
would not really have been possible without Basia's indefatigable input!

I owe an inestimable debt of gratitude to the many radiographers who

uncomplainingly put up with my requests for 'better' pictures, which I could use
in my book! I therefore specifically wish to thank the Departments of Radiology
at Whipps Cross University Hospital, London, King George Hospital, Holly
House Hospital and the BUPA Reding Hospital, all in Essex and the Department
of Radiology at (BMI) Chelsfield Park Hospital in Kent (UK) .

Finally it is a pleasure to acknowledge the help of my good friends

Dr. Bill Cohen, Dr. Olaf Rohof, Dr. J.R. Wedley, Dr. Jan van Zundert and
Dr. Martine Puylaert who provided me with some excellent X-ray pictures.
Their help is specifically acknowledged in the relevant sections where their
pictures have been used .

I am also grateful to my pain fellow, Dr. Serge Nikolic who helped me with
the proof-reading of my draft manuscript.

Charles A. Gauci

London, UK
March 2004
INTRODUCTION TO 2N°. EDITION (2008)

The First Edition of Manual of RF Techniques, which was printed in 2004

proved to be a worldwide success and was sold out by late 2007.

In view of its popularity, it was also published in Korean and Spanish Editions.

My publishers, Flivopress, suggested that due to the demand we should publish

a second edition.

I have therefore taken the opportunity to expand the book.

My good friend Professor Eric Cosman together with his son, Dr. Eric Cosman
jnr, both world-renowned scientists, experts on the physics of radiofrequency
and designers of the Cosman ® RF Lesion Generator, have very kindly added
to the chapter on the physics of pulsed RF.

I have also added some information on the physics of Cooled RF, developed
by the Baylis Medical Company Inc®.

There is also a section on Pulsed Dose, a feature unique to the NeuroTherm ®

1000 and 1 100 RF lesion generators.

The pictures of the main Radiofrequency machines in the Physics section have
been upgraded .

I have added sections on L5 Dorsal Root Ganglion PRF and on Sacral Nerve
PRF.

I have also modified the sections on Sphenopalatine RF /PRF , on Trigeminal RF

and on Percutaneous Cervical Cordotomy.

I have removed the section on RF Annuloplasty, substituting it with a more

general description of intervertebral disc needling together with a description
of the technique of Cooled RF for treating Discogenic pain (Biacuplasty}. The
technique of using Cooled RF for treating sacroiliac joint pain is also included .
The chapter on Sympathetic procedures now also contains the technique of
RF /PRF of the Ganglion lmpar (Walther's Ganglion) .

In view of the fact that pulsed radiofrequency is now also being used to treat
many different targets, including peripheral ones, I have included additional
material on 'target' location in both the upper and lower limbs . I have also
included sections on the Atlanto-Occipital and Lateral Atlantoaxial joints and
on the Glossopharyngeal Nerve.

As always, I am indebted to my graphic designer, Basia Jankowiak for her

skill, patience and invaluable advice.

I trust that this second edition will be as popular as the original version and
will be of use to interventional pain physicians around the world .

Charles A. Gauci
London, UK
June 2008
A
THE PHYSICS OF RADIOFREQUENCY
& PULSED RADIOFREQUENCY

Section 1: Dr. Eric R. Cosman Jr., Dr. Charles A. Gauci &

Prof. Eric R. Cosman Sr. 16
Section 2 : Prof. Eric R. Cosman Sr. & Dr. Eric R. Cosman Jr. 37
Section 3: Prof. Alex Cahana, Prof. Phillippe Richebe' & Dr. Cyril Rivat 43
16 THE PHYSICS OF RADIOFREQUENCY
& PULSED RADIOFREQUENCY
SECTION 1

DR. ERIC R. COSMAN JR. MEng PhD, DR. CHARLES A. GAUCI MD

FRCA FIPP FFPMRCA AND PROF. ERIC R. COSMAN SR. PhD

Radiofrequency (RF)Iesioning refers to the delivery of high frequency electrical

current in the RF range (=500 kHz) to patient tissue via an RF electrode to
induce a biological effect, such as the thermal destruction of nerves that
carry painful impulses. RF methods used in pain management today can be
subdivided by the following broad characteristics, each of which involves
different physical and clinical considerations.

