Improving the Safety and Efficacy of Bimodal Electric Tissue Ablation · 2012. 11. 7. · size of ablations in RP-BEA was significantly smaller compared to RFA and BETA (9.1mm vs

1

Improving the Safety and Efficacy of

Bimodal Electric Tissue Ablation

Thesis submitted to The University of Adelaide for the degree of

Master of Surgery

By

Leong Ung TIONG, MB.BS. (Adelaide)

Discipline of Surgery

School of Medicine

The University of Adelaide

Supervisors:

Professor Guy J. Maddern, PhD, FRACS (Principle Supervisor)

Professor Peter Hewett, MBBS, FRACS (Co-Supervisor)

2

Table of Contents

Page Number

Title 1

Table of Contents 2

Thesis Abstract 6

Statement of Declaration 8

Acknowledgements 9

Abbreviations 11

1. Introduction 13

2. Radiofrequency Ablation 15

2.1. History of Radiofrequency Ablation 15

2.2. Principles and Mechanisms of Action 15

2.3. Biological Effects of Hyperthermic Therapy 16

2.4. Radiofrequency Ablation Generators 17

2.5. Radio-Imaging in Radiofrequency Ablation 18

2.5.1. Pre-Ablation Imaging 18

2.5.2. Intra-Ablation Imaging 18

2.5.3. Post-Ablation Imaging 19

2.5.3.1. U

Ultrasonography 19

2.5.3.2. C

Computed Tomography 20

2.5.3.3. M

Magnetic Resonance Imaging 20

3

2.5.3.4. P

Positron Emission Tomography 21

2.6. Complications After Radiofrequency Ablation 21

2.6.1. Haemorrhagic Complications 21

2.6.2. Abdominal Infections 22

2.6.3. Biliary Tract Injury 23

2.6.4. Hepatic Vascular Injury 24

2.6.5. Liver Failure 25

2.6.6. Visceral Organ Injury 25

2.6.7. Skin Burns 27

2.6.8. Tumour Seeding 27

2.6.9. Miscellaneous 28

2.7. A Systematic Review of Survival and Disease Recurrence after

Radiofrequency Ablation for Hepatocellular Carcinoma 29

2.8. A Systematic Review of Survival and Disease Recurrence after

Radiofrequency Ablation for Hepatic Metastases 50

3. Electrolysis and Electrochemical Therapy 66

3.1. Animal Experiments 67

3.1.1. Tissue Temperature 67

3.1.2. 3 Water Content 67

3.1.3. 3 Elemental Concentrations 68

3.1.4. 3 Tissue pH 68

3.1.5. 3 Gas Productions 68

3.1.6. 3 Cellular Histological Changes 68

4

3.1.7. 3 Volume of Tissue Ablation 69

3.1.8. 3 Safety 69

3.2. Human Studies 70

3.3. Modifications and Innovations 70

3.4. Problems in Electrochemical Therapy 71

4. Bimodal Electric Tissue Ablation 72

4.1. Early Experimental Results 72

5. Rational for Current Research 76

5.1. Experiment 1: Does Bimodal Electric Tissue Ablation really work by

increasing tissue hydration? 76

5.2. Experiment 2: Where is the optimum place to put the anode in Bimodal

Electric Tissue Ablation? 77

5.3. Experiment 3: Can Bimodal Electric Tissue Ablation be incorporated into

the Cool-Tip RF System? 77

Experiment 1: Bimodal Electric Tissue Ablation – Effect of Reversing the

Polarity of the Direct Current on the Size of Ablation. 79

Experiment 2: Bimodal Electric Tissue Ablation – Ablation Size when the

Anode is Placed on the Peritoneum and the Liver. 94

Experiment 3: BETA compared to standard Radiofrequency Ablation using

the Cool-Tip RF System (Covidien, ValleyLab). 110

6. Area for Future Research 128

7. Conclusions 129

Appendix 1 131

Appendix 2 133

Appendix 3 135

5

Appendix 4 136

Appendix 5 137

Appendix 6 142

Appendix 7 146

Appendix 8 147

Appendix 9 149

Appendix 10 153

Appendix 11 158

Appendix 12 159

References 160

6

Thesis Abstract

Introduction:

Bimodal electric tissue ablation (BETA) is a new method of ablation, which combines the

process of electrolysis with radiofrequency ablation (RFA) to increase the size of tissue

ablations. The cathode of the electrolytic circuit is connected to the radiofrequency (RF)

electrode to increase the surrounding tissue hydration. This allows the RFA process to

continue for a longer period of time and therefore produce larger ablations. Previous research

has shown that BETA could produce larger ablations compared to standard RFA and that it

did not produce any significant short or long-term complications. The studies described here

aim to increase the knowledge on how BETA works to facilitate its translation into clinical

practice to treat liver tumours.

Materials & Methods

The first study tested whether BETA really acts by increasing the hydration of tissues around

the RF electrode. This was achieved by reversing the polarity of the electrolytic circuit, which

theoretically would produce smaller ablations compared to standard RFA. The second study

assessed where would be the best location (skin, parietal peritoneum or liver) for the anode of

the electrolytic circuit during a BETA process. The third experiment determined whether the

principle of BETA could be incorporated into the Cool-Tip RF system, which uses internally-

cooled electrodes (ICEs).

Results

The duration of ablation when the polarity of the electrolytic circuit was reversed (called

reversed polarity bimodal electric ablation, or RP-BEA) were significantly shorter compared

to standard RFA and BETA (48s vs. 148s and 84s respectively, p=0.004). Consequently the

size of ablations in RP-BEA was significantly smaller compared to RFA and BETA (9.1mm

vs. 13.4mm and 11.6mm, p=0.001). The second experiment showed that the size of ablations

were significantly larger when the anode of the electrolytic circuit was placed on the

peritoneum or the liver, compared to when it was placed on the skin (19.7mm and 17.9mm

7

vs. 12.4mm, p<0.001). Lastly, the third experiment showed that the principle of BETA could

be incorporated into the Cool-Tip RF system to produce significantly larger ablations

compared to standard RFA alone (23.1mm vs. 20.1mm, p<0.001).

Discussion

The results from this study confirmed the theory that BETA increases ablation size due to the

effects of increased tissue hydration around the RF electrode. The increased hydration delays

tissue desiccation during an ablation, thus allowing the process to continue for longer periods

of time, therefore producing larger ablations. The efficacy of BETA depends on good

electrical conductivity between the cathode and the anode of the DC circuit. Results from the

second study showed that BETA works best when the anode of the electrolytic circuit was

placed deep to the skin as the stratum corneum consisted of a layer of anucleated cells which

have high electrical resistivity. Lastly, BETA could be incorporated into the Cool-Tip RF

system (Covidien, ValleyLab), which is one of the popular RFA generators in the market.

This means that BETA could be readily incorporated into existing RF generators, therefore

facilitating its translation into the clinical settings.

8

Statement of Declaration

This work contains no material which has been accepted for the award of any other degree or

diploma in any University or other tertiary institution to Dr. Leong Ung TIONG and, to the

best of my knowledge and belief, contains no material previously published or written by

another person, except where due reference has been made in the text. I give consent to this

copy of my thesis when deposited in the University Library, being made available for loan

and photocopying, subject to the provisions of the Copyright Act 1968. The author

acknowledges that copyright of published works contained within this thesis (as listed below)

resides with the copyright holder(s) of those works. I also give permission for the digital

version of my thesis to be made available on the web, via the University’s digital research

repository, the Library catalogue, the Australasian Digital Theses Program (ADTP) and also

through web search engines, unless permission has been granted by the University to restrict

access for a period of time.

1. Tiong LU, Finnie JW, Field JBF, Maddern GJ. Bimodal Electric Tissue Ablation

(BETA) – Effect of Reversing the Polarity of the Direct Current on the Size of

Ablation (published online in the Journal of Surgical Research - 08 February 2011

(10.1016/j.jss.2011.01.013))

2. Tiong LU, Finnie JW, Field JBF, Maddern GJ. Bimodal Electric Tissue Ablation

(BETA): A study on Ablation Size when the Anode is placed on the Peritoneum and

the Liver (published online in the Journal of Surgical Research - 28 February 2011

(10.1016/j.jss.2011.01.061))

3. Tiong LU, Maddern GJ. A Systematic Review of Survival and Disease Recurrence

after Radiofrequency Ablation for Hepatocellular Carcinoma (published in the British

Journal of Surgery, September 2011; Vol 98 (9): 1210-1224)

4. Tiong LU, Field JBF, Maddern GJ. Bimodal Electric Tissue Ablation (BETA)

compared to the Cool-Tip RFA System. (accepted for publication by the Australian and New

Zealand Journal of Surgery)

Leong Ung TIONG (6/10/2011)

9

Acknowledgements

I am thankful for the scholarships provided by the University of Adelaide (Australian

Postgraduate Award) and the Royal Australasian College of Surgeons (WG Norman

Research Fellowship) to make this research possible.

I am also most grateful to the following individuals who have provided invaluable assistance

to me during the course of my research. My work would never have been completed without

their contribution, and therefore my sincerest thanks to them.

First and foremost I would like to thank my supervisors Professor Guy Maddern and

Professor Peter Hewett who have been most supportive and encouraging during the course of

my research. Their constant mentorship ensured I stayed on track while allowing me a great

degree of independence to carry out my research work.

Dr. Martin Bruening for his support enabling me to work part-time at the Department of

Surgery (The Queen Elizabeth Hospital) to keep in touch with clinical practice.

Dr. Christopher Dobbins, who is a friend and a colleague, has been a valuable source of

information from the start of the project to the end. He was generous with advice and

suggestions to help me overcome the various obstacles common in surgical research. He

shared various tips on how to survive the transition from clinical practice into the world of

surgical research, which made it a less daunting experience for me.

The research staff at the Department of Surgery at The Queen Elizabeth Hospital: Ms.

Brooke Sivendra, Ms. Lisa Leopardi, Ms. Sheona Page and Ms. Sandra Ireland who have all

been invaluable to me as the ‘go to people’ whenever I had any administrative issues

concerning my research. Their kind assistance with the complex process of animal research

ethics applications was most appreciated.

10

Mr. Matthew Smith and the staff at the animal research laboratory (The Queen Elizabeth

Hospital) were also very accommodating and helpful with my research work especially with

anaesthetizing the research animals and providing post-operative care to them.

Dr. John Field from the Faculty of Health Sciences (University of Adelaide) provided vital

statistical support for this research project. Dr. John Finnie from the Institute of Medical and

Veterinary Services (IMVS) Adelaide provided histopathological support for this research.

Both provided their time and assistance willingly and freely, for which I am grateful.

Many thanks to Dr. Christopher Lauder who kindly taught me the various surgical procedures

and research techniques involving the research animals, and to Dr. Andy Strickland for

‘brain-storming’ with me.

Last but not least my utmost appreciation to my family for the support and encouragement

that kept me going through this research project.

Dr. Leong Ung TIONG

11

Abbreviations:

AC – alternating current

AFP – alpha-fetoprotein

ASA - American Society of Anaesthesiologists

BETA – bimodal electric tissue ablation

CLM – colorectal liver metastasis

CT – computed tomography

DC – direct current

ECT – electrochemical therapy

FDG – Fludeoxyglucose

HAI – hepatic artery infusion

HCC – hepatocellular carcinoma

HIFU – high intensity focused ultrasound

ICE – internally cooled electrode

IVC – inferior vena cava

LITT – laser interstitial thermal therapy

MCT – microwave coagulation therapy

MRI – magnetic resonance imaging

PAI – percutaneous acetic acid injection

12

PE – perfused electrode

PEI – percutaneous ethanol injection

PET – positron emission tomography

RCT – randomized controlled trials

RF – radiofrequency

RFA – radiofrequency ablation

RNA – ribonucleic acid

RP-BEA – reversed polarity bimodal electric ablation

TACE – trans-arterial chemo-embolization

US - ultrasonography

13

1. Introduction

Radiofrequency ablation (RFA) is currently one of the most popular thermal ablative

therapies for un-resectable liver cancers[1]. It uses alternating electric current (AC) at high

frequencies (200-1200 kHz) to generate thermal energy which causes coagulative necrosis of

the targeted tissues[1]. Besides liver cancers, RFA has also been used successfully to treat

other solid organ tumours including those of the lungs, kidneys, adrenals and the skeleton[2].

Electrochemical therapy (ECT) is another ablative therapy used around the world to treat

various malignancies. It uses a low energy direct electric current (DC) to drive an electrolytic

process at its two electrodes, the anode and the cathode, to produce various cytotoxic

chemicals which cause cellular necrosis[3, 4].

Both RFA and ECT have the advantage of being minimally invasive, with low risks of

morbidity or mortality[3-7]. However the efficacy of RFA, defined as the ability to

completely ablate a tumour, is limited by the small ablation size achievable. This leads to

higher local disease recurrence and lower survival rates compared to surgical resection[8].

ECT on the other hand, has the disadvantage of requiring a long period of time, up to several

hours to administer[9], which may not be practical in the current busy hospital practice.

Therefore, clinical outcomes after RFA and ECT for hepatic malignancies are still inferior

compared to curative surgery[3, 4, 10].

Recently a group of researchers have introduced a new local ablative therapy combining RFA

and ECT, which is called bimodal electric tissue ablation (BETA)[11-14]. BETA uses the

hydration effect produced at the cathode during ECT to enhance the efficacy of thermal

ablations produced by radio-frequency (RF) generators. Their research showed that BETA

was able to produce larger ablations compared to standard RFA[11, 14].

As BETA is a relatively new innovation, there were several questions that needed to be dealt

with before this technology could be introduced into the clinical setting. Firstly it had not

14

been proven that the capability of BETA to produce larger ablations was indeed due to the

increased tissue hydration secondary to the electrochemical reactions of the DC. Secondly,

there was the question of where would be the best placement of the anode, which in previous

research had been shown to cause local tissue injury. Lastly, it was not known whether the

principle of BETA could be incorporated into other types of RF generators in the market

besides the RF 3000 generator (Boston Scientific) used in previous studies.

This research consists of a series of animal experiments performed to answer the above

questions. A literature review on RFA and ECT was conducted, followed by a more detailed

discussion on BETA including its experimental results to date. Lastly the experimental

procedures were described and the results discussed followed by a concluding summary on

this new and promising technique of bimodal electric tissue ablation.

15

2. Radiofrequency Ablation (RFA)

2.1. History of RFA

The pioneering work on RFA was reported by the French scientist d’Arsonval in 1891[15].

d’Arsonval discovered that AC with frequencies over 250 kHz was able to produce heat in

living tissues without causing neuromuscular excitation. This led to the invention of

electrocautery and medical diathermy in the early 1900s[16-18]. Clark reported the use of

RFA to treat breast and skin cancers in 1911[17] and a decade later Cushing and Bovie

developed the “Bovie knife” to treat brain tumours[19]. In 1976 Organ reported the

interactions between AC and biological tissues[20]. He showed that AC at low power causes

ionic agitations in adjacent tissues, which subsequently produced heat by friction. Since then

RFA has been used for a multitude of conditions including cardiac arrhythmias and malignant

tumours in various parts of the body. McGahan et al and Rossi et al, from the United States

and Italy respectively, were two groups of researchers who first reported the use of RFA to

treat liver tumours in 1990[21, 22].

2.2. Principles and Mechanisms of Action

RFA refers to the use of AC that oscillates at high frequencies (300-500 kHz) to destroy

biological tissues[1, 23, 24]. In principle a closed-loop circuit is created by placing a

generator, a dispersive or grounding pad, a patient and a needle electrode in series. The

grounding pad is usually placed on the patient’s thigh, while the needle electrode is inserted

into the centre of the lesion to be ablated. When the RF generator is activated, an alternating

electrical field is generated within the patient between the grounding pad and the needle

electrode. The grounding pad must be sufficiently larger than the needle electrode, thus

ensuring that the electrical current is concentrated around the needle electrode. As biological

tissues have higher electrical resistance than the metal electrodes, the electrical currents will

cause ionic agitation within the cells adjacent to the needle electrode as they attempt to

follow the changes in directions of the alternating current. The ionic agitation will produce

frictional heat that subsequently destroys the tissue if the temperature rises to an adequate

level.

16

2.3. Biological Effects of Hyperthermic Therapy

The mechanism of tissue destruction in RFA is due to thermal coagulative necrosis. Tissue

temperature during RFA can increase up to 100°C. The volume of tissue destruction by

coagulative necrosis in RFA is governed by the temperature[25]. A model to describe the

distribution of heat in biological tissue known as the “Bioheat Equation” was described by

Pennes[26] in 1948, and subsequently simplified by Goldberg[27] to:

Coagulation necrosis = energy deposited x local tissue interactions - heat lost

Cellular homeostasis can maintain normal function at temperatures up to 40°C. At higher

temperatures (42-45°C), cells become more susceptible to injury e.g. chemotherapy or

radiotherapy[28]. However, even when exposed to these temperatures for prolonged periods

of time, viable cells could still be observed[28]. Irreversible cellular damage occurred when

cells were exposed to a temperature of 46°C for 60 minutes[29]. Exposure to temperatures

beyond 50-52°C will shorten the time required to cause lethal cell injury exponentially[30].

As temperatures reach 60-100°C, coagulation of protein and cellular death is near

instantaneous[27, 31, 32]. Temperatures greater than 100°C causes intra- and extra-cellular

water to boil, vaporize and the surrounding tissue to carbonize. The resultant gas and charred

tissues act as electrical insulators preventing further heat deposition. Hence the optimal target

temperature to achieve and maintain is between 50-100°C[33, 34].

Besides protein coagulation, thermal energy also produces vascular changes characterized by

microvascular cell swelling and disruption, intravascular thrombosis, and neutrophil

adherence to venular endothelium. A few experimental RFA studies also demonstrated

secondary anticancer immunity due to activation of tumour-specific T-lymphocytes[35].

These secondary effects may explain the ongoing tissue necrosis after cessation of RFA.

17

2.4. RFA Generators

There are a variety of different RFA generators available commercially, each with slightly

different configurations. Two of the more popular RF generators, which were used in this

research project, were the Cool-Tip RF System (Covidien, formerly ValleyLab) and the RF

3000 system (Boston Scientific). Each machine has its own ablation algorithm, monitoring

systems and needle electrodes.

1. Cool-Tip RF System (Covidien) – This generator is capable of producing 200 watts of

energy at 480 kHz. It uses internally cooled electrodes (ICEs), which come in the

straight single needle applicator or cluster systems (3 single electrodes spaced 5mm

apart and grouped equidistantly in a triangle). A peristaltic pump is used to circulate

chilled saline throughout the needle electrode. This reduces the tissue temperature

immediately adjacent to the electrode to prevent premature charring/desiccation. Each

ablation is started with baseline tissue impedance measurement and internal cooling

of the needle electrode for 1 minute before RFA[36], followed by maximal power

ablation. The generator continuously monitors tissue impedance and temperature

during an ablation process. When tissue impedance rises more than 10 Ohms (Ω)

above baseline, the ablation process is automatically paused for 15 seconds before the

generator delivers anymore energy[36]. The generator shuts itself automatically after

12-15 minutes[37]. The intermittent pauses when tissue impedance increases allows

gases adjacent to the electrode to dissipate while the internal cooling with saline

minimizes tissue charring, hence improving the delivery of energy to surrounding

tissues.

2. RF 3000 System (Boston Scientific) – This generator is also capable of producing up

to 200 watts of energy at 480 kHz. It uses expandable needles with an umbrella

configuration. The expandable needles increase the surface area of the electrode in

contact with the liver tissues, thus increasing the size of ablations. This machine

requires the operator to set the power output manually in a stepwise incremental

manner to avoid early tissue boiling and charring[37]. Power output is stopped

automatically when “roll-off” occurs, defined as when tissue impedance rises

significantly and prevent further conduction of electricity. After a brief pause, a

second cycle of RFA is started at a lower power setting. The whole ablative process

finishes when the second roll-off occurs. The expandable electrode system can

achieve ablation sizes between 3-5cm[24].

18

2.5. Radio-Imaging in RFA

Radio-imaging plays a critical role in the field of local tumour ablation throughout the whole

treatment process. In the pre-operative setting, radio-imaging is used to estimate the size of

the lesion and the anatomical location. Intra-operatively, it is used to guide electrode

placement and real-time monitoring of the ablative process. Finally, it is used to assess the

efficacy of tumour ablation post-procedure to ensure that the entire tumour is destroyed.

2.5.1. Pre-ablation Imaging

Various imaging modalities (ultrasonography (US), computed tomography (CT) and

magnetic resonance imaging (MRI)) can be used in the pre-operative setting based on

availability, operators’ preference and experience, and the individual characteristics of the

patients and lesions. However almost all patients these days will receive contrast-enhanced

CT or MRI scans to allow accurate tumour localization and volume calculations. Contrast

enhanced scans are also useful for comparison with the post-operative scans to detect residual

un-ablated tumour tissue.

2.5.2. Intra-ablation Imaging

Ultrasonography (US) is one of the most popular modalities used because it is inexpensive,

easy to use and safe. US has a major role in guiding the placement of the electrodes

regardless of whether RFA is conducted percutaneously, laparascopically or intra-

operatively. However real-time monitoring of the ablative process using conventional B-

mode US is unreliable as it can potentially under- or over-estimates the completeness of

tumour ablation. This is because the hyperechoic focus observed around the distal electrode

tip is a result of gas micro-bubbles from the vaporization of intracellular water in the heated

liver tissue, instead of the coagulated tissue per se[38]. Boehm et al reported that any fatty

tissue surrounding a tumour quickly becomes hyper-echogenic during RFA which makes

visual monitoring of the actual tumour during ablation impossible[39].

19

2.5.3. Post-ablation Imaging

Repeat scans should be performed soon after RFA (between 1-4 weeks), to detect any

residual tumour so that re-treatment can be planned, and then at regular intervals afterwards

(every 3-6 months) to detect any progressive or new tumours[40]. The general consensus on

the imaging feature that suggests complete tumour ablation is the disappearance of previously

seen vascular enhancement on contrast enhanced imaging[41]. However, radio-pathologic

correlation study have shown that contrast-enhanced CT/MRI is accurate to only within 2-3

mm[27]. These techniques are limited by their spatial resolution in detecting small foci of

peripheral tumour which are potential sources of tumour recurrence[24]. Therefore, all RFA

should include a 1 cm ablative margin of normal tissues to ensure complete eradication of

malignant tumour[24].

2.5.3.1. Ultrasonography

Conventional US is reported to be unreliable in assessing therapeutic efficacy of RFA, and is

difficult to use for assessing tumours in the hepatic dome. Raman et al[42] studied the radio-

pathologic correlation of US in RFA, and found that early US is poorly correlated with and

tends to under-estimate the true size of the ablated lesion. RFA produces an echogenic cloud

on US that quickly dissipates when the procedure is terminated, leaving a predominantly

hypoechoic lesion with a smaller central echogenic nidus[27]. New technology such as colour

and power Doppler US have improved their efficacy, but it is still an inadequate

discriminator of ablated versus viable tissues [27]. Micro-bubble US contrast agents have

been used to differentiate between perfused and non-perfused tissue and allow more an

accurate detection of residual tumour after RFA in both hepatocellular carcinomas (HCC) and

liver metastases[43]. Recent research has seen the development of contrast-enhanced

wideband harmonic gray-scale sonography – which further improved the colour and power of

Doppler US by cancelling signals from stationary tissues to show only signals generated by

microbubble contrast agents[44]. This has enabled the examination of tumour perfusion flow

and significantly increased the accuracy of US in the detection and characterization of liver

lesions[45]. Meloni et al reported a study which found contrast-enhanced pulse inversion

harmonic sonography more sensitive than contrast-enhanced power Doppler sonography in

the detection of residual tumour (83.3% vs. 33.3%, p<0.05)[46]. In a study comparing

contrast-enhanced gray scale harmonic US and contrast enhanced CT performed within 1

20

month after RFA for HCC, Choi et al reported equal efficacy between the two modalities

[47]. However, contrast enhanced axial imaging (CT or MRI) is still considered the most

sensitive modality, and hence the gold standard, in assessing RFA efficacy for patients with

HCC [46].

2.5.3.2. Computed tomography

Multi-phasic helical CT has been shown to accurately differentiate between ablated and

viable residual tumour[41]. Cha et al[38] compared CT vs. US, and reported that un-

enhanced CT had the best correlation to pathologic size (r = 0.74), followed by contrast-

enhanced CT (r= 0.72) and sonography (r= 0.56). Contrast enhanced CT performed best in

characterizing the shape of the lesion, but tends to over-estimate the ablated zone because of

the ischaemic areas peripheral to the ablated lesion. Ablated tissues appeared as

homogenously hypo-attenuating area having well defined borders. In early scans taken soon

after RFA, the volume or size of the ablation should be equal to the pre-procedure scans, or

ideally larger to achieve the 1 cm ablative margin. These early scans may also show a rim of

hyper-attenuation around the ablated lesion during the arterial phase which corresponds to an

inflammatory reaction to the thermal damage seen at histopathologic examination[48]. This

hyperaemic rim, which gradually dissipates with time, may limit the detection of residual

tumour tissue in the periphery.

2.5.3.3. Magnetic Resonance Imaging

Un-enhanced T1- and T2-weighted MRI after RFA produces heterogeneous signal intensity

within the ablated lesion[27]. This variability in signal intensity throughout the ablated region

is most likely caused by an uneven evolution of the necrotic area and the host response to

thermal damage ablated tissues appearing as areas with low signal intensity on T2-weighted

spin-echo images. Therefore contrast-enhanced MRI is recommended to assess therapeutic

efficacy of RFA. Viable tumour cells produce moderately hyper-intense signals on T2-

weighted images associated with corresponding enhancement on contrast-enhanced T1-

weighted images[49]. Coagulation necrosis appears as a markedly hypo-intense area with loss

of gadolinium enhancement on dynamic post-contrast scans[50]. Any viable residual tumours

show the typical and similar signal intensity and enhancement compared to the pre-RFA

scans. Similar to CT scans, the rim of enhancement surrounding the ablated tumour

corresponding to inflammatory reactions can be observed. However in contrast to CT, this

enhancement may persist up to several months after ablation. A new technology currently in

21

evaluation is the use of “heat-sensitive” sequences to monitor the ablation procedure in real-

time[51].

2.5.3.4. Positron Emission Tomography

Functional imaging with FDG radionuclide scanning has been gaining popularity. The avid

uptake of fluorine-18-labelled deoxyglucose (18F-FDG) by tumour tissue has been used to

accurately detect residual disease by positron emission tomography (PET)[52]. A concern

with PET scans is the possibility of false-positive results as the inflammatory cells and tissues

after RFA can display signals similar to tumour tissues. However, a study by Khandani et

al[53] showed that an early PET scan (within 48 hours of RFA) infrequently showed

inflammatory uptake. They concluded that early PET after RFA might be useful by indicating

macroscopic tumour-free margin as total photopenia and macroscopic residual tumour as

focal uptake. Donckier et al[54] reported PET to be more accurate in detecting residual

tumour tissue compared to contrast-enhanced helical CT.

2.6. Complications after RFA

RFA has been shown to be a safe procedure in various studies published in the literature. Its

morbidity (2.2-10.6%) and mortality rates (0.3-1.4%) are much lower compared to surgical

resection[5, 6], therefore making RFA a very useful option for patients who have multiple co-

morbidities or at high surgical risks[6]. Obviously the risks are much greater if RFA is used

during or in combination with surgical resection. The overall mortality and morbidity rates

have been reported to be 7.5% and 50-60% respectively[6]. As RFA becomes increasingly

popular, several large series have been published reporting the complications encountered

after RFA of hepatic tumours[5, 6, 55, 56].

2.6.1. Haemorrhagic Complications

Bleeding is one of the most common complications following radio-frequency treatment of

liver tumours. The mechanisms involved include coagulopathy as a result of underlying

hepatic impairment such as cirrhosis[57], mechanical trauma from the needle electrode

during placement, and thermal injury to adjacent hepatic vessels.

22

Mulier et al[57] reported a total of 60 out of 3670 (1.6%) cases of abdominal bleeding, of

which 0.7% were intra-peritoneal, 0.5% sub-capsular, 0.2% intra-hepatic while the rest were

abdominal wall and non-specific haemorrhage (0.2%). Akahane et al[55], in a study of 1000

RF treatments for 2140 lesions in 664 patients, reported a rate of 0.2% for haemorrhage

requiring transfusion. De Baere et al[5] reported a 0.3% rate of sub-capsular haemorrhage in

their study involving 312 patients who had a total of 350 procedures. In a large multi-centre

trial in Italy involving 2,320 patients with 3,554 lesions, Livraghi et al[6] reported that the

rate of peritoneal bleeding requiring intervention was 0.3%. In the Korean Study Group of

RFA involving 1139 patients, the prevalence of haemorrhage was 0.46% [58].

Several key precautions have been identified to reduce the risk of bleeding after RFA. Image-

guided electrode placement is mandatory for accurate tumour targeting and to avoid large

vessels [56]. Cauterization of the electrode track has also been shown to reduce the risk of

haemorrhage [59]. In a review by Mulier et al[57], none of the 214 patients who had

cauterization of their electrode track experienced haemorrhage, compared to 10 of 1036 (1%)

of patients who did not have cauterization and bled.

2.6.2. Abdominal Infections

Abdominal infections are usually the result of enteric bacterial contamination after a RF

treatment. Factors increasing the risk of abdominal infections include abnormal biliary tract

anatomy (e.g. bilio-enteric fistula or anastomosis) leading to bacterial colonization and a

compromised immune system (e.g. type 2 diabetes mellitus).

Mulier et al[57] reported a total of 42 abdominal infections in 3670 patients (1.1%), of which

34 (0.9%) were hepatic abscesses. Four patients died as a result of sepsis; two from hepatic

abscesses, one from peritoneal Staphylococcus aureus infection and one from septic ascites.

de Baere et al[5] reported a 2% rate of hepatic abscess in his study involving 350 procedures

in 312 patients. In this study, all three patients who had bilio-enteric anastomoses developed

hepatic abscesses. In the Italian multi-centre study, six (0.3%) cases of intra-hepatic

23

abscesses were identified, of which two were diabetic and three had bilio-enteric anastomoses

[6]. Choi et al[60] reported that hepatic abscesses developed after 13 ablations in 13 patients

out of a total of 751 procedures (1.7%). Their analysis revealed that three factors were

associated with significantly higher rates of hepatic abscesses; pre-existing biliary

abnormality (p = 0.0088), tumour with retention of iodized oil from previous transcatheter

arterial chemoembolization (OR=3.381, p = 0.040), and treatment with an internally cooled

electrode system (OR=12.434, p = 0.016). In the multi-centre Korean study, hepatic abscess

was the most common complications with a prevalence of 0.66%[58].

Early diagnosis of abdominal infections can be challenging, as patients often experience low-

grade temperature and mild leukocytosis as part of the post-ablation syndrome. Several

reports indicated that the fever, associated with post-ablation syndrome, usually lasts 1-9

days[56]. Therefore, one should be suspicious of an infective process if fever persists for

longer than two weeks[56]. The commonest microorganisms that have been identified in

abscesses after hepatic ablation include Escherichia coli, Clostridium perfringens,

Streptococcus D and Enterococcus[61]. Treatment modalities include percutaneous aspiration

and antibiotics; a logical choice would be amoxycillin plus clavulanate that is active against

these microorganisms.

2.6.3. Biliary Tract Injury

The main bile ducts are protected by the heat-sink effect of the portal vein and the hepatic

artery that run along-side them[62]. However biliary tract injury can occur when the blood

flow to the liver is decreased by Pringle’s manoeuvre, portal vein thrombosis or vascular

injury. Aggressive heating of central hepatic tumours adjacent to the porta hepatis to

overcome the heat-sink effect of the larger vessels can also damage the biliary tract[58].

Previous studies reported that only bile ducts adjacent to small (<3mm) thrombosed blood

vessels are destroyed[63].

In Mulier’s review, a total of 38 out of 3670 (1%) patients experience biliary tract

complications, of which 18 (0.5%) were biliary strictures and 7 (0.2%) were bilomas[57]. In

24

another study, Kim et al[64] reported that bile duct changes occurred in 69 of 571 (12%)

treatments and 66 of 389 (17%) patients. The average time interval to the discovery of bile

duct change was 1.6 months, and 69 of the patients (87%) had no progression of the injury

[64]. All the bile duct changes noted in this study occurred peripheral to or within the

ablation zones. The most common biliary tract changes seen were upstream biliary tract

dilatation peripheral to the ablation zone (57 patients or 82.6%) followed by biloma (four

patients or 5.8%) [64]. Eight patients (11.4%) had both features on follow-up imaging scans

[64]. In the Italian study involving 2320 patients, biliary tract strictures occurred in six

(0.3%) patients including one patient who needed a stent, and three (0.1%) patients developed

bilomas of which one required drainage [6]. The Korean study of 1139 patients reported three

bilomas (0.20%) and one biliary tract stricture (0.07%) [58].

Two methods have previously been described to prevent biliary tract injury during RFA of

central tumours. The first method involved the prophylactic insertion of a biliary stent [65],

while the second method involved cooling the biliary ducts with chilled saline [66]. There are

concerns however, that these methods might introduce bacterial contamination into the biliary

tract resulting in infective complications [56].

2.6.4. Hepatic Vascular Injury

Vascular thrombosis after RFA occurred most commonly in small vessels <4mm[67] whereas

vessels >4mm were usually spared because of the “heat sink” effect of blood flow[63].

However, thrombosis can occur in hepatic vessels >4mm if blood flow is reduced, for

example by the Pringle’s manoeuvre[68], or in someone who has poor hepatic reserve[57].

Traumatic injury of the hepatic vessels can also occur from insertion of the electrodes.

Mulier’s review reported 22 (0.6%) cases of hepatic vascular damage, of which nine were

portal vein thrombosis, two hepatic vein thrombosis, nine hepatic artery damage and two

unspecified hepatic infarction[57]. Three out of the nine portal vein thromboses resulted in

death[57]. They found that RFA with the Pringle’s manoeuvre increased the risk of portal

vein thrombosis compared to RFA without the Pringle’s manoeuvre (2.1% versus 0.2%,)[57].

25

One patient with hepatic artery damage had extensive hepatic infarction resulting in

death[57]. De Baere et al reported 11 (3%) cases of vascular thromboses after RFA – five

hepatic vein, three segmental portal vein and three portal trunk (all patients with portal trunk

thromboses passed away)[5]. They found a significantly higher rate of portal vein thrombosis

when RFA was performed in combination with the Pringle’s manoeuvre in cirrhotic (two of

five patients) compared to non-cirrhotic livers (0 of 54 patients) (p<0.00001)[5]. Livraghi et

al reported nine patients who developed arterioportal shunt discovered incidentally on

follow-up CT scans, and one patient who developed portal hypertension, portobiliary fistula,

hemobilia, phlebitis, and acute thrombosis, with portal venous cavernous transformation[6].

Akahane reported a 0.4% rate of portal vein thrombosis[55].

2.6.5. Liver Failure

Liver failure is a rare but serious complication of RFA. The common causes of hepatic failure

reported in the literature are portal vein thrombosis and excessive ablation[56].

Mulier reported 29 (0.8%) patients who developed hepatic failure, of which seven (0.2%)

were fatal and 22 (0.6%) were mild[57]. Four of the fatal cases were secondary to central

vascular thrombosis, and the other three due to over-estimation of liver reserve[57]. de Baere

reported one case of fatal liver failure after radiofrequency treatment combined with a right

hemi-hepatectomy[5]. The Italian study reported three (0.1%) cases of rapid hepatic

decompensation (all in HCC) with one resulting in death, and 11 (0.5%) patients, all with

liver cirrhosis, experienced transient hepatic decompensation[6]. The Korean study reported

only one (0.09%) case of hepatic failure[58].

2.6.6. Visceral Organ Injury

RFA has been reported to cause iatrogenic injury to various intra-abdominal organs and

structures such as the gallbladder, small and large bowel, stomach, kidneys, diaphragm and

the abdominal wall. Recognized risk factors for visceral organ injury include the use of high

power RF generator and prolonged ablation time. Ablation of sub-capsular tumours or a

central tumour abutting vital structures also increases the risk of iatrogenic thermal injury.