• Waveform / Set Temperature:

o Thermal RF (TRF): The sustained tissue temperature exceeds 42 oc
grossly. A continuous RF (CRF) waveform and tissue temperatures in the
range of 70-90 oc are typical. The clinical objective is gross thermal
nerve ablation. This category includes "cooled RF" methods, where
the electrode is internally cooled, but induced tissue temperatures are
neurolytic.
o Pulsed RF (PRF): The tissue temperature is held at or below 42 oc on
average . RF is delivered in short high-intensity bursts, so that the RF
electric field strength is increased without gross heating. The clinical
objective is neural modification by electric and thermal fields (Cosman
and Cosman, 2005), but the pain relief mechanism remains under
scientific investigation, as described later on in this book by Cahana et al.

• Electrode Polarity:
o Monopolar RF: Current passes between a needle electrode and a large-
area reference ground pad. RF current intensities are highest near
the needle electrode's uninsulated tip. In Monopolar Thermal RF, an
ellipsoidal heat lesion is generated (fig. 1). With proper full adhesion of
the ground pad to the skin, current densities are low over the pad's large
area, and thus nearby tissue is not typically elevated to lesion levels.
o Bipolar RF: Current passes between two needle-electrode tips and the
current density is high at both locations. Thus, in Bipolar Thermal RF, a
heat lesion is generated near both tips . When parallel tips are brought
close together, the electric field is focused between the tips and a large
"strip" lesion is formed (fig. 6a) .

MANUAL OF RF TECHNIQUES
 THE PHYSICS OF RADIOFREQUENCY 17
& PULSED RADIOFREQUENCY
SECTION 1

Monopolar Thermal RF is the

most common and basic form of
RF treatment and has been used
widely in pain management and
neurosurgery since the earliest
RF generators were built by B. J.
Cosman, S. Aronow, and 0. A.
Wyss in the early 1950's (Sweet
and Mark, 1953; Cosman and
Cosman, 1974; Cosman and
Cosman, 1984). In the 1990's,
monopolar Pulsed RF was
introduced by Sluijter, Cosman,
Rittman, and van Kleef (1998)
and is used where conventional
Thermal RF is contraindicated
(e.g. neuropathic pain) or
could be potentially hazardous
(e .g. DRG lesioning). Bipolar
37 44 51 58 55 ·c
Thermal RF between parallel
electrodes has been used in pain Fig. 1 Monopolar Thermal RF: Electric field
management for the last decade (above); Steady-state tissue temperatures (below)
(Ferrante et al., 2001; Burnham and the heat lesion boundary (black)
et al., 2007), but only recently
has the large size of bipolar RF
lesions been fully appreciated (Cosman and Gonzalez, 2011 ). A pioneering
application of Bipolar Pulsed RF has been reported, and this was in the
treatment of carpal tunnel syndrome pain (Ruiz-Lopez, 2008).

In one author's clinical experience (CAG), there are some basic rules which
should be followed in RF lesioning. Thermal RF should be used only for
treatment of nociceptive pain. RF should not be used in patients with marked
psychological overlay and/ or drug dependency. RF should not be used in
patients with total body pain. You should ensure that the patient has realistic
expectations since the total abolition of pain may not be possible . You should
exhaust all other non-destructive forms of treatment first and achieve unequivocal
benefit from preliminary prognostic blocks.

MANUAL OF RF TECHNIQUES
18 THE PHYSICS OF RADIOFREQUENCY
& PULSED RADIOFREQUENCY

SECTION 1

MONOPOLAR THERMAL RF

Using standard equipment, the steps for monopolar RF lesioning in the spine
typically include the following steps :

1. Place the Ground Pad on the skin near the treatment site .
2. Place the RF Cannula percutaneously near the target nerve.
3 . Stimulate: The RF Electrode delivers sensory and motor nerve stimulation
to ensure that the cannula's tip is near the target nerve and distant from
non-target nerves .
4 . Inject anesthetic through the cannula to prevent pain during lesioning.
5 . Lesion : The electrode delivers RF current to the cannula's tip and the nearby
nerve(s) are lesioned with temperature control.

The RF cannula is typically a hollow 22G, 21 G, 20G, 18G, or 16G needle that
is fully insulated except at the tip. The cannula 's hollow interior accepts either
(a) a stilette to make the cannula solid for insertion, (b) injected fluid anesthetics
and steroids, or (c) a 28G thermocouple (TC) electrode for tip temperature
measurement and delivery of stimulation and RF currents . In some applications,
such as cordotomy, DREZ, brain , and even spinal lesioning, the electrode and
cannulae are integrated into a single device . X-ray guidance is typically used
to position the cannula nearby the target nerve by reference to bony landmarks .
Once positioned , the cannula 's stilette is removed and is replaced by the
electrode . The operator then seeks the nerve by sensory stimulation, which are
low-voltage electrical pulses delivered at 50 Hz (pulses per second) . A stronger
sensory response at a lower voltage indicates the cannula's tip is closer to the
nerve . In the clinical experience of one author (CAG), the cannula needs to
be within 3 mm of the nerve in order to create an adequate heat lesion and a
stimulation level of at most 0.6 V is indicative of this .

The operator should always ensure that the cannula/ electrode is not
dangerously close to any motor nerve in the vicinity of the sensory nerve he/she
is trying to lesion. To accomplish this, low-frequency motor stimulation pulses
are delivered at 2 Hz. In the clinical experience of one author (CAG), if no
muscle twitch in the territory of the nerve is noted at twice the voltage strength
necessary to ach ieve sensory stimulation, it can be safely assumed that there
are no motor paths within 3 mm of the needle, and that consequently, there

MANUAL OF RF TECHNIQUES
 THE PHYSICS OF RADIOFREQUENCY 19
& PULSED RADIOFREQUENCY
SECTION 1

is no risk of damage to any motor nerve . When working on spinal nerves,

e.g. medial branches of posterior primary rami, one should not worry about
localized contractions close to the area of needle insertion; one is concerned
with motor twitches at more distant sites, e.g. the arm or the leg.

When the operator is satisfied that the needle is safely in position, RF current
is delivered to the electrode and cannula. Frictional heating occurs near the
cannula's uninsulated tip due to tissue electrolytes being pulled to and fro by
the RF current alternating at approximately 500 kHz (500,000 cycles per
second) . While heating occurs only in the tissue and not within the electrode,
within a few seconds of sustained RF heating, the temperature measured in
the electrode/cannula's tip registers the maximum tissue temperature (Cosman
and Cosman, 2003; Cosman, 2010; fig. 1) . This occurs due to coherent
heat diffusion into the electrode tip from all sides . This maximum temperature
can be directly controlled by the operator. It must be cautioned that for cooled
RF, where the electrode is cooled by internally circulating water, the electrode
does not measure the maximum tissue temperature; rather, the maximum tissue
temperature occurs at a variable location remote of the electrode and can
far exceed the temperature measured within or nearby the electrode (Wright,
2007). As the current is applied at the destructive levels typical of Thermal RF,
a well-circumscribed heat lesion appears . It will grow until a steady-state is
reached; at this point, the passage of current only maintains the temperature .
Little further spread takes place at the edge of the lesion, since (a) the electric
field and rate of heating decreases with distance from the electrode, and (b) the
rate of RF heating within the lesion volume is roughly balanced by the rate of
heat diffusion into the surrounding tissue, heat diffusion into the electrode shaft,
and blood-flow cooling.
The heat lesion is shaped like a match head (fig. 1) and is commonly defined
as the tissue regions for which the temperature exceeds 45-50 oc for at
least 20 seconds (Brodkey, 1964; Dieckmann, 1965; Smith, 1981; Cosman
and Cosman, 1974 and 1984) . Though permanent neurological damage
occurs when tissue is exposed to temperatures exceeding 42 oc over longer
durations (Cosman, Cosman, Bove, 2009), for practical purposes, when we
talk about lesion size, we mean the volume of tissue within the 45 °C isotherm
(fig. 2). According to Abou-Sherif et al. (2003), thermal RF produces the
following effects in the rat sciatic nerve at 6-8 weeks: Wallerian degeneration
in all nerve fibres, physical disruption of the basal laminae, focal disruption
of the perineurium, degranulation of mast cells, recruitment of exogenous