26

Percutaneous RFA has a higher risk compared to either laparascopic or open RFA[56],

especially if the patient has abdominal adhesions from previous abdominal surgery.

Mulier et al reported a total of 19 (0.5%) cases of visceral organ injury among 3670 patients,

all of which occurred in patients who received percutaneous RFA; five cholecystitis, five

diaphragmatic burns, two colonic burns, one gastric burn, one jejunal burn, two renal burns,

two abdominal burns and one non-specified burn[57]. In the Italian report, major visceral

thermal injury occurred in seven (0.3%) patients – six colonic perforations (four who had

previous bowel resection), and one cholecystitis[6]. Ten (0.4%) patients had minor

complications secondary to visceral thermal injury – six asymptomatic gallbladder wall

thickening, three thickening of the diaphragm and one direct damage to renal tissue without

clinical sequelae[6]. The Korean study reported three (0.3%) cases of complications which

could be attributed to thermal injury – one diaphragmatic injury, one gastric ulcer and one

renal infarction[58]. In the study reported by de Baere, there was one (0.3%) case of colonic

perforation, which resulted in death[5]. Akahane et al reported three (0.5%) cases of

iatrogenic thermal injury from RFA out of 664 patients – one each for gastric, duodenal and

colonic perforation[55].

Awareness of the risk of iatrogenic thermal injury to intra-abdominal organs is critical to safe

RFA. Some authors contraindicated RFA of tumours closer than one cm to other intra-

abdominal organs[69]. They proposed that RFA in these cases be performed either by

laparascopic or open surgery. Several studies in animal models reported that full thickness

burns of the stomach and the small and large intestines could occur if the edge of the RFA

lesion was less than one cm from the surface of the liver[70]. Therefore sub-capsular tumours

should be approached via either laparascopic or open surgery, where these organs can be

separated from the liver[71]. Another method which has been investigated was peritoneal

saline instillation to create an artificial insulating barrier between the surface of the liver and

adjacent structures[72].

27

2.6.7. Skin Burns

Reports of dispersive pad skin burns are increasing with the introduction of high-power RF

generators. Most cases are superficial first and second degree burns, but third degree burns do

occur as well[57]. Specific precautions and care must be taken when placing the dispersive

pads to ensure there is good contact between the skin and the dispersive grounding pad. It is

now recognized that multiple dispersive pads are required to minimize the risk of skin burns,

especially when using high-power RF generators or if the RFA continued for a prolonged

period of time. Goldberg’s experimental study on animals investigated the variables affecting

safe dispersive grounding system for RFA and concluded that up to four 100 cm2 dispersive

pads should be used instead of one[73]. These pads should be placed at equi-distance from

the electrode and with the long edge facing the active electrode.

Dispersive pad skin burns occur in 0.6% in the review by Mulier[57], 0.5% in Akahane’s[55]

report, 0.2% in Rhim’s[58] Korean study, 0.2% in Livraghi’s[6] Italian study and 1.4% in de

Baere’s[5] study.

2.6.8. Tumour Seeding

Several theories responsible for tumour seeding have been proposed. Cancerous cells could

be deposited along the insertion track by the electrode itself during removal, or spread by

bleeding which occurred as a complication of the procedure[57]. Sudden increases in the

intra-tumoral pressure, which might happen during RFA[57] or when an interstitial saline

infusion RF system is used[74], can force cancerous cells into the vascular systems[75].

Pre-procedure tumour biopsy was also found to increase the risk of tumour seeding[76].

Llovet reported a 12.5% rate of electrode track seeding, and identified several factors

associated with this phenomenon - no cauterization of the electrode tract upon removal,

poorly differentiated tumour cells, and a perpendicular approach to sub-capsular tumours[75].

Other researchers have reported a much lower rate of tumour seeding – 0.2%[57], 0.5%[6],

0.6%[77] and 2.8%[78].

28

A practical step to reduce the risk of tumour seeding includes cauterizing the electrode track

during its removal after RFA. Llovet et al cauterized all electrode tracks except for four sub-

capsular tumours, and tumour seeding occurred in all of them[75]. Similarly, radio-imaging

support is necessary to ensure accurate electrode placement to prevent multiple re-

positioning. If the electrode has to be re-positioned to enable a complete ablation, then each

electrode track should be cauterized.

2.6.9. Miscellaneous Complications

Numerous other complications and adverse events had been reported following RFA albeit in

a very small number of cases. In Mulier’s review, there were 0.8% pulmonary complications

(including pneumo-haemothorax and pleural effusion), 0.4% cardiac complications

(arrhythmias, myocardial infarction and cardiac failure), 0.2% coagulopathy, 0.1% renal

failure, 0.2% myoglobinaemia/myoglobinuria and 0.1% hormonal complications (carcinoid

crisis, hyperglycaemia and Addisonian crisis)[57]. Other rare complications that have been

reported include renal infarction[58], sepsis[78], transient ischemic attack[78], cardiac

arrest[6], hypoxaemia[5], haemoperitoneum[78], central hyperthermia[57], brachial

plexopathy[57], and gastrointestinal tract bleeding[57].

29

Title of Thesis: Improving the Safety and Efficacy of Bimodal Electric Tissue Ablation (p.31)

Student Name: Dr. Tiong, LU

CHAPTER 2.7

A Systematic Review of Survival and Disease Recurrence after Radiofrequency Ablation for Hepatocellular Carcinoma

Leong Tiong (MBBS), Guy Maddern (FRACS, PhD)

Department of Surgery, The Queen Elizabeth Hospital

University of Adelaide, SA

Australia

British Journal of Surgery - September 2011; Volume 98 (9): 1210-1224

a1172507

Text Box

NOTE: This article was published as: "Systematic review and meta-analysis of survival and disease recurrence after radiofrequency ablation for Hepatocellular Carcinoma"

30

Title of Thesis: Improving the Safety and Efficacy of Bimodal Electric Tissue Ablatio


Statement of Authorship

Title of Paper: A Systematic Review of Survival and Disease Recurrence after Radiofrequency Ablation for Hepatocellular Carcinoma British Journal of Surgery – September 2011; Volume 98 (9): 1210-1224

Dr. Leong Ung Tiong (Candidate)

Performed literature search, data collection and wrote the manuscript.

I hereby certify that the statement of contribution is accurate.

Professor Guy Maddern

Supervised the development of work, helped in data interpretation, manuscript evaluation and acted as the corresponding author.

I hereby certify that the statement of contribution is accurate and I give permission for the inclusion of the paper in the thesis

31

2.7. Systematic Review of Survival and Disease Recurrence after Radiofrequency

Ablation for Hepatocellular Carcinoma

Introduction

Hepatocellular carcinoma (HCC) is the 5th most common cause of cancer in the world and

the 3rd most common cause of cancer related-death[10, 79, 80]. Surgical resection and liver

transplantation are the only curative options available for these patients, with 5-year survival

rates between 36-70% and 60-70% respectively[81-84]. However only 10-20% of these

patients have resectable disease[85-87]. Factors precluding surgery include extra-hepatic

metastases, vascular invasion, high-risk anatomic location, excessive size or number of

lesions, insufficient remnant liver to support life, or co-morbid conditions[10, 88]. Lack of

liver donors compounds the problem. If untreated, the median survival for these patients is 6-

12 months[89, 90], with few surviving beyond 3 years[91]. Systemic chemotherapy can

increase the median survival of patients with un-resectable HCC to approximately 14

months[92]. However chemotherapy is toxic with unpleasant side effects and less than ideal

disease control capabilities.

There has been a surge of interest in local ablative therapy for un-resectable liver cancers

worldwide in the past 2 decades, which includes cryotherapy, percutaneous ethanol injection

(PEI), percutaneous acetic acid injection (PAI), laser induced thermal therapy (LITT), high-

intensity focused ultrasound (HIFU), microwave ablation and radiofrequency ablation (RFA).

Among these, RFA has been the most widely investigated therapeutic option for un-

resectable liver cancers[1]. It has been shown in numerous large series that RFA is safe, with

minimal morbidity and mortality. RFA has also been shown to achieve satisfactory local

response rate, with >80% complete ablation in most studies[93]. It also significantly

improves overall survival when compared to other modalities e.g. chemotherapy or PEI[86].

General consensus guidelines from North America and Japan where RFA has been used

extensively for HCC recommend that RFA be used for ≤3 HCC which are ≤3cm in diameter

[86, 94, 95].

32

A major drawback of RFA is the high disease recurrence rate seen in patients who received

this treatment. This could have an adverse effect on patient survival, and is the main reason

why RFA is considered inferior to surgery for the treatment of resectable disease. Early RFA

results were limited by the small ablation size achievable, and the lack of sensitive radio-

imaging modalities to assess treatment response.

Intense research over the last 2 decades has produced impressive results. Higher-powered

radiofrequency generators[96] and modifications to the electrodes[97-99] have enabled

ablation sizes up to 6-7 cm in diameter in animal models. Whereas only lesions <3 cm were

treatable with RFA in the past, physicians now can ablate tumours up to 5 cm and still

achieve a 0.5-1 cm ablative margin[100-104]. These advances have opened the door to more

patients who previously were considered “untreatable” and whose options were only

palliation or chemotherapy.

The majority of reports in the literature are case-series, with few randomized controlled trials

comparing RFA to other interventions especially surgical resection. One reason for this is that

the long term outcomes after RFA for liver cancers are considered inferior to resection;

therefore randomizing patients with resectable liver cancers to RFA would be un-ethical. This

review aims to examine the survival and disease recurrence rates after RFA for HCC over the

past decade.

Methods:

A literature search was conducted using Medline (Jan 2000 – week 3 Nov 2010), EMBASE

(Jan 2000 – week 49 2010), Cochrane Central Register of Controlled Trials (Jan 2000 - 4th

Quarter 2010), Cochrane Database of Systematic Reviews (2005 to November 2010),

Cochrane Methodology Register (Jan 2000 - 4th Quarter 2010), Database of Abstracts of

Reviews of Effects (Jan 2000 - 4th Quarter 2010) as per the search terms in Table 1 without

language restriction.

33

1. Catheter Ablation/ or radiofrequency ablation

2. RFA

3. hepatocellular carcinoma or Carcinoma, Hepatocellular/

4. primary liver cancer

5. 1 or 2

6. 3 or 4

7. 5 and 6

8. limit 7 to (comment or editorial or letter or meta analysis or "review")

9. metastas* or Neoplasm Metastasis/

10. 7 not 8 not 9

11. limit 10 to (humans and yr="2000 -Current") Table 1. Search terms used for literature search

Inclusion criteria for studies were as follows (1) Participants – patients with HCC. Patients

who received other therapies (e.g. liver resection, PEI, chemotherapy etc.) prior to RFA for

their HCC were included as data in this area is lacking. (2) Intervention – RFA with any of

the commercially available RFA generators or needle designs. (3) Comparative interventions

– surgical resection, chemotherapy and/or other ablative treatment e.g. PEI, MCT, LITT. (4)

Outcomes data (measured from the time of intervention) including survival rates (overall

median survival, median survival at 1-, 3-, and 5-years, and median disease free survival),

and disease recurrence rates (calculated per patient when data available). Three types of

disease recurrences were recorded; ablation site (tumour recurrence at the site of ablation),

intra-hepatic (tumour recurrence in the liver away from the site of ablation), and extra-hepatic

(tumour recurrence outside the liver). (5) Types of study – randomized controlled trials,

quasi-randomized controlled trials and non-randomized comparative studies were included in

the review. In addition meeting abstracts and each article’s bibliography identified above

were cross-referenced for relevant publications. Only articles reporting survival and/or

disease recurrence >12 months were included in this review. Articles which reported a

combination of RFA with other treatment modalities (e.g. surgery, chemotherapy, other local

34

ablative therapy) were also included. (6) Exclusion criteria – articles were excluded if the

outcome data could not be clearly attributed to each specific intervention (e.g. RFA vs.

resection) or disease (e.g. HCC vs. liver metastases). Meta-analysis, review articles, case

series, letters/comments and editorials were also excluded. Methodological qualities of all

RCTs were assessed using both the Cochrane Collaboration’s tool[105] for assessing risk of

bias and the Jadad scoring system[106].

Studies Adequate Sequence Generation

Allocation Concealment

Blinding (observer)

Blinding (patient)

Adequate follow-up

Jadad Score

Lencioni (2003)[107]

Yes Yes NP NP Yes 3

Lin (2004)[108] Yes Yes NP NP Yes 3

Lin (2005)[109] Yes Yes NP NP Yes 2

Shiina (2005)[110]

Yes NR NP NP Yes 3

Shibata (2006)[111]

NR Yes NP NP NR 1

Ferrari (2007)[112]

Yes NR NP NP NR 2

Zhang (2007)[103]

Yes Yes NP NP Yes 3

Brunello (2008)[113]

Yes Yes NP NP Yes 3

Cheng (2008)[114]

Yes Yes NP NP Yes 3

Yang (2008)[115] NR NR NP NP NR 1

Morimoto (2010)[116]

Yes NR NP NP NR 2

Chen (2006)[117] Yes NR NP NP Yes 3

Table 2. Assessment of Bias in RCTs included. NR=not reported, NP=not possible

35

There were 5 RCT comparing RFA to PEI for HCC which were pooled together for a meta-

analysis using the RevMan 5.1 software[118]. The data were analyzed using the random

effect model of Dersimonian and Laird[119]. The results were reported as pooled risk ratios

with 95% confidence interval. Heterogeneity between studies was assessed using χ2 test with

significance set at p<0.100[120]. The patients in the other studies were too heterogenous for

any meaningful meta-analysis.

Results

A total of 43 articles were included in this review including 12 RCT and 31 non-randomized

comparative studies (Figure 1). The 12 RCTs in this review had moderate methodological

quality, with a mean Jadad score of 2.5 (range 1-3; Table 2). Ten trials described appropriate

methods of generating the sequence of randomization[103, 107-110, 112-114, 116, 117],

while 7 reported methods of allocation concealment[103, 107-109, 111, 113, 114]. Four trials

did not report loss to follow-up[111, 112, 115, 116]. Due to the differences in the nature of

the interventions studied in the RCTs, double blinding was virtually impossible. Patients

treated with RFA could be broadly divided into 2 groups; “un-resectable HCC” and

“resectable HCC”. The patient survival and disease recurrence rates were shown in

Appendices 1-8.

36

Figure. 1 Quorum chart

1. Outcomes after RFA for Un-Resectable HCC

There were 30 comparative studies published in the past 10 years which reported survival and

disease recurrence rates after RFA (used in various combinations with PEI or TACE) for

patients with un-resectable HCC. In some studies RFA was used in combination with surgery

for patients whose disease was otherwise not treatable by resection alone.

Potentially relevant studies identified in the literature search and screened for retrieval (n=1990)

Studies retrieved for more detailed evaluation (n=1628)

Potentially appropriate studies to be included in the systematic review (n=266)

Studies included in the systematic review (n=43)

12 Randomized controlled trials

30 comparative studies

1 case series

362 articles excluded – duplicates

1362 articles excluded – failed inclusion/exclusion criteria or not related to RFA for HCC after reading title/abstract

223 articles excluded - failed inclusion/exclusion criteria after reading full text.

37

1.1. RFA vs. Resection

1.1.1. Within Milan Criteria (Appendices 5 & 6)

There were 16 non-randomized studies comparing RFA to resection for the treatment of

HCC. Eight of these articles included only patients within Milan criteria[121-128]. These

patients were treated with RFA instead of resection because of: (1) patient preferences, (2)

severe co-morbidities, and (3) insufficient post-operative hepatic remnant. The total number

of patients was 928 in the RFA and 708 in the resection group. Median tumour size ranged

from 1.8-2.1 (mean 2.4-3.65) cm in the RFA and 2.0-2.7 (mean 2.5-4.0) cm in the resection

group. Median disease free survival rates at 1-, 3-, and 5-years in the RFA group were 78-

83%[124, 128], 36-59%[122, 124, 128], and 17-25%[122, 124, 128]. The corresponding

figures for the resection group were 80-83%[124, 128], 47-64%[122, 124, 128], and 22-

38%[122, 124, 128]. Median overall survival rates at 1-, 3-, and 5-yr in the RFA groups were

96-100%[121, 123-126, 128], 53-92%[121-126, 128], and 41-77%[122, 124, 126-128], and

in the resection group were 91-99%[121, 123-126, 128], 57-92%[121-126, 128], and 54-

80%[122, 124, 126-128]. Ablation site and intra-hepatic disease recurrence rates in the RFA

group were 7-24%[121, 123, 124, 126, 127] and 28-68%[121, 123, 126, 128], while those in

the resection group were 0-10%[121, 123, 124, 126, 127] and 33-51%[121, 123, 126, 128].

All 8 studies showed no significant differences in overall survival rates between the RFA and

resection groups. However patients treated with resection had significantly lower local

disease recurrence rates[126, 127, 129], and higher disease free survival[124, 128].

1.1.2. Outside Milan Criteria (Appendices 5 & 6)

Eight articles included patients outside the Milan criteria in their comparison between RFA

(n=797 patients) and resection (n=712 patients)[100, 130-136]. Because of larger numbers

and sizes of tumours, 4 of the studies combined RFA with TACE[130, 132, 135, 136]. The

median tumour sizes were 3.0-4.6 cm and 4.6-7.4 cm in the RFA and resection group,

respectively. In 2 studies, tumours size was >3cm in more than 70% of patients[100, 134].

Median overall survival rates at 1-, 3-, and 5-year in the RFA group were 78-98%[100, 131-

38

134, 136], 33-94%[100, 131-134, 136], and 20-75%[100, 131-133, 135, 136]. The

corresponding figures in the resection group were 75-97%[100, 131-134, 136], 64-93%[100,

131-134, 136], and 31-81%[100, 131-133, 135, 136]. Median overall survival was 28-51

months and 37-57 months in the RFA and resection groups, respectively[100, 130, 133].

Median disease free survival was 16-25 months and 36-53 months in the RFA and resection

groups, respectively[100, 132]. The ablation site, intra-hepatic and extra-hepatic disease

recurrence rates were 3-15%[131, 133, 134, 136], 25-59%[131-134, 136] and 12-21%[131,

132] in the RFA group; compared to 0-2%[133, 136], 28-37%[132, 133, 136] and 13%[132]

in the resection group.

Three studies found that patients treated with resection had better overall and disease

free survival compared to those treated with RFA. The survival benefits, however, were

generally limited to patients with Child-Pugh A cirrhosis and single HCC >3 cm[100, 134,

135].

1.2. RFA vs. PEI for Un-Resectable HCC[103, 107-110, 113] (Appendices 1 & 2)

Five RCTs compared RFA (n=354 patients) to PEI (n=347 patients) for the treatment of un-

resectable HCC. The mean tumour diameter was 2.42-2.9 cm and 2.25-2.8 cm in the 2 groups

respectively. Meta-analysis of these trials showed that patients treated with RFA had better 1-

and 3-yr overall survival than those treated with PEI (Fig. 2 & 3). RFA was associated with

significantly better disease free survival rates at 1-and 3-yr; 74-86% and 37-43%, compared

to the PEI group; 61-77% and 17-21% respectively[107-109]. Disease recurrence rates at the

ablation site were significantly lower in the RFA group (2-14%) compared to the PEI group

(11-35%)[107-110].

39

Figure 2. RFA vs. PEI for Un-Resectable HCC (Survival at 1-Year)

Figure 3. RFA vs. PEI for Un-Resectable HCC (Survival at 3-Years)

In another RCT Zhang et al[103] compared the efficacy of RFA + PEI versus RFA alone in

treating un-resectable HCC. A total of 67 patients received RFA + PEI where absolute

alcohol was injected into the tumour followed by RFA, whereas in the 2nd group 66 patients

received RFA only. The RFA + PEI group had significantly better overall survival with 1-, 2-

, 3-, 4-, and 5-year survival being 95.4%, 89.2%, 75.8%, 63.3%, and 49.3%, compared to the

RFA only group; 89.6%, 68.7%, 58.4%, 50.3% and 35.9% respectively (p<0.05). Local

tumour progression rates were also significantly lower in the RFA + PEI group compared to

the RFA only group (6.1% vs. 20.9%, p=0.01). Sub-group analyses revealed that RFA + PEI

Study or Subgroup

Brunello 2008

Lencioni 2003

Lin 2004

Lin 2005

Shiina 2005

Total (95% CI)

Total events

Heterogeneity: Tau² = 0.00; Chi² = 1.61, df = 4 (P = 0.81); I² = 0%

Test for overall effect: Z = 2.27 (P = 0.02)

Events

4

0

11

11

4

30

Total

70

52

52

62

118

354

Events

10

2

15

16

7

50

Total

69

50

52

62

114

347

Weight

13.6%

1.9%

36.7%

36.1%

11.7%

100.0%

M-H, Random, 95% CI

0.39 [0.13, 1.20]

0.19 [0.01, 3.91]

0.73 [0.37, 1.44]

0.69 [0.35, 1.36]

0.55 [0.17, 1.84]

0.62 [0.41, 0.94]

RFA PEI Risk Ratio Risk Ratio

M-H, Random, 95% CI

0.01 0.1 1 10 100

Favours experimental Favours control

Study or Subgroup

Brunello 2008

Lin 2004

Lin 2005

Shiina 2005

Total (95% CI)

Total events

Heterogeneity: Tau² = 0.02; Chi² = 7.04, df = 3 (P = 0.07); I² = 57%

Test for overall effect: Z = 2.41 (P = 0.02)

Events

52

34

24

46

156

Total

70

52

62

118

302

Events

52

46

36

63

197

Total

69

52

62

114

297

Weight

31.2%

28.5%

16.8%

23.4%

100.0%

M-H, Random, 95% CI

0.99 [0.81, 1.20]

0.74 [0.59, 0.92]

0.67 [0.46, 0.97]

0.71 [0.53, 0.93]

0.79 [0.65, 0.96]

RFA PEI Risk Ratio Risk Ratio

M-H, Random, 95% CI

0.01 0.1 1 10 100

Favours experimental Favours control

40

improved overall survival of patients with tumours between 3.1-5.0 cm in diameter, but not

for tumours ≤3.0 cm or 5.1-7.0 cm.

1.3. RFA vs. TACE (Appendices 3 & 4)

Three RCTs compared RFA to RFA + TACE. The data could not be pooled due to the

heterogeneity of patient populations and inclusion/exclusion criteria.

Cheng et al[114] conducted a RCT comparing RFA + TACE (n=96) vs. RFA-only (n=100)

vs. TACE-only (n=95) for patients with up to 3 HCC ≤7.5cm in diameter. The average largest

tumour diameter in the 3 groups was approximately 5.0 cm, while duration of follow-up was

35.8, 24.6 and 25.4 months respectively. Complete tumour response, overall survival,

disease-free survival and disease recurrence were significantly better in the RFA + TACE

group compared to either RFA or TACE alone.

Yang et al[115] randomized 78 patients to RFA (n=12, median size 5.2 cm), TACE (n=11,

median size 6.4 cm), RFA + TACE (n=24, median size 6.6 cm) and RFA + TACE + lentinan

fungal abstract (n=31, median size 6.5 cm). Patients in the last group had significantly better

median survival (28 months) and lower ablation site and intra-hepatic disease recurrence rates

(18%) compared to the others.

Morimoto et al[116] randomized patients with single HCC 3.1-5.0 cm to RFA (n=18, mean

size 3.7 cm) or RFA + TACE (n=19, mean size 3.6 cm). After a mean follow-up of 31

months, the RFA + TACE group had significantly lower disease recurrence rate at the

ablation site than the RFA group (6% vs. 39%, p=0.012). There were however no significant

differences in the median survival rates at 1- and 3-years.

41

Two retrospective comparative studies also showed that the combination therapy of RFA +

TACE produced significantly longer overall and disease free survival, and lower disease

recurrence rates compared to RFA alone[137, 138].

Only 2 studies compared RFA to TACE alone. Chok et al[139] compared 51 patients treated

with RFA to 40 patients receiving TACE and found no significant differences in overall

survival rate at 1- and 2-years, or median disease free survival. On the other hand Murakami

et al[140] reported that patients treated with RFA (n=105) had lower rates of disease

progression/recurrence compared to TACE (n=133).

1.4. RFA vs. LITT (Appendices 5 & 6)

Ferrari et al[112] randomized patients to RFA (n=40) or LITT (n=41) for single HCC ≤4cm

or up to 3 HCC ≤3cm. Mean tumour size of the tumours in the 2 groups were 2.67 cm and

2.89 cm respectively. No significant differences between the 2 groups were found in ablation

site and intra-hepatic disease recurrence rates, median disease free survival, or median

survival rate at 1-yr, 3-yr, and 5-years. Sub-group analysis, however, showed that Child-Pugh

A patients (HR 0.18, p=0.017) and those with tumour ≤2.5cm (HR 0.18, p=0.018) had better

survival rates when treated with RFA compared to LITT.

1.5. RFA vs. MCT (Appendices 5 & 6)

Three observational studies compared RFA (n=171 patients) to MCT (n=151 patients). Mean

tumour sizes were 1.6-2.6 cm in the RFA, and 1.7-2.6 cm in the MCT group. No significant

differences between the 2 groups were found in disease recurrence, disease free survival or

overall survival rates in 2 studies[141, 142], but one study[143] reported that patients treated

with RFA had significantly higher median survival rates at 1-, 3- and 4-years compared to

MCT (100%, 70% and 70% vs. 89%, 49%, 39%, p=0.018).

42

1.6. RFA vs. RFA + Interferon (Appendices 5 & 6)

A matched case-control study compared RFA + interferon therapy (n=43 patients, median

tumour size 1.8 cm) for patients with Child-Pugh A cirrhosis and up to 3 HCC ≤3cm to RFA

(n=84 patients, median tumour size 1.5 cm) alone[94]. Interferon therapy was started after

confirmation of complete response to RFA, and continued for a median duration of 4.7 years.

Patients in the RFA only group received conventional anti-inflammatory therapy consisting

of ursodeoxycholic acid or strong neo-minofagen C. Five year overall survival rate was

significantly higher in the RFA + interferon group compared to RFA-only group (83% vs.

66%, p=0.004), and lower intra-hepatic disease recurrence rates were lower (56% vs. 71%,

p=0.04).

In another study interferon maintenance therapy after RFA in patients with HCC and HCV

positive RNA conferred better overall survival rate (5-yr; 90% vs. 70%, p=0.0181), and

maintained Child-Pugh A classification for a longer period of time (37 vs. 32 months,

p=0.0025) compared to those patients not taking the drug[127].

2. Outcomes after RFA for Resectable HCC (Appendices 7 & 8)

One RCT[117] and three comparative articles[144-146] were identified. As data on RFA for

resectable HCC is lacking, a recently published large series[137] was also included to provide

a comprehensive evidence review with a total of 680 patients.

2.1. Resectable 1st episode HCC (Appendices 7 & 8)

Chen et al[117] randomized patients to RFA (n=90) or surgery (n=90) for single resectable

Child-Pugh A HCC ≤5.0 cm in diameter. Fifty-two percent and 47% of the patients had

tumours ≤3cm in the RFA and resection groups respectively. Nineteen (21%) patients

withdrew consent for RFA post-randomization, and received surgical resection instead.

Analyses of RFA vs. resection including and excluding these 19 patients showed that both

modalities produced comparable overall and disease-free survival rates. Similar outcomes

were achieved regardless of tumour diameter (≤3.0 cm or 3.1-5.0 cm). Surgical resection was

43

associated with higher morbidity rates (55.6% vs. 4.2%, p<0.05), and longer hospital stay

(19.7 days vs. 9 days, p<0.05) compared to RFA.

Two comparative studies looked at the same topic. Montorsi et al[146] compared 40 patients

who had surgical resection to 58 patients who received RFA for a single HCC nodule <5.0

cm in diameter. Baseline patient characteristics were comparable between the 2 groups, and

the average follow-up period was approximately 2 years in both groups. Complete response

after RFA was achieved in 55 patients (95%); 2 patients required TACE and 1 had a

subsequent resection. The RFA group had higher rates of intra-hepatic tumour recurrence

compared to the resection group (35% vs. 30%, p=0.018), but there were no statistically

significant differences in 1-, 2-, 3- and 4-year survival rates (p=0.139).

Abu-Hilal et al[144]compared resection versus RFA in patients with resectable uni-focal

HCC <5.0 cm in diameter and found no significant differences in 1-, 2-, and 5-year overall

survival (p=0.302). However median disease free survival was longer in the resection group

compared to RFA (35 vs. 10 months, p=0.028). Local tumour recurrence was significantly

higher in the RFA group (30% vs. 4%, p=0.001). Multivariable analyses showed that RFA

was associated with reduced overall (HR=4, p=0.014) and disease-free survival (HR=2.3,

p=0.022)

In 224 patients with Child-Pugh A cirrhosis and resectable single HCC ≤5.0 cm with no

extra-hepatic or vascular invasion managed with RFA as first line treatment, median disease

free survival was 48 months[137]. The median overall survival and disease free survival rates

at 5-and 10-years were 60%and 34%, and 36%and 18% respectively.

2.2. Resectable Recurrent HCC (Appendices 7 & 8)

Liang et al[145]compared percutaneous RFA (n=66) versus repeat resection (n=44) for

recurrent technically resectable HCC. Inclusion criteria were; <3 lesions, <5.0 cm diameter,

no other treatment apart from previous hepatectomy for HCC, no evidence of tumour

44

invasion into major portal vein/hepatic vein branches, and Child-Pugh A/B cirrhosis.

Complete response was achieved in 65 patients (98%) who received RFA; 1 patient was

treated with TACE after 2 failed ablation attempts. Four patients in the resection group

received TACE; 2 for ruptured tumour during surgery and 2 for inadequate resection margin.

No significant differences in overall survival, disease free survival or disease recurrence rates

were found between the 2 interventions. Repeat resection was associated with more major

complications compared to RFA (68% vs. 3%, p<0.005). No significant difference between

the survivals of patients treated with repeat hepatectomy or RFA for recurrent tumors ≤3 cm

(p=0.62) or >3 cm (p=0.57) was found.

3. Techniques and equipment

3.1. Comparison of percutaneous and laparoscopic/open RFA (Appendices 5 & 6)

Khan et al[101] compared percutaneous to “surgical RFA” in 228 patients with up to 3 HCC

≤5.0 cm. Percutaneous RFA was performed in 117 patients, while 111 patients had “surgical

RFA” (open=91 patients, laparoscopic=20 patients). More patients in the “surgical RFA”

group had cirrhosis (95% vs. 74%, p<0.001), and higher AFP levels (776ng/ml vs. 193ng/ml,

p=0.05). Thirty-six percent of patients in the percutaneous RFA group had previous liver

resection compared to 19% in the “surgical RFA” group (p=0.03). Both approaches had

similar overall, disease free survival and disease recurrence rates for tumours ≤3 cm.

However for tumours >3 cm, “surgical RFA” had significantly better median survival rates

than percutaneous RFA at 1-year (92% vs. 81%, p=0.03) and 3-years (68% vs. 42%, p=0.03)

respectively.

3.2. RFA generator (Appendices 5 & 6)

In the only RCT available, Shibata et al[111] compared the internally-cooled electrode (Cool-

Tip RF system, Radionics) to the expandable LeVeen electrode (RF 2000 generator, Boston

Scientific) in 74 patients with up to 3 HCC ≤3 cm. Multiple electrode insertions were used to

treat tumours >2.5 cm. No significant differences were found in complete ablation rates,

overall or disease-free survival, or disease recurrence rates between the 2 groups.

45

Lin et al[147] prospectively compared 100 patients with up to 3 HCC ≤4 cm in diameter

treated with 4 different RF generators and their respective electrodes which can ablate an area

of 3.0-5.0 cm (25 patients per group, mean tumour size 2.6 cm). Complete tumour response

rates were 91% in the RF 2000 group and 97% in the other 3 groups (p=ns). There were no

significant differences in disease recurrence, median overall survival or disease free survival

rates at 1- and 2-years between groups.

Seror et al[148] compared RFA for HCC during 2 different periods of time; from 2000-2002

forty-five patients were treated with internally-cooled electrodes (Cool-tip; Radionics/Tyco,

Burlington, Massachusetts), and from 2002-2004 forty-four patients were treated with the

perfused electrode (PE) (Berchtold/Integra, Tuttlingen, Germany). Only patients with Child-

Pugh A/B cirrhosis and up to 3 HCC ≤3.0 cm were included in the study, with no significant

baseline differences in patient or tumour characteristics. Complete ablation rate was 96% in

both groups, but only 15% of tumours treated with internally-cooled electrodes required

multiple RF applications to achieve complete ablations, compared to 72% of tumours treated

with the PE (p<0.00005). Treatment with the PE was also associated with higher rates of

intra-hepatic disease recurrence compared to the ICEs (64% vs. 31%, p<0.01). Median

overall survival at 1-year, 2-years, and ablation site disease recurrence rates were not

significantly different between the 2 groups.

Discussion

This review showed that RFA could achieve good clinical outcomes for un-resectable HCC.

A meta-analysis of 5 RCTs showed that RFA was better than PEI, with higher overall and

disease-free survival rates. Data on RFA compared to LITT or MCT were inconclusive, with

some studies reporting no significant differences between these ablative modalities, while

others showed better results after RFA. A combination of RFA + TACE has been shown to

be superior compared to RFA only therapy. More recently some clinicians have started using

RFA for resectable early HCC within Milan criteria (≤3 HCC < 3cm) and produced

comparable results as surgical resection. Lastly a comparison of different RFA electrodes and

generators showed no significant differences in disease-free or overall survivals.

46

Current RFA capabilities are limited by the size of coagulation that can be achieved in one

RFA application, which leads to incomplete treatment response and consequently higher rates

of tumour recurrence. The inability to completely ablate “larger” tumours, or tumours in high

risk locations[149] e.g. adjacent to large hepatic vessels or in sub-capsular areas compounds

the problem and these are adverse prognostic factors for tumour recurrences. The high local

tumour recurrence rates could have a negative influence on patient survival in the long term,

and is one of the main reasons why RFA is associated with inferior outcomes compared to

surgical resection[150]. In a large series by Kim et al, the ablation site recurrence rates

significantly increased from 0% (when the ablation margin around the tumour was >3 mm) to

6%, 19%, and 23% when the ablation margin was 2-3 mm, 1-2 mm and <1 mm

respectively[151].

There has been a clear evolution in the use of RFA for hepatic malignancies in the past

decade. In the early 2000s, the use of RFA was limited by the size (≤3.0 cm) and number

(<3) of lesions. The rate of successful complete ablation of a tumour, as measured by non-

enhancement of the tumour during contrast-enhanced CT scan, is mainly dependent on its

size[131]. The local tumour recurrence rate after RFA can be up to 40-50%[152, 153], and is

directly related to the incapability to completely ablate a larger lesion. Rhim et al reported a

complete ablation rate of 96.7% for HCC and a 5-year survival rate of 58%[154]. When used

for larger tumours, the complete ablation rate and the long-term outcomes predictably

deteriorated. Chen et al[155]reported an overall complete response rate of 95% after RFA for

hepatic malignancies, and found that the success rate fell to 85 % for tumours >3.5 cm. The

response rates were also lower when tumours were adjacent to the gallbladder (86.3%) and

the bowels (83.3%) respectively.

With better equipment and understanding, the indication for RFA continues to expand, while

maintaining satisfactory outcomes. Clinicians are now commonly using RFA to treat hepatic

tumours >3.0 cm in size, with some even treating tumours >5.0 cm with satisfactory

results[100, 103, 156]. The number of lesions has ceased to become an absolute

47

contraindication to RFA[157]. A multi-modal approach to the treatment of HCC (e.g. RFA +

PEI or TACE) has improved the efficacy of RFA for tumours >3.0 cm[103, 114-116].

RFA has its own distinct advantages compared to surgical resection of HCC. It is minimally

invasive and has much lower rates of morbidity and mortality compared to surgery. Most of

the RFA are performed as day procedures. In addition, RFA is a versatile tool which has

proven to be very useful as it can be performed percutaneously, laparoscopically, and/or in

combination with surgical resection. The combination of RFA and surgical resection provides

a curative option to many patients who previously had inoperable tumours (e.g. bilobar

disease)[158-160].