MANUAL OF RF TECHNIQUES
20 THE PHYSICS OF RADIOFREQUENCY
& PULSED RADIOFREQUENCY

SECTION 1

LESION WIDTH VERSUS ELECTRODE

INSULATED SHAFT Tl P DIAMETER AND TEMPERATURE
14 ELECTRODE
TIP DIAM .
~ 12
1,6 MM
l!J 10 (16G)
lii
z
0
e
lii I.I-1.2MM
~ 6 (19 G)
w
~ 4
w
> 0 .25 MM
~ 2 (31 G)
<t
~ 0
30 40 so 60 70 eo so 100
t
TIP TEMPERATURE ( •c) BOILING
POINT

Fig. 2 Monopolar Thermal RF lesion Zone Fig. 2a Post-mortem monopolar thermal RF lesion
and the 45 °C isotherm . Adapted from width around the electrode shaft for different
Cosman and Cosman ( 1984) electrode diameters/gauges and tip temperatures .
Adapted from Cosman et al. (1988)

macrophages, local muscle necrosis, delayed axonal regeneration and

prolonged changes in the microvascular bed (vascular stasis) with extravasation
of erythrocytes, this latter resembling the ischemic changes of re-perfusion
injury.

The heat lesion extends maximally around the shaft of the cannula, with a
diameter that ranges from 2-1 0 mm depending on the cannula's diameter/
gauge, the tip temperature, and lesion time (fig. 2a). The lesion extends
1-2 mm both ahead of the tip and up the shaft, yielding a total length 2-3
mm longer than the tip length (Cosman and Cosman, 1984). Because of this
geometry, many physicians prefer 'para llel' /'side on' cannula placement for
monopolar thermal RF lesioning, so that the nerve is positioned at the side
of the cannula tip where the lesion extends maximally. In the alternative
'perpendicular'/'point-on' approach, the nerve is placed directly ahead of
the cannula tip, thus exposing a smaller volume of the nerve to neurolytic
temperatures.

MANUAL OF RF TECHNIQUES
 THE PHYSICS OF RADIOFREQUENCY 21
& PULSED RADIOFREQUENCY
SECTION 1

For a given electrode/cannula

tip temperature, if lesion size is LESION SIZE VERSUS LESION TIME
plotted against exposure time,
it will be observed that the size ~
'2
increase is relatively linear over :::::1
the early part of the curve, but ~ /Size at Thermal Equilibrum
then begins to slow as the steady :5e¥ ---- - --------------------~
-.::.;
-- ::.;::
-= - ~~

state is approached (fig. 3) . For ....

~
electrode/cannula of the sizes
used in pain management, the
~ ---------
00
steady-state lesion size is not
reached until 30-90 seconds after
the tip temperature reaches its
set value. Thus, the tip should be
20 30 40 50 60
held at the desired temperature Exposure Time in Seconds
for this duration of time to ensure
that the lesion has reached its Fig. 3 Schematic plot of thermal RF lesion size
full spread for that temperature . vs . exposure time to RF current. Adapted from
The steady-state lesion size (fig. Cosman and Cosman (1974)
4) is strongly influenced by the

T1

Permanent
00 T2
-.a...
Q)

45.0
Lesion

_________l ______ _ Zone

...t'O
Q)
42.5
I
I I
of Reversibility
c. I I
E I I I

{E. 37.0
c. ---------+~---r-:--------------- ~~~~erature
j::
Small Lesion (T2) ,
Large Lesion (T1) Distance from Electrode Tip
(arbitrary units)

Fig. 4 Effect of tip temperature on RF lesion size . Adapted from Cosman and Cosman (1974)

MANUAL OF RF TECHNIQUES
22 THE PHYSICS OF RADIOFREQUENCY
& PULSED RADIOFREQUENCY
SECTION 1

tip temperature and electrode/cannula diameter (fig. 2a) . All other things
being equal, a larger heat lesion will be produced by a larger electrode tip
and a higher tip temperature (assuming that boiling does not shut down RF
current flow). Additionally, several factors can affect lesion size and dynamics,
including variations in tissue densities, proximity to bone, proximity to CSF
(especially in Trigeminal lesions), blood vessels, etc.

It is advisable to keep the tissue temperature below boiling ( 100 oq. Boiling
can lead to uncontrolled gas discharges, burning steam that travels up the
electrode's shaft to the skin, irregular lesion geometry, and charring at the
electrode tip.
In one author's clinical practice (CAG), the lesion temperature is held below 85
0
( to give a broad temperature margin relative to 100°C.

The resistance to the flow of electrical current from the tip of the cannula, the
impedance, can be measured and should be observed by the operator. A very
high impedance, or open circuit, can indicate that the electrode or g round pad
is not in proper contact with the patient, or that the cables are disconnected .
A rising, high impedance can also indicate tissue is boiling at the cannula 's
tip, since electrical current cannot easily traverse boiling gas bubble; this is an
important safety check in case the temperature sensor is broken or misplaced
outside the cannula's tip (Cosman, 201 0) . A very low impedance, or short
circuit, can indicate a failure of the RF equipment, or direct contact between
the electrode and the ground pad or contact with a large metallic implant.
Impedance can also be of use in certain procedures since it can indicate the
tissue type in which the cannula's tip is positioned . For example, during a
percutaneous cordotomy, the impedance will be 400 Q when the tip is in the
extradural tissues, fall to 200 Q as the needle tip enters the CSF, and then rise
to over 800 Q as the needle tip enters the spinal cord. When working in the
intervertebral disc, the impedance is usually very high in the outer annulus,
falling to less than 200 Q in the nucleus pulposus .

For facet denervations, some physicians use "pole needles". These are non-
temperature-monitoring, tissue-piercing electrodes with integrated, flexible, fluid
injection lines. They are used when it is felt that the electrode position must not
be perturbed through stimulation, injection, and lesioning. Typically, 20 Volts
is applied with the expectation of producing an 80 oc heat lesion . However,
in vivo clinical experiment shows that the tip temperature is not consistently

MANUAl OF RF TECHNIQUES
 THE PHYSICS OF RADIOFREQUENCY 23
& PULSED RADIOFREQUENCY

SECTION 1

80 °(, but rather can range from values less than 80 oc

to those exceeding
boiling (Buijs et al., 2004; Gultuna, Aukes, van Gorp, Cosman, 2011 ). As
such, when pole needles are used, one should halt RF delivery if an impedance
rise is observed that indicates tissue boiling; and when precise lesion control is
required, one should use temperature-monitoring injection electrodes .
Four standard radiofrequency lesion generators in common use around the
world are shown in fig. 5.

(a) The Cosman G4 Four-Electrode RF Generator b) NeuroTherm NT2000 RF lesion generator

(c) Kimberly Clark Pain Management System (d) DIROS OWL URF-3AP Multi-Lesion

Fig. 5 RF Generators

MANUAL OF RF TECHNIQUES
24 THE PHYSICS OF RADIOFREQUENCY
& PULSED RADIOFREQUENCY
SECTION 1

BIPOLAR THERMAL RF

Whereas a monopolar configuration drives RF current between an electrode's

exposed tip and a distant ground pad, a bipolar configuration drives RF current
between two nearby electrode tips . As bipolar electrode tips are brought closer
together, the resulting thermal lesion shape transitions from that of two volumes
surrounding each tip separately, to that of a single volume connecting the tips
(fig. 6a). The connected geometry and larger total lesion volume are strongly
influenced by a focusing of the electric and current density fields between
closely-spaced electrode tips. Bipolar electrodes can be arranged collinearly or
in parallel, but parallel arrangements produce the largest lesion size increases
(Cosman, Nashold, Ovelman-Levitt, 1984) . Important features of parallel
bipolar heat lesions include:

• Large: Bipolar RF lesions are larger than cooled RF lesions as used in pain
management (fig. 6b; fig. 6c, left) . The size of one bipolar RF lesion is
roughly that of three conventional monopolar RF lesions placed side by side
(fig. 6c, right).
• Conformal: Bipolar RF applied to closely-spaced electrode tips produces
heat lesions shaped like a rounded brick, also known as a "strip lesion".
To conform to anatomical constraints, the width and length of the strip can
be adjusted nearly independently of each other and the lesion depth (fig.
6a). As such, a large lesion can be produced without unnecessary damage
to healthy tissue and with reduced risk to sensitive structures . This is not

Spacing =10 mm 12mm 15mm

~~.i. I . 1- •
: ,, ••. ;.'~ •

. '(~- . . .. ·. ·' '

'
'
·.· ,_.'•·.$
. ',
. .: . -
.
~ .• f .

. . . .. . · ..
.
. .
I . .