Furthermore RFA has been used as a “bridging therapy” for patients with HCC while

awaiting liver transplant. Due to scarcity of organ donors, there is a high patient dropout rate

(10-30%) while waiting for liver transplantation[82, 161, 162]. Several studies have found

that pre-transplant treatment with RFA can reduce the dropout rate to 10-20%[82, 163]. In a

prospective study by Mazzaferro et al, RFA was found to be a safe and effective bridging

therapy to liver transplantation as there was no rapid HCC deterioration, tumour seeding or

vascular invasion during the pre-transplant period[164].

Another area where RFA would be useful is for the treatment of recurrent HCC[138, 145,

165]. Several studies have shown that the intra-hepatic tumour recurrence rate after resection

for HCC could be as high as 70% at 5-years[166-169]. Although repeat resection could

provide an effective treatment, it is limited to only 10-30% of the patients[170-173].

Some clinicians are concerned by the much higher tumour recurrence rates, and lower disease

free and overall survival rates in patients who received RFA compared to surgical resection.

Currently there are few “head to head” comparisons between RFA versus surgery in

technically resectable HCC. The majority of the literature available reported results where

RFA was used to treat “un-resectable” tumours which were, most of the time, associated with

advanced disease (e.g. Child-Pugh B/C HCC, or bilobar tumours) or the patient was too sick

48

to undergo surgery. These are adverse prognostic factors which could have a negative

influence on the patients’ outcomes, and therefore comparing RFA to surgery in these

different groups of patients is akin to comparing “apples to oranges”.

The capability of RFA to completely ablate a tumour is the most important principle

underlying its recent success in achieving survival parity with resection, albeit with the strict

criterion that the tumour size is 3.0 cm or less. As the results of this review show, there are no

significant differences in survival rates between RFA and resection for HCC within Milan

criteria[121-128]. When RFA was used for tumours outside Milan criteria, there were

significantly lower overall and disease free survival rates compared to resection[100, 134,

135].

There are now at least 5 reports[117, 137, 144-146], including 1 RCT[117], where RFA was

used to treat small resectable HCC in a carefully selected group of patients (early HCC within

Milan criteria). The results from these reports showed comparable overall survivals between

RFA and surgery, although there is a significantly higher tumour recurrence rate in the

former. Whether the higher tumour recurrence rates have any effect on the overall well-being

and health related quality of life of patients remains to be investigated. Tumour recurrences

under these circumstances can generally be re-treated, which might explain the comparable

overall survival rates between RFA and resection found in these reports.

As research progress continues in the field of RFA, there is little doubt that its indication for

use will broaden to include resectable HCC in the near future. However this progress must be

based on solid evidence from randomized controlled trials.

Addendum

Since the publication of this review article in the British Journal of Surgery, it has been

brought to the authors’ attention that one of the randomized controlled trials included in this

systematic review has since been retracted by its publisher[114]. The reasons for the

49

retraction was because of concerns regarding the validity of the study[174]. The authors were

not aware that this article has been retracted as it was retrieved for inclusion in this systematic

review before the notice of retraction was issued.

The authors subsequently re-analyzed the data of this systematic review, excluding the

retracted article, and found that this did not change the main findings or the conclusion of this

paper. Therefore no significant changes were made to this systematic review.

50

2.8. Systematic Review of Survival and Disease Recurrence after Radiofrequency

Ablation for Hepatic Metastases

Introduction

Liver cancer is the fifth most common malignancy worldwide and the third most common

cause of cancer related deaths[10]. Colorectal liver metastasis (CLM) is the most common

cause of secondary liver metastases where approximately 50% patients with colorectal cancer

will develop liver metastases, 25% as synchronous[175] and 25% as metachronous[176]

disease. Surgical resection is the only curative options available for these patients, with 5-

year survival as high as 58%[159, 177]. However, only 20% of these patients have resectable

disease[88]. Factors precluding surgery include extra-hepatic metastases, high risk anatomic

location, excessive size or number of lesions, in-sufficient remnant liver to support life, or co-

morbid conditions[10, 88]. If untreated, the median survival for these patients is 6-12

months[89, 90], with few surviving beyond 3 years[91]. Adjuvant chemotherapy can increase

the median survival to approximately 20 months. However, chemotherapy is toxic with

unpleasant side effects and less than ideal disease control capabilities.

There has been a surge of interest in local ablative therapy for un-resectable liver cancers

worldwide in the past 2 decades, which includes cryotherapy, PEI, LITT, HIFU, MCT and

RFA. Among these, RFA has been the most widely investigated therapeutic option for un-

resectable liver cancers[1]. It has been shown in numerous large series that RFA is safe, with

minimal morbidity and mortality. RFA has also been shown to achieve a satisfactory local

response rate, with >80% complete response (defined as negative contrast-enhanced CT scan

post-RFA) in most studies[93]. It also significantly improves overall survival when compared

to other modalities e.g. chemotherapy or PEI[86].

A major drawback of RFA is the high local tumour recurrence rate seen in patients who

received this treatment. This could have an adverse effect on patient survival, and is the main

reason why RFA is considered inferior to surgery for the treatment of resectable liver

51

cancers. Early RFA results were limited by the small ablation size achievable, and the lack of

sensitive radio-imaging modalities to assess treatment response.

Intense research over the last two decades has produced impressive results. Higher-powered

radiofrequency generators and modifications to the electrodes can produce ablation sizes up

to 6-7 cm in diameter. Whereas only lesions <3 cm were treatable with RFA in the past,

physicians now can ablate tumours up to 5 cm and still achieve a 0.5-1 cm ablative margin

[100, 103, 156]. These advances have opened the door to more patients who previously were

deemed “un-treatable” and whose options were only palliation or chemotherapy. Recent

results are showing that the gap between RFA and surgery for liver cancer is narrowing.

This systematic review aims to examine the results of RFA for hepatic metastases over the

past decade in terms of patient survival and disease recurrence.

Methods

A literature search was conducted using Medline (Jan 2000 – week 3 Nov 2010), EMBASE

(Jan 2000 – week 49 2010), Cochrane Central Register of Controlled Trials (Jan 2000 - 4th

Quarter 2010), Cochrane Database of Systematic Reviews (2005 to November 2010),

Cochrane Methodology Register (Jan 2000 - 4th Quarter 2010), Database of Abstracts of

Reviews of Effects (Jan 2000 - 4th Quarter 2010) as per the search terms in Table 3 without

language restriction.

52

1. Catheter Ablation/ or radiofrequency ablation.mp.

2. RFA.mp.

3. Liver Neoplasms/ or Neoplasm Metastasis/ or liver metastas*.mp.

4. hepatic metastas*.mp.

5. 1 or 2

6. 3 or 4

7. 5 and 6

8. limit 7 to (comment or editorial or letter or meta analysis or "review")

9. hepatocellular carcinoma.mp. or Carcinoma, Hepatocellular/

10. 7 not 8 not 9

11. limit 10 to (humans and yr="2000 -Current")

Table 3. Search terms used for literature search

Inclusion criteria were as follows: (1) Participants – individuals with hepatic metastases of

any origin. (2) Intervention – RFA with any generator or needle designs. (3) Comparative

interventions – surgical resection, chemotherapy and/or other ablative treatment e.g. PEI,

cryoablation, MCT, LITT, HIFU. (4) Outcome data (measured from the time of intervention)

includes survival rates (overall median survival, median survival at 1-, 3-, and 5-years, and

median disease free survival), and disease recurrence rates (calculated per patient). Three

types of disease recurrences were recorded; ablation site (tumour recurrence at the site of

ablation), intra-hepatic (tumour recurrence in the liver away from the site of ablation), and

extra-hepatic (tumour recurrence outside the liver). (5) Types of study – randomized

controlled trials, quasi-randomized controlled trials and non-randomized comparative studies

were included in the review. Case series reporting more than 50 patients receiving RFA were

also included to provide a comprehensive evidence summary of the outcomes after RFA for

hepatic metastases. In addition meeting abstracts and each article’s bibliography identified

above were cross-referenced for relevant publications. Only articles reporting survival and/or

disease recurrence were included in this review. Articles which reported a combination of

RFA with other treatment modalities (e.g. surgery, chemotherapy, other local ablative

53

therapy) were also included. (6) Exclusion criteria – articles were excluded if the outcome

data could not be clearly attributed to each specific intervention (e.g. RFA vs. resection) or

disease (e.g. HCC vs. liver metastases). Meta-analysis, review articles, letters/comments and

editorials were also excluded.

Results

A total of 39 articles were identified and included in this review (Figure 4 – Quorum chart),

of which only 10 were comparative studies. Twenty-nine articles reported RFA for CLM, 8

for various liver metastases, while one article each was identified for neuroendocrine and

breast cancer liver metastases. No RCT of RFA for hepatic metastases was identified. No

meta-analysis could be performed due to the heterogeneity in treatment modalities and patient

populations. The patients who were treated with RFA could be broadly divided into 2 groups;

“un-resectable hepatic metastases” (Appendices 9 & 10) and “resectable hepatic metastases”

(Appendices 11 & 12).

54

Figure 4. Quorum chart

Outcomes after RFA for Un-Resectable Hepatic Metastases (Appendices 9 & 10)

There were 34 articles which reported the results of RFA for liver metastases. Analysis of

these articles showed 3 distinct ways of RFA utilization; RFA-only, RFA + resection, and

RFA + chemotherapy.

Potentially relevant studies identified in the literature search and screened for retrieval (n=1591)

Studies retrieved for more

detailed evaluation

(n=1319)

Potentially appropriate studies to be included in the systematic review (n=149)

(n=149)

Studies included in the systematic review (n=39)

29 – colorectal liver metastases

1 – neuroendocrine liver metastases

1 – breast cancer liver metastases

8 – various liver metastases

272 articles excluded- duplicates

1170 articles excluded – failed inclusion/exclusion criteria or not related to RFA for hepatic metastases after reading title/abstract

110 articles excluded - failed inclusion/exclusion criteria after reading full text.

55

RFA for Un-resectable Hepatic Metastases [102, 155, 178-200]

Twenty-five studies reported the outcomes after RFA for liver metastases, involving a total of

2446 patients. The median largest tumour diameter in the studies ranged from 1.2-3.7 (mean

1.5-5.2) cm, whereas median follow-up was between 14-42 (mean 17-33.2) months. The

median survivals reported in 15 studies were between 25-52 months[178, 182, 183, 185-191,

195-198, 201]. The 1-, 3-, and 5-year median survival rates were 72.5-96% [102, 153, 155,

179, 180, 183, 184, 189, 190, 198, 199, 202], 25.1-68%[102, 153, 155, 156, 179, 180, 183,

184, 189, 190, 199, 200, 202], and 5-48%[156, 179, 189-191, 197, 198, 200, 202]

respectively. The rates of ablation site, intra-hepatic distant and extra-hepatic disease

recurrence were 9.7-47.2%[153, 155, 179, 181, 184, 185, 187, 190, 191, 194-200, 203], 9-

62%[179-181, 190, 191, 195, 197, 200], and 5-54%[179, 180, 190, 191, 195, 197, 199, 200]

respectively.

Recently Gillams et al[188] published the largest series of RFA for CLM (median size 3.5

cm) in 2009 involving 309 patients. One hundred and fifteen patients (37%) had extra-hepatic

disease, while 292 (94.5%) patients had chemotherapy with no/partial response. Forty-eight

(15.5%) patients had previous liver resection. The patients were stratified into 2 groups based

on the number and size of tumours; group 1: ≤5 tumours ≤5 cm, and group 2: >5 tumours >5

cm. The overall median survival, 3-, and 5-yr median survival were 58%, 26%, and 39

months for group 1, and 29%, 5%, and 25 months for group 2 respectively (p<0.05). The

authors found that the number/size of tumour and the presence of extra-hepatic disease were

significant risks for worse survival in both uni- and multi-variate analysis. Sub-group analysis

56

showed that patients with ≤3 tumours <3.5 cm had the best outcome, with 5-year survival rate

of 33%.

Berber et al[197] compared laparoscopic RFA (n=68) to resection (n=90) in a group of

patients with solitary CLM. Median follow-up was 23 and 33 months for the 2 groups

respectively. The sizes of the tumours in the 2 groups were similar (3.7 cm vs. 3.8 cm,

p=0.9). All patients treated with laparoscopic RFA had un-resectable disease; and 26 of them

had extra-hepatic disease. No peri-operative mortality was reported. The complication rates

were 2.9% in the RFA group and 31.1% in the resection group. The authors reported a

median survival of 24 months for RFA patients with extra-hepatic disease, 34 months for

RFA patients without extra-hepatic disease, and 57 months for patients who had resection

(p<0.0001). Median disease free survival was 9 months in the RFA group versus 30 months

in the resection group (p<0.0001). There was no significant difference in the 5-year median

survival rate between the RFA and the resection group (30% vs. 40%, p=0.35) however. A

Cox proportional hazards model analysis showed that larger tumour size (>30mm vs.

<30mm, HR=1.6, p<0.0008) is a risk for worse outcome, while the type of intervention (RFA

vs. resection, HR=1.24, p=0.16) is not. A sub-group analysis of ASA I-II patients without

extra-hepatic disease who had RFA compared to those who had resection showed no

significant difference in median survival (49 months vs. 59 months, p=0.9).

Hur et al[200] retrospectively analyzed 67 patients with single CLM treated with either

resection (n=42) or RFA (n=25). Median size of the tumours was 2.6 and 2.5 cm in the

resection and RFA groups respectively. They reported that overall the disease recurrence and

survival rates were significantly better in the resection group compared to RFA. However

57

sub-group analysis showed that for patients with tumours <3 cm, there were no significant

differences in the overall survival or disease free survival between resection and RFA.

In an article published in 2005 Berber et al[195] examined 53 patients who had 192 “unusual

tumours” (cancers other than HCC, colorectal or neuroendocrine liver metastases) exclusive

to the liver treated with laparoscopic RFA. The majority of the patients had sarcoma (n=18)

or breast cancer (n=10). Disease recurrence at the site of ablation was 17% after a mean

follow-up of 24 months. The overall median survival was 33 months for the whole group.

Based on these results, the authors concluded that patients with “liver-exclusive disease” are

suitable candidates for RFA.

In 2007 Mazzaglia et al[198] reported the results of laparoscopic RFA for neuroendocrine

liver metastases. This is the largest case series to date involving neuroendocrine liver

metastases in 63 patients (384 tumours, mean size 2.3 cm). Nearly half of the patients (49%)

received medical and/or radiation therapy and 38% had extra-hepatic disease. Fifty-seven

percent of the patients were symptomatic pre-operatively. One week after RFA treatment,

92% of these patients reported at least partial symptom relief, and 70% had significant or

complete relief. After a mean follow-up of 33.6 months, 6.3% of the patients had disease

recurrence at the ablation site. Median survival was 46.8 months, whereas 1-, 2- and 5-yr

median survivals were 91%, 77% and 48% respectively.

Meloni et al[190] reported a series of 52 patients (87 tumours, mean size 2.5 cm) who were

treated with RFA for breast cancer liver metastases. Only patients with <5 tumours ≤5 cm

58

were included in the study. Ninety percent of the patients had no or partial response to

chemotherapy and/or hormonal therapy. Overall median survival after RFA was 29.9 months,

whereas median survival rates at 1-, 3-, and 5-years were 68%, 43% and 27% respectively.

Disease recurrence rates at the ablation site, intra-hepatic and extra-hepatic were 25%, 53%

and 54% respectively after a median follow-up of 19.1 months.

RFA + Regional/Systemic therapy for Liver Metastases [204, 205]

Scaife et al[204] reported a prospective series of 50 patients with colorectal liver metastases

(median largest diameter 2 cm) who received RFA in conjunction with hepatic artery infusion

chemotherapy (HAI) of continuous-infusion floxuridine (0.1 mg/kg days 1–7) and bolus

fluorouracil (12.5 mg/kg days 15, 22, and 29). Sixty-two percent of the patients completed

the full course of the chemotherapy. Post-operative morbidity and mortality rates were 18%

and 2% respectively. After a median follow-up period of 20 months, 32% of patients were

disease-free. The rates of disease recurrence at the ablation site, intra-hepatic and extra-

hepatic were 10%, 30% and 48% respectively. Although there were 31 patients who received

resection at the same time of RFA, this did not significantly affect the disease recurrence

rates.

Machi et al[205] reported the use of RFA in 100 patients with un-resectable CLM (mean

diameter 3 cm), either as first-line treatment (n=55) followed by chemotherapy or as second-

line intervention after failed chemotherapy (n=45) which consisted of fluorouracil plus

leucovorin and/or irinotecan. The overall median survival was 28 months, and 1-, 3-, and 5-

years median survival were 90%, 42%, and 30.5% respectively. Uni-variate analysis showed

59

that RFA had significantly better median survival when used as first-line therapy compared to

it being used as 2nd-line therapy (48 vs. 22 months, p=0.0001).

In the article by Siperstein et al[206] RFA was used to treat 234 patients with un-resectable

CLM (mean diameter 3.9 cm), and who displayed disease progression despite chemotherapy.

The median overall survival was 24 months, and 3-, and 5-years median survival were 20.2%

and 18.4% respectively. They reported better median survival for patients with ≤3 versus >3

tumours (27 vs. 17 months, p=0.0018), and whose tumour diameter was ≤3 cm versus >3 cm

(28 vs. 20 months, p=0.07). Twenty four percent of the patients had extra-hepatic disease

during the first ablation, although this did not affect median survival compared to those

without extra-hepatic disease (20% vs. 25%, p=0.34). The types of chemotherapy regimens

(5-FU-leucovorin vs. FOLFOX/FOLFIRI vs. bevacizumab) also did not affect survival

(p=0.11).

RFA + Resection for Hepatic Metastases [158, 159, 207-210]

Six studies reported the outcomes after RFA was used together with surgical resection in 442

patients. The median tumour diameter ablated was between 1.0-2.5 cm, and median follow-

up between 21-27.6 months. The median survival was 36-45.5 months[207, 211], whereas the

1- and 3-year median survival rates were 83-92% [207, 210], and 30-47% [207, 208, 210,

212] respectively. The disease recurrence rates at the ablation, intra-hepatic and extra-hepatic

sites were 2.3-17.4% [207, 209, 211, 212], 10.3-60.7% [207-209, 211, 212] and 30.2-46.2%

[207, 208, 211, 212] respectively.

60

Pawlik et al[158] published the outcomes of RFA combined with hepatic resection in 172

patients with multi-focal hepatic malignancies (72.1% CLM) which were considered to be

un-resectable by conventional standards. Both procedures were performed during 1 operation

where 387 tumours were resected and 350 tumours ablated. After a median follow-up period

of 21.3 months, the rates of local tumour progression, intra-hepatic distant recurrence and

extra-hepatic metastases were 2.3%, 38.8% and 53% respectively. The median overall

survival was 45.5 months. The overall mortality and morbidity rates were 2.3% and 19.8% in

this series. Patients with CLM had worse survival compared to non-CLM (median survival 37

vs. 59 months, p=0.03). The authors found that tumours >3 cm had adverse effects on

survival (HR=1.85, p=0.04).

Abdalla et al[159] reported 418 patients with CLM who received 1 of 4 different treatment

modalities; hepatic resection (n=190), RFA + resection (n=101), RFA only (n=57) or

chemotherapy only (n=70). RFA-only treatment was associated with significantly higher

rates of local tumour progression and intra-hepatic distant recurrence when compared to RFA

+ resection or hepatic resection (p<0.001). However there was no significant differences in

the rates of extra-hepatic metastases between the 3 groups (p=ns). Hepatic resection provided

significantly better outcomes when compared to the other treatment groups in terms of

overall and recurrence-free survival. When compared to hepatic resection in a multi-variate

analysis, treatment by RFA + resection (HR 2.14, p=0.004) or RFA only (HR 2.79,

p<0.0001) were risk factors for decreased overall survival. The authors also analyzed the

results of RFA + resection, RFA-only versus chemotherapy-only which might be the more

comparable groups considering that the patients in these groups technically had “un-

resectable” tumours. Both the RFA + resection and RFA only groups had significantly better

61

results when compared to the chemotherapy group with a median 4-year overall survival rate

being 36%, 22% and 8% respectively (p=0.002).

More recently Gleisner et al[160] published their results of patients who had hepatic

resection (n=192), RFA + resection (n=55) or RFA-only (n=11) therapy for CLM. The

authors found that the patients who underwent resection had the best overall and disease-free

survival. The median overall survival for the resection group versus the RFA + resection

group was 73.4 months and 38.1 months respectively (p<0.001). The median disease-free

survival was 19.5 months versus 10.2 months respectively (p<0.001).

In contrast, in the article published by Leblanc et al[209], there were no statistically

significant differences in median survival at 2-years between patients with liver metastases

who received RFA + resection (n=28) versus those who had resection (n=37) only (68% vs.

83%, p=ns). There were also no significant differences in median disease-free survival

between the 2 groups of patients (12 vs. 18 months, p=ns).

Outcomes after RFA for Resectable Hepatic Malignancies (Appendices 11 & 12)

Three articles were identified involving a total of 245 patients. Two articles[213, 214]

involved only patients with CLM, and 1 article[215] with mixed hepatic malignancies.

62

Resectable First-Episode Hepatic Metastases

Livraghi et al[215] used percutaneous RFA to treat 88 patients with 134 resectable colorectal

liver metastases. Inclusion criteria were: age ≤75 years, lesion numbers ≤3, and size ≤4 cm in

diameter. Eighty percent of the patients had received chemotherapy, and 24% had previous

hepatic metastasectomy prior to RFA. Complete response rate was achieved in 53 (60%)

patients. After a median follow-up of 33 months, 40% of patients developed local tumour

recurrence whereas 10% and 6.8% developed new intra-hepatic and extra-hepatic recurrence

respectively.

Otto et al[214] was the first to report the results of percutaneous RFA for first episode

resectable colorectal liver metastases compared to surgical resection. As part of their

institutional clinical pathway, patients who developed colorectal liver metastases within 12

months of their colorectal surgery were treated preferentially with RFA. Exclusion criteria for

RFA were; tumour diameter >5 cm, number of lesions >5, superficial lesions, or lesions in

proximity to large vessels or bile ducts. Patients who were previously treated with liver

resection, ablative therapy or portal vein embolization, or who received down-staging

chemotherapy were also excluded from analysis. There were 28 patients in the RFA group,

and 82 in the surgical resection group, with an average tumour diameter of 3 cm (range: 1–5

cm) and 5 cm (range: 1–14 cm) respectively. The complete response rate after the first RFA

was 100%. The authors reported that the patients in the RFA group had significantly higher

rate of local tumour recurrence (32% vs. 4%, p<0.001), but similar rates of intra-hepatic

(50% vs. 34%, p=0.179) and extra-hepatic tumour recurrence (32% vs. 37%, p=0.820)

respectively. However most patients with local tumour recurrence in the RFA group were

amenable to further intervention compared to the resection group (50% vs. 27%, p=0.012),

therefore leading to similar rates of estimated 5-year survival (48% vs. 51%, p=0.961).

63

sectable Recurrent Hepatic Metastases

Elias et al[213] used percutaneous RFA to treat 47 patients (107 tumours) with resectable

hepatic tumour recurrences after previous hepatectomy. This article was included in this

systematic review although the number of patients was less than 50 (as per inclusion criteria)

because it is only one of the 3 articles available in the literature where RFA was used to treat

resectable disease. Therefore its data would be of significant value to clinicians treating

hepatic malignancies. In the article only patients with <5 lesions and maximal tumour

diameter <3.5 cm were included in the study. Twenty-nine (62%) patients had CLM and 5

(11%) had HCC. The average tumour diameter and follow-up period were 2.1 cm and 14.4

months respectively. The authors reported a mortality rate of 2% (n=1, portal vein

thrombosis) and a morbidity rate of 9%. Following the first RFA 26 patients developed a

second recurrence after an average of 5.5 months in the liver of which 18 were amenable to

repeat RFA. Six of the 18 patients developed a third recurrence after 3.4 months of which 4

were treated with repeat RFA. The rate of local tumour progression, intra-hepatic distant

recurrence and extra-hepatic metastases were 31.9%, 21.3% and 31.9% respectively. The

median overall survival rates at 1- and 2-years were 88% and 55% respectively. The authors

compared this cohort to a matched group of patients who received repeat resection in the

same institution of which the survival rates were 84% and 60% respectively. The authors

suggested that RFA could be an alternative to repeat resection for hepatic tumour recurrences

considering its low morbidity and mortality, and the similar rates of overall survival between

the 2 therapies. The study only reported short term results up to 2 years however.

Discussion

Current RFA capabilities are limited by the size of coagulation that can be achieved which

leads to incomplete ablations and consequently higher rates of local tumour recurrence. The

inability to completely ablate “larger” tumours, or tumours in high risk locations (e.g.

adjacent to large hepatic vessels or in sub-capsular areas) compounds the problem and are

adverse prognostic factors for local tumour recurrences. In addition most hepatic

malignancies are irregular in shapes which mean part of them may escape ablation. For these

larger and irregularly shaped tumours, multiple electrode insertions and ablations are usually

required to produce complete necrosis of the whole lesion including a 1 cm ablation margin,

which is not always easy to accomplish leading to incomplete ablation.

64

Some clinicians are concerned over the much higher rates of local tumour recurrences and

lower disease-free and overall survival in patients who received RFA compared to surgical

resection. It should be noted that there are few “head to head” comparisons between RFA

versus surgery for resectable hepatic malignancies. The majority of the literature available

reported results where RFA was used to treat “un-resectable” tumours which were, most of

the time, associated with advanced disease (e.g. Childs-Pugh B/C HCC, or bilobar tumours)

or the patient was too sick to undergo surgery. These are adverse prognostic factors which

could have a negative influence on the patients’ outcomes, and therefore comparing RFA to

surgery in these different groups of patients is akin to comparing “apples to oranges”.

One intrinsic flaw of RFA is the high local tumour recurrence rate which can be up to 30-

40% [153], which is related to the capability to completely ablate a lesion. The complete

ablation rate of RFA, as measured by non-enhancement of the tumour during contrast-

enhanced CT scan, is directly related to the size of the tumour. For tumours where a higher

rate of complete ablation could be achieved, the 5-yr median survival rates could be up to

68.5% [216]. When used for larger tumours, the complete ablation rate and the long term

outcomes predictably became worse. In the article published by Chen et al [155] who

reported an overall complete ablation rate of 95% after RFA for hepatic malignancies, the

authors found that the success rate fell to 85 % for tumours >3.5 cm. The complete ablation

rates were also lower when the tumours were adjacent to the gallbladder (86.3%) and the

bowels (83.3%) respectively. The high local tumour recurrence rates and its adverse influence

on patients’ survival in the long term is the main reason why RFA is inferior compared to

surgical resection. Several studies have shown that when limited to tumours <3 cm, there

were no statistically significant differences in disease-free or overall survival rates between

RFA and resection[217, 218].

The evolution in the RFA technology has been phenomenal in the past decade. In the early

2000s, the use of RFA was limited by the size (≤3 cm) and number (<5) of lesions. With

better equipment and understanding, the indication for RFA continues to expand while

maintaining satisfactory outcomes. Ahmad et al [181] examined patient outcomes when they

65

were treated with a first generation RFA needle electrode (3 cm ablation diameter) compared

to a newer needle design (5 cm ablation diameter). Baseline patient and disease burden

(tumour numbers and sizes) characteristics were similar between the 2 groups of patients.

Their results showed that after a median follow-up of 26.2 months, the patients treated with

the newer RFA electrode had better disease free survival (16 vs. 8 months, p<0.01) and lower

rates of disease recurrence at the ablation site (5.2% vs. 17.4%, p<0.04). Clinicians are now

commonly using RFA to treat hepatic tumours >3 cm in size, with some even treating

tumours >5 cm with satisfactory results [100, 103, 156]. The number of lesions has ceased to

become an absolute contraindication to RFA.

RFA has its own distinct advantages compared to surgical resection of hepatic malignancies.

It is minimally invasive when performed percutaneously and has much lower rates of

morbidity and mortality compared to surgery. Most of the RFA are performed as day

procedures with patients leaving the hospital on the same day. In addition, RFA is a versatile

tool which has proven to be very useful to the hepatic surgeon as it can be performed

percutaneously, laparoscopically, or in combination with surgical resection. The combination

of RFA and surgical resection provides a curative option to patients who have inoperable

tumours by conventional standards (e.g. bilobar disease) [158-160, 207, 209, 210].

There is now at least 1 report [214] where RFA was compared to surgery for resectable

colorectal liver metastases. This report involved a carefully selected group of patients with

strict inclusion and exclusion criteria. The authors found that there was no statistically

significant difference in 3-yr median survival (67% vs. 60%, p=0.93) between the 2 groups.

However there are several potential biases to consider in this paper. Firstly the treatment

protocol used by the author is part of their clinical pathway to treat patients with CLM, and

not a randomized trial. Secondly the size of the tumours were significantly larger in the

surgery compared to the RFA group (5 cm vs. 3 cm, p=0.004), and lastly there were only 28

patients in the RFA group. Nevertheless this article has increased the evidence that perhaps

the time for a randomized controlled trial comparing RFA to surgery for resectable hepatic

malignancies has arrived.

66

3. Electrolysis and Electrochemical Therapy (ECT)

Electrolysis is the passage of a direct electric current through an ionic substance in a suitable

solvent, resulting in chemical reactions at the electrodes and separation of materials. This

process is used for a variety of purposes including treating malignant tumours in humans,

which is known as ECT. Various types of electrodes have been used and reported in the

literature, the most common of which is made of platinum[11, 219-221]. The low energy DC

polarizes the two electrodes, causing electron transfer from the cathode to the anode. The

anode will attract negatively charged ions (e.g. Cl-) while the cathode will attract positively

charged ions (e.g. Na+ and K+). The main chemical reactions occurring at the anode are[4,

222, 223]:

2 Cl- Cl2 + 2e-

2 H2O O2 + 4H+ + 4e-

There are several possible outcomes from a combination of the various ions above. The Cl2

and O2 can be liberated as gases. The H+ can react with Cl- to form HCl which increases the

acidity of the surrounding tissue. Finally all three by-products (H+, O- and Cl-) can combine

to form HOCl (hypochlorous acid). As a result the pH of the tissue around the anode becomes

more acidic [224-226].

The main chemical reaction at the cathode is [4]:

2H2O + 2e → H2 + 2OH-

The hydrogen is liberated as a gas which is evident as rigorous bubbling, whereas the sodium

ions combine with hydroxyl ions (OH-) to form NaOH. This makes the tissue pH more basic

[224-226].

The products of the chemical reactions above are responsible for the cellular necrosis seen in

ECT. Species produced at the anode and cathode are mainly transported to the surrounding

tissue by diffusion due to concentration gradients, and by migration (charged species) due to

the potential gradient.

67

Chlorine, a powerful oxidant, and hypochlorous acid (HOCl) can both cause lethal injuries to

the surrounding cells [221]. The hydrogen ions released from the hydrolysis of water

molecules decrease the tissue pH, causing complete cellular necrosis when the pH is less than

six[222, 223]. The tissues around the cathode become more basic as a result of the liberated

hydroxyl (OH-) species. Tissue necrosis is complete when the pH is more than nine[222,

223].

Apart from the chemical insults described above there is evidence that other mechanisms of

cellular injury are actively involved, for example disturbances in blood flow and oxygenation

to the tumour. Jarm et al inserted ECT electrodes into healthy tissue on opposite ends, and 5

mm away from a mice fibrosarcoma tumour model[227]. The distance between the electrodes

and the tumour edge prevented the toxic chemicals from directly affecting the neoplastic

cells. They found that low level DC (0.6mA for 60 minutes) resulted in damaged or occluded

blood vessels at the insertion sites of the electrodes, leading to severe reduction in tissue

perfusion, extensive extravasation of blood cells and focal areas of necrosis in the tumours.

3.1. Animal Experiments

3.1.1. Tissue Temperature

Thermal energy has been shown to have no role in the cellular necrosis seen in ECT. Baxter

et al studied the relationship between ECT and tissue temperature in rat (2-4 mA) and porcine

(20-50 mA) livers [225]. The tissue temperature in the rat liver remained the same, but the

temperature in the pig liver increased by 4.2 degrees (p<0.01) after ECT. However the

average tissue temperature in the pig liver was 45.2 degrees at the end of ECT, which would

not be high enough to cause cellular necrosis by hyperthermia.

3.1.2. Water Content

Studies in animals have shown that water moves from the anode to the cathode. Li et al found

that the water content was 76% around the cathode, compared to 71% at the anode and 73%

68

in an untreated part of a canine liver model after 124 coulombs of DC was passed between

two platinum electrodes[228, 229].

3.1.3. Elemental Concentrations

The different polarity between the anode and the cathode will cause movements of the

different elemental ions. Negatively charged ion e.g. Cl- will move towards the anode,

whereas positively charged ions e.g. Na+, K+ will move towards the cathode [228, 229]. The

concentrations of multivalent ions e.g. Cu2+, Mg2+ and Ca2+ did not change much, because

their larger size is inversely proportional to the velocity of their movements [228, 229].

3.1.4. Tissue pH

Various chemical reactions occur around the anode and the cathode during ECT. The

elements H+, Cl- and O- around the anode can react to forms two types of acids; HCl and

HOCl. The element Na+ can react with OH- to form the alkali, NaOH. Tissue pH around the

anode becomes acidic, whereas that around the cathode becomes basic. The changes in tissue

pH were thought to be the main mechanism of cellular destruction during ECT. A study

published by Finch et al reported that tissue pH can be used to reliably monitor tissue

ablation in a porcine liver model[222]. Total cellular necrosis was observed when tissue pH

was less than six or more than nine[222]. Tissue pH during ECT could be as low as 2.1 at the

anode, and as high as 12.9 at the cathode[228, 229].

3.1.5. Gas Production

The main gas produced at the anode is chlorine, whereas that at the cathode, is hydrogen[228,

229].

3.1.6. Cellular Histological Changes

Histopathologic study showed marked dehydration of the hepatocytes around the anode with

pyknotic nucleus and small or absent cytoplasm[230]. The tissues around the cathode showed

cellular oedema, nuclear and cytoplasmic swelling with occasional disruption of the plasma

membranes.

69

Histological studies in liver models showed a sharp demarcation between necrotic and normal

hepatic tissues. Necrotic cells appear as featureless, eosinophilic hepatocytes lacking

glycogen vacuoles[222, 231]. There was total destruction of cellular membrane, cytoplasmic

structures and the nuclei[232]. This zone of coagulative necrosis was surrounded by a rim of

actively proliferating cells (which included fibrobroblast and biliary ductules) with

neutrophilic infiltration[222, 231]. The necrotic area is gradually replaced by fibrotic tissues

and the scar contracts as healing occurs[231]. It was observed that ECT produced an area of

wedge-shaped infarct, with the apex at the site of the electrode placement and the base

extending towards the edge of the liver. These wedge-shaped infarcts were likely secondary

to vessel thrombosis induced during the ECT, and could be seen at both the anode and the

cathode sites[219].

3.1.7. Volume of Tissue Ablation

There is a linear relationship between the volume of liver tissue destroyed and the amount of

DC energy administered[219, 226]. The higher the ECT dose (measured in Coulombs) which

is given, the larger the volume of tissue destruction. The volume of tissue destruction was

found to be greater at the anode compared to the cathode[219].

3.1.8. Safety

Long-term studies in pigs showed that ECT was well tolerated and produced no major

adverse effects on the liver functions. The liver enzymes, aspartate transaminase, alanine

transaminase and gamma-glutamyltransferase, were elevated after ECT but returned to

baseline level after one week[233].

ECT was found to be remarkably safe around blood vessels. Wemyss-Holden et al studied

the effect of ECT around major vasculatures by inserting the electrodes into, and adjacent to

the hepatic veins of six pigs before administering 100 coulombs of DC[231]. No major

bleeding complications were encountered, and despite gas bubbles entering the hepatic veins

and inferior vena cava IVC, all animals recovered well post-operatively.