Fig. 6a Bipolar Lesion Size for 20 gauge, 10 mm tip length , 90 °C, 3 minutes and increasing
spacing: Strip 12 x 15 x 8 mm 3 (left), Strip 10 x 17 x 5 mm 3 (middle) , Two Ellipsoids 12 x 7 x 7
mm 3 (right)

MANUAL OF RF TECHNIQUES
 THE PHYSICS OF RADIOFREQUENCY 25
& PULSED RADIOFREQUENCY

SECTION 1

possible for monopolar lesions

around a cylindrical electrode
since the lesion width and
depth are the same.
• Connected strip lesions : By
TE
leap-frogging electrodes E
(Ferrante et al., 2001), brick-
like strip lesions can be placed
side-by-side without gaps to
produce an elongated lesion
j_
zone that has consistent
height and thickness (fig 6c,
middle; fig 6d). This is 1 - - -- 22 mm --~
not possible for cooled and
conventional monopolar RF Fig. 6b Bipolar Heat Lesion Size is 15x 22 x
without positioning electrodes 8 mm 3 for 18 gauge, 15 mm tip length , 15 mm
very close together. spacing , 90 °C, 3 minutes
• Robust: Strip lesions can be
generated reliably for parallel tip spacings of 10 mm, tip temperature 90 °C,
and lesion time 3 minutes . Perturbations of these geometric and RF parameters
do not substantially affect lesion size (Cosman and Gonzalez, 201 1; fig. 6a) .
The tip temperature and lesion time used for bipolar RF are greater than those
used for monopolar RF since it is desired that larger heat lesions are formed.

Cooled Monopolar (18 ga, 4 mm tip) Bipolar (20 ga, 10 mm tip) Monopolar (20 ga, 10 mm tip)

E
E
~

Sacral
Blopolar
Surface
Omm
Spacing

10mm 12mm 10mm 12 mm 4. 2 mm

Fig. 6c Comparison of bipolar RF lesion size with that of cooled and conventional monopolar RF

MANUAL OF RF TECHNIQUES
26 THE PHYSICS OF RADIOFREQUENCY
& PULSED RADIOFREQUENCY
SECTION 1

Fig. 6d Palisade Sacroiliac Joint Denervation

As an example, all these features are illustrated by the RF palisade approach

to sacroiliac joint (SU) denervation (fig. 6d). In this approach, 4 to 5 large
bipolar RF lesions are placed side-by-side like bricks in wall to traverse the
region between the dorsal sacral foramina and SU line in which sacral lateral
branch nerves form the SU's dorsal innervation . While each lesion is large in
the inferior-superior direction, its depth is constrained in the left-right direction,
thus reducing the risk of damage to the sacral nerve roots. Because lesion size
is robust to variations in tip spacing, and because adjacent lesions overlap, the
total lesion zone has a consistent thickness and height from the sacral surface .

Bipolar RF lesions of the sizes shown in (fig. 6a) have been used successfully
in pain management (Ferrante et al., 2001; Burnham et al., 2007; Cosman
and Gonzalez, 2011). Ex vivo experiments by Cosman and Gonzalez (2011)
document further flexibility in the size and shape of bipolar lesions. Indeed,
bipolar lesions with dimensions exceeding 2 em can be readily created with
standard RF equipment (fig. 6b) . As for all RF lesioning, before the clinical use
of novel bipolar configurations, a physician must consult lesion-size studies to
determine whether that configuration is appropriate for the target anatomy. The
proximity of target nerves to non-target nerves, blood vessels, skin surface, and
other sensitive structures imposes an upper bound on the safe size of any heat
lesion, especially in the spine.

MANUAL OF RF TECHNIQUES
 THE PHYSICS OF RADIOFREQUENCY 27
& PULSED RADIOFREQUENCY

SECTION 1

MONOPOLAR PULSED RF

While making a radiofrequency lesion in the standard thermal RF mode, the

tissue which surrounds the tip of the electrode is exposed to a concentrated
electric field that induces tissue heating (fig. 1}. The electric field (E-field)
intensity decreases precipitously
with distance from the tip, falling
V (volts)
to a low level at distances beyond
the extent of a typical heat lesion
(Cosman and Cosman, 2005) . 40r-
30
Since the high temperatures within 20-
the heat lesion volume reliably 10 -
induce cellular death, it is assumed or-.-----.---~.----.

8
-10- ~ TIME (seconds)
that the E-field per se has little or no

~E~~5~
clinical effect in Thermal RF.

The introduction of Pulsed RF

(Siuijter et al., 1998) was
motivated by the desire to expose 10mcs
nerves to high electric fields without
gross neurodestructive heating, Fig. 7 Pul sed RF
so as to reduce the risk of RF
treatment in sensitive anatomy such
as the DRG. In the mid-1990's, CRF Waveform
u::
Cosman and Sluijter modified
a standard lesion generator to
deliver radiofrequency voltage
Ii :li!

bursts at a repetition rate of 2

Hz. Since each burst is only 20 PRF Waveform 600KHz
Frequency
msec long, the intervening inactive 1/Pulse Rate Pulse Width

period 480 msec allows heat to u::

dissipate into the surrounding

tissue after exposure to the electric
field (fig. 7 & fig. 7a}. As
such, the RF voltage, and thus the
--'

II
..........,_.
rn :li!

Burst of RF RFOff

E-field strength, can be increased

while holding the electrode tip Fig.7a; Schematic RF waveforms for CRF and
temperature at or below 42°C, a PRF. (Parameters and limes not to scale.)

MANUAL OF RF TECHNIQUES
28 THE PHYSICS OF RADIOFREQUENCY
& PULSED RADIOFREQUENCY
SECTION 1

level assumed not to produce gross neurodestructive effects (fig. 7b). Cosman
and Cosman (2005) have shown that tissue around the electrode shaft is
broadly exposed to high-intensity E-fields without substantial heating . They also
showed that the very intense electric fields at electrode's pointed tip cause "hot
flashes" during each RF burst. The full details of this physical geometry is given
later on in this book, but some salient points are :

• Ahead of the tip: Within =0.2mm of the electrode point, temperature spikes
into the neurolytic range and above the measured tip temperature during
each burst of RF (fig. 7d). At larger distances and between RF bursts, the
temperature does not substantially exceed that of the electrode tip. While
the electric field is maximal within =0.2mm of the electrode point, it falls
off very quickly with distance ahead of the tip, so that beyond =0.2mm, its
magnitude is smaller ahead of the tip than it is lateral to the shaft (fig 7c).

Electric Field and Temperature During a Pulse

200.000 45

:§
2:. 100.000 41
!![ - jEjPath1

50.000

0.0 0.5 1.0

Distance from Electrode (mm)

Fig. 7c E- and T-fields during the first PRF pulse

for V(RF)=45 V and Pulse Width = 20 msec .

6,000 96,500 187,000 V/m

Fig. 7b {Top) Schematic E-field patterns.

(Bottom) E calculated in tissue for a 22 #
electrode at V(RF) = 45V.

MANUAL OF RF TECHNIQUES
 THE PHYSICS OF RADIOFREQUENCY 29
& PULSED RADIOFREQUENCY

SECTION 1

• Around the shaft: Temperature

does not substantially exceed
the measured tip temperature.
The electric field falls off slowly
with distance and exposes
tissue to electrical forces that
are high in biological terms
and that appear to produce a
disruptive effect (Erdine, Bilir,
Cosman, Cosman, 2009); as
such, its range of influence is
broader around the shaft than
ahead of the tip (fig. 7c &
fig. 7e). 0 100 200 300 400 500 600
Time (ms)
In typical Pulsed RF practice,
the generator is set to target Fig. 7d Hot flashes during a PRF pulse
Pulse Voltage=45 V, Pulse
Width=20 msec, and Pulse
Rate=2 Hz. The generator then
automatically adjusts either the
Pulse Voltage, the Pulse Width,
or less commonly the Pulse Rate
to maintain the temperature at
or below 42 °C for 120 seconds.
Sluijter {personal communication)
further recommends that the tissue
impedance be reduced by the
injection of about 1 ml of local
anaesthetic or normal saline. This
approach is supported by finite-
element calculations of the electric
field that assume directional saline
spread toward nerve (Cosman
and Cosman, 2005a). Dr. Bill
Cohen (personal communication) Fig. 7e E-fields dominate overT-fields in PRF.
also advocates saline injection The opposite is true for CRF

MANUAL OF RF TECHNIQUES