70

3.2. Human Studies

There is relatively sparse data on the clinical use of ECT in human beings, with no

randomized controlled trials reported in the literature. Most of the data originates from China,

with three case series published in 1994. Xin et al reported the use of ECT in 388 patients

with various types of tumours, both benign and malignant[234]. However due to the

heterogeneity of the study population, the results were not easily interpretable or applicable to

clinical practice. Wang et al used ECT to treat 74 patients with HCC ranging from 3-20 cm

with a 1-year survival rate of 33%[235]. Lao et al treated 50 patients with HCC ranging from

3.5-21 cm with a 1-year survival rate of 69%[236]

3.3. Modifications/Innovations

Lin et al investigated the effect of saline injection on increasing the efficacy of ECT[237]. He

compared the size of tissue ablation after injection of water, 0.9%, 3% or 26% saline versus

no injection during ECT in an ex-vivo porcine liver model. It was postulated that the

interstitial saline injection would lower electrical impedance and allow more electrical

current to pass through to the target tumour. In addition the increased water and electrolytes

concentration would enhance the electrochemical reactions. They found that the volume of

tissue destruction was 8.1 times greater in the 26% saline group compared to control.

Several studies have found that placing the electrodes at least 2 cm apart produced

significantly bigger volumes of tissue destruction compared to if the electrodes were placed

closer to each other [221, 226, 238]. It was postulated that placing the electrodes near each

other would result in mixing of the electrochemical products to re-form water and sodium

chloride, therefore reducing the cytotoxic effects of ECT.

71

3.4. Problems in ECT

The main problem encountered in ECT is that each treatment takes a long time to complete as

it depends on the diffusion of various cytotoxic chemicals to produce cellular necrosis. Hinz

et al took an average of 31 minutes to ablate of volume of 1.5cm3 of liver using ECT,

whereas ablation using RF took only four minutes[232]. In addition, they had to use four

electrodes (two cathodes and two anodes) for the ECT, compared to a single electrode

insertion for RFA[232]. In a case report published by Fosh et al, ECT for a 4.2 cm x 4.2 cm x

2.6 cm HCC took 288 minutes to complete[9]. In the case series reported by Wang et al and

Lao et al, the ECT treatment took between 1.5 to five hours to complete[235, 236].

72

4. Bimodal Electric Tissue Ablation (BETA)

Bimodal electric tissue ablation (BETA) is a new method of local ablative therapy utilizing

the electrochemical reactions of the DC to enhance the efficacy of RFA. BETA utilizes the

hydration effect produced at the cathode in the DC circuit to enhance the efficacy of thermal

ablations produced by RF generators.

It had been shown that during electrolysis, water was attracted to the cathode evident as local

tissue swelling around this electrode[7, 239]. Early BETA studies in porcine livers showed

that the liver would appear congested, and fluid would ooze out from its surface while

microscopically, the cells exhibited marked intra-cellular swelling compared to standard

RFA[12]. This property of the cathode is utilized in combination with standard RFA with the

aim of postponing tissue charring and carbonization around the RF electrode which are the

limiting factors in producing bigger ablations. This hydration effect allows RFA to continue

for a longer period of time, therefore producing larger ablation zones. Increasing the water

content in the tissue around the active electrode also improves electrical conduction allowing

thermal energy to be distributed more evenly throughout the whole tumour which is to be

ablated.

4.1. Early Experimental Results

In 2007 Cockburn et al[11] published a study investigating the effect of applying increasing

amounts of DC before and during RFA of porcine liver. Nine volts of DC was applied for

increasing duration of time (0, 30, 60, 90, 120, 300, 600, 900 seconds) before the RF 3000

generator was started at 20W, and both the DC and RF circuit was then allowed to run

simultaneously until roll-off. This new setup produced larger ablations when compared to

standard RFA (p<0.001).

In 2008 Dobbins et al published a study comparing the ablative size of BETA compared to

standard RFA using a 3.5 cm multi-tined LeVeen electrode[14]. Nine volts of DC was

provided for 15 minutes, after which the RF 3000 generator was switched on to provide 80W

73

of power. Both the DC and RF current were allowed to run until roll-off occurred; defined as

impedance higher than 700Ω or power output less than 5W. BETA was shown to produce

significantly larger ablations compared to standard RFA (27.78 mm vs. 49.55 mm, p<0.001).

The average treatment duration in BETA was significantly longer compared to standard RFA

(1115 seconds vs. 249 seconds, p<0.001).

Dobbins et al subsequently investigated the long term morbidity and pathological features of

BETA in a pig liver model[12]. Each procedure was started with 9V of DC for five minutes,

after which the RF circuit was started with 20W power output and both circuits allowed to

run simultaneously. Six ablations were produced in each of the 10 pigs used in the study,

after which two pigs were euthanatized at two days, two weeks, two months and four months

respectively. The pigs had their bloods taken at different time periods depending on their

survival, and upon sacrifice, their internal organs were examined for any abnormalities. The

authors found no significant changes in haemoglobin, total white cell count, creatinine,

albumin, bilirubin, alkaline phosphatase, alanine transaminase or gamma-glutamyltransferase

compared to baseline. There was a transient rise in serum aspartate transaminase, alkaline

phosphatase and C-reactive protein in the immediate post-operative period which normalized

after two days. The only complication reported was the occurrence of local tissue injury at the

site of the anode which manifested as full thickness skin necrosis. BETA was noted to

produce coagulative necrosis which healed from the periphery of the lesion towards the

centre. This feature was similar to that produced by standard RFA and electrochemical

therapy.

In all animal experiments performed so far, BETA had proven to be safe, except for the full

thickness skin necrosis at the site where the anode was placed. This was not unexpected

considering previous experiments involving ECT had shown that various cytotoxic chemicals

were produced at the anode, including acidic hydrogen ions and chlorine. Chlorine reacts

with water to form hypochlorous acid, chloride and hydrogen ions. As a result of these

reactions, the pH in the vicinity of the anode drops to around 1-2 with lethal consequences to

the surrounding cells and tissues. Such a complication is clearly unacceptable in humans and

before BETA could be used in the clinical setting, its safety feature needs to be improved

further.

74

Dobbins et al proceeded to investigate an alternative method to use as the anode with the aim

of preventing skin injury[13]. They hypothesized that by increasing the surface area of the

anode, it will reduce the current density in the adjacent tissue, therefore reducing the risk of

local tissue injury. He replaced the scalpel blade used in the earlier studies with a dispersive

grounding pad similar to the ones used for electro-surgical units. This had the advantage of

being easily available, and could be conveniently placed on the skin which was attractive

considering that many RFA procedures are carried out percutaneously. Dobbins et al

compared the severity of tissue injury occurring at the anode (scalpel blade versus dispersive

pad) and also the diameter of the ablation achieved, with standard RFA as the control. They

reported only mild skin erythema in three out of the six pigs where the dispersive pads were

used as the anode. These changes resolved completely in all three animals after 48 hours.

Post-mortem histopathologic examination showed that tissues at the site of dispersive pads

placement showed no significant changes compared to controls. Full thickness skin necrosis

was observed in all animals where scalpel blades were used. The ablation size was largest

when scalpel blades were used compared to dispersive pads (2.5 cm vs. 1.8 cm, p<0.001). A

possible explanation for this observation was that the outer skin of the pigs was very thick,

and a poor conductor of electricity. Therefore, the electrical resistance to the flow of DC was

greater when a dispersive pad was used as the anode on the skin compared to a scalpel blade

inserted subcutaneously. As a result, less water would accumulate around the cathode leading

to earlier charring and desiccation of tissue. This meant roll-off would have occurred sooner,

producing smaller ablation sizes. The ablations using dispersive pads were however, still

significantly larger when compared to standard RFA (1.8 cm vs. 1.533 cm, p<0.001).

Continuous effort is being made to further improve the efficacy of BETA. Recently a multi-

national group of researchers published an article on dose optimization study on BETA in an

ex-vivo bovine liver model. Tanaka et al[240] used similar circuitry modifications as

described by Dobbins et al, namely a DC generator attached to a Boston Scientific RF 3000

generator to allow ECT and RFA to run separately but also concurrently. The cathode was

connected to a 3 cm LeVeen RFA electrode via a 100mH inductor. The return electrode of

the RF generator and the positive electrode of the DC generator were attached to a metallic

basin into which the liver was placed. Electrolysis was performed for 15 minutes with three

75

different voltage settings (2.2V, 4V and 9V) prior to any RFA. After that the RFA was started

while allowing the DC energy to flow continuously. The RFA was performed using three

different protocols: (1) stepwise increase pattern where the RFA was started at 40W and

increased by 10W every 30 seconds up to 80W; (2) 40W fixed without increase in power; and

(3) 80W fixed without increase in power. The procedure was continued until roll-off occurred

twice.

The authors found that pre-treatment with 4.5V or 9V DC combined with RFA using either

the 40W fixed or step-wise increase protocol produced ablation volumes nearly twice as large

as the control or the 2.2V group (p=0.009). However, there were no significant differences in

ablation sizes when comparing the 4.5V to the 9V groups. The duration of RFA was

significantly shorter in the 40W step-wise increase protocol compared to the fixed 40W

protocol (296s vs. 423s, p=0.028) in the 4.5V DC group. There were no significant

differences in the duration of ablation when comparing the 4.5V to the 9V DC group.

In summary, the step-wise increase RF protocol produced ablation volumes comparable to

the 40W fixed protocol. The latter however, took a significantly longer time to produce. In

addition pre-treatment with a DC set at 4.5V produced ablation volumes comparable to a 9V

DC, but the total amperage applied was approximately half as much. These observations led

the authors to conclude that a combination of a step-wise increase RF protocol with a DC

current of 4.5V is the optimum BETA setting to increase coagulation volume and, at the same

time, minimize procedure duration in ex-vivo bovine liver.

76

5. Rationale for the Current Research

5.1. Experiment 1: Does BETA really work by increasing tissue hydration?

It is not known exactly how BETA works. It has not been definitively proven yet that the

capability of BETA to produce larger ablations was indeed due to the increased tissue

hydration secondary to the DC. It is also not known how much the DC can increase the

hydration of tissues adjacent to the cathode. Existing data shows that electrolysis only

increased the water content at the cathode by approximately 3%[228, 239]. However, the DC

was run for a significantly longer period of time (69 minutes[228], 48 hours [239]) compared

to what was utilized in BETA. This “pre-treatment with DC” in BETA must not be prolonged

or it will make the whole procedure too time-consuming for clinical use.

One simple method to investigate the above question, whether BETA produced larger

ablations by increasing the tissue moisture, is to reverse the polarity of the DC. The anode,

instead of the cathode, is attached to the RF electrode and its effect on tissue ablation studied.

There are several different reactions that occur at the anode compared to the cathode besides

the net movement of water molecules from the former to the later. Due to the different

polarity, cations (e.g. Na+ and K+) will move towards the cathode, whereas anions (Cl-) move

towards the anode[228, 229]. Chloride can be liberated as chlorine gas, or react with H+

and/or O- to form HCL or HOCL respectively. These acids greatly reduce the tissue pH and is

one of the main mechanisms of cellular destruction seen in ECT. At the cathode the water

molecule is broken down to liberate hydrogen gas, and the by-product hydroxyl ions react

with sodium ions to form the alkali sodium hydroxide. Therefore, the tissue pH at the cathode

rises which will also cause cellular destruction, although to a lesser extent compared to the

anode.

Apart from the net movement of water molecules from the anode to the cathode, all the other

reactions and by-products of electrolysis should not have any effect on the RFA. Therefore, if

the hypothesis put forward by Cockburn et al and Dobbins et al is correct, then reversing the

polarity of BETA would produce smaller ablations compared to standard RFA or BETA as

the anode would cause tissue desiccation.

77

5.2. Experiment 2: Where is the optimum place to put the anode?

One of the important questions in BETA is where to place the anode. Dobbins et al reported

full thickness skin necrosis and abscess formation when the anode is placed in the

subcutaneous tissues[12]. This problem was solved by attaching the anode to the skin using

electrosurgical dispersive pads[13]. However the ablations produced were significantly

smaller due to the higher electrical impedance in the skin. Therefore a new option is required

where the anode can be placed to maximize the benefit of BETA, and yet minimize the local

tissue injury.

Evidence in the literature suggests that the tissue with the highest electrical impedance in the

body is the skin, more specifically the stratum corneum of the epidermal layer[241]. This

would explain the observation that placing the anode on the skin surface resulted in smaller

ablations. It was postulated that bypassing the skin layer and putting the anode below it

would overcome this problem.

In the second experiment, the peritoneum and the liver were studied as alternate locations to

place the anode. ECG dots were used instead of a needle electrode to increase the surface

area in contact with the tissues with the aim of minimizing local tissue injury.

5.3. Experiment 3: Can BETA be incorporated into the Cool-Tip RF System?

Previous research on BETA using the RF 3000 System (Boston Scientific) has shown that it

could significantly increase the duration and size of ablations compared to standard RFA. The

RF 3000 System has the “roll-off” as its end-point, which means that the ablation stops

automatically when the tissue impedance has risen too high to allow further electrical

conductance. Another popular RF system on the market is the Cool-Tip RF System

(Covidien), which uses time as its end-point during an ablation. The manufacturer’s

recommended protocol suggests switching the generator to the “impedance mode” to provide

maximum power for 12 minutes ablation. In this “impedance mode” the generator

continuously monitors tissue impedance, and will stop power output for 15 seconds when the

impedance rises more than 10 Ohms above baseline values. This, in addition to the internal

78

cooling of the electrode using chilled-saline, minimizes tissue charring and allows better

energy distribution throughout the whole tumour.

It was not known whether the principle of BETA could be incorporated into the Cool-Tip RF

System (Covidien) to increase the size of ablations compared to standard RFA only, which

was the objective of the third experiment.

79

Title of Thesis: Improving the Safety and Efficacy of Bimodal Electric Tissue Ablation


CHAPTER 5:

Experiment 1

Bimodal Electric Tissue Ablation (BETA) – Effect of Reversing the Polarity of the Direct Current on the Size of Ablation

Tiong LU (MBBS)‡, Finnie JW (BVSc, PhD, FRCVS)*, Field JBF (PhD, AStat)†, Maddern GJ (PhD, MS, MD, FRACS)‡

‡Department of Surgery, The Queen Elizabeth Hospital, Adelaide, Australia

*SA Pathology, Institute of Medical and Veterinary Science, Adelaide, Australia

†University of Adelaide Faculty of Health Sciences & Basil Hetzel Institute, Adelaide, Australia

Journal of Surgical Research 2011 – accepted paper

80




Title of Paper: Bimodal Electric Tissue Ablation (BETA) – Effect of Reversing the Polarity of the Direct Current on the Size of Ablation Journal of Surgical Research 2011 – accepted paper

Dr. Tiong, LU (Candidate)

Planned and performed experiment, data collection and analysis, and prepared the manuscript.


Dr. Finnie, JW

Performed histo-pathological analysis of specimens


81



Dr. Field, JBF

Performed power calculation for sample size and statistical analysis on the experimental data obtained


Prof. Maddern, GJ



82

Experiment 1: Bimodal Electric Tissue Ablation (BETA) – Effect

of Reversing the Polarity of the Direct Current on the Size of

Ablation

Introductions

Radiofrequency ablation (RFA) is increasingly used to treat liver tumours (e.g. HCC and

liver metastases) especially in cases where the tumours are technically un-resectable[89, 242-

245]. One major limitation of this technique is the excessively high rate of local tumour

progression which can be up to 30-40%[153, 246]. This limitation is related to the

incapability of RFA to achieve complete ablation of a liver tumour[194, 247]. It has been

shown in numerous studies that successful complete ablations in smaller sized tumours (≤30

mm) were associated with lower rates of local tumour progression (<10%)[107, 123, 159,

248, 249]. Livraghi et al[216] published a case series in 2008 where they reported a total of

218 patients with single HCC ≤20 mm in diameter who were treated with RFA. They

reported a sustained complete ablation rate of 97.2% after a median follow-up of 31 months,

and 5-year survival rate of 68.5%. When the size of the tumour increases, the success rates of

complete ablation reduce therefore leading to a high local disease recurrence rate[250].

Multiple RF generator and electrode modifications have been made to produce larger

ablations to overcome this problem, however none have yet to prove 100% effective. High-

powered generator capable of delivering higher energy output [96, 249], cold-saline perfused

needle electrode [251], saline enhanced electrode [250, 252], and multi-tined needle electrode

[253] are some of the examples of recent innovations in the field of RFA.

More recently the combination of DC with RFA to increase the size of tissue ablation was

introduced[11-14]. This is in a sense a combination of electrolysis and RFA, although the

underlying principle is slightly different from the conventional electrolytic therapy.

83

The process of electrolytic therapy to treat liver tumours has been investigated extensively

[220, 221, 224-226, 238]. The anode was conventionally inserted into the centre of the

tumour to be ablated as it produces larger area of necrosis compared to the cathode. The

mechanisms of cellular necrosis have been attributed to various processes including changes

in tissue pH levels [222, 230], release of toxic gases (chlorine and hydrogen), and the

occlusion of vessels feeding the tumour [227].

During the electrolytic process, it was noted that the tissues surrounding the anode would

become desiccated, while those surrounding the cathode would become oedematous with

water [239, 254]. Therefore, there is a net movement of the water molecules to the tissues

adjacent to the cathode. It was this “hydrating” property of the cathode that forms the

underlying principle in BETA.

Cockburn et al and Dobbins et al postulated that increasing the hydration of the liver tissues

around the active RF electrode would reduce the tissue temperature during ablation[11-14].

This would delay tissue desiccation and allow the ablation process to continue for a longer

period of time therefore, produce larger ablations. This combination of the cathode belonging

to a DC circuit together with a RF electrode is called bimodal electric tissue ablation (BETA),

and it has been shown to produce significantly larger ablations compared to standard RFA.

It is not known exactly how BETA works, and it has not yet been proven that the capability

of BETA to produce larger ablations was indeed due to the increased tissue hydration

secondary to the DC. It is also not known how much the DC can increase hydration of tissues

adjacent to the cathode. Existing data has shown that electrolysis only increases the water

content at the cathode by approximately 3% [228, 229]. However, that DC was run for a

significantly longer period of time (69 minutes [228], 48 hours [239]) compared to what was

utilized in BETA. This “pre-treatment with DC” in BETA must not be too long or else it will

make the whole procedure too time-consuming for clinical use.

84

One simple method to investigate the above question, whether BETA works by increasing the

tissue moisture, is to reverse the polarity of the DC. The anode, instead of the cathode, is

attached to the RF electrode and its effect on tissue ablation studied. There are several

different reactions that occur at the anode compared to the cathode besides the net movement

of water molecules from the former to the later. Due to the different polarity, cations (e.g.

Na+ and K+) will move towards the cathode, whereas anions (Cl-) move towards the anode

[228, 229]. The chloride can be liberated as chlorine gas, or react with H+ and/or O- to form

HCL or HOCL respectively. These acids greatly reduce the tissue pH and this is one of the

main mechanisms of cellular destruction seen in electrolytic therapy. At the cathode the water

molecule is broken down to liberate hydrogen gas, and the by-product hydroxyl ions react

with sodium ions to form the alkali sodium hydroxide. Therefore, the tissue pH at the cathode

rises which causes cellular destruction, although to a lesser extent compared to the anode.

Apart from the net movement of water molecules from the anode to the cathode, all the other

reactions and by-products of electrolysis should not have any effect on the RFA. Therefore, if

the hypothesis put forward by Cockburn and Dobbins is correct, then reversing the polarity of

BETA will produce smaller ablations compared to standard RFA or BETA as the anode will

cause tissue desiccation.

This study aims to investigate the size of ablation when the polarity of DC is reversed,

namely the anode is combined with the RF electrode. This new combination is abbreviated to

RP-BEA (reversed polarity bimodal electric ablation) throughout the rest of the thesis to

distinguish it from standard RFA and BETA.

Materials and Methods

This study was performed in the animal laboratory at The Queen Elizabeth Hospital

(Adelaide) using domestic female white pigs each weighing approximately 50 kg. All

animals were admitted to the experimental facility a minimum of two days before the

experiment for acclimatization. The animals were housed in individual pens, maintained at 23

+/- 1ºC, at ambient humidity. Lighting was artificial, with a 12-hour on/off cycle. The air

85

exchange rate and airflow speed complied with the Australian code of practice for the care

and use of experimental animals. The pigs were fed and watered ad libitum (standard grower

diet of 0.7 g of available lysine per mega-Joule of digestible energy, with a digestible energy

content of 14 MJ/kg). Water quality was suitable for human consumption.

Preoperatively, the pigs were fasted for 12 hours. Each pig was sedated with an intra-

muscular injection of ketamine (0.5 mg/ kg). General anaesthetic was induced and maintained

using of 1.5% isofluorane mixed in oxygen. An endotracheal tube was placed to maintain the

airway and a temperature probe was placed inside the endotracheal tube to monitor core

temperature of the animal. The pig was placed on a warming pad in the base of its cradle to

assist in temperature homeostasis. An oxygen saturation probe was placed on the pigs tongue

to monitor oxygen saturations. Throughout the procedure, recordings for pulse, temperature,

oxygen saturations, end-tidal carbon dioxide levels and cardiac rhythm was monitored. The

pig received 0.9% normal saline solution through an intravenous line throughout the course

of the procedure.

The abdomen was cleaned with iodine solution and square-draped with sterile towels. A mid-

line incision was made from the xiphi-sternum to the umbilicus. The falciform ligament was

divided and the liver mobilized inferiorly. The porcine liver exhibits deep fissures that divide

it into left lateral and medial and right lateral and medial lobes. Additionally, the short

quadrate lobe and the caudate process are present centrally[255]. All experimental

procedures were carried out in the liver tissue thick enough to accommodate the whole

ablation. The surrounding organs were protected and packed away with moist gauze packs.

Three different ablations were carried out on each pig as follows:

1. Standard RFA only

2. BETA-skin – with the anode attached to the skin using ECG dots

3. RP-BEA – with the cathode attached to the skin using ECG dots

86

A Boston Scientific RF 3000 generator was used to provide the radiofrequency energy.

Aluminium rods measuring 40x2 mm were used as the RF electrode and were inserted 20

mm into the liver tissue. The grounding pad for the RF 3000 generator was placed on the

inner thigh of the animals’ hind-leg.

A generic AC/DC adaptor was used to provide the DC. In BETA the cathode of the AC/DC

adaptor was connected to the RF electrode wiring using a 1 mH inductor. This inductor

allows the flow of the DC into the radiofrequency circuit, but prevents the leakage of

alternating current from the RF 3000 generator into the DC circuit. Therefore the needle

electrode of the RF 3000 generator and the cathode of the DC circuit are one and the same.

The anode of the DC circuit was attached to the skin using a standard ECG diagnostic

electrode (ECG dots). In RP-BEA the polarity of the DC circuit was reversed; therefore the

anode was connected to the RF electrode, and the cathode attached to the skin via ECG dots.

For the first procedure, only the RF energy was delivered to the liver tissue. The RF 3000

generator was started to deliver 40 watts of power until “roll-off” occurred – defined as when

power output was less than 5 watts or tissue impedance was more than 700 Ohm. The total

ablation time for each procedure was recorded. For the second and third procedure 9 volts of

DC was provided to the liver tissue for 10 minutes before the RF 3000 generator was

switched on. This 10 minutes of 9V DC is called the “pre-treatment” phase. Thereafter both

electrical circuits were allowed to run concurrently until “roll-off” occurred.

A thermocouple was inserted 10 mm into the liver tissue along the RF electrode track, and

temperature recordings were taken at the following 4 different time-points:

1. Baseline – before the start of any procedures

2. Pre-treatment – after 10 minutes of DC, before the start of RFA

3. Highest temperature achieved during ablation

4. End temperature – when “roll-off” occurred

87

Upon completion of the procedure the abdomen was inspected for any signs of haemorrhage

or injuries to the liver and the surrounding organs. The abdominal wall was closed in layers

using 1/0 Maxon for the fascia and 3/0 Caprosyn for the skin. The wound edges were

infiltrated with 20 ml of 0.5% bupivacaine and the anaesthesia was reversed. A single dose of

noracillin (0.3 mg/kg) was given intra-muscularly. The pigs were also provided with intra-

muscular injections of buprenorphine (0.5 mg/kg) 12-hourly, and ketoprofen (3mg/kg) daily.

Once self-ventilation recommenced, the endotracheal tube was removed and the pigs returned

to an individualized, warm pen and closely monitored for signs of distress until it was awake

and able to stand. Food and water was supplied so the pigs could recommence eating as soon

as they wished.

At 48 hours, the pigs were sacrificed under anaesthesia by lethal intravenous injection of

sodium pentobarbitone. The liver and the surrounding organs were inspected for signs of

injury or haemorrhage. After death skin biopsies were taken from the areas where the ECG

dots were placed and put in 10% buffered formalin. The livers were then harvested and the

ablation zones resected in their entirety. The axial diameter (parallel to the electrode insertion

track) and two transverse diameters (at right angles to each other) were measured between the

white zones of the ablation. The liver specimens were then placed in 10% buffered formalin

and subjected to histological examination under haematoxylin and eosin stain. Photographs

were taken of the areas where the ECG dots were placed before and after the procedure, and

at 48 hours when the animals were sacrificed.

Animal research ethics approval was obtained from the Institute of Medical and Veterinary

Service (IMVS) and the University of Adelaide animal ethics committee. The study

conformed with the Code of Practice for the Care and Use of Animals for Scientific Purposes

2004 and the South Australian Prevention of Cruelty to Animals Act 1985.

88

Results

Ten pigs were used in this study and all tolerated the procedures well and survived 48 hours

until euthanasia. There were no signs of haemorrhage or injury to the surrounding organs

when the abdomen was re-opened to harvest the liver.

The study outcome measures are shown in Table 4. There was a statistically, but not

clinically, significant difference in the baseline temperature of the liver tissue between the

RFA compared to the BETA and RP-BEA groups (37.5°C vs. 37.9°C and 37.8°C, p<0.001).

Ten minutes of DC at 9 V did not produce statistically significantly changes in the tissue

temperature in the BETA or the RP-BEA group when compared to each other (37.8°C and

37.9°C respectively, p=0.11), or to their baseline temperature. The highest tissue temperature

recorded during the ablation process in the RFA, BETA and RP-BEA groups were 87.1°C,

73.3°C and 71.6°C respectively; the differences did not achieve statistical significance

(p=0.07). Similarly, there were no statistically significant differences in the end temperature

(when ablation “rolled-off”) between the 3 groups (p=0.18).

The duration of ablation was significantly longer in the RFA and BETA group compared to

the RP-BEA groups (148s and 84s and 48s, p=0.004).

The sizes of ablations were smaller in all three dimensions in RP-BEA compared to standard

RFA (Table 4). The transverse diameter A & B, and the axial diameter in RP-BEA were

12.5mm, 9.1mm and 18.1mm; which were significantly smaller than those produced by

standard RFA (15.8mm, 12.4mm and 22.3mm respectively; p<0.05). The transverse diameter

B and the axial diameter were also significantly smaller in RP-BEA compared to BETA-skin

(9.1mm vs. 11.6mm, p=0.001; and 18.1mm vs. 21.4mm, p=0.006).

89

Standard RFA BETA-

skin RP-BEA p-value

Temperature

Baseline (°C) 37.5 a 37.9 b 37.8 b <0.001

Pre-treatment (°C) n/a 37.8 37.9 0.11

Highest (°C) 87.1 73.3 71.6 0.07

End (°C) 78.5 67.6 68.2 0.18

Duration of Ablation (seconds) 148 a 84 a 48 b 0.004

Size of ablation (mm)

Transverse diameter A 15.8 a 13.2 a,b 12.5 b 0.04

Transverse diameter B 13.4 a 11.6 a 9.1 b 0.001

Axial diameter 22.3 a 21.4 a 18.1 b 0.006

Table 4. Tissue temperature, duration of ablation and size of ablation in the Standard RFA, BETA-skin and RP-

BEA groups respectively.

In four animals (Pig 1-4) small skin ulcers (2-3 mm) were noted after RP-BEA where the

cathode was placed on the skin (Figure 5). In Pig 4 the skin ulcer healed completely after 48

hours at euthanasia and was no longer macroscopically evident. Microscopic examination

showed variable extent of coagulation necrosis of the epidermal layer. Some sections showed

only intra-epidermal necrosis of stratum spinosum with intact stratum corneum and stratum

basale. Other sections however showed extensive coagulation necrosis of the whole

epidermal layer down to the upper dermis. There was minimal inflammatory reaction in the

dermal layer. After the BETA-skin procedure, there were no macroscopic or microscopic

changes to the skin where the anode was placed (Figure 5.a).

Macroscopic examination of the RP-BEA liver specimens showed cylindrical lesions clearly

demarcated from the surrounding viable tissues (Fig. 6.a). Immediately adjacent to the

90

electrode track was a rim of tissue of pale discoloration corresponding to an area of

coagulation necrosis. Surrounding this area of coagulation necrosis is a thin envelope (1-2

mm) of hyperaemic zone where viable cells could still be found. Microscopic examination

revealed a central haemorrhagic wound track with widespread disruption of hepatic cords and

individualisation of degenerate and necrotic hepatocytes (Fig. 6.b). In more severely injured

parts of the wound track, there was coagulation necrosis of hepatocytes with preservation

only of cellular outlines. At the periphery of the track, there was sometimes a mild to marked

neutrophilic reaction and, in many wound tracks, a severe necrotising vasculopathy,

sometimes attended by thrombosis. The mentioned features are similar to those seen in

standard RFA and BETA-skin liver specimens (Fig. 7 & 8).

Figure 5 (a) & (b)

Figure 6 (a) & (b)

91

Figure 7 (a) & (b)

Figure 8 (a) & (b)

Discussion

BETA and RP-BEA work in a similar way to RFA, which uses thermal energy to cause

cellular coagulation necrosis. The main purpose of combining the cathode of the DC circuit

with the RF electrode is to increase tissue hydration which will delay premature tissue

desiccation, therefore allowing the ablation process to continue for a longer period of time

and produce larger ablations.

92

It can be inferred from the results of this experiment that it was the hydrating effect of the DC

at the cathode that improved the efficacy of BETA, therefore leading to the larger ablation

size compared to standard RFA. Reversing the polarity of the DC, as in RP-BEA, desiccated

the liver tissues, causing “roll-off” to occur earlier compared to standard RFA.

An observation in this study worth noting is that the sizes of ablations were similar in

standard RFA and BETA-skin. This is in contradiction to the results obtained by Dobbins et

al who reported that BETA produced larger ablations than standard RFA (18 mm vs. 15.33

mm, p=0.001)[13]. There were some differences between their study protocol and the current

one. Dobbins et al ran 9V of DC for 15 minutes, compared to 10 minutes in this study. In

addition they set the RF 3000 generator to deliver 80 W of energy, compared to 40 W in this

study. Lastly, Dobbins et al used a much larger electro-surgical grounding pad, as compared

to the ECG dots used here. These different protocols could explain the conflicting results

described above.

There were no differences in the temperature profiles between the three ablation groups

investigated. Previous studies on electrolysis found minimal changes in tissue temperature of

approximately 4 °C[225]. In the current study, although the p<0.001 for baseline temperature,

the difference was only 0.4 °C and therefore not clinically significant. After 10 minutes of

DC at 9V, the pre-treatment tissue temperature essentially remained unchanged compared to

baseline levels. There was no pre-treatment temperature for the RFA group as DC was not

used. Although the average highest- and end-temperatures were much higher (14-16 °C) in

the RFA compared to the BETA and RP-BEA group, the differences were not statistically

significant. The reason for this observation is not clear. One possible explanation is that the

DC interfered with the circuitry of the thermocouple, which measures temperature based on

electrical conductivity. However, the thermocouple was always checked before and after each

procedure and found to be functioning properly. The thermocouple was always inserted 10

mm into the liver tissue along the same track as the RF electrode. However there is always

the possibility that the location and distance of the thermocouple from the RF electrode was

different across the study groups.

93

It is also not clear why or how the first four pigs developed skin ulcers where the cathode was

attached to the skin using ECG dots during RP-BEA. Histological examination under H&E

stains showed features of coagulation necrosis involving mostly the superficial epidermal

layer, but with some focal areas of full epidermal necrosis. The process of electrolysis itself

can cause tissue injury, mainly due to increased pH levels from the accumulation of sodium

hydroxide. Conventional electrolytic therapy takes a much longer time to cause cellular

injury, usually in the order of several hours. In this study, the whole process of RP-BEA took

on average only 10.8 minutes (10 minutes pre-treatment with DC + average ablation duration

48 seconds). A leakage of the alternating current from the RF 3000 generator into the DC

circuit leading to thermal injury is also a possibility, although less likely for several reasons.

Firstly, no skin injury was observed where the anode was attached to the skin using similar

ECG dots in the BETA group. Secondly and most importantly, these skin ulcers were only

observed in the first four pigs. If the fault was indeed due to electrical leakage, one would

expect to see the skin ulcers in all 10 pigs.

In conclusion this study showed that RP-BEA (which combines the anode of a DC circuit to

the RF electrode) leads to a shorter duration of ablation and smaller ablation size compared to

standard RFA and BETA. The anode desiccated the tissues adjacent to it, leading to the

observations as described above. Therefore the theory that BETA (which combines the

cathode of a DC circuit to the RF electrode) increases ablation size due to the effects of

increased tissue hydration around the RF electrode is correct.

94



CHAPTER 5:

Experiment 2

Bimodal Electric Tissue Ablation (BETA): A study on Ablation Size when the Anode is placed on the Peritoneum and the Liver

Tiong LU (MBBS)‡, Finnie JW (BVSc, PhD, FRCVS)*, Field JBF (PhD, AStat)†, Maddern GJ (PhD, MS, MD, FRACS)‡


*SA Pathology, Institute of Medical and Veterinary Science, Adelaide, Australia


Journal of Surgical Research 2011 – accepted paper

95




Title of Paper: Bimodal Electric Tissue Ablation (BETA): A study on Ablation Size when the Anode is placed on the Peritoneum and the Liver Journal of Surgical Research 2011 – accepted paper




Dr. Finnie, JW

Performed histo-pathological analysis of specimens


96



Dr. Field, JBF



Prof. Maddern, GJ



97

Experiment 2: Bimodal Electric Tissue Ablation (BETA) – A

Study on Ablation Size When the Anode is placed on the

Peritoneum and the Liver

Introduction

Surgical resection is the gold standard treatment for resectable liver cancers e.g. HCC and

liver metastases. However, only 20% of liver cancers are amenable to surgical resection[88,

256, 257]. RFA is a technique that is increasingly used to treat un-resectable liver tumours. It

is a minimally invasive therapy with low morbidity and mortality rates, and can be performed

percutaneously in a “day-surgery” setting. However the long term outcomes of RFA for liver

tumours are inferior to surgical resection due to the high local tumour recurrence rates. This

is related to the incapability of RFA to achieve complete ablation of the whole tumour,

especially when the size of the tumour is >3 cm[185, 258, 259].

Numerous modifications have been made to both the RF generator and the electrode design to

increase the size of tissue ablation achievable. One recent discovery is BETA which

combines the cathode of a DC circuit to the RF electrode to increase the size of tissue

ablation[11-14]. The cathode will increase the hydration of the tissues around it which will

delay tissue desiccation and “roll-off” during an ablation. Therefore, it allows the ablation

process to continue for a longer period of time resulting in larger ablations.

BETA is still a new technique in the field of ablative therapy, therefore, its safety and

efficacy profile needs to be ensured before its use can be translated into the clinical setting.

One of the problems with BETA identified in a previous study was the tissue injury

associated with the positive electrode (anode). In their animal studies, Dobbins et al attached

the anode to a scalpel blade which was inserted into the subcutaneous tissue which

subsequently resulted in a full thickness skin necrosis[12]. In retrospect this was not

unexpected considering that in previous experiments involving electrolytic therapy, various

cytotoxic chemicals were shown to be produced at the anode including acidic hydrogen ions

98

and chlorine[224-226]. Chlorine reacts with the hydrogen ions and water to form

hydrochloric and hypochlorous acid[224-226]. As a result of these reactions, the pH in the

vicinity of the anode drops to around 1-2 with lethal consequences to the surrounding

cells[224-226]. A complication such as this is clearly unacceptable in humans.

Dobbins et al proceeded to investigate an alternative method to use the anode with the aim of

preventing skin injury[13]. They hypothesized that increasing the surface area of the anode it

will reduce the current density per centimetre square of tissue, thereby reducing the risk of

local tissue injury. They replaced the scalpel blade with a dispersive grounding pad similar to

the ones used for electro-surgical units. This had the advantage of being easily available, and

could be conveniently placed on the skin which was attractive considering that many ablative

therapies are carried out percutaneously. Dobbins et al compared the severity of tissue injury

occurring at the anode (scalpel blade versus dispersive pad), and also the diameter of the

ablations achieved. Standard RFA were also performed as controls. They reported mild skin

erythema in three out of the six pigs where the dispersive pads were used as the anode. These

changes resolved completely in all three animals, and at 48 hours during post-mortem

examination the tissue at the site where the dispersive pads were used showed no changes

compared to controls. Full thickness skin necrosis was observed in all animals where scalpel

blades were used. However the size of ablation was significantly smaller when the dispersive

pads were used compare to the scalpel blade (1.8 cm vs. 2.5 cm, p<0.001). A possible

explanation for this observation was that the outer skin of the pigs was very thick, and is a

poor conductor of electricity. This leads to greater resistance to the flow of direct current

when the dispersive pad was placed on the skin compared to the scalpel blade inserted

subcutaneously. Therefore less water would accumulate in the tissues around the cathode

leading to earlier tissue desiccation and roll-off, and resulting in smaller ablation sizes.

Thus an alternative solution is required which maximizes the benefits of the DC, while

minimizing the complications. This study investigated two alternative sites to place the

anode, one on the internal abdominal wall (parietal peritoneum) and another on the liver, in

order to improve the efficacy and safety profile of BETA.

99

Materials and Methods


(Adelaide) using domestic female white pigs each weighing approximately 50 kg. All


experiment for acclimatization. The animals were housed in individual pens maintained at

23±1 ºC at ambient humidity. Lighting was artificial with a 12-hours on/off cycle. The air


and use of experimental animals. The pigs were fed and watered ad libitum (standard grower

diet of 0.7 g of available lysine per mega-Joule of digestible energy, with a digestible energy

content of 14 MJ/kg). Water quality was suitable for human consumption.

Pre-operatively, the pigs were fasted for 12 hours. Each pig was sedated with an intra-


using of 1.5% isoflurane mixed in oxygen. An endotracheal tube was placed to maintain the

airway and a temperature probe was placed inside the endotracheal tube to monitor the core


assist in temperature homeostasis. A pulse oximeter was placed on the pigs tongue to monitor

oxygen saturations. Throughout the procedure temperature, oxygen saturations, end-tidal

carbon dioxide levels, heart rate and cardiac rhythm were monitored. The pig received 0.9%

normal saline solution through an intravenous line throughout the course of the procedure.

The abdomen was cleaned with iodine solution and square-draped with sterile towels. A mid-

line incision was made from the xiphi-sternum to the umbilicus. The falciform ligament was

divided and the liver mobilized inferiorly. The porcine liver exhibits deep fissures that divide

it into left lateral and medial and right lateral and medial lobes. Additionally, the short

quadrate lobe and the caudate process are present centrally[255]. All experimental procedures

were carried out in the liver tissue thick enough to accommodate the whole ablation. The

surrounding organs were protected and packed away with moist gauze packs.

Four different ablation configurations were carried out in each pig as follows:

100

1. Standard RFA only.

2. BETA-skin – with the anode attached to the skin using ECG dots. The anode was

always placed on the left side and 10 cm away from the midline incision.

3. BETA-peritoneum - with the anode attached to the parietal peritoneum using ECG

dots. The anode was always placed on the internal abdominal wall on the left side and 10 cm

away from the midline incision.

4. BETA-liver - with the anode attached to the liver using ECG dots. The anode was

always placed on the surface of the liver away from the RF electrode.

A Boston Scientific RF 3000 generator was used to provide the radiofrequency energy.

Aluminium rods measuring 4 x 0.2 cm were used as the RF electrode and were inserted 2 cm

into the liver tissue. The grounding pad for the RF 3000 generator was placed on the inner

thigh of the animals’ hind-leg.

A generic AC/DC adaptor was used to provide the DC. In BETA the cathode of the AC/DC

adaptor was connected to the RF electrode wiring using a 1 mH inductor. This inductor

allows the flow of the DC into the RF circuit, but prevents the leakage of alternating current

from the RF 3000 generator into the DC circuit. Therefore the needle electrode of the RF

3000 generator and the cathode of the DC circuit are one and the same. The anode of the DC

circuit was attached to the three different places as described above using a standard ECG

diagnostic electrode (ECG dots).

For the first procedure, only the RF energy was delivered to the liver tissue. The RF 3000

generator was started to deliver 40 watts of energy until “roll-off” occurred – defined as when

power output was less than 5 watts or tissue impedance was more than 700 Ohm. The total

ablation time for each procedure was recorded. For the second, third and fourth procedure, 9

volts of DC was provided to the liver tissue for 10 minutes before the RF 3000 generator was

switched on. This 10 minutes of 9V DC is called the “pre-treatment” phase. Then both

electrical circuits were allowed to run concurrently until “roll-off” occurred.

101

A thermocouple was inserted 1 cm into the liver tissue along the RF electrode track, and

temperature recordings were taken at the following 4 different time-points:

1. Baseline – before the start of any procedures

2. Pre-treatment – after 10 minutes of DC, before the start of RFA

3. Highest temperature achieved during ablation

4. End temperature – when “roll-off” occurred

Upon completion of the procedure the abdomen was inspected for any signs of haemorrhage

or injuries to the liver and the surrounding organs. The abdominal wall was closed in layers

using 1/0 Maxon for the fascia and 3/0 Caprosyn for the skin. The wound edges were

infiltrated with 20 mL of 0.5% bupivacaine and the anaesthesia was reversed. A single dose

of noracillin (0.3 mg/kg) was given intra-muscularly. The pigs were also provided with intra-

muscular injections of buprenorphine (0.5 mg/kg) 12-hourly, and ketoprofen (3mg/kg) daily.

Once self-ventilation recommenced, the endotracheal tube was removed and the pigs returned

to an individualized, warm pen and closely monitored for signs of distress until awake and

able to stand. Food and water was supplied so the pigs could recommence eating as soon as

they wished.

At 48 hours, the pigs were sacrificed under anaesthesia by lethal intravenous injection of

sodium pentobarbitone. The liver and the surrounding organs were inspected for signs of

injury or haemorrhage. After death biopsy specimens were taken from the spots where the

anodes were placed (skin, parietal peritoneum and the liver) and put in 10% buffered

formalin. The livers were then harvested and the ablation zones resected in their entirety. The

axial diameter (parallel to the electrode insertion track) and two transverse diameters

(perpendicular to each other) were measured between the white zones of the ablation. The

liver specimens were then fixed in 10% buffered formalin for examination under

haematoxylin and eosin (H&E) stain. Photographs were taken of the areas where the ECG

102

dots were placed before and after the procedure, and at 48 hours when the animals were

sacrificed.


Service (IMVS) and the University of Adelaide animal ethics committee. The study



Results

Ten pigs were used in this study and all tolerated the procedures well and survived 48 hours

until euthanasia. There were no signs of haemorrhage or injury to the surrounding organs

when the abdomen was re-opened to harvest the liver.

The study results can be seen in Table 5. The baseline temperatures of the liver tissue were

essentially the same between all the groups although there were statistically, but not

clinically, significant differences. Ten minutes of 9V DC did not produce significant changes

in the tissue temperature in the BETA-skin, BETA-peritoneum or the BETA-liver group

when compared to one another (37.8°C, 38.6°C and 38.5°C respectively, p=0.11), or to their

baseline temperature. The highest tissue temperature recorded during the ablation process in

the RFA, BETA-skin, BETA-peritoneum and the BETA-liver groups were 87.1°C, 73.3°C

and 88.7°C and 84.7°C respectively; the differences did not achieve statistical significance

(p=0.21). Similarly there were no statistically significant differences in the end temperature

(when ablation “rolled-off”) between the four study groups (p=0.39).

103

RFA BETA-

skin

BETA-

peritoneum

BETA-

liver p-value

Temperature

Baseline (°C) 37.5 a 37.9 c 37.5 a,b 37.8 bc 0.002

Pre-treatment (°C) n/a 37.8 38.6 38.5 0.11

Highest (°C) 87.1 73.3 88.7 84.7 0.21

End (°C) 78.5 67.6 79.5 78.2 0.39

Duration of ablation

(seconds) 154 a,b 84 a 220 b 214 b 0.006

Size of ablation (mm)

Transverse diameter A 15.8 b 13.2 a 20.8 c 18.5 c <0.001

Transverse diameter B 13.4 a 11.6 a 18.5 b 17.3 b <0.001

Average transverse

diameter 14.6 b 12.4 a 19.7 c 17.9 c <0.001

Axial diameter 22.3 21.4 24.5 23.7 0.09

Table 5. Each variable was examined using a randomised block analysis of variance with pigs as blocks.

Duration was analysed on a log scale. The table gives the mean for each treatment. The mean followed by the

same letter are not significantly different from each other (at p=0.05, using Fisher’s protected least significant

differences).

The duration of ablation in the BETA-peritoneum and BETA-liver groups was 220 seconds

and 214 seconds respectively, which were significantly longer than the BETA-skin group (84

seconds). The duration of ablation in the standard RFA group (154 seconds) was not

significantly different from those in the BETA-skin, BETA-peritoneum or BETA-liver

groups.

104

The transverse diameter A and B in the BETA-peritoneum and BETA-liver groups were

significantly larger when compared to the RFA and BETA-skin groups (p<0.001). The

average transverse diameter in the BETA-peritoneum, BETA-liver, RFA and BETA-skin

groups were 19.7 mm, 17.9 mm, 14.6 mm and 12.4 mm respectively (p<0.001). The axial

diameter in the BETA-peritoneum and BETA-liver groups were also larger compared to the

RFA and BETA-skin groups, although the differences did not reach statistical significance

(p=0.09).

Macroscopic and Microscopic Findings

The gas bubbling was most vigorous during BETA-peritoneum and BETA-liver compared to

BETA-skin. There was no gas bubbling during RFA as no DC energy was provided. This

indicated that the electrolytic process was most active when the anode was placed on the

peritoneum and the liver compared to the skin.

Macroscopic and microscopic examinations of the skin specimens where the anode was

placed in the BETA-skin group showed no signs of local tissue injury (Figure 9).

On the internal abdominal wall where the anode was placed on the peritoneum (BETA-

peritoneum), there was a circular area of erythema which persisted up to 48 hours when the

liver was harvested. The circular shape corresponded to the ECG dots used. Microscopic

examination showed focal coagulation necrosis involving the serosal lining of mesothelium

and sub-mesothelial layer of connective tissue, attended by minor haemorrhage, fibrin

deposition, a mild mixed inflammatory infiltrate (chiefly neutrophils), and early fibrovascular

granulation tissue formation (Figure 10)

There was a similar discoid area of purplish discoloration on the liver where the anode was

placed in the BETA-liver group. This discoloration however was no longer visible

macroscopically at 48 hours. Under microscopic examination the discoid area of

discoloration corresponded to an extensive, but of variable severity, area of coagulation

105

necrosis of the collagenous hepatic (Glisson’s) capsule with a mixed neutrophilic and

lymphoplasmacytic infiltrate and early fibroblastic invasion. The underlying liver

parenchyma was normal (Figure 11).

(a) (b) (c)

Figure 9. Morphology of the skin pre-operatively (a) compared to post-operatively (b) The brown discoloration

in the picture on the right was from iodine solution (c) Microscopic examination showing normal skin.

(a) (b) (c)

Figure 10. BETA-peritoneum (a) A circular area of erythema and inflammation was evident on the internal

abdominal wall immediately after procedure, and (b) 48 hours later (c) H&E (x4 magnification) showed

coagulation necrosis of the peritoneal serosa and superficial submesothelial connective tissue.

106

(a) (b) (c)

Figure 11. Beta-liver (a) A similar circular area of inflammation on the liver, the anode was placed on the

surface of the liver away from the RF electrode (b) 48 hours later the inflammation was no longer visible

macroscopically, but H& E examination (c) showed coagulation necrosis involving the liver capsule with

sparing of the underlying liver parenchyma.

Discussion

The ECG dots used in this study worked well as the anode of the DC circuit. It conducted

electricity well and avoided the unnecessary trauma of inserting an electrode into the animal

tissues. However it was difficult to stick the ECG dots onto the abdominal wall or the liver

because of the “wetness” of those surfaces. A pack was used to hold the ECG dots against the

peritoneal and liver surfaces. Therefore it might be impractical to be used in humans in the

clinical setting.

Rigorous gas bubbling, a sign of DC activity, could be seen during the pre-treatment phase

and was more active in the BETA-peritoneum and BETA-liver groups compared to the

BETA-skin group. This observation correlated with previous experiments showing better

electrical conductivity in the peritoneum and the liver tissues compared to the skin[260].

The results from this study showed that the more “active” the DC was, the larger the ablation

size. Better electrical conductivity led to more rigorous electrolytic reactions which meant

that there was more net movement of the water molecules from the anode to the cathode. The

relatively higher tissue hydration in the BETA-peritoneum and BETA-liver groups compared

to the BETA-skin and RFA groups meant that the ablation process could proceed for a

significantly longer period of time before “roll-off” occurred. Consequently larger ablations

107

were obtained in the BETA-peritoneum and BETA-liver groups compared to the latter (Table

5).

It was observed during the course of the animal study that the distance between the cathode

and the anode might affect the size of ablation produced. During BETA-liver the anode (ECG

dot) was placed on the surface of the liver away from the RF electrode (to which the cathode

was attached). It was noted that when the anode was placed on the opposite surface of the

liver to the RF electrode, the duration of ablation would be shorter and the ablation size

relatively smaller compared to when the anode was placed on a separate liver lobe. In

addition the axial and the average transverse diameter in BETA-liver were slightly smaller

when compared to BETA-peritoneum in this study (Table 5). Therefore putting the cathode

and the anode too closely together could negate the benefits of BETA. When the anode was

placed in close proximity to the cathode, the distance between the two electrodes might be

too small for any meaningful transport of the water molecules.

We found that the anode still produced localized tissue injury when attached to the

peritoneum or the liver using ECG dots. There was visible local tissue inflammation after the

ablation process, with evidence of coagulation necrosis under microscopic H&E examination.

The extent of the injury however was superficial and not as severe as those in previous

studies[12, 13].

Placing the anode on the peritoneum may be undesirable as it produced localised coagulation

necrosis of the superficial epithelium. The extent of injury was not as severe as the full

thickness skin necrosis as seen in previous studies[12, 13]. Nevertheless any peritoneal injury

can induce adhesions which can cause complications such as bowel obstructions. In addition,

the peritoneal surface is not accessible to place the anode during percutaneous RFA.

The liver could be an ideal place to put the anode. During conventional electrolytic therapy,

the anode will induce small vessel thrombosis with resultant wedge ischaemia/infarct in the

liver tissues distally[227]. Therefore the anode could be inserted into a proximal location

108

relative to the tumour to be ablated. This will induce thrombosis of the vessels feeding the

tumour causing ischaemia, and may have a synergistic effect with the subsequent RFA. The

distance between the electrodes must not be too close; otherwise the main benefit of the DC

to increase tissue hydration around the RF electrode is lost. A second option is to insert the

anode into a different liver lobe to where the tumour is located. This gives the distance

required for effective tissue hydration at the cathode. These two options would limit the

iatrogenic injury to the liver only. There would be coagulation necrosis at the anode, but the

amount of tissue involved would be minimal as the whole BETA process will take

significantly less time than the conventional electrolytic therapy. In addition the liver has a

large functional reserve and excellent regenerative capability.

An alternative option is to induce artificial ascites in the intra-abdominal compartment using

0.9% normal saline solution, and immerse the anode in this solution. Artificial ascites using

normal saline solution has been employed when RFA was used to treat

superficial/subcapsular tumour in close proximity to surrounding organs e.g. bowels and

stomach[261, 262]. This method appeared to be well tolerated with minimal morbidity. The

artificial ascites act as an intermediary medium between the anode and the biological tissues.

The presence of sodium chloride in the solutions greatly facilitates electrical conduction. In

addition the larger surface contact area between the saline solution and the biological tissues

minimizes any adverse effects normally seen at the anode, as the toxic chemicals produced

are diluted in the saline solution.

The idea of using an electrosurgical grounding pad attached to the skin as the anode is very

attractive as it cheap and readily available. The downside of this was the fact that the ablation

sizes produced were not as large compared to when the anode was placed intra-abdominally.

The hypothesis was that the skin has a high electrical resistivity, therefore minimizing the

hydrating effect at the cathode. It is possible that the pig skin is thicker than human skin,

therefore causing higher electrical resistivity. As BETA has never been tested in humans, it is

not known what the effect of using an electrosurgical grounding pad on the human skin as the

anode would be. Therefore further research is required to investigate how to maximize the

efficacy of the DC with the anode attached to the skin using an electrosurgical grounding pad.

109

There are several possible methods to reduce the electrical resistivity of the skin. One is to

wet the skin which will greatly facilitate electrical conduction[263]. The risk with this

obviously is the possibility of causing electrical burns. Another method to improve electrical

conductivity of the skin is to scrap the superficial layers of the skin off. Previous studies have

shown that the resistivity to DC and AC resides almost exclusively in the stratum

corneum[263, 264]. The stratum corneum is only approximately 15-20 μm thick[265, 266]

and consists of anucleated cells which contain only 15% water[241]. Subsequent layers of the

epidermis contain 70% water and therefore have electrical resistivity similar to internal

organs[241]. One group of researchers found that combing the hair scraps off the superficial

layer of the scalp greatly facilitates measurement of the brainwaves activity during an

electroencephalography (EEG)[267]. Therefore simple methods, e.g. applying sticky tape to

the skin and stripping it off multiple times before putting on the anode, may improve

electrical conductivity. However further studies are required to ensure that such methods do

not cause unwanted side effects such as skin irritation, or increasing the skin’s susceptibility

towards electrochemical injury.

In conclusion, BETA produced larger tissue ablations compared to standard RFA, and hence

could be used to treat larger tumours more effectively and potentially reduce the tumour

recurrence rates. The efficacy of BETA depends on ensuring good electrical conductivity

between the cathode and the anode of the DC circuit. Research so far has shown that BETA

works best when the anode is placed deep into the skin layer as the stratum corneum consists

of a layer of anucleated cells which have high electrical resistivity. The liver could be the

ideal location to place the anode as it has excellent electrical conductivity, therefore ensuring

maximum tissue hydration around the cathode to produce the largest ablations possible.

Future studies should investigate the effect of the distance between the cathode and the anode

on the size of tissue ablation in BETA. There might be an optimum distance between the two

electrodes which will produce the largest tissue ablation.

110



CHAPTER 5:

Experiment 3

Bimodal Electric Tissue Ablation (BETA) compared to the Cool-Tip RFA System

Tiong LU (MBBS)‡, Field JBF (PhD, AStat)†, Maddern GJ (PhD, MS, MD, FRACS)‡



Australian and New Zealand Journal of Surgery 2011 – accepted paper

111




Title of Paper: Bimodal Electric Tissue Ablation (BETA) compared to the Cool-Tip RFA System Australian and New Zealand Journal of Surgery 2011 – accepted paper




Dr. Field, JBF



112



Prof. Maddern, GJ



113

Experiment 3: Bimodal Electric Tissue Ablation Compared to

Standard Radiofrequency Ablation Using the Cool-Tip RF

System

Introduction

RFA is currently one of the most popular ablative therapies used for the treatment of un-

resectable liver malignancies, including HCC and secondary liver metastases[1, 85, 268]. The

most important aim of RFA is to achieve complete ablation of a tumour. However the

fundamental problem with the existing RFA technology is the limited ablation size that is

achievable, leading to incomplete tumour ablation. Currently some RFA equipment is

capable of creating a lesion up to 5 cm in size during a single application[269-271]. This is

only adequate for treating a tumour 3 cm in size, with a 1 cm ablative margin to ensure

complete eradication of any cancerous cells. In reality the ability to achieve complete ablation

including a safety margin of 1 cm is usually more complicated. Numerous factors such as

tumour size >3 cm, irregularly shaped tumours, tumours adjacent to vital structures e.g.

bowel or gallbladder, and the “heat sink” effect from major vessels can all affect the complete

ablation rates[149, 155]. For these reasons, RFA is associated with higher rates of local

disease recurrence compared to other curative treatment such as resection or liver

transplantation for HCC[153, 217, 218].

One of the latest developments in the field of RFA is BETA[11-14]. BETA incorporates the

process of electrolysis (which uses direct current) into RFA to increase the size of tissue

ablation. The cathode of a DC circuit is attached to the RF electrode so that both the RF and

DC energy can be administered at the same time. The electrochemical reactions at the

cathode (which is attached to the RF electrode) attract water molecules to the tissues

surrounding it[7, 239]. The increased tissue hydration was postulated to delay the process of

desiccation during RFA, which prolongs the duration of ablation therefore producing larger

ablations compared to standard RFA[11, 14].

114

Previous research on BETA used a 5-15 minutes pre-treatment phase where 9V of DC was

used to increase tissue hydration before the RF generator was started[12-14]. Thereafter both

electrical circuits were allowed to run at the same time until “roll-off” occurred. Dobbins et al

used the RF 3000 generator (Boston Scientific) with multi-tined LeVeen needle electrodes,

and produced significantly larger ablations compared to standard RFA[14].

The RF 3000 generator (Boston Scientific) is an impedance based machine. This means that

an ablation process typically continues until “roll-off” occurs; defined as when the tissue

electrical impedance increases dramatically which precludes any further electrical

conduction. The RF generator then automatically stops any further power output. Therefore

BETA can improve the efficacy of the RF 3000 system (Boston Scientific) by delaying tissue

desiccation and prolonging the ablation process resulting in larger ablations.

Another popular RF system in the market is the Cool-Tip RF system (Covidien) which uses

internally-cooled electrodes (ICE) for ablation. This system circulates chilled saline to the tip

of the electrode needle, which lowers the temperature of the tissue immediately adjacent to it.

In addition, the generator has an “impedance mode” which makes it capable of continuously

monitoring the tissue impedance during an ablation. When it senses that the tissue impedance

has increased by ≥15 Ohms from baseline level, the power output will be ceased temporarily.

These intermittent pauses allow gases adjacent to the electrode to dissipate while the internal

cooling with chilled-saline minimizes tissue charring, hence improving the delivery of energy

to surrounding tissues[36]. Data from the manufacturer’s website claims that an ICE with a 3

cm exposed tip can produce a lesion 3.6 x 3.1 x 3.7 cm in diameter in ex-vivo bovine livers

after 12 minutes of ablation. The endpoint of an ablation process using the Cool-Tip RF

system is based on time (12 minutes), rather than upon “roll-off”. Therefore whether the

principle of BETA can be applied to the Cool-Tip RF system (Covidien) to improve its

efficacy is unknown and has never been tested before.

Besides this, all previous research using in-vivo animal models had run the DC circuit before

(for 5-15 minutes) and during RFA[11-14]. It was not known whether applying the DC only

during the pre-treatment phase, followed by just the RFA (named ECT/RFA in this thesis),

could produce similar results compared to BETA. This information has important practical

115

and financial implications. If the 2 methods have similar efficacy, it is possible to use the

existing equipment (ECT and RFA are widely used all over the world) and start treating

patients more effectively without the need to spend large sums of money or time to develop a

new BETA machine.

Therefore, the purpose of this experiment is two-fold:

1. To investigate whether the principle of BETA can be applied to the Cool-Tip RF

system (Covidien) to produce larger ablations compared to standard RFA.

2. To compare the size of ablations produced in ECT/RFA (defined as 10 minutes of

pre-treatment with 9V DC followed by RFA only) to standard RFA and BETA.

Methods


(Adelaide) using domestic female white pigs each weighing approximately 40-50 kg. All


experiment for acclimatization. The animals were housed in individual pens maintained at

23±1ºC at ambient humidity. Lighting was artificial with a 12-hours on/off cycle. The air


and use of experimental animals. The pigs were fed and watered ad libitum.

Pre-operatively, the pigs were fasted for 12 hours. Each pig was sedated with an intra-


using 1.5% isoflurane mixed with oxygen. An endotracheal tube was placed to maintain the

airway and a temperature probe was placed inside the endotracheal tube to monitor the core


assist in temperature homeostasis. A pulse oximeter was placed on the pigs tongue to monitor

oxygen saturations. Throughout the procedure temperature, oxygen saturations, end-tidal

carbon dioxide levels, heart rate and cardiac rhythm was monitored.

116

The abdomen was cleaned with iodine solution and square-draped with sterile towels. A right

subcostal incision was made to expose the liver. The falciform ligament was divided and the

liver mobilized infero-medially. The porcine liver exhibits deep fissures that divide it into left

lateral and medial and right lateral and medial lobes. Additionally, the short quadrate lobe

and the caudate process were present centrally[255].

All experimental procedures were carried out in the liver tissues thick enough to

accommodate the whole ablation. The surrounding organs were protected and packed away

with moist gauze packs. Three types of ablation setup were tested:

1. Standard RFA

2. BETA – 9V of DC for 10 minutes, then the RF generator was switched on and both

circuits operated simultaneously

3. ECT/RFA – 9V of DC for 10 minutes, then followed by RFA only

A Cool-Tip RF Ablation System (Covidien) capable of producing 200 watts of energy at 480

kHz was used in this study. The parameters displayed on the panel of the generator included

tissue temperature and impedance, electrical current, power output and time. The generator

comes with a peristaltic pump that circulates chilled saline through the needle electrodes. The

Cool-Tip RF Ablation System (Covidien) has an automatic feedback algorithm which

continuously monitors tissue impedance and adjust power output to maximize energy

delivery. When tissue impedance rises more than 15 Ohm above baseline during an ablation,

the process automatically pauses for 20 seconds before the generator delivers any more

energy[36]. The intermittent pauses when tissue impedance increased allowed gases adjacent

to the electrode to dissipate while the internal cooling with chilled-saline minimizes tissue

charring, hence improving the delivery of energy to surrounding tissues. The grounding pad

for the RFA generator was attached to the inner hind-leg of the animal.

117

Two 20 cm internally cooled electrodes (ICEs) with 3 cm exposed tips were used in this

study. Each of the electrodes had insulated electrical wire tubing, as well as two extra plastic

tubing to circulate chilled saline throughout the needle electrode.

A generic AC/DC adaptor was used to provide the DC. In BETA and ECT/RFA the cathode

of the AC/DC adaptor was connected to the RF electrode wiring using a 100 mH inductor.

This inductor allowed the flow of the DC into the radiofrequency circuit, but prevented the

leakage of alternating current from the RF 3000 generator into the DC circuit. Therefore the

needle electrode of the Cool-Tip RF system (Covidien) and the cathode of the DC circuit

were one and the same. The anode of the DC circuit was attached to aluminium rods inserted

into the subcutaneous tissue. A new aluminium rod was used and inserted at a different

location during each BETA and ECT/RFA procedure.

For standard RFA, only the Cool-Tip RF system (Covidien) was used. The water pump was

started first to circulate the chilled saline (<10ºC) throughout the needle electrode. Once the

temperature of the electrode tips dropped <10ºC, they were inserted into the liver and the

ablation process started as per the manufacturer’s protocol. The generator was switched to the

impedance mode and the timer set to 6 minutes. The power output was set to 100 watts each

time.

In BETA and ECT/RFA, 9 volts of DC was provided to the liver tissue for 10 minutes before

the Cool-Tip RF system (Covidien) was switched on. This 10 minutes of 9V DC was called

the “pre-treatment” phase. After this period of pre-treatment, the RFA generator was started

and both circuits allowed to run concurrently in BETA. IN ECT/RFA however, the DC

circuit was switched off after the 10 minutes pre-treatment phase, and only the RF generator

was used as described above.

The ICEs have sensors at their tips which were capable of measuring tissue temperature and

impedance. Therefore measurements of these 2 parameters were performed at 3 time-points

during each procedure:

1. Baseline – before the start of any procedure.

118

2. Pre-treatment phase – after the delivery of DC energy, but before RFA. The RF and

DC generators and the peristaltic pump were switched off temporarily to ensure accurate

tissue temperature and impedance measurement.

3. End – after 6 minutes of RFA the maximum tissue temperature and impedance were

recorded.

Once all ablations were completed, the animals were euthanized using intravenous injections

of sodium pentobarbitone. The livers were then harvested and the sizes of the ablations were

measured in three dimensions. The axial diameter (parallel to the electrode insertion track)

and two transverse diameters (perpendicular to each other) were measured between the white

zones of the ablation.


Science (IMVS) and the University of Adelaide animal ethics committees. The study



Results

A total of 12 pigs were used in this study. Twelve ablations (RFA=4, BETA=4, ECT/RFA=4)

in three pigs were performed in a pilot study to determine the optimum ablation settings to

use for this experiment. RFA was initially conducted according to the manufacturer’s

protocol which sets the generator to deliver 100W of power for 12 minutes in the impedance

mode. However this setting was “too powerful” to use in our in-vivo liver model using 50kg

pigs. It was observed that some of the ablations would extend to both the anterior and

posterior surface of the liver, therefore making comparison of ablation sizes between the

groups impossible. As our animal laboratory had limited capacity to accommodate pigs

>50kg, we chose to modify the ablation protocol by reducing the duration of ablations to 6

minutes. Therefore the ablation setting used in this study was 100W of RF energy for 6

minutes in the impedance mode. Forty-four ablations (RFA=14, BETA=16, ECT/RFA=14) in

119

9 pigs were performed using this modified RFA protocol. None of the pigs died prematurely.

The experimental parameters measured are displayed in Table 6.

BETA ECT/RFA RFA p-value

Diameter (mm)

Transverse A 23.1 (a) 20.1 (b) 17.4 (c) <0.001

Transverse B 21.1 (a) 18.9 (b) 16.6 (c) <0.001

Axial 37.3 36.3 35.4 0.78

Impedance (Ω)

Baseline 75.6 (a) 74.4 (a) 84.7 (b) 0.004

Pre-treatment 68.2 67.4 - 0.30

End 62.4 (a) 63.1 (a) 74.2 (b) 0.001

Temperature (ºC)

Baseline 38.5 38.5 38.4 0.12

Pre-treatment 38.9 39.1 - 0.23

End 68.9 64.9 59.5 0.08

Table 6. The data was analysed using analysis of variance for unbalanced data in Genstat 13 th edition (VSN

International, UK) to remove the effects of pigs and replicates within pigs from the treatment comparisons. The

table shows means for each treatment with the significance level for the treatment comparison from the analysis

of variance. Where the significance level is less than 0.05, significance of means is indicated: means followed

by the same alphabetical letter are not significantly different at p=0.05.

The ablations achieved using BETA were significantly larger compared to ECT/RFA and

RFA. The mean transverse diameter A was 23.1 vs. 20.1 vs. 17.4 mm (p<0.001), whereas the

mean transverse diameter B was 21.1 vs. 18.9 vs. 16.6 mm (p<0.001) respectively. The mean

120

axial diameter was also larger in the BETA group compared to ECT/RFA and RFA, although

the differences were not statistically significant (37.3 vs. 36.3 vs. 35.4 mm, p=0.78).

The baseline mean liver tissue impedance was significantly higher in the RFA group

compared to BETA and ECT/RFA (84.7 vs. 75.6 vs. 74.4 Ohm, p<0.004). A similar

observation was made at the end of the ablation process (74.2 vs. 62.4 and 63.1 Ohm,

p<0.001). After 9V of DC was provided for 10 minutes in the BETA and ECT/RFA groups,

the mean liver tissue impedance was reduced to an average of 68.2 and 67.4 Ohm

respectively. There was no significant difference in the reduction of tissue impedance

between the 2 groups after the pre-treatment phase.

There were no significant differences in baseline tissue temperature between the three groups.

The tissue temperature essentially remained the same after 10 minutes of pre-treatment with

9V DC. The end tissue temperature was higher in the BETA and ECT/RFA groups compared

to the RFA group, although the differences were not statistically significant. The average end

tissue temperature in all three groups was ≥60ºC, which would be enough to cause

instantaneous cellular necrosis.

On one occasion involving standard RFA in Pig 5, a loud “popping” sound was heard during

the ablation (Figure 12). It was later discovered that the inferior surface of the liver had

“fractured” and bled quite profusely. The tear in the liver tissue, which measured 1 cm in

length, was likely caused by a high intra-tumoral pressure created by the ablation process. It

was hypothesized that the ablated liver tissue was more fragile and brittle, and could not

withstand the pressure built-up leading to the haemorrhagic fracture. Fortunately on this

occasion the procedure was the final ablation in the animal, and it was euthanized as per

protocol.

121

Figure 12. Fractured liver tissue after standard RFA associated with the “popping sound” phenomenon.

Fig. 13 BETA lesion

122

Fig 14. ECT/RFA lesion

Fig 15. RFA lesion

123

Discussion

The principle of BETA involves the use of electrolysis to improve the efficacy of RFA to

produce larger ablations. The electrochemical reactions from the DC attract water molecules

to the cathode, which is attached to the RF electrode. The increased tissue hydration will

delay tissue desiccation during RFA, therefore allowing the ablation process to continue for

longer periods of time to produce larger ablations.

In a way this process is not dissimilar to the mechanism of perfused electrodes currently used

in some RFA generators. These perfused electrodes infuse sterile saline into the tissue

interstitium before and during an ablation[74, 271, 272]. The saline infusion increases the

tissue hydration and the ionic concentration around the tissue to be ablated, and this improves

the electrical conductivity[273]. This allows the thermal energy to be distributed more

uniformly throughout the whole volume of tumour tissue to be ablated [274, 275]. Increased

tissue hydration reduces the risk of tissue desiccation adjacent to the electrode and allowed

the ablative process to continue for a longer duration of time [274, 275]. All these effects

worked together to produce larger ablations. In addition, when saline is infused into the

interstitial tissue, it acts as an extension of the metal electrode forming a “virtual” or “liquid”

electrode which has a larger surface area than the metal electrode. Previous research has

shown that the diameter of ablation was proportional to the surface area of the electrode,

hence this “liquid electrode” may produce larger ablations[74]. This perfused electrode

system is not without flaws in its concept. Infusion of saline at a high rate has been shown to

spread irregularly into the tissue and to leak along the electrode track, causing iatrogenic

thermal injury to distant structures [36, 39, 252]. Several authors have raised the possibility

that the saline contaminated with tumour cells may leak along the electrode track and cause

tumour seeding[37, 74]. Another concern is that saline infusion may cause an increase in

intra-tumoral pressure, therefore forcing tumour cells into the circulation causing distant

tumour seeding[37, 74]. The difference between the perfused electrode system and BETA

however, is that water molecules are “sucked” to the tissues surrounding the RF electrode,

instead of being infused into the interstitium. Therefore the risk of tumour seeding due to

high intra-tumoral pressure, or iatrogenic viscera/vessels/ducts injury from the hot saline is

not present in BETA.

124

The main finding of this study was that the principles of BETA could be incorporated into the

Cool-Tip RF system (Covidien) using the ICEs to increase the size of tissue ablations. The

results demonstrated the mean transverse diameter A & B produced in BETA (23.1 and 21.1

mm) were significantly larger than those in RFA (17.4 and 16.6 mm) (p<0.001). The axial

diameter was also larger in BETA compared to RFA, although the difference was not

significant (37.3mm vs. 35.4mm, p=0.78).

BETA was also proven to be more effective than ECT/RFA (where DC was only provided

for 10 minutes in the pre-treatment phase). This suggested that the beneficial effect of the DC

continued even during the RFA process. This study showed that the mean transverse diameter

A (23.1 mm vs. 20.1 mm) & B (21.1 mm vs. 18.9 mm) in BETA were significantly larger

than ECT/RFA (p<0.001). The mean axial diameter was also larger in BETA although it was

not statistically significant (37.3 mm vs. 36.3 mm, p=0.78). ECT/RFA, however, produced

significantly larger ablations compared to standard RFA. In summary, ECT/RFA increased

the size of ablation by approximately 2.5 mm compared to standard RFA, while BETA

increased it by 5 mm.

The duration of the RFA (12 minutes as per the manufacturer’s recommendation) had to be

modified to 6 minutes because the porcine livers in this study were not large enough to

accommodate the full ablations. This change applied to all three study groups and should not

biased the results in any way.

It was noted that BETA and ECT/RFA produced significantly larger ablations compared to

RFA using the Cool-Tip RF System (Covidien) despite the same duration of ablations in all

three groups. It was discovered that during each of the 6 minutes ablation, the RFA group

would “roll-off” an average of four times compared to two times in BETA and ECT/RFA.

Therefore despite the same duration of ablations in each group, the flow of energy was

actually significantly more during BETA and ECT/RFA which could explain the larger

ablations in the latter groups. Another possibility is that the increased tissue hydration around

the RF electrode allowed a more uniform and improved delivery of energy to the liver, thus

producing larger ablations.

125

The reason for the differences in the baseline tissue impedance measured between the RFA

and the other treatment groups was not clear. The order of the experiment (RFA, BETA,

ECT/RFA) performed in each pig was random to minimize any bias. It was unlikely to be due

to interference from the DC energy, as the tissue impedance was always measured with the

DC generator switched off.

The electrochemical reactions from the DC increased the tissue hydration around the RF

electrode. Besides delaying tissue desiccation and prolonging the ablation process, the

increased hydration also lowered tissue electrical impedance. The Cool-Tip ICE has a sensor

at the tip of the needle was used to measure the electrical impedance in the tissue before and

after the 10 minutes of pre-treatment with 9V DC. The data showed that the mean baseline

electrical impedance was reduced by approximately seven Ohm in both the BETA and the

ECT/RFA groups.

The tissue impedance in all three study groups dropped significantly after ablation, which is

contrary to what was expected. The tissue impedance was expected to be significantly higher

as they became desiccated which then prevented further conduction of electrical energy,

therefore resulting in roll-off of the ablation. One possible explanation is that there could be

blood seeping into the needle track, which would result in the low impedance reading. The

impedance of the ablated tissues, on the other hand, was likely to be much higher than the

baseline values.

This study also showed that the beneficial effect of the DC is not due to additional thermal

energy. The tissue temperatures before and after the 10 minutes of 9V DC were essentially

unchanged. The tissue temperatures measured using the ICEs showed that they were all

≥60ºC at the end of ablation, enough to cause instantaneous cellular necrosis.

126

The data from the current and previous research have shown that BETA can be readily

incorporated into existing RF systems such as the RF 3000 RFA System (Boston Scientific)

and the Cool-Tip RF System (Covidien). RFA and electrochemical therapy is widely used

around the world for the treatment of un-resectable liver cancers. Therefore it would not be

difficult to assimilate these two technologies to create BETA without spending enormous

amounts of time or money.

In addition, both procedures were proven to be safe with minimal morbidity and mortality

risks. The safety features of BETA have also been elucidated in animal research. Data from

Dobbins et al showed that apart from a transient rise in serum liver enzymes and

inflammatory markers, there were no long term adverse effects when BETA was tested in

pigs[12]. As the safety and efficacy of BETA has been confirmed in animal experiments, it

might be time to take a step further and bring this technology into human study.

During the course of this experiment in the 5th pig, an unexpected complication of RFA was

encountered. A loud “popping” sound was heard during the ablation process, followed by the

discovery of a “fractured” liver surface with profuse bleeding. This “popping sound” has

been described in the literature and was attributed to the high intra-tumoral pressure created

by the ablation process. In one report the incidence of the “popping sound” phenomenon was

as high as 58%[276]. In the cardiovascular literature, this phenomenon has been associated

with major complications such as ventricular wall rupture[277]. However there is no report

yet of a liver fracture or a bleeding complication as a result of this “popping”. Clinicians need

to be aware of this potentially disastrous complication especially when RFA is used

percutaneously in a day procedure setting. Under such circumstances it would be easy to miss

a liver fracture, leading to a major haemorrhagic complication.

In conclusion, this study has shown that BETA increases the size of ablation by

approximately 5 mm using the Cool-Tip RF System (Covidien) with the ICEs. The benefit of

the DC extended into the RFA phase, and therefore it should be continued for the whole

treatment duration. Providing the DC only during the pre-treatment phase (9V DC for 10

minutes in this study) also produced significantly larger ablations compared to standard RFA,

127

although the benefit is less compared to BETA. The principle of BETA works by attracting

the water molecules to the tissues surrounding the cathode, which is attached to the RF

electrode. The increased tissue hydration improves energy distribution, delays tissue

desiccation and allows the ablation process to continue for longer periods of time and

therefore produce larger ablations.

128

6. Area for Future Research

Future studies should investigate what is the optimum duration and voltage of the ECT to use

to achieve the maximum liver hydration, therefore producing the largest ablations possible.

The ECT settings used in the previous and current research (9 volts for 10 or 15 minutes)

were arbitrarily chosen, and may not be the best. The duration of the ECT must not be too

long, or it will make BETA impractical to use in the current busy hospital settings.

Another area of research is whether the principle of BETA could be incorporated into other

thermal ablative therapy such as MCT or LITT. These ablative technologies also have the

problem of premature tissue desiccation which BETA may help to overcome.

Lastly it will be very useful to have a custom-built BETA machine, which combines both the

DC and the RF circuit into one. Currently the electrical insulators of the RF electrodes have

to be removed to attach the DC circuit to them. An inductor was used to allow the DC to flow

into the RF circuit, but not vice versa. This method is crude and not suitable for clinical

human trials as the exposed electrical wirings pose an occupational health and safety risks. A

custom-built BETA machine would overcome this problem and facilitate a step further

towards human trials.

129

7. Conclusion

With better knowledge and equipment, the clinical outcomes after RFA for liver tumours are

improving. Systematic reviews of the literature showed that RFA for un-resectable liver

tumours could achieve good outcomes. Some centres around the world have started to use

RFA to treat resectable liver tumours in a carefully selected group of patients. Early data

from these studies showed that the results were favourable.

The critical factor in RFA is its ability to completely ablate a tumour. Current RFA

equipment is only capable of ablating a 3 cm tumour with a 1 cm ablative margin. Tumours

larger than 3 cm, or multi-focal tumours, are risk factors for incomplete ablation leading to

higher local disease recurrence rates and reduced survivals.

BETA is a new local ablative therapy that has been shown in previous experiments to

produce significantly larger ablations compared to standard RFA. The research projects

described here added further knowledge in this field

The first experiment demonstrated that the ability of BETA to produce larger ablations was

due to the increased tissue hydration from the electrolytic process. The polarity of BETA was

reversed, and the anode was attached to the RF electrode instead of the cathode. This new

arrangement, called reversed polarity bimodal electric ablation (RP-BEA), was shown to

produce shorter duration of ablation and smaller ablation size compared to standard RFA and

BETA. The anode desiccated the tissues adjacent to the electrode, therefore leading to earlier

roll-off and smaller ablations.

The second experiment showed that the efficacy of BETA was significantly better when the

anode of the DC circuit was placed below the skin layer. In the experiment the size of BETA

ablations was compared to standard RFA with the anode placed at different locations (skin,

peritoneum, and liver). The results showed that ablation size was largest when the anode was

placed on the peritoneum and the liver. The liver could be the ideal location to place the

130

anode as it has excellent electrical conductivity, therefore ensuring maximum tissue hydration

around the cathode to produce the largest ablations possible. In addition the liver has a huge

functional reserve and excellent regenerative capability to tolerate the local tissue injury

associated with the electrolytic reactions at the anode.

The third experiment showed that the principle of BETA could be incorporated into the Cool-

Tip RF System (Covidien), which is another popular RF system on the market. BETA could

produce significantly larger ablations compared to standard RFA using the internally-cooled

electrodes. Therefore BETA can be readily translated into the clinical setting using existing

equipment as both RFA and ECT are widely used around the world. In addition it was also

shown that the benefit of the DC extended into the RFA phase, and therefore it should be

continued for the whole treatment duration. Providing the DC only during the pre-treatment

phase (ECT/RFA - 9V DC for 10 minutes in this study) also produced significantly larger

ablations compared to standard RFA, although the benefit is less compared to BETA.

In summary BETA is a new innovation in the field of local ablative therapy which has shown

promising results. Research in animal liver models demonstrated that BETA could be readily

incorporated into existing RF generators on the market to produce significantly larger

ablations compared to standard RFA. This can improve the efficacy of RFA in treating larger

liver tumours, minimizing local disease recurrence rates and increasing survival. Data from

previous and the current study suggested that it might be time to extend research in this area

into human clinical trials.

131

Appendix 1: Survival Rates – RCTs comparing RFA vs. PEI for Un-Resectable HCC

Study Treatment

Patients (tumours

)

Median size

(mean) in mm

Median

follow-up

(mean) in

months

Tumours

Median survival rate at 1 year (%)

Median survival rate at 3

years (%)


years (%)

Median survival (months

)

Disease free

survival (months)

Lencioni

(2003)[107

]

RFA 52 (71) (28) (22.9)

HCC

100 2yr=98

% n/a n/a

1yr^=86%

,

2yr^=64%

PEI 50 (73) (28) (22.4 ) 96 2yr=88

% n/a n/a

1yr=77%,

2yr=43%

Lin

(2004)[108

]

RFA 52 (29) (24.5)

HCC

≤4cm

90^ 74^ n/a n/a

1yr=78%^

,

2yr=59%^

,

3yr=37%^

PEI 52 (28) (23.8) 85 50 n/a n/a 1yr=61%, 2yr=42%, 3yr=17%

High dose

PEI 53 (28) (24.1) 88 55 n/a n/a

1yr=63%, 2yr=45%, 3yr=20%

Lin

(2005)[109

]

RFA 62 (25) (28)

HCC ≤30

mm

93^ 74^ n/a n/a

1yr=74%^

,

2yr=60%^

,

3yr=43%^

PEI 62 (23) (27) 88 81 n/a n/a

1yr=70%,

2yr-41%,

3yr=21%

PAI 63 (23) (27) 90 53 n/a n/a

1yr=71%,

2yr=43%,

3yr=23%

Shiina

(2005)[110

]

RFA 118

≤2cm

(38%)

>2cm

(62%)

37.2

HCC

≤3cm

n/a n/a 4yr=74%

^ n/a n/a

PEI 114

≤2cm

(50%),

>2cm

(50%)

34.8 n/a n/a 4yr=57% n/a n/a

Non-naïve

RFA 345 (26) 27.6 92 62 38 n/a n/a

Brunello

(2008)[113

]

RFA 70 (24.2) 26.1 Child-

Pugh

A/B, ≤3

HCC ≤30

mm

n/a 59 n/a n/a n/a

PEI 69 (22.5) 25.3 n/a 57 n/a n/a n/a

132

The tables were presented according to the year of article publication, and the name of the first author in

alphabetical order. The percentage numbers of survival and disease recurrence rates were rounded to the nearest

figure.

^statistically significant differences compared to other groups

133

Appendix 2: Recurrence Rates – RCTs comparing RFA vs. PEI for Un-Resectable HCC

Study Treatment Patients (tumours)

Median size

(mean) in mm

Median follow-up (mean) in months

Tumours Recurrence site

Recurrence rate (%)

Lencioni (2003)

[107]

RFA 52 (71) (28) (22.9)

HCC

Ablation site 5^

Intra-hepatic 24

Extra-hepatic 0

PEI 50 (73) (28) (22.4 )

Ablation site 26

Intra-hepatic 22

Extra-hepatic 0

Lin (2004)[108]

RFA 52 (29) (24.5)

HCC ≤4cm

Ablation site 14^

Intra-hepatic 31

Extra-hepatic 0

PEI 52 (28) (23.8)

Ablation site 35

Intra-hepatic 37

Extra-hepatic 0

High dose PEI 53 (28) (24.1)

Ablation site 24

Intra-hepatic 32

Extra-hepatic 0

Lin (2005)

[109]

RFA 62 (25) (28)

HCC ≤30 mm

Ablation site 13^

Intra-hepatic 30

PEI 62 (23) (27) Ablation site 35

Intra-hepatic 35

PAI 63 (23) (27) Ablation site 29

Intra-hepatic 36

Shiina (2005)

[110]

RFA 118

≤2cm

(38%),

>2cm

(62%)

37.2

HCC ≤3cm

Ablation site 2^

Intra-hepatic 63

Extra-hepatic 2

PEI 114

≤2cm

(50%),

>2cm

(50%)

34.8

Ablation site 11

Intra-hepatic 64

Extra-hepatic 4

Brunello

(2008)[113]

RFA 70 (24.2) 26.1 Child-Pugh

A/B, ≤3 HCC

≤30 mm

Intra-hepatic 46

PEI 69 (22.5) 25.3 Intra-hepatic 51

134

The tables were presented according to the year of article publication, and the name of the first author in

alphabetical order. The percentage numbers of survival and disease recurrence rates were rounded to the nearest

figure.


135

Appendix 3: Survival Rates – RCTs comparing RFA vs. RFA + TACE for Un-

Resectable HCC


Median size

(mean) in mm

Median follow-

up (mean)

in months

Tumours

Median survival rate at 1 year (%)

Median survival rate at 3 years

(%)


(%)

Median survival (months)

Disease free

survival (months)

Cheng

(2008)[114]

TACE 95 (49.2) (25.4)

≤3 HCC ≤7.5cm

74 32 13 24 n/a

RFA 100 (49.8) (24.6) 67 32 8 22 n/a

RFA +

TACE 96 (49.6) (35.8) 83^ 55^ 31^ 37^ n/a

Yang

(2008)[138]

RFA 12 52 n/a

HCC

58 n/a n/a 19 n/a

TACE 11 64 53 n/a n/a 15 n/a

RFA +

TACE 24 66 68 n/a n/a 22 n/a

RFA +

TACE +

Lentinan

31 65 81 n/a n/a 28^ n/a

Morimoto

(2010)[116]

RFA 18 (37) (32) Single

HCC 3.1-

5cm

89 80 n/a n/a n/a

RFA +

TACE 19 (36) (30) 100 93 n/a n/a n/a

The tables were presented according to the year of article publication, and the name of the first author in alphabetical order. The percentage

numbers of survival and disease recurrence rates were rounded to the nearest figure.


136

Appendix 4: Recurrence Rates – RCTs comparing RFA vs. RFA + TACE for Un-

Resectable HCC


Median size

(mean) in mm



Recurrence rate (%)

Cheng (2008)

[114]

TACE 95 (49.2) (25.4)

≤3 HCC ≤7.5cm

Ablation site 15

Intra-hepatic 53

Extra-hepatic 13

RFA 100 (49.8) (24.6)

Ablation site 16

Intra-hepatic 54

Extra-hepatic 11

TACE + RFA 96 (49.6) (35.8)

Ablation site 4^

Intra-hepatic 48

Extra-hepatic 7

Yang

(2008)[138]

RFA 12 52

n/a HCC

Ablation site +

Intra-hepatic 35^

TACE 11 64 Ablation site +

Intra-hepatic 46^

RFA + TACE 24 66 Ablation site +

Intra-hepatic 29

RFA + TACE

+ Lentinan 31 65

Ablation site +

Intra-hepatic 18

Morimoto

(2010)[116]

RFA 18 (37) (32) Single HCC

3.1-5cm

Ablation site 39^

RFA + TACE 19 (36) (30) Ablation site 6




137

Appendix 5: Survival Rates after RFA for Un-Resectable HCC

Study Treatment Patients (tumour

s)

Median size

(mean) in mm

Median

follow-up

(mean) in

months

Tumours

Median

survival rate

at 1 year (%)


(%)


years (%)

Median survival (month

s)

Disease free

survival (months)

Vivarelli

(2004) [134]

RFA 79

>30

mm

(72%)

(15.6) HCC 78^ 33^ n/a n/a

1yr=79%

^,

3yr=50%

^

Resection 79

>30

mm

(73%)

(28.9) Resectable

HCC 83 65 n/a n/a

1yr=60%,

3yr=20%

Cho (2005)

[121]

RFA 99 (31) (23) Child-Pugh

A, ≤3 HCC

≤50 mm

96 80 n/a n/a n/a

Surgery 61 (34) (21.9) 98 77 n/a n/a n/a

Hong

(2005) [123]

RFA 55 (24) 22.7 Single

HCC ≤4cm

100 73 n/a n/a n/a

Resection 93 (25) 25.5 98 84 n/a n/a n/a

Lu

(2005)[141]

RFA 53 (72) (26) (24.8) ≤5 HCC

≤8cm

72 38 4yr=24% 27 17

MCT 49 (98) (25) (25.1) 82 51 4yr=37% 33 16

Maluccio

(2005)[132]

RFA +

TACE 33 40 22

Single HCC <7cm

97 77 56 n/a 25.1

Resection 40 46 23 81 70 58 n/a 53.1

Ogihara

2005 [133]

RFA 40 46 16 Single

HCC

78 58 39 51 n/a

Resection 47 74 16 75 65 31 47 n/a

Xu 2005

[142]

RFA 84 (26) 19.7

(24.1)

≤5 HCC

≤8cm n/a n/a n/a

23 6

MCT 53 35 9

Chok

(2006)[139]

RFA 51 30 19 <4 HCC ≤5

cm

82 2yr=72

% n/a n/a 10

TACE 40 33 23 80 2yr=58

% n/a n/a 10

Shibata*

2006 [111]

RFA –

cooled-tip 38 (41) (17.5) (21)

≤3 HCC

≤30mm

100 94 n/a n/a

1yr=47%,

2yr=34%,

3yr=34%

RFA – 36 (42) (19.7) (28) 94 77 n/a n/a 1yr=44%,

2yr=22%,

138

expandable 3yr=22%

Ferrari*

(2007)[112]

RFA 40 (50) (26.7) n/a 1 HCC

≤4cm, or

≤3 HCC

≤3cm

92 61

55

months=4

1

n/a 17.8

Laser 41 (45) (28.9) n/a 89 57 23 n/a 15.5

Helmberger

(2007) [130]

Resection 52

≤5cm

(71%),

>5 cm

(29%)

n/a

Resectable

HCC,

Child-Pugh

score 5-6

n/a n/a n/a 37^ n/a

RFA +

TACE 44

≤5cm

(100%) n/a

Unresectab

le HCC

n/a n/a n/a 45^ n/a

TACE 107

≤5cm

(68%),

>5 cm

(32%)

n/a n/a n/a n/a 13 n/a

Tamoxifen 21

≤5cm

(52%),

>5 cm

(48%)

n/a n/a n/a n/a 6 n/a

Khan (2007)

[101]

Percutaneo

us RFA 92 (19) (19)

HCC ≤30

mm

91 71 n/a n/a 1yr=52%,

3yr=33%

Surgical

RFA 63 (22) (19) 89 57 n/a n/a

1yr=52%,

3yr=22%

Percutaneo

us RFA 25 (36) (18)

HCC >30 mm

81^ 42^ n/a n/a 1yr=29%,

3yr=0%

Surgical

RFA 48 (39) (18) 92 68 n/a n/a

1yr=54%,

3yr=19%

Kudo

(2007)[94]

RFA 84 15 58.8 Child-Pugh

A, ≤3 HCC

≤3cm

n/a n/a 66^ n/a n/a

RFA +

Interferon 43 18 61.2 n/a n/a 83 n/a n/a

Lin 2007

[249]

RF 2000 25 (34) (25) 21

≤3 HCC

≤40 mm

87 2yr=73

% n/a n/a 1yr=77%, 2yr=55%

RF 3000 25 (35) (26) 22 88 2yr=75

% n/a n/a 1yr=80%, 2yr=56%

RITA 25 (31) (26) 22 89 2yr=76

% n/a n/a 1yr=79%, 2yr=55%

Cool-Tip 25 (33) (27) 22 90 2yr=78

%

n/a n/a 1yr=79%, 2yr=54%

Lupo (2007)

[125]

RFA 60 (36.5) (27) Single

HCC 3-

5cm

96 53 32 n/a n/a

Resection 42 (40) (31.3) 91 57 43 n/a n/a

Takahashi

(2007)[127]

RFA 171 21

36.7

Child-Pugh

A, 1 HCC

<5cm, or

≤3 HCC

<3cm

n/a n/a 77 n/a 23^

Resection 53 25 n/a n/a 70 n/a 25

139

Zhang*

(2007) [103]

RFA 67

≤30

mm

(46.3%

), 31-

50 mm

(32.8%

), 51-

70 mm

(28.9%

)

(32.2)

1 HCC

≤7cm, or

≤3 HCC

≤3cm

90 58 36 n/a n/a

RFA + PEI 66

≤30

mm

(44%),

31-50

mm

(37.9%,

51-70

(18.1%

)

(35.5) 95^ 76^ 49^ n/a n/a

Guglielmi

(2008) [100]

RFA 109

≤30

mm

(30%),

31-60

mm

(70%)

(23)

≤3 HCC

≤6cm

83 42 20 28^ 16

Resection 91

≤30

mm

(34%),

31-60

mm

(66%)

(32) 84 64 48 57 36

Hiraoka

(2008) [122]

RFA 105 n/a n/a Child-Pugh

A/B, 1

HCC <3cm

n/a 88 59 n/a 3yr=59%, 5yr=25%

Resection 59 n/a n/a n/a 91 59 n/a 3yr=64%, 5yr=22%

Lam (2008)

[131]

RFA 273

(357) 30 24 HCC 89 60 38 n/a n/a

Resection 240 n/a 35 ≤4 HCC

≤8cm 90 77 68 n/a n/a

Seror

(2008)[148]

RFA –

internally

cooled

45 (54) (21.4) (29.8)

≤3 HCC ≤3cm

90 2yr=87

% n/a n/a n/a

RFA _

saline

perfused

44 (54) (21.1) (17.7) 87 83% n/a n/a n/a

Yamagiwa

(2008)[135]

Resection 101 n/a

33.6 Resectable

HCC n/a n/a 59^ n/a

5yr=32^

%

RFA + 115 24.3 Child-Pugh n/a n/a 72^ n/a 5yr=14^

140

TACE A/B, ≤5

HCC ≤5

cm

%

PEI +

TACE 43 42.8 n/a n/a 41^ n/a 5yr=4%

TACE 86 20.3 n/a n/a 15^ n/a 5yr=5%

Yamakado

(2008) [136]

RFA +

TACE 104 (25) (37)

HCC

98 94 75 n/a

1yr=92%,

3yr=64%,

5yr=27%

Surgery 62 (27) (38) 97 93 81 n/a

1yr=89%,

3yr=69%,

5yr=26%

Kobayashi

(2009)[124]

RFA 209 18^

39.6

Child-Pugh

A cirrhosis

≤3 HCC

≤3cm

99 87 5yr=75%,

7yr=65% n/a

1yr=83%,

3yr=42%

^,

5yr=17%

^,

7yr=6%^

Resection 199 20 97 90 5yr=79%,

7yr=62% n/a

1yr=83%,

3yr=51%,

5yr=37%,

7yr=23%

Ohmoto

(2009)[143]

RFA 34 (37) (16) (26.2) HCC ≤2cm

100^ 70^ 4yr=70%^ n/a n/a

MCT 49 (56) (17) (40) 89 49 4yr=39% n/a n/a

Santambrog

io

(2009)[126]

Lap RFA 74 (26.6) (38.2) Single Child-Pugh

A HCC <5cm

88 66 41 n/a n/a

Resection 78 (29.1) (36.2) 93 85 54 n/a n/a

Ueno

(2009)[128]

RFA 155 20 (36.8) 1 HCC <5cm, or ≤3 HCC

<3cm

98 92 63 n/a

1yr=78%^, 3yr=36%^, 5yr=20%^,

Resection 123 27 (35) 99 92 80 n/a 1yr=80%, 3yr=47%, 5yr=38%,

Yang

(2009)[138]

RFA 37 (38)

22 Recurrent HCC after resection

74 51 28 37 n/a

TACE 35 (36) 66 39 20 31 n/a

RFA +

TACE 31 (35) 89 65 44 52 n/a

Hiraoka

(2010) [278]

RFA 63 20.9 23

HCC

within

Milan

criteria, age

≥75

n/a 83 50 n/a n/a

RFA 143 20.7 30.5

HCC

within

Milan

criteria, age

<75

n/a 78 58 n/a n/a

Peng

(2010)[279] RFA 120

≤5cm

(73%), (34.8)

1 HCC

≤7cm, or 89^ 64^ 42^ n/a

1yr=76%^,

3yr=47%^,

141

>5cm

(27%)

≤3 HCC

≤3cm

5yr=30%^

RFA +

TACE 120

≤5cm

(71%),

>5cm

(29%)

(36.5) 93 75 50 n/a 1yr=90%, 3yr=63%, 5yr=42%




*Randomized Controlled Trial

142

Appendix 6: Recurrences Rates after RFA for Un-Resectable HCC


Median size (mean) in mm

Median follow-

up (mean)

in months


Recurrence rate (%)

Vivarelli (2004)

[134] RFA 79 >30mm (72%) 15.6 HCC

Ablation site 15

Intra-hepatic 33

Cho (2005) [121]

RFA 99 (31) (23) Child-Pugh A,

≤3 HCC ≤50

mm

Ablation site 18

Intra-hepatic 28

Surgery 61 (34) (21.9) Ablation site 10

Intra-hepatic 33

Hong (2005)

[123]

RFA 55 (24) 22.7 Single HCC

≤40 mm

Ablation site 7^

Intra-hepatic 51

Resection 93 (25) 25.5

Ablation site 0

Intra-hepatic 45

Lu (2005)[141]

RFA 53 (72) (26) (24.8)

≤5 HCC ≤8cm

Ablation site 21

Intra-hepatic 69

MCT 49 (98) (25) (25.1) Ablation site 12

Intra-hepatic 76

Maluccio

(2005)[132]

RFA + TACE 33 40 22

1 HCC <7cm

Ablation site

+ Intra-

hepatic

42

Extra-hepatic 21

Resection 40 46 23

Resection

site + Intra-

hepatic

35

Extra-hepatic 13

Ogihara 2005

[133]

RFA 40 46 16

Single HCC

Ablation site 10

Intra-hepatic 25

Resection 47 74 16 Ablation site 2

Intra-hepatic 28

Ferrari*

(2007)[112]

RFA 40 (50) (26.7) n/a 1 HCC ≤4cm,

or ≤3 HCC

≤3cm

Ablation Site 15

Intra-hepatic 4

Laser 41 (45) (28.9) n/a Ablation Site 23

143

Intra-hepatic 9

Khan (2007)

[101]

Percutaneous

RFA 92 (19) (19)

HCC ≤30 mm

Ablation site 13

Intra-hepatic 26

Extra-hepatic 3

Surgical RFA 63 (22) (19)

Ablation site 10

Intra-hepatic 38

Extra-hepatic 11

Percutaneous

RFA 25 (36) (18)

HCC >30 mm

Ablation site 8

Intra-hepatic 52

Extra-hepatic 12

Surgical RFA 48 (39) (18)

Ablation site 13

Intra-hepatic 35

Extra-hepatic 6

Kudo (2007)[94]

RFA 84 15 58.8 Child-Pugh A,

≤3 HCC ≤3cm

Ablation site 4

Intra-hepatic 71^

RFA +

Interferon 43 18 61.2

Ablation site 3

Intra-hepatic 56

Lin 2007 [249]

RF 2000 25 (34) (25) 21

≤3 HCC ≤40

mm

Ablation Site 12

Intra-hepatic 24

RF 3000 25 (35) (26) 22 Ablation Site 8

Intra-hepatic 32

RITA 25 (31) (26) 22 Ablation Site 8

Intra-hepatic 32

Cool-Tip 25 (33) (27) 22 Ablation Site 8

Intra-hepatic 28

Murakami

(2007)[140]

RFA 105 (109) (16)

22.4

1 HCC ≤5cm,

or ≤3HCC

≤3cm

Ablation Site 1yr=24%^,

2yr=40%^

TACE 133 (173) (17) Ablation Site 1yr=37%,

2yr=51%

Takahashi

(2007)[127]

RFA 171 21

36.7

Child-Pugh A,

1 HCC <5cm,

or ≤3 HCC

<3cm

Ablation site 17^

Resection 53 25 Resection

site 0

Zhang* (2007)

[103]

RFA 67

≤30 mm (46%),

31-50 mm

(33%), 51-70

mm (21%)

(32.2)

1 HCC ≤7cm,

or ≤3 HCC

≤3cm

Ablation site 21^

Intra-hepatic 39

Extra-hepatic 9

RFA + PEI 66

≤30 mm (44%),

31-50 mm

(38%), 51-70

mm (18%)

(35.5)

Ablation site 6

Intra-hepatic 33

Extra-hepatic 12

Lam (2008)

[131] RFA 273 (357) 30 24 HCC

Ablation site 13

Intra-hepatic 59

Extra-hepatic 12

144

Seror

(2008)[148]

RFA –

internally

cooled

45 (54) (21.4) (29.8)

≤3 HCC ≤3cm

Ablation site 1yr=9%, 2yr=11%

Intra-hepatic 1yr=19%,

2yr=31%^

RFA _ saline

perfused 44 (54) (21.1) (17.7)

Ablation site 1yr=11%,

2yr=15%

Intra-hepatic 1yr=37%,

2yr=64%

Yamakado

(2008) [136]

RFA + TACE 104 (25) (37)

HCC

Ablation site 3

Intra-hepatic 33

Surgery 62 (27) (38)

Ablation site 0

Intra-hepatic 37

Intra-hepatic 18

Kobayashi

(2009)[124]

RFA 209 18^

39.6

Child-Pugh A

cirrhosis ≤3

HCC ≤3cm

Ablation site 9^

Resection 199 20 Ablation site 1

Ohmoto

(2009)[143]

RFA 34 (37) (16) (26.2)

HCC ≤2cm

Ablation site

1yr=9%^,

2yr=9%^,

3yr=9%^,

4yr=9^%

Intra-hepatic

1yr=28%,

2yr=52%,

3yr=65%,

4yr=65%

MCT 49 (56) (17) (40)

Ablation site

1yr=13%,

2yr=16%,

3yr=19%,

4yr=19%

Intra-hepatic

1yr=35%,

2yr=62%,

3yr=72%,

4yr=78%

Santambrogio

(2009)[126]

Lap RFA 74 (26.6) (38.2) Single Child-Pugh A HCC

<5cm

Ablation site 24^

Intra-hepatic 68^

Resection 78 (29.1) (36.2) Ablation site 6

Intra-hepatic 51

Ueno

(2009)[128]

RFA 155 20 (36.8)

1 HCC <5cm, or ≤3 HCC

<3cm

Ablation site

+ Intra-

hepatic 61

Resection 123 27 (35)

Ablation site

+ Intra-

hepatic 42

Yang

(2009)[138]

RFA 37 (38)

22 Recurrent HCC after resection

Intra-hepatic 43 TACE 35 (36) Intra-hepatic 57

RFA + TACE 31 (35) Intra-hepatic 51^

Peng(2010)[279]

RFA 120 ≤5cm (73%),

>5cm (27%) (34.8) 1 HCC ≤7cm,

or ≤3 HCC

≤3cm

Ablation site 4

Intra-hepatic 46^

Extra-hepatic 3

RFA + TACE 120 ≤5cm (71%),

>5cm (29%) (36.5)

Ablation site 3

Intra-hepatic 28

145

Extra-hepatic 8




*Randomized Controlled Trial

146

Appendix 7: Survival Rates after RFA for Resectable HCC

Study Treatment

Patients (tumours

)

Median size

(mean) in mm

Median

follow-up

(mean) in

months

Tumours


year (%)


years (%)


years (%)

Median survival (months

)

Disease Free

Survival (months)

Montorsi

(2005)

[146]

RFA 58 n/a (25.7) Single

HCC

<50 mm

85 61 4yr=45% n/a n/a

Resection 40 n/a (22.4) 84 73 4yr=61% n/a n/a

Abu-Hilal

(2008)

[144]

RFA 34 30 30

HCC

≤5cm

meeting

Milan

criteria

83 2yr=62

% 57

Not

achieved

at time

of report

10^

Resection 34 38 43 91 2yr=81

% 56 74 35

Chen*

(2006)

[117]

RFA

90 (90) –

19 had

resection

≤3 cm

(37 pt),

3.1-5

cm (34

pt)

(27.9)

Child-

Pugh A

single

cirrhosis

<5cm

94 69 4yr=66% n/a n/a

Resection

90 (90) +

ethanol

injection

in 2

patients

≤3 cm

(42 pt),

3.1-5

cmm

(48 pt)

(29.2) 93 73 4yr=64% n/a n/a

Liang

(2008)

[145]

RFA 66 (88)

≤3 cm

(44 pt),

>3 cm

(22 pt)

21 ≤3

recurrent

HCC <

5cm

77 49 40 n/a n/a

Resection 44 (55)

≤3 cm

(26 pt),

>3 cm

(18 pt

33 79 45 28 n/a n/a

Peng

(2010)[137

]

RFA 224 25 (44.1)

Single

Child-

Pugh A

HCC

≤5cm

5yr=60

%

7yr=55

%

10yr=34

% 76.1

48; 5yr=36%, 7yr=29%, 10yr=18

%

147



^Statistically significant differences compared to other groups

148

Appendix 8: Recurrences Rates after RFA for Resectable HCC


Median size

(mean) in mm



Recurrence rate (%)

Montorsi (2005)

[146]

Lap RFA 58 n/a (25.7) Single HCC

<5 cm

Ablation site 19^

Intra-hepatic 35

Resection 40 n/a (22.4) Resection site 0

Intra-hepatic 30

Abu-Hilal

(2008) [144]

RFA 34 30 30 HCC ≤5cm

meeting Milan

criteria

Ablation site 30^

Intra-hepatic 30

Resection 34 38 43 Resection site 4

Intra-hepatic 57

Liang (2008)

[145]

RFA 66 (88)

≤3 cm (44

pt), >3 cm

(22 pt)

21

≤3 recurrent

HCC < 5cm

Ablation site 8

Intra-hepatic 65

Extra-hepatic 5

Resection 44 (55)

≤3 cm (26

pt), >3 cm

(18 pt)

33

Resection site 7

Intra-hepatic 75

Extra-hepatic 5

Peng

(2010)[137] RFA 224 25 (44.1)

Single Child-

Pugh A HCC

≤5cm

Ablation site 13

Intra-hepatic 49

Extra-hepatic 1



^Statistically significant differences compared to other groups *Randomized Controlled Trial

149

Appendix 9: Survival after RFA for Un-Resectable Liver Metastases

Study Treatment

Number

of

Patients

(tumour

s)

Media

n size

(mm)

Median

follow-

up

(month

s)

Types of

liver

metastas

es

Media

n

surviv

al rate

at 1

year

(%)

Media

n

surviv

al rate

at 3

years

(%)

Media

n

surviv

al rate

at 5

years

(%)

Overall

median

surviva

l

(month

s)

Median

disease

free

survival

(months

)

Solbiati

(2001)[153]

Percutaneo

us RFA 117 (n/a) 26 n/a CLM 93 46 n/a 36 12

Ianitti (2002)[102] Percutaneo

us RFA 52

52

(mean) 20 CLM 87 50 n/a n/a n/a

Pawlik (2003)[158]

RFA +

resection

172

(737) 18 21.3

Mets

(124

CLM)

n/a n/a n/a 45.5 n/a

Abdalla

(2004)[159]

RFA 57 (110) 25 21 CLM n/a 37 4yr:

22% n/a n/a

RFA +

resection 101 25 21 CLM n/a 43

4yr:

36% n/a n/a

Resection 190 25 21 CLM n/a 73^ 58^ n/a n/a

Chemo 70 25 21 CLM n/a 14 4yr:

8% n/a n/a

Gillams (2004)

[179]

Percutaneo

us RFA 167 (n/a)

39

(mean)

17

(mean) CLM 91 40 17 32 n/a

Berber (2005)[195] Lap RFA 53 (192) 31

(mean)

24

(mean)

“Unusual

” Mets n/a n/a n/a 33 n/a

Berber (2005)[196] Lap RFA 135 41

(mean) n/a CLM n/a n/a n/a 28.9 6

Chen (2005) [155] Percutaneo

us RFA

134

(333)

41

(mean) n/a Mets 75.3 25.1 n/a n/a n/a

Elias (2005)[207] RFA +

Resection 63 (351)

13

(15) 27.6 CLM 92 47 n/a 36

1yr=92

%,

2yr=55

%,

3yr=27

%

150

Navarra (2005)

[180]

RFA (+

resection in

12 pt)

57 (297) n/a 18.1 Mets (38

CLM) 72.5 52.5 n/a n/a n/a

Ahmad

(2006)[181]

First gen

probe 21

37.4

(mean) 26.2 CLM n/a n/a n/a n/a 16^

Newer gen

probe 31

34.7

(mean) 26.2 CLM n/a n/a n/a n/a 8

Amersi (2006)

[182]

RFA (+

resection in

majority of

pt)

74 35.6

(mean)

33.2

(mean) CLM n/a n/a n/a 29.7 n/a

Chen (2006)[199] RFA 127

(195) (39) 14 Mets 80 31 n/a n/a n/a

Hildebrand

(2006)[183] RFA 81 35 21.2 Mets 89 38 n/a 27 n/a

Jakobs (2006)[184] Percutaneo

us RFA 68 (183)

22.8

(mean)

21.4

(mean) CLM 96 68 n/a n/a n/a

Machi (2006) [205] RFA +

chemo

100

(507) 30 24.5 CLM 90 42 30.5 28 n/a

van Duijhoven

(2006) [185]

RFA (+

resection in

29 pt)

87 (199) 29 25 CLM n/a n/a n/a 27.8 15

Mazzaglia(2007)[1

98] Lap RFA 63 (384)

23

(mean)

32.4

(mean)

Neuro-

endocrine

liver mets

91 2yr=77 48 46.8 9

Siperstein (2007)

[206]

Lap RFA

(after failed

chemo)

234 39

(mean) 24 CLM n/a 20.2 18.4 24 n/a

Sorensen (2007)

[186]

RFA (+

resection in

25 and

chemo in 6

pt)

102

(332)

22

(mean) 23.6 CLM 96 64 44 52 n/a

Berber (2008)[197] Lap RFA 68 37 23 (27) Solitary

CLM

n/a n/a 30 24-34^ 9^

Resection 90 38 33 (41) n/a n/a 40 57 30

Blusse (2008)[187] RFA 87 n/a n/a CLM n/a n/a n/a 25 13

151

Gleisner

(2008)[160]

RFA 11 25 n/a CLM n/a 72.7 n/a n/a 3yr=

7.4%

RFA +

Resection 55 25 n/a CLM n/a 44.9^ n/a n/a

3yr=

34%

Resection 192 35 n/a CLM n/a 74.1 n/a n/a 3yr=

40%^

Leblanc

(2008)[209]

RFA 34 10 36

CLM

(n=18);

non-CLM

(n=16)

2-yr=75 n/a n/a n/a 24

RFA +

Resection 28 10 25

CLM

(n=16);

non-CLM

(n=12)

2-yr=68

n/a n/a n/a 12

Resection 37 21^ 29

CLM

(n=26);

non-CLM

(n=11)

2-yr=83

n/a n/a n/a 18

Veltri (2008)[189] RFA 122

(199)

25

(29)

18.8

(24.2) CLM 79 38 22 31.5 n/a

Gillams (2009)

[156]

RFA 192 n/a n/a

Number

of CLM

≤5, and

diameter

≤50 mm

n/a 58^ 26^ 39^ n/a

RFA 117 n/a n/a

Number

of CLM

>5, and

diameter

>50 mm

n/a 29 5 25 n/a

Hur (2009)[200]

RFA 25 25

(25) 42

Single

CLM

n/a 60 26 n/a n/a

Resection 42 26

(28) n/a 70 50 n/a n/a

Meloni

(2009)[190] RFA 52 (87)

25

(mean) 19.1

Breast Ca

Liver

Mets 68 43 27 29.9 n/a

Reuter (2009)[191]

RFA (+

chemo in 7

pt)

66 32

(mean) 20 CLM n/a n/a 21 27 12.2^

152

CLM-colorectal liver metastases; Mets-metastases; DFS-disease free survival

^Statistically significant differences between group(s)

Resection

(+ chemo

in 18 pt)

126 53

(mean) 20 CLM n/a n/a 23 36.4 31.1

Vyslouzil

(2009)[210]

RFA +

Resection

+ chemo

23 n/a n/a CLM 83 30 n/a n/a n/a

RFA +

chemo 31 n/a n/a CLM 87 26 n/a n/a n/a

Resection 136 n/a n/a CLM 91 58 n/a n/a n/a

153

Appendix 10: Tumour Recurrences after RFA for Un-Resectable Liver Metastases

Included articles Treatment

Number

of

patients

(tumours)

Median

size

(mm)

Median

follow-

up

(months)

Types of

liver

metastases

Recurrence

(ablation

site, intra-

hepatic or

extra-

hepatic)

Recurrence rate

(%)

Solbiati (2001) [153] RFA 117 (179) 26 n/a CLM Ablation site 39

Bleicher (2003) [194] RFA 59 25 11

(mean) CLM Ablation site 18.3

Pawlik (2003) [158]

RFA + resection 172 (737) 18 21.3

Mets (124

CLM)

Ablation site 2.3

Intra-hepatic 22

Extra-hepatic 30.2

Scaife (2003)[204] RFA + HAI 50 20 20 CLM

Ablation site 10

Intra-hepatic 30

Extra-hepatic 48

Abdalla

(2004) [159]

RFA 57 (110) 25 21 CLM

Ablation site 9^

Intra-hepatic 44

Extra-hepatic 40

RFA + resection 101 25 21 CLM

Ablation site 5

Intra-hepatic 28

Extra-hepatic 37

154

Resection only 190 25 21 CLM

Ablation site 2

Intra-hepatic 11^

Extra-hepatic 41

Elias (2004) [203]

RFA 88 (227) 12 (15) 27.6 Mets Ablation site 14.8

Wedge resection 64 (99) 10 (14) 27.6 Mets Resection site 10.9

Anatomical

hepatectomy 40 (40) 42 (44.2) 27.6 Mets Resection site 12.5

Gillams (2004) [179] RFA 167 (n/a) 39

(mean) 17 CLM

Ablation site 15.9

Intra-hepatic 33

Extra-hepatic 49

Berber (2005)[195] Lap RFA 53 (192) 31

(mean)

24

(mean)

“Unusual”

Mets Ablation site 17

Berber (2005)[196] Lap RFA 135 412

(mean) n/a CLM

Ablation site 46

Intra-hepatic 53

Extra-hepatic 41

Chen (2005) [155] RFA 134 (333) 41

(mean) n/a Liver Mets Ablation site 10.5

Chiou (2005)[192] RFA 69 (109) 29

(mean)

22.4

(mean) CLM Ablation site 13.8

Elias (2005)[207] RFA +

Resection 63 (351) 13 (15) 27.6 CLM

Ablation site 17.4

Intra-hepatic 42.2

Extra-hepatic 46.2

Navarra (2005) [180] RFA +/-

resection 57 (297) n/a 18.1

Mets (38

CLM)

Intra-hepatic 9

Extra-hepatic 5

Ahmad (2006)[181]

First gen probe 21 37.4

(mean) 26.2 CLM

Ablation site 38^

Intra-hepatic 62

Newer gen

probe 31

34.7

(mean) 26.2 CLM

Ablation site 9.7

Intra-hepatic 52

Chen (2006)[199] RFA 127 (195) (39) 14 Mets Ablation site 15

155

Intra-hepatic 54

Jakobs (2006)[184] RFA 68 (183) 22.8

(mean)

21.4

(mean) CLM Ablation site 18

Machi (2006) [205] RFA + chemo 100 (507) 30 24.5 CLM Ablation site 6.7

van Duijhoven (2006)

[185]

RFA +/-

resection

87 (199) 29 25 CLM Ablation site 47.2

Mazzaglia

(2007)[198] RFA 63 (384)

23

(mean)

32.4

(mean)

Neuro-

endocrine

liver mets

Ablation site 11

Siperstein (2007)

[206]

RFA (after

failed chemo) 234

39

(mean) 24 CLM

Ablation site 18; Median time to

recurrence=6 mth

Intra-hepatic Median time to

recurrence=9 mth

Extra-hepatic Median time to

recurrence=10 mth

Berber (2008)[197]

Lap RFA 68 37 23 (27)

Solitary CLM

Ablation site 16

Intra-hepatic 57

Extra-hepatic 49

Resection 90 38 33 (41)

Ablation site 2

Intra-hepatic 24

Extra-hepatic 30

Blusse (2008)[187] RFA 87 n/a n/a CLM Ablation site 46

Gleisner (2008)[160]

RFA +

Resection 55 25 n/a CLM

Intra-hepatic

recurrence at

1yr

10.3

Extra-hepatic

recurrence at

1yr

40.6

Resection 192 35 n/a CLM

Intra-hepatic

recurrence at

1yr

2^

Extra-hepatic

recurrence at 12.8^

156

1yr

RFA 11 25 n/a CLM

Intra-hepatic

recurrence at

1yr

41.3

Extra-hepatic

recurrence at

1yr

21.2

Leblanc (2008)[209]

RFA 34 10 36

CLM (n=18);

non-CLM

(n=16)

Ablation site 5.9

Intra-hepatic 41

RFA +

Resection 28 10 25

CLM (n=16);

non-CLM

(n=12)

Ablation site 3.6

Intra-hepatic 60.7

Resection 37 21^ 29

CLM (n=26);

non-CLM

(n=11)

Intra-hepatic 54

Hur (2009)[200]

RFA 25 25 (25)

42 Single CLM

Ablation site 28

Intra-hepatic 32

Extra-hepatic 12

Resection 42 26 (28)

Ablation site 10

Intra-hepatic 14

Extra-hepatic 24

Meloni (2009)[190] RFA 52 25

(mean) 19.1

Breast Ca

Liver Mets

Ablation site 25

Intra-hepatic 53

Extra-hepatic 54

Reuter (2009)[191]

RFA (+ chemo

in 7 pt) 66

32

(mean) 20 CLM

Ablation site 17^

Intra-hepatic 33^

Extra-hepatic 35

Resection (+

chemo in 18 pt) 126

53

(mean) 20 CLM

Ablation site 2

Intra-hepatic 14

157

Extra-hepatic 33


HAI-hepatic arterial infusion of chemotherapy

158

Appendix 11: Survival after RFA for Resectable Liver Metastases

Included

articles Treatment

Number

of

patients

(tumours)

Median

size

(mm)

Median

follow-

up

(months)

Types of

liver

metastases

Median

survival

rate at

1 year

(%)

Median

survival

rate at

3 years

(%)

Median

survival

rate at

5 years

(%)

Overall

median

survival

(months)

Disease

free

survival

(months)

Elias

(2002)

[213]

RFA 47 (107) 21

(mean) 14.4

Recurrent

hepatic

malignancies

(29 CLM)

88 n/a n/a n/a 9

Otto

(2010)

[214]

RFA 28 30 814 days CLM n/a 67 n/a Beyond

day 1352

203

days^

Resection 82 50 644 days CLM n/a 60 44 1694

days 416 days


159

Appendix 12: Tumour Recurrences after RFA for Resectable Liver Metastases

Included

articles Treatment

Patients

(tumours)

Median

size (mm)

Median

follow-up

(months)

Types of liver

metastases

Recurrence

site (ablation

zone, intra-

hepatic or

extra-

hepatic)

Recurrence

rate (%)

Elias (2002)

[213] RFA 47 (107) 21 (mean) 14.4

Recurrent hepatic

malignancies (29

CLM)

Ablation site 31.9

Intra-hepatic 21.3

Extra-hepatic 31.9

Livraghi

(2003) [215] RFA 88 (134) 21 (mean) 33 CLM

Ablation site 40

Intra-hepatic 10

Extra-hepatic 6.8

Otto (2010)

[214]

RFA 28 30 814 days CLM

Ablation site 32^

Intra-hepatic 50

Extra-hepatic 32

Resection 82 50 644 days CLM

Ablation site 4

Intra-hepatic 34

Extra-hepatic 37


160

References

1. Buscarini E, Savoia A, Brambilla G, Menozzi F, Reduzzi L, Strobel D, et al. Radiofrequency thermal ablation of liver tumors. Eur Radiol. 2005 May;15(5):884-94. 2. Brown DB. Concepts, considerations, and concerns on the cutting edge of radiofrequency ablation. J Vasc Interv Radiol. 2005 May;16(5):597-613. 3. Nilsson E, von Euler H, Berendson J, Thorne A, Wersall P, Naslund I, et al. Electrochemical treatment of tumours. Bioelectrochemistry. 2000 Feb;51(1):1-11. 4. Gravante G, Ong SL, Metcalfe MS, Bhardwaj N, Maddern GJ, Lloyd DM, et al. Experimental application of electrolysis in the treatment of liver and pancreatic tumours: Principles, preclinical and clinical observations and future perspectives. Surg Oncol. 2009 Dec 31. 5. de Baere T, Risse O, Kuoch V, Dromain C, Sengel C, Smayra T, et al. Adverse events during radiofrequency treatment of 582 hepatic tumors. AJR Am J Roentgenol. 2003 Sep;181(3):695-700. 6. Livraghi T, Solbiati L, Meloni MF, Gazelle GS, Halpern EF, Goldberg SN. Treatment of focal liver tumors with percutaneous radio-frequency ablation: complications encountered in a multicenter study. Radiology. 2003 Feb;226(2):441-51. 7. Wemyss-Holden SA, Berry DP, Robertson GS, Dennison AR, De La MHP, Maddern GJ. Electrolytic ablation as an adjunct to liver resection: Safety and efficacy in patients. ANZ J Surg. 2002 Aug;72(8):589-93. 8. Wong SL, Mangu PB, Choti MA, Crocenzi TS, Dodd GD, 3rd, Dorfman GS, et al. American Society of Clinical Oncology 2009 clinical evidence review on radiofrequency ablation of hepatic metastases from colorectal cancer. J Clin Oncol. 2010 Jan 20;28(3):493-508. 9. Fosh BG, Finch JG, Anthony AA, Lea MM, Wong SK, Black CL, et al. Use of electrolysis for the treatment of non-resectable hepatocellular carcinoma. ANZ J Surg. 2003 Dec;73(12):1068-70. 10. Lau WY, Lai EC. The current role of radiofrequency ablation in the management of hepatocellular carcinoma: a systematic review. Ann Surg. 2009 Jan;249(1):20-5. 11. Cockburn JF, Maddern GJ, Wemyss-Holden SA. Bimodal electric tissue ablation (BETA) - in-vivo evaluation of the effect of applying direct current before and during radiofrequency ablation of porcine liver. Clin Radiol. 2007 Mar;62(3):213-20. 12. Dobbins C, Brennan C, Wemyss-Holden S, Cockburn J, Maddern G. Bimodal electric tissue ablation-long term studies of morbidity and pathological change. J Surg Res. 2008 Aug;148(2):251-9. 13. Dobbins C, Brennan C, Wemyss-Holden SA, Cockburn J, Maddern GJ. Bimodal electric tissue ablation: positive electrode studies. ANZ J Surg. 2008 Jul;78(7):568-72. 14. Dobbins C, Wemyss-Holden SA, Cockburn J, Maddern GJ. Bimodal electric tissue ablation-modified radiofrequency ablation with a le veen electrode in a pig model. J Surg Res. 2008 Jan;144(1):111-6. 15. d’Arsonval M. Action physiologique des courantsalternatifs. CR Soc Biol. 1891;43:283-6. 16. Beer E. Removal of neoplasms of the urinary bladder: a new method employing high frequency (oudin) currents through a cauterizing cystoscope. JAMA. 1910;54:1768-9. 17. Clark W. Oscillatory desiccation in the treatment of accessible malignant growths and minor surgical conditions. J Adv Therap. 1911;29:169-83. 18. Clark W, Morgan J, Asnia E. Electrothermic methods in treatment of neoplasms and other lesions with clinical and histological observations. . Radiology. 1924;2:233-46. 19. Cushing H, Bovie W. Electrosurgery as an aid to the removal of intracranial tumors Surg Gynecol Obstet. 1928;47:751-84. 20. Organ LW. Electrophysiologic principles of radiofrequency lesion making. Appl Neurophysiol. 1976;39(2):69-76. 21. McGahan JP, Browning PD, Brock JM, Tesluk H. Hepatic ablation using radiofrequency electrocautery. Invest Radiol. 1990 Mar;25(3):267-70.

161

22. Rossi S, Fornari F, Pathies C, Buscarini L. Thermal lesions induced by 480 KHz localized current field in guinea pig and pig liver. Tumori. 1990 Feb 28;76(1):54-7. 23. Nahum Goldberg S, Dupuy DE. Image-guided radiofrequency tumor ablation: challenges and opportunities--part I. J Vasc Interv Radiol. 2001 Sep;12(9):1021-32. 24. Ni Y, Mulier S, Miao Y, Michel L, Marchal G. A review of the general aspects of radiofrequency ablation. Abdom Imaging. 2005 Jul-Aug;30(4):381-400. 25. Cosman ER, Nashold BS, Ovelman-Levitt J. Theoretical aspects of radiofrequency lesions in the dorsal root entry zone. Neurosurgery. 1984 Dec;15(6):945-50. 26. Pennes HH. Analysis of tissue and arterial blood temperatures in the resting human forearm. J Appl Physiol. 1948 Aug;1(2):93-122. 27. Goldberg SN, Gazelle GS, Compton CC, Mueller PR, Tanabe KK. Treatment of intrahepatic malignancy with radiofrequency ablation: radiologic-pathologic correlation. Cancer. 2000 Jun 1;88(11):2452-63. 28. Seegenschmiedt MH, Brady LW, Sauer R. Interstitial thermoradiotherapy: review on technical and clinical aspects. Am J Clin Oncol. 1990 Aug;13(4):352-63. 29. Larson TR, Bostwick DG, Corica A. Temperature-correlated histopathologic changes following microwave thermoablation of obstructive tissue in patients with benign prostatic hyperplasia. Urology. 1996 Apr;47(4):463-9. 30. Goldberg SN, Gazelle GS, Halpern EF, Rittman WJ, Mueller PR, Rosenthal DI. Radiofrequency tissue ablation: importance of local temperature along the electrode tip exposure in determining lesion shape and size. Acad Radiol. 1996 Mar;3(3):212-8. 31. Zervas NT, Kuwayama A. Pathological characteristics of experimental thermal lesions. Comparison of induction heating and radiofrequency electrocoagulation. J Neurosurg. 1972 Oct;37(4):418-22. 32. Thomsen S. Pathologic analysis of photothermal and photomechanical effects of laser-tissue interactions. Photochem Photobiol. 1991 Jun;53(6):825-35. 33. Haines DE, Verow AF. Observations on electrode-tissue interface temperature and effect on electrical impedance during radiofrequency ablation of ventricular myocardium. Circulation. 1990 Sep;82(3):1034-8. 34. Nath S, Haines DE. Biophysics and pathology of catheter energy delivery systems. Prog Cardiovasc Dis. 1995 Jan-Feb;37(4):185-204. 35. Wissniowski TT, Hansler J, Neureiter D, Frieser M, Schaber S, Esslinger B, et al. Activation of tumor-specific T lymphocytes by radio-frequency ablation of the VX2 hepatoma in rabbits. Cancer Res. 2003 Oct 1;63(19):6496-500. 36. Denys AL, De Baere T, Kuoch V, Dupas B, Chevallier P, Madoff DC, et al. Radio-frequency tissue ablation of the liver: in vivo and ex vivo experiments with four different systems. Eur Radiol. 2003 Oct;13(10):2346-52. 37. Mulier S, Ni Y, Miao Y, Rosiere A, Khoury A, Marchal G, et al. Size and geometry of hepatic radiofrequency lesions. Eur J Surg Oncol. 2003 Dec;29(10):867-78. 38. Cha CH, Lee FT, Jr., Gurney JM, Markhardt BK, Warner TF, Kelcz F, et al. CT versus sonography for monitoring radiofrequency ablation in a porcine liver. AJR Am J Roentgenol. 2000 Sep;175(3):705-11. 39. Boehm T, Malich A, Goldberg SN, Reichenbach JR, Hilger I, Hauff P, et al. Radio-frequency tumor ablation: internally cooled electrode versus saline-enhanced technique in an aggressive rabbit tumor model. Radiology. 2002 Mar;222(3):805-13. 40. Solbiati L, Goldberg SN, Ierace T, Livraghi T, Meloni F, Dellanoce M, et al. Hepatic metastases: percutaneous radio-frequency ablation with cooled-tip electrodes. Radiology. 1997 Nov;205(2):367-73. 41. McGahan JP, Dodd GD, 3rd. Radiofrequency ablation of the liver: current status. AJR Am J Roentgenol. 2001 Jan;176(1):3-16.

162

42. Raman SS, Lu DS, Vodopich DJ, Sayre J, Lassman C. Creation of radiofrequency lesions in a porcine model: correlation with sonography, CT, and histopathology. AJR Am J Roentgenol. 2000 Nov;175(5):1253-8. 43. Hosoki T, Mitomo M, Chor S, Miyahara N, Ohtani M, Morimoto K. Visualization of tumor vessels in hepatocellular carcinoma. Power Doppler compared with color Doppler and angiography. Acta Radiol. 1997 May;38(3):422-7. 44. Harvey CJ, Blomley MJ, Eckersley RJ, Heckemann RA, Butler-Barnes J, Cosgrove DO. Pulse-inversion mode imaging of liver specific microbubbles: improved detection of subcentimetre metastases. Lancet. 2000 Mar 4;355(9206):807-8. 45. Ding H, Kudo M, Onda H, Suetomi Y, Minami Y, Maekawa K. Contrast-enhanced subtraction harmonic sonography for evaluating treatment response in patients with hepatocellular carcinoma. AJR Am J Roentgenol. 2001 Mar;176(3):661-6. 46. Meloni MF, Goldberg SN, Livraghi T, Calliada F, Ricci P, Rossi M, et al. Hepatocellular carcinoma treated with radiofrequency ablation: comparison of pulse inversion contrast-enhanced harmonic sonography, contrast-enhanced power Doppler sonography, and helical CT. AJR Am J Roentgenol. 2001 Aug;177(2):375-80. 47. Choi D, Lim HK, Lee WJ, Kim SH, Kim YH, Lim JH. Early assessment of the therapeutic response to radio frequency ablation for hepatocellular carcinoma: utility of gray scale harmonic ultrasonography with a microbubble contrast agent. J Ultrasound Med. 2003 Nov;22(11):1163-72. 48. Rhim H, Dodd GD, 3rd. Radiofrequency thermal ablation of liver tumors. J Clin Ultrasound. 1999 Jun;27(5):221-9. 49. Lim HS, Jeong YY, Kang HK, Kim JK, Park JG. Imaging features of hepatocellular carcinoma after transcatheter arterial chemoembolization and radiofrequency ablation. AJR Am J Roentgenol. 2006 Oct;187(4):W341-9. 50. Dromain C. Follow-up imaging of liver tumorstreated using percutaneous radio frequency (RF) therapy with helical CT and MR imaging (abstr). Radiology. 1999;213(P):382. 51. Merkle EM, Boll DT, Boaz T, Duerk JL, Chung YC, Jacobs GH, et al. MRI-guided radiofrequency thermal ablation of implanted VX2 liver tumors in a rabbit model: demonstration of feasibility at 0.2 T. Magn Reson Med. 1999 Jul;42(1):141-9. 52. Ruers TJ, Langenhoff BS, Neeleman N, Jager GJ, Strijk S, Wobbes T, et al. Value of positron emission tomography with [F-18]fluorodeoxyglucose in patients with colorectal liver metastases: a prospective study. J Clin Oncol. 2002 Jan 15;20(2):388-95. 53. Khandani AH, Calvo BF, O'Neil BH, Jorgenson J, Mauro MA. A pilot study of early 18F-FDG PET to evaluate the effectiveness of radiofrequency ablation of liver metastases. AJR Am J Roentgenol. 2007 Nov;189(5):1199-202. 54. Donckier V, Van Laethem JL, Goldman S, Van Gansbeke D, Feron P, Ickx B, et al. [F-18] fluorodeoxyglucose positron emission tomography as a tool for early recognition of incomplete tumor destruction after radiofrequency ablation for liver metastases. J Surg Oncol. 2003 Dec;84(4):215-23. 55. Akahane M, Koga H, Kato N, Yamada H, Uozumi K, Tateishi R, et al. Complications of percutaneous radiofrequency ablation for hepato-cellular carcinoma: imaging spectrum and management. Radiographics. 2005 Oct;25 Suppl 1:S57-68. 56. Rhim H. Complications of radiofrequency ablation in hepatocellular carcinoma. Abdom Imaging. 2005 Jul-Aug;30(4):409-18. 57. Mulier S, Mulier P, Ni Y, Miao Y, Dupas B, Marchal G, et al. Complications of radiofrequency coagulation of liver tumours. Br J Surg. 2002 Oct;89(10):1206-22. 58. Rhim H, Yoon KH, Lee JM, Cho Y, Cho JS, Kim SH, et al. Major complications after radio-frequency thermal ablation of hepatic tumors: spectrum of imaging findings. Radiographics. 2003 Jan-Feb;23(1):123-34; discussion 34-6. 59. Pritchard WF, Wray-Cahen D, Karanian JW, Hilbert S, Wood BJ. Radiofrequency cauterization with biopsy introducer needle. J Vasc Interv Radiol. 2004 Feb;15(2 Pt 1):183-7.

163

60. Choi D, Lim HK, Kim MJ, Kim SJ, Kim SH, Lee WJ, et al. Liver abscess after percutaneous radiofrequency ablation for hepatocellular carcinomas: frequency and risk factors. AJR Am J Roentgenol. 2005 Jun;184(6):1860-7. 61. de Baere T, Roche A, Amenabar JM, Lagrange C, Ducreux M, Rougier P, et al. Liver abscess formation after local treatment of liver tumors. Hepatology. 1996 Jun;23(6):1436-40. 62. Patterson EJ, Scudamore CH, Owen DA, Nagy AG, Buczkowski AK. Radiofrequency ablation of porcine liver in vivo: effects of blood flow and treatment time on lesion size. Ann Surg. 1998 Apr;227(4):559-65. 63. Hansen PD, Rogers S, Corless CL, Swanstrom LL, Siperstien AE. Radiofrequency ablation lesions in a pig liver model. J Surg Res. 1999 Nov;87(1):114-21. 64. Kim SH, Lim HK, Choi D, Lee WJ, Kim MJ, Lee SJ, et al. Changes in bile ducts after radiofrequency ablation of hepatocellular carcinoma: frequency and clinical significance. AJR Am J Roentgenol. 2004 Dec;183(6):1611-7. 65. Wood TF, Rose DM, Chung M, Allegra DP, Foshag LJ, Bilchik AJ. Radiofrequency ablation of 231 unresectable hepatic tumors: indications, limitations, and complications. Ann Surg Oncol. 2000 Sep;7(8):593-600. 66. Dominique E, El Otmany A, Goharin A, Attalah D, de Baere T. Intraductal cooling of the main bile ducts during intraoperative radiofrequency ablation. J Surg Oncol. 2001 Apr;76(4):297-300. 67. Lu DS, Raman SS, Vodopich DJ, Wang M, Sayre J, Lassman C. Effect of vessel size on creation of hepatic radiofrequency lesions in pigs: assessment of the "heat sink" effect. AJR Am J Roentgenol. 2002 Jan;178(1):47-51. 68. Takada Y, Kurata M, Ohkohchi N. Rapid and aggressive recurrence accompanied by portal tumor thrombus after radiofrequency ablation for hepatocellular carcinoma. Int J Clin Oncol. 2003 Oct;8(5):332-5. 69. Solbiati L. Radiofrequency thermal ablation of liver metastases. Bartolozzi C, Lencioni R, editors: Springer; 1999. 70. Buczkowski A, Scudamore C, Patterson E, editors. Safety study of hepatic radiofrequency ablation in an animal model. 13th Annual Meeting of the Academy of Surgical Research; 1997; Texas. 71. Curley SA, Izzo F, Delrio P, Ellis LM, Granchi J, Vallone P, et al. Radiofrequency ablation of unresectable primary and metastatic hepatic malignancies: results in 123 patients. Ann Surg. 1999 Jul;230(1):1-8. 72. Ohmoto K, Yamamoto S. Percutaneous microwave coagulation therapy using artificial ascites. AJR Am J Roentgenol. 2001 Mar;176(3):817-8. 73. Goldberg SN, Solbiati L, Halpern EF, Gazelle GS. Variables affecting proper system grounding for radiofrequency ablation in an animal model. J Vasc Interv Radiol. 2000 Sep;11(8):1069-75. 74. Miao Y, Ni Y, Mulier S, Wang K, Hoey MF, Mulier P, et al. Ex vivo experiment on radiofrequency liver ablation with saline infusion through a screw-tip cannulated electrode. J Surg Res. 1997 Jul 15;71(1):19-24. 75. Llovet JM, Vilana R, Bru C, Bianchi L, Salmeron JM, Boix L, et al. Increased risk of tumor seeding after percutaneous radiofrequency ablation for single hepatocellular carcinoma. Hepatology. 2001 May;33(5):1124-9. 76. Takamori R, Wong LL, Dang C, Wong L. Needle-tract implantation from hepatocellular cancer: is needle biopsy of the liver always necessary? Liver Transpl. 2000 Jan;6(1):67-72. 77. Livraghi T. Tumor dissemination after radiofrequency ablation of hepatocellular carcinoma. Hepatology. 2001 Sep;34(3):608-9; author reply 10-1. 78. de Sio I, Castellano L, De Girolamo V, di Santolo SS, Marone A, Del Vecchio Blanco C, et al. Tumor dissemination after radiofrequency ablation of hepatocellular carcinoma. Hepatology. 2001 Sep;34(3):609-10; author reply 10-1.

164

79. Bruix J, Sherman M, Llovet JM, Beaugrand M, Lencioni R, Burroughs AK, et al. Clinical management of hepatocellular carcinoma. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol. 2001 Sep;35(3):421-30. 80. El-Serag HB. Hepatocellular carcinoma: an epidemiologic view. J Clin Gastroenterol. 2002 Nov-Dec;35(5 Suppl 2):S72-8. 81. Figueras J, Jaurrieta E, Valls C, Ramos E, Serrano T, Rafecas A, et al. Resection or transplantation for hepatocellular carcinoma in cirrhotic patients: outcomes based on indicated treatment strategy. J Am Coll Surg. 2000 May;190(5):580-7. 82. Llovet JM, Fuster J, Bruix J. Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology. 1999 Dec;30(6):1434-40. 83. Margarit C, Escartin A, Castells L, Vargas V, Allende E, Bilbao I. Resection for hepatocellular carcinoma is a good option in Child-Turcotte-Pugh class A patients with cirrhosis who are eligible for liver transplantation. Liver Transpl. 2005 Oct;11(10):1242-51. 84. Bigourdan JM, Jaeck D, Meyer N, Meyer C, Oussoultzoglou E, Bachellier P, et al. Small hepatocellular carcinoma in Child A cirrhotic patients: hepatic resection versus transplantation. Liver Transpl. 2003 May;9(5):513-20. 85. Ikai I, Kudo M, Arii S, Omata M, Kojiro M, Sakamoto M, et al. Report of the 18th follow-up survey of primary liver cancer in Japan. Hepatology Research. 2010 November;40 (11):1043-59. 86. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology. 2005 Nov;42(5):1208-36. 87. Ryder SD. Guidelines for the diagnosis and treatment of hepatocellular carcinoma (HCC) in adults. Gut. 2003 May;52 Suppl 3:iii1-8. 88. Scheele J, Stang R, Altendorf-Hofmann A, Paul M. Resection of colorectal liver metastases. World J Surg. 1995 Jan-Feb;19(1):59-71. 89. Pereira PL. Actual role of radiofrequency ablation of liver metastases. Eur Radiol. 2007 Aug;17(8):2062-70. 90. McCarter MD, Fong Y. Metastatic liver tumors. Semin Surg Oncol. 2000 Sep-Oct;19(2):177-88. 91. Lehnert T, Golling M. [Indications and outcome of liver metastases resection]. Radiologe. 2001 Jan;41(1):40-8. 92. Abou-Alfa GK, Johnson P, Knox JJ, Capanu M, Davidenko I, Lacava J, et al. Doxorubicin plus sorafenib vs doxorubicin alone in patients with advanced hepatocellular carcinoma: a randomized trial. JAMA. 2010 Nov 17;304(19):2154-60. 93. Scudamore CH, Lee SI, Patterson EJ, Buczkowski AK, July LV, Chung SW, et al. Radiofrequency ablation followed by resection of malignant liver tumors. Am J Surg. 1999 May;177(5):411-7. 94. Kudo M, Sakaguchi Y, Chung H, Hatanaka K, Hagiwara S, Ishikawa E, et al. Long-term interferon maintenance therapy improves survival in patients with HCV-related hepatocellular carcinoma after curative radiofrequency ablation. A matched case-control study. Oncology. 2007;72 Suppl 1:132-8. 95. Shigeki A, Michio S, Michiie S, Mitsuo S, Takashi K, Shuichiro S. Evidence-Based Clinical Practice Guideline for Hepatocellular Carcinoma (revised version) (in Japanese). Hepatology Research. 2010;40:667-85. 96. Solazzo SA, Ahmed M, Liu Z, Hines-Peralta AU, Goldberg SN. High-power generator for radiofrequency ablation: larger electrodes and pulsing algorithms in bovine ex vivo and porcine in vivo settings. Radiology. 2007 Mar;242(3):743-50. 97. Schmidt D, Trubenbach J, Brieger J, Koenig C, Putzhammer H, Duda SH, et al. Automated saline-enhanced radiofrequency thermal ablation: initial results in ex vivo bovine livers. AJR Am J Roentgenol. 2003 Jan;180(1):163-5.

165

98. Lee JM, Han JK, Kim SH, Shin KS, Lee JY, Park HS, et al. Comparison of wet radiofrequency ablation with dry radiofrequency ablation and radiofrequency ablation using hypertonic saline preinjection: ex vivo bovine liver. Korean J Radiol. 2004 Oct-Dec;5(4):258-65. 99. Lee JM, Han JK, Kim SH, Choi SH, An SK, Han CJ, et al. Bipolar radiofrequency ablation using wet-cooled electrodes: an in vitro experimental study in bovine liver. AJR Am J Roentgenol. 2005 Feb;184(2):391-7. 100. Guglielmi A, Ruzzenente A, Valdegamberi A, Pachera S, Campagnaro T, D'Onofrio M, et al. Radiofrequency ablation versus surgical resection for the treatment of hepatocellular carcinoma in cirrhosis. J Gastrointest Surg. 2008 Jan;12(1):192-8. 101. Khan MR, Poon RT, Ng KK, Chan AC, Yuen J, Tung H, et al. Comparison of percutaneous and surgical approaches for radiofrequency ablation of small and medium hepatocellular carcinoma. Arch Surg. 2007 Dec;142(12):1136-43; discussion 43. 102. Iannitti DA, Dupuy DE, Mayo-Smith WW, Murphy B. Hepatic radiofrequency ablation. Arch Surg. 2002 Apr;137(4):422-6; discussion 7. 103. Zhang YJ, Liang HH, Chen MS, Guo RP, Li JQ, Zheng Y, et al. Hepatocellular carcinoma treated with radiofrequency ablation with or without ethanol injection: a prospective randomized trial. Radiology. 2007 Aug;244(2):599-607. 104. Livraghi T, Goldberg SN, Lazzaroni S, Meloni F, Ierace T, Solbiati L, et al. Hepatocellular carcinoma: radio-frequency ablation of medium and large lesions. Radiology. 2000 Mar;214(3):761-8. 105. Higgins JP, Green S. Cochrane Handbook for systematic reviews of interventions. Chichester: John Wiley & Sons; 2008. 106. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996 Feb;17(1):1-12. 107. Lencioni RA, Allgaier HP, Cioni D, Olschewski M, Deibert P, Crocetti L, et al. Small hepatocellular carcinoma in cirrhosis: randomized comparison of radio-frequency thermal ablation versus percutaneous ethanol injection. Radiology. 2003 Jul;228(1):235-40. 108. Lin SM, Lin CJ, Lin CC, Hsu CW, Chen YC. Radiofrequency ablation improves prognosis compared with ethanol injection for hepatocellular carcinoma <=4 cm. Gastroenterology. 2004 Dec;127 (6):1714-23. 109. Lin SM, Lin CJ, Lin CC, Hsu CW, Chen YC. Randomised controlled trial comparing percutaneous radiofrequency thermal ablation, percutaneous ethanol injection, and percutaneous acetic acid injection to treat hepatocellular carcinoma of 3 cm or less. Gut. 2005 Aug;54(8):1151-6. 110. Shiina S, Teratani T, Obi S, Sato S, Tateishi R, Fujishima T, et al. A randomized controlled trial of radiofrequency ablation with ethanol injection for small hepatocellular carcinoma. Gastroenterology. 2005 Jul;129(1):122-30. 111. Shibata T, Maetani Y, Isoda H, Hiraoka M. Radiofrequency ablation for small hepatocellular carcinoma: prospective comparison of internally cooled electrode and expandable electrode. Radiology. 2006 Jan;238(1):346-53. 112. Ferrari FS, Megliola A, Scorzelli A, Stella A, Vigni F, Drudi FM, et al. Treatment of small HCC through radiofrequency ablation and laser ablation. Comparison of techniques and long-term results. Radiologia Medica. [Comparative Study]. 2007 Apr;112(3):377-93. 113. Brunello F, Veltri A, Carucci P, Pagano E, Ciccone G, Moretto P, et al. Radiofrequency ablation versus ethanol injection for early hepatocellular carcinoma: A randomized controlled trial. Scand J Gastroenterol. 2008;43(6):727-35. 114. Cheng BQ, Jia CQ, Liu CT, Fan W, Wang QL, Zhang ZL, et al. Chemoembolization combined with radiofrequency ablation for patients with hepatocellular carcinoma larger than 3 cm: a randomized controlled trial. JAMA. 2008 Apr 9;299(14):1669-77.

166

115. Yang P, Liang M, Zhang Y, Shen B. Clinical application of a combination therapy of lentinan, multi-electrode RFA and TACE in HCC. Advances in Therapy. [Randomized Controlled Trial]. 2008 Aug;25(8):787-94. 116. Morimoto M, Numata K, Kondou M, Nozaki A, Morita S, Tanaka K. Midterm outcomes in patients with intermediate-sized hepatocellular carcinoma: A randomized controlled trial for determining the efficacy of radiofrequency ablation combined with transcatheter arterial chemoembolization. Cancer. 2010 01 Dec;116 (23):5452-60. 117. Chen MS, Li JQ, Zheng Y, Guo RP, Liang HH, Zhang YQ, et al. A prospective randomized trial comparing percutaneous local ablative therapy and partial hepatectomy for small hepatocellular carcinoma. Ann Surg. 2006 Mar;243(3):321-8. 118. . p. Review Manager (RevMan) [Computer program]. Version 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011. 119. DerSimonian R, Laird N. Meta-analysis in clinical trials. . Control Clin Trials. 1986;7:177-88. 120. Higgins J, Thompson S. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002(21):1539-2558. 121. Cho CM, Tak WY, Kweon YO, Kim SK, Choi YH, Hwang YJ, et al. [The comparative results of radiofrequency ablation versus surgical resection for the treatment of hepatocellular carcinoma.]. Korean J Hepatol. 2005 Mar;11(1):59-71. 122. Hiraoka A, Horiike N, Yamashita Y, Koizumi Y, Doi K, Yamamoto Y, et al. Efficacy of radiofrequency ablation therapy compared to surgical resection in 164 patients in Japan with single hepatocellular carcinoma smaller than 3 cm, along with report of complications. Hepato-Gastroenterology. [Comparative Study]. 2008 Nov-Dec;55(88):2171-4. 123. Hong SN, Lee SY, Choi MS, Lee JH, Koh KC, Paik SW, et al. Comparing the outcomes of radiofrequency ablation and surgery in patients with a single small hepatocellular carcinoma and well-preserved hepatic function. J Clin Gastroenterol. 2005 Mar;39(3):247-52. 124. Kobayashi M, Ikeda K, Kawamura Y, Yatsuji H, Hosaka T, Sezaki H, et al. High serum des-gamma-carboxy prothrombin level predicts poor prognosis after radiofrequency ablation of hepatocellular carcinoma. Cancer. [Research Support, Non-U.S. Gov't]. 2009 Feb 1;115(3):571-80. 125. Lupo L, Panzera P, Giannelli G, Memeo M, Gentile A, Memeo V. Single hepatocellular carcinoma ranging from 3 to 5 cm: radiofrequency ablation or resection? HPB (Oxford). 2007;9(6):429-34. 126. Santambrogio R, Opocher E, Zuin M, Selmi C, Bertolini E, Costa M, et al. Surgical resection versus laparoscopic radiofrequency ablation in patients with hepatocellular carcinoma and Child-Pugh class a liver cirrhosis. Annals of Surgical Oncology. 2009 Dec;16(12):3289-98. 127. Takahashi S, Kudo M, Chung H, Inoue T, Nagashima M, Kitai S, et al. Outcomes of nontransplant potentially curative therapy for early-stage hepatocellular carcinoma in Child-Pugh stage A cirrhosis is comparable with liver transplantation. Digestive Diseases. [Comparative Study]. 2007;25(4):303-9. 128. Ueno S, Sakoda M, Kubo F, Hiwatashi K, Tateno T, Baba Y, et al. Surgical resection versus radiofrequency ablation for small hepatocellular carcinomas within the Milan criteria. Journal of Hepato-Biliary-Pancreatic Surgery. 2009;16 (3):359-66. 129. Hong SN, Lee S-Y, Choi MS, Lee JH, Koh KC, Paik SW, et al. Comparing the outcomes of radiofrequency ablation and surgery in patients with a single small hepatocellular carcinoma and well-preserved hepatic function. Journal of Clinical Gastroenterology. [Comparative Study]. 2005 Mar;39(3):247-52. 130. Helmberger T, Dogan S, Straub G, Schrader A, Jungst C, Reiser M, et al. Liver resection or combined chemoembolization and radiofrequency ablation improve survival in patients with hepatocellular carcinoma. Digestion. 2007 Aug;75 (2-3):104-12. 131. Lam VW, Ng KK, Chok KS, Cheung TT, Yuen J, Tung H, et al. Risk factors and prognostic factors of local recurrence after radiofrequency ablation of hepatocellular carcinoma. J Am Coll Surg. 2008 Jul;207(1):20-9.

167

132. Maluccio M, Covey AM, Gandhi R, Gonen M, Getrajdman GI, Brody LA, et al. Comparison of survival rates after bland arterial embolization and ablation versus surgical resection for treating solitary hepatocellular carcinoma up to 7 cm. J Vasc Interv Radiol. 2005 Jul;16(7):955-61. 133. Ogihara M, Wong LL, Machi J. Radiofrequency ablation versus surgical resection for single nodule hepatocellular carcinoma: long-term outcomes. HPB (Oxford). 2005;7(3):214-21. 134. Vivarelli M, Guglielmi A, Ruzzenente A, Cucchetti A, Bellusci R, Cordiano C, et al. Surgical resection versus percutaneous radiofrequency ablation in the treatment of hepatocellular carcinoma on cirrhotic liver. Ann Surg. 2004 Jul;240(1):102-7. 135. Yamagiwa K, Shiraki K, Yamakado K, Mizuno S, Hori T, Yagi S, et al. Survival rates according to the Cancer of the Liver Italian Program scores of 345 hepatocellular carcinoma patients after multimodality treatments during a 10-year period in a retrospective study. Journal of Gastroenterology & Hepatology. [Comparative Study]. 2008 Mar;23(3):482-90. 136. Yamakado K, Nakatsuka A, Takaki H, Yokoi H, Usui M, Sakurai H, et al. Early-stage hepatocellular carcinoma: radiofrequency ablation combined with chemoembolization versus hepatectomy. Radiology. 2008 Apr;247(1):260-6. 137. Peng ZW, Zhang YJ, Chen MS, Lin XJ, Liang HH, Shi M. Radiofrequency ablation as first-line treatment for small solitary hepatocellular carcinoma: long-term results. European Journal of Surgical Oncology. [Research Support, Non-U.S. Gov't]. 2010 Nov;36(11):1054-60. 138. Yang W, Chen MH, Wang MQ, Cui M, Gao W, Wu W, et al. Combination therapy of radiofrequency ablation and transarterial chemoembolization in recurrent hepatocellular carcinoma after hepatectomy compared with single treatment. Hepatology Research. 2009;39 (3):231-40. 139. Chok KS, Ng KK, Poon RTP, Chi ML, Yuen J, Wai KT, et al. Comparable survival in patients with unresectable hepatocellular carcinoma treated by radiofrequency ablation or transarterial chemoembolization. Archives of Surgery. 2006;141 (12):1231-6. 140. Murakami T, Ishimaru H, Sakamoto I, Uetani M, Matsuoka Y, Daikoku M, et al. Percutaneous radiofrequency ablation and transcatheter arterial chemoembolization for hypervascular hepatocellular carcinoma: rate and risk factors for local recurrence. Cardiovascular & Interventional Radiology. [Comparative Study]. 2007 Jul-Aug;30(4):696-704. 141. Lu MD, Xu HX, Xie XY, Yin XY, Chen JW, Kuang M, et al. Percutaneous microwave and radiofrequency ablation for hepatocellular carcinoma: A retrospective comparative study. Journal of Gastroenterology. 2005 Nov;40 (11):1054-60. 142. Xu HX, Lu MD, Xie XY, Yin XY, Kuang M, Chen JW, et al. Prognostic factors for long-term outcome after percutaneous thermal ablation for hepatocellular carcinoma: a survival analysis of 137 consecutive patients. Clin Radiol. 2005 Sep;60(9):1018-25. 143. Ohmoto K, Yoshioka N, Tomiyama Y, Shibata N, Kawase T, Yoshida K, et al. Comparison of therapeutic effects between radiofrequency ablation and percutaneous microwave coagulation therapy for small hepatocellular carcinomas. Journal of Gastroenterology and Hepatology. 2009 February;24 (2):223-7. 144. Abu-Hilal M, Primrose JN, Casaril A, McPhail MJ, Pearce NW, Nicoli N. Surgical resection versus radiofrequency ablation in the treatment of small unifocal hepatocellular carcinoma. J Gastrointest Surg. 2008 Sep;12(9):1521-6. 145. Liang HH, Chen MS, Peng ZW, Zhang YJ, Zhang YQ, Li JQ, et al. Percutaneous radiofrequency ablation versus repeat hepatectomy for recurrent hepatocellular carcinoma: a retrospective study. Ann Surg Oncol. 2008 Dec;15(12):3484-93. 146. Montorsi M, Santambrogio R, Bianchi P, Donadon M, Moroni E, Spinelli A, et al. Survival and recurrences after hepatic resection or radiofrequency for hepatocellular carcinoma in cirrhotic patients: a multivariate analysis. J Gastrointest Surg. 2005 Jan;9(1):62-7; discussion 7-8. 147. Abitabile P, Hartl U, Lange J, Maurer CA. Radiofrequency ablation permits an effective treatment for colorectal liver metastasis. Eur J Surg Oncol. 2007 Feb;33(1):67-71.

168

148. Seror O, N'Kontchou G, Tin-Tin-Htar M, Barrucand C, Ganne N, Coderc E, et al. Radiofrequency Ablation with Internally Cooled versus Perfused Electrodes for the Treatment of Small Hepatocellular Carcinoma in Patients with Cirrhosis. Journal of Vascular and Interventional Radiology. 2008 May;19 (5):718-24. 149. Kang TW, Rhim H, Kim EY, Kim YS, Choi D, Lee WJ, et al. Percutaneous radiofrequency ablation for the hepatocellular carcinoma abutting the diaphragm: assessment of safety and therapeutic efficacy. Korean Journal of Radiology. [Comparative Study]. 2009 Jan-Feb;10(1):34-42. 150. Ng KK, Poon RT, Lo CM, Yuen J, Tso WK, Fan ST. Analysis of recurrence pattern and its influence on survival outcome after radiofrequency ablation of hepatocellular carcinoma. J Gastrointest Surg. 2008 Jan;12(1):183-91. 151. Kim YS, Lee WJ, Rhim H, Lim HK, Choi D, Lee JY. The minimal ablative margin of radiofrequency ablation of hepatocellular carcinoma (> 2 and < 5 cm) needed to prevent local tumor progression: 3D quantitative assessment using CT image fusion. American Journal of Roentgenology. 2010 September;195 (3):758-65. 152. Kim SH, Lim HK, Choi D, Lee WJ, Kim SH, Kim MJ, et al. Percutaneous radiofrequency ablation of hepatocellular carcinoma: effect of histologic grade on therapeutic results. AJR. 2006 May;American Journal of Roentgenology. 186(5 Suppl):S327-33. 153. Solbiati L, Livraghi T, Goldberg SN, Ierace T, Meloni F, Dellanoce M, et al. Percutaneous radio-frequency ablation of hepatic metastases from colorectal cancer: long-term results in 117 patients. Radiology. 2001 Oct;221(1):159-66. 154. Rhim H, Lim HK, Kim YS, Choi D, Lee WJ. Radiofrequency ablation of hepatic tumors: lessons learned from 3000 procedures. J Gastroenterol Hepatol. 2008 Oct;23(10):1492-500. 155. Chen MH, Yang W, Yan K, Gao W, Dai Y, Wang YB, et al. Treatment efficacy of radiofrequency ablation of 338 patients with hepatic malignant tumor and the relevant complications. World J Gastroenterol. 2005 Oct 28;11(40):6395-401. 156. Gillams AR, Lees WR. Five-year survival in 309 patients with colorectal liver metastases treated with radiofrequency ablation. Eur Radiol. 2009 May;19(5):1206-13. 157. Li W, Ma K, Cheng M, Chui K, Chan P, Chu W, et al. Radiofrequency ablation for hepatocellular carcinoma: a survival analysis of 117 patients. ANZ Journal of Surgery. 2010;80:714–21. 158. Pawlik TM, Izzo F, Cohen DS, Morris JS, Curley SA. Combined resection and radiofrequency ablation for advanced hepatic malignancies: results in 172 patients. Ann Surg Oncol. 2003 Nov;10(9):1059-69. 159. Abdalla EK, Vauthey JN, Ellis LM, Ellis V, Pollock R, Broglio KR, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg. 2004 Jun;239(6):818-25; discussion 25-7. 160. Gleisner AL, Choti MA, Assumpcao L, Nathan H, Schulick RD, Pawlik TM. Colorectal liver metastases: recurrence and survival following hepatic resection, radiofrequency ablation, and combined resection-radiofrequency ablation. Arch Surg. 2008 Dec;143(12):1204-12. 161. Yao FY, Bass NM, Nikolai B, Merriman R, Davern TJ, Kerlan R, et al. A follow-up analysis of the pattern and predictors of dropout from the waiting list for liver transplantation in patients with hepatocellular carcinoma: implications for the current organ allocation policy. Liver Transpl. 2003 Jul;9(7):684-92. 162. Freeman RB, Edwards EB, Harper AM. Waiting list removal rates among patients with chronic and malignant liver diseases. Am J Transplant. 2006 Jun;6(6):1416-21. 163. Fisher RA, Maluf D, Cotterell AH, Stravitz T, Wolfe L, Luketic V, et al. Non-resective ablation therapy for hepatocellular carcinoma: effectiveness measured by intention-to-treat and dropout from liver transplant waiting list. Clin Transplant. 2004 Oct;18(5):502-12. 164. Mazzaferro V, Battiston C, Perrone S, Pulvirenti A, Regalia E, Romito R, et al. Radiofrequency ablation of small hepatocellular carcinoma in cirrhotic patients awaiting liver transplantation: a prospective study. Ann Surg. 2004 Nov;240(5):900-9.

169

165. Choi D, Lim HK, Rhim H, Kim YS, Yoo BC, Paik SW, et al. Percutaneous radiofrequency ablation for recurrent hepatocellular carcinoma after hepatectomy: Long-term results and prognostic factors. Annals of Surgical Oncology. 2007 Aug;14 (8):2319-29. 166. Minagawa M, Makuuchi M, Takayama T, Kokudo N. Selection criteria for repeat hepatectomy in patients with recurrent hepatocellular carcinoma. Ann Surg. 2003 Nov;238(5):703-10. 167. Poon RT, Fan ST, Lo CM, Ng IO, Liu CL, Lam CM, et al. Improving survival results after resection of hepatocellular carcinoma: a prospective study of 377 patients over 10 years. Ann Surg. 2001 Jul;234(1):63-70. 168. Fong Y, Sun RL, Jarnagin W, Blumgart LH. An analysis of 412 cases of hepatocellular carcinoma at a Western center. Ann Surg. 1999 Jun;229(6):790-9; discussion 9-800. 169. Makuuchi M, Takayama T, Kubota K, Kimura W, Midorikawa Y, Miyagawa S, et al. Hepatic resection for hepatocellular carcinoma -- Japanese experience. Hepatogastroenterology. 1998 Aug;45 Suppl 3:1267-74. 170. Poon RT, Fan ST, Lo CM, Liu CL, Wong J. Intrahepatic recurrence after curative resection of hepatocellular carcinoma: long-term results of treatment and prognostic factors. Ann Surg. 1999 Feb;229(2):216-22. 171. Matsuda M, Fujii H, Kono H, Matsumoto Y. Surgical treatment of recurrent hepatocellular carcinoma based on the mode of recurrence: repeat hepatic resection or ablation are good choices for patients with recurrent multicentric cancer. J Hepatobiliary Pancreat Surg. 2001;8(4):353-9. 172. Kakazu T, Makuuchi M, Kawasaki S, Miyagawa S, Hashikura Y, Kosuge T, et al. Repeat hepatic resection for recurrent hepatocellular carcinoma. Hepatogastroenterology. 1993 Aug;40(4):337-41. 173. Arii S, Monden K, Niwano M, Furutani M, Mori A, Mizumoto M, et al. Results of surgical treatment for recurrent hepatocellular carcinoma; comparison of outcome among patients with multicentric carcinogenesis, intrahepatic metastasis, and extrahepatic recurrence. J Hepatobiliary Pancreat Surg. 1998;5(1):86-92. 174. DeAngelis C, Fontanarosa P. Retraction: Cheng B-Q, et al. Chemoembolization combined with radiofrequency ablation for patients with hepatocellular carcinoma larger than 3 cm: a randomized controlled trial. JAMA. 2008;299(14):1669-1677. JAMA [serial on the Internet]. 2009; 301(18). 175. Jessup JM, McGinnis LS, Steele GD, Jr., Menck HR, Winchester DP. The National Cancer Data Base. Report on colon cancer. Cancer. 1996 Aug 15;78(4):918-26. 176. Weiss L, Grundmann E, Torhorst J, Hartveit F, Moberg I, Eder M, et al. Haematogenous metastatic patterns in colonic carcinoma: an analysis of 1541 necropsies. J Pathol. 1986 Nov;150(3):195-203. 177. Choti MA, Sitzmann JV, Tiburi MF, Sumetchotimetha W, Rangsin R, Schulick RD, et al. Trends in long-term survival following liver resection for hepatic colorectal metastases. Ann Surg. 2002 Jun;235(6):759-66. 178. Solbiati L, Livraghi T, Goldberg SN, Ierace T, Meloni F, Dellanoce M, et al. Percutaneous radio-frequency ablation of hepatic metastases from colorectal cancer: Long-term results in 117 patients. Radiology. 2001;221 (1):159-66. 179. Gillams AR, Lees WR. Radio-frequency ablation of colorectal liver metastases in 167 patients. Eur Radiol. 2004 Dec;14(12):2261-7. 180. Navarra G, Ayav A, Weber JC, Jensen SL, Smadga C, Nicholls JP, et al. Short- and-long term results of intraoperative radiofrequency ablation of liver metastases. Int J Colorectal Dis. 2005 Nov;20(6):521-8. 181. Ahmad A, Chen SL, Kavanagh MA, Allegra DP, Bilchik AJ. Radiofrequency ablation of hepatic metastases from colorectal cancer: are newer generation probes better? The American surgeon. 2006 Oct;72 (10):875-9.

170

182. Amersi FF, McElrath-Garza A, Ahmad A, Zogakis T, Allegra DP, Krasne R, et al. Long-term survival after radiofrequency ablation of complex unresectable liver tumors. Arch Surg. 2006 Jun;141(6):581-7; discussion 7-8. 183. Hildebrand P, Kleemann M, Roblick UJ, Mirow L, Birth M, Leibecke T, et al. Radiofrequency-ablation of unresectable primary and secondary liver tumors: results in 88 patients. Langenbecks Archives of Surgery. 2006 Apr;391(2):118-23. 184. Jakobs TF, Hoffmann RT, Trumm C, Reiser MF, Helmberger TK. Radiofrequency ablation of colorectal liver metastases: Mid-term results in 68 patients. Anticancer Research. 2006 Jan;26 (1 B):671-80. 185. van Duijnhoven FH, Jansen MC, Junggeburt JM, van Hillegersberg R, Rijken AM, van Coevorden F, et al. Factors influencing the local failure rate of radiofrequency ablation of colorectal liver metastases. Ann Surg Oncol. 2006 May;13(5):651-8. 186. Sorensen SM, Mortensen FV, Nielsen DT. Radiofrequency ablation of colorectal liver metastases: long-term survival. Acta Radiol. 2007 Apr;48(3):253-8. 187. Blusse Van Oud-Alblas M, Fioole B, Jansen MC, Van Duijnhoven FH, Van Hillegersberg R, Rijken AM, et al. Radiofrequency ablation of colorectal metastases to the liver: Results since the first application in the Netherlands. [Dutch]. Nederlands Tijdschrift voor Geneeskunde. 2008 12 Apr;152 (15):880-6. 188. Gillams AR, Lees WR. Five-year survival in 309 patients with colorectal liver metastases treated with radiofrequency ablation. European Radiology. 2009;19 (5):1206-13. 189. Veltri A, Sacchetto P, Tosetti I, Pagano E, Fava C, Gandini G. Radiofrequency ablation of colorectal liver metastases: Small size favorably predicts technique effectiveness and survival. CardioVascular and Interventional Radiology. 2008 September;31 (5):948-56. 190. Meloni MF, Andreano A, Laeseke PF, Livraghi T, Sironi S, Lee Jr FT. Breast cancer liver metastases: US-guided percutaneous radiofrequency ablation - Intermediate and long-term survival rates. Radiology. 2009 December;253 (3):861-9. 191. Reuter NP, Woodall CE, Scoggins CR, McMasters KM, Martin RCG. Radiofrequency Ablation vs. Resection for hepatic colorectal metastasis: Therapeutically equivalent? Journal of Gastrointestinal Surgery. 2009 March;13 (3):486-91. 192. Chiou YY, Chou YH, Chiang JH, Wang HK, Chang CY. Percutaneous ultrasound-guided radiofrequency ablation of colorectal liver metastases. Chinese Journal of Radiology. 2005 Jun;30 (3):153-8. 193. Elias D, Baton O, Sideris L, Matsuhisa T, Pocard M, Lasser P. Local recurrences after intraoperative radiofrequency ablation of liver metastases: a comparative study with anatomic and wedge resections. Annals of surgical oncology : the official journal of the Society of Surgical Oncology. 2004 May;11 (5):500-5. 194. Bleicher RJ, Allegra DP, Nora DT, Wood TF, Foshag LJ, Bilchik AJ. Radiofrequency ablation in 447 complex unresectable liver tumors: lessons learned. Ann Surg Oncol. 2003 Jan-Feb;10(1):52-8. 195. Berber E, Ari E, Herceg N, Siperstein A. Laparoscopic radiofrequency thermal ablation for unusual hepatic tumors: operative indications and outcomes. Surgical Endoscopy. 2005 Dec;19(12):1613-7. 196. Berber E, Pelley R, Siperstein AE. Predictors of survival after radiofrequency thermal ablation of colorectal cancer metastases to the liver: a prospective study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2005 1 Mar;23 (7):1358-64. 197. Berber E, Tsinberg M, Tellioglu G, Simpfendorfer CH, Siperstein AE. Resection versus laparoscopic radiofrequency thermal ablation of solitary colorectal liver metastasis. Journal of Gastrointestinal Surgery. 2008 November;12 (11):1967-72. 198. Mazzaglia PJ, Berber E, Milas M, Siperstein AE. Laparoscopic radiofrequency ablation of neuroendocrine liver metastases: a 10-year experience evaluating predictors of survival. Surgery. 2007 Jul;142 (1):10-9.

171

199. Chen MH, Wei Y, Yan K, Gao W, Dai Y, Huo L, et al. Treatment strategy to optimize radiofrequency ablation for liver malignancies. Journal of Vascular & Interventional Radiology. 2006 Apr;17(4):671-83. 200. Hur H, Ko YT, Min BS, Kim KS, Choi JS, Sohn SK, et al. Comparative study of resection and radiofrequency ablation in the treatment of solitary colorectal liver metastases. American Journal of Surgery. 2009 June;197 (6):728-36. 201. Gillams AR, Lees WR. Radio-frequency ablation of colorectal liver metastases in 167 patients. European Radiology. 2004 Dec;14 (12):2261-7. 202. Sorensen SM, Mortensen FV, Nielsen DT. Radiofrequency ablation of colorectal liver metastases: long-term survival. Acta radiologica (Stockholm, Sweden : 1987). 2007 Apr;48 (3):253-8. 203. Elias D, Baton O, Sideris L, Matsuhisa T, Pocard M, Lasser P. Local recurrences after intraoperative radiofrequency ablation of liver metastases: a comparative study with anatomic and wedge resections. Ann Surg Oncol. 2004 May;11(5):500-5. 204. Scaife CL, Curley SA, Izzo F, Marra P, Delrio P, Daniele B, et al. Feasibility of adjuvant hepatic arterial infusion of chemotherapy after radiofrequency ablation with or without resection in patients with hepatic metastases from colorectal cancer. Annals of surgical oncology : the official journal of the Society of Surgical Oncology. 2003 May;10 (4):348-54. 205. Machi J, Oishi AJ, Sumida K, Sakamoto K, Furumoto NL, Oishi RH, et al. Long-term outcome of radiofrequency ablation for unresectable liver metastases from colorectal cancer: evaluation of prognostic factors and effectiveness in first- and second-line management. Cancer J. 2006 Jul-Aug;12(4):318-26. 206. Siperstein AE, Berber E, Ballem N, Parikh RT. Survival after radiofrequency ablation of colorectal liver metastases: 10-year experience. Ann Surg. 2007 Oct;246(4):559-65; discussion 65-7. 207. Elias D, Baton O, Sideris L, Boige V, x00E, rie, et al. Hepatectomy plus intraoperative radiofrequency ablation and chemotherapy to treat technically unresectable multiple colorectal liver metastases. Journal of Surgical Oncology. [Clinical Trial]. 2005 Apr 1;90(1):36-42. 208. Gleisner AL, Choti MA, Assumpcao L, Nathan H, Schulick RD, Pawlik TM. Colorectal liver metastases: recurrence and survival following hepatic resection, radiofrequency ablation, and combined resection-radiofrequency ablation. Archives of surgery (Chicago, Ill. 2008 Dec;: 1960). 143 (12):1204-12. 209. Leblanc F, Fonck M, Brunet R, Becouarn Y, Mathoulin-Pelissier S, Evrard S. Comparison of hepatic recurrences after resection or intraoperative radiofrequency ablation indicated by size and topographical characteristics of the metastases. European Journal of Surgical Oncology. 2008 Feb;34 (2):185-90. 210. Vyslouzil K, Klementa I, Stary L, Zboril P, Skalicky P, Dlouhy M, et al. Radiofrequency ablation of colorectal liver metastases. [German]. Zentralblatt fur Chirurgie. 2009 Apr;134 (2):145-8. 211. Pawlik TM, Izzo F, Cohen DS, Morris JS, Curley SA. Combined resection and radiofrequency ablation for advanced hepatic malignancies: Results in 172 patients. Annals of Surgical Oncology. 2003;10 (9):1059-69. 212. Abdalla EK, Vauthey J-N, Ellis LM, Ellis V, Pollock R, Broglio KR, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Annals of Surgery. [Comparative Study]. 2004 Jun;239(6):818-25; discussion 25-7. 213. Elias D, De Baere T, Smayra T, Ouellet JF, Roche A, Lasser P. Percutaneous radiofrequency thermoablation as an alternative to surgery for treatment of liver tumour recurrence after hepatectomy. Br J Surg. 2002 Jun;89(6):752-6.

172

214. Otto G, Duber C, Hoppe-Lotichius M, Konig J, Heise M, Pitton MB. Radiofrequency ablation as first-line treatment in patients with early colorectal liver metastases amenable to surgery. Ann Surg. 2010 May;251(5):796-803. 215. Livraghi T, Solbiati L, Meloni F, Ierace T, Goldberg SN, Gazelle GS. Percutaneous radiofrequency ablation of liver metastases in potential candidates for resection: the "test-of-time approach". Cancer. 2003 Jun 15;97(12):3027-35. 216. Livraghi T, Meloni F, Di Stasi M, Rolle E, Solbiati L, Tinelli C, et al. Sustained complete response and complications rates after radiofrequency ablation of very early hepatocellular carcinoma in cirrhosis: Is resection still the treatment of choice? Hepatology. 2008 Jan;47(1):82-9. 217. Berber E, Tsinberg M, Tellioglu G, Simpfendorfer CH, Siperstein AE. Resection versus laparoscopic radiofrequency thermal ablation of solitary colorectal liver metastasis. J Gastrointest Surg. 2008 Nov;12(11):1967-72. 218. Hur H, Ko YT, Min BS, Kim KS, Choi JS, Sohn SK, et al. Comparative study of resection and radiofrequency ablation in the treatment of solitary colorectal liver metastases. Am J Surg. 2009 Jun;197(6):728-36. 219. Griffin DT, Dodd NJ, Zhao S, Pullan BR, Moore JV. Low-level direct electrical current therapy for hepatic metastases. I. Preclinical studies on normal liver. Br J Cancer. 1995 Jul;72(1):31-4. 220. Samuelsson L, Jonsson L, Lamm IL, Linden CJ, Ewers SB. Electrolysis with different electrode materials and combined with irradiation for treatment of experimental rat tumors. Acta Radiol. 1991 Mar;32(2):178-81. 221. Samuelsson L, Olin T, Berg NO. Electrolytic destruction of lung tissue in the rabbit. Acta Radiol Diagn (Stockh). 1980;21(4):447-54. 222. Finch JG, Fosh B, Anthony A, Slimani E, Texler M, Berry DP, et al. Liver electrolysis: pH can reliably monitor the extent of hepatic ablation in pigs. Clin Sci (Lond). 2002 Apr;102(4):389-95. 223. von Euler H, Nilsson E, Olsson JM, Lagerstedt AS. Electrochemical treatment (EChT) effects in rat mammary and liver tissue. In vivo optimizing of a dose-planning model for EChT of tumours. Bioelectrochemistry. 2001 Nov;54(2):117-24. 224. Hagedorn R, Fuhr G. Steady state electrolysis and isoelectric focusing. Electrophoresis. 1990 Apr;11(4):281-9. 225. Baxter PS, Wemyss-Holden SA, Dennison AR, Maddern GJ. Electrochemically induced hepatic necrosis: the next step forward in patients with unresectable liver tumours? Aust N Z J Surg. 1998 Sep;68(9):637-40. 226. Robertson GS, Wemyss-Holden SA, Dennison AR, Hall PM, Baxter P, Maddern GJ. Experimental study of electrolysis-induced hepatic necrosis. Br J Surg. 1998 Sep;85(9):1212-6. 227. Jarm T, Cemazar M, Steinberg F, Streffer C, Sersa G, Miklavcic D. Perturbation of blood flow as a mechanism of anti-tumour action of direct current electrotherapy. Physiol Meas. 2003 Feb;24(1):75-90. 228. Li K, Xin Y, Gu Y, Xu B, Fan D, Ni B. Effects of direct current on dog liver: possible mechanisms for tumor electrochemical treatment. Bioelectromagnetics. 1997;18(1):2-7. 229. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med. 1999 Mar 11;340(10):745-50. 230. von Euler H, Olsson JM, Hultenby K, Thorne A, Lagerstedt AS. Animal models for treatment of unresectable liver tumours: a histopathologic and ultra-structural study of cellular toxic changes after electrochemical treatment in rat and dog liver. Bioelectrochemistry. 2003 Apr;59(1-2):89-98. 231. Wemyss-Holden SA, de la MHP, Robertson GS, Dennison AR, Vanderzon PS, Maddern GJ. The safety of electrolytically induced hepatic necrosis in a pig model. Aust N Z J Surg. 2000 Aug;70(8):607-12. 232. Hinz S, Egberts JH, Pauser U, Schafmayer C, Fandrich F, Tepel J. Electrolytic ablation is as effective as radiofrequency ablation in the treatment of artificial liver metastases in a pig model. J Surg Oncol. 2008 Aug 1;98(2):135-8.

173

233. Wemyss-Holden SA, Robertson GS, Dennison AR, Vanderzon PS, Hall PM, Maddern GJ. A new treatment for unresectable liver tumours: long-term studies of electrolytic lesions in the pig liver. Clin Sci (Lond). 2000 May;98(5):561-7. 234. Xin YL. Advances in the treatment of malignant tumours by electrochemical therapy (ECT). Eur J Surg Suppl. 1994(574):31-5. 235. Wang HL. Electrochemical therapy of 74 cases of liver cancer. Eur J Surg Suppl. 1994(574):55-7. 236. Lao YH, Ge TG, Zheng XL, Zhang JZ, Hua YW, Mao SM, et al. Electrochemical therapy for intermediate and advanced liver cancer: a report of 50 cases. Eur J Surg Suppl. 1994(574):51-3. 237. Lin XZ, Jen CM, Chou CK, Chou CS, Sung MJ, Chou TC. Saturated saline enhances the effect of electrochemical therapy. Dig Dis Sci. 2000 Mar;45(3):509-14. 238. Samuelsson L, Jonsson L. Electrolytic destruction of tissue in the normal lung of the pig. Acta Radiol Diagn (Stockh). 1981;22(1):9-14. 239. Nordenstrom B. Biologically Closed Electric Circuits: Clinical, Experimental and Theoretical Evidence for an Additional Circulatory System. Stockholm: Nordic Medical Publications; 1983. 240. Tanaka T, Isfort P, Bruners P, Penzkofer T, Kichikawa K, Schmitz-Rode T, et al. Optimization of Direct Current-Enhanced Radiofrequency Ablation: An Ex Vivo Study. Cardiovasc Intervent Radiol. 2010 Jan 22. 241. Marks R. The stratum corneum barrier: the final frontier. J Nutr. 2004 Aug;134(8 Suppl):2017S-21S. 242. Cho YK, Kim JK, Kim MY, Rhim H, Han JK. Systematic review of randomized trials for hepatocellular carcinoma treated with percutaneous ablation therapies. Hepatology. 2009 Feb;49(2):453-9. 243. McGrane S, McSweeney SE, Maher MM. Which patients will benefit from percutaneous radiofrequency ablation of colorectal liver metastases? Critically appraised topic. Abdom Imaging. 2008 Jan-Feb;33(1):48-53. 244. Rhim H. Review of asian experience of thermal ablation techniques and clinical practice. Int J Hyperthermia. 2004 Nov;20(7):699-712. 245. Sutherland LM, Williams JA, Padbury RT, Gotley DC, Stokes B, Maddern GJ. Radiofrequency ablation of liver tumors: a systematic review. Arch Surg. 2006 Feb;141(2):181-90. 246. Kim SH, Lim HK, Choi D, Lee WJ, Kim MJ, Kim CK, et al. Percutaneous radiofrequency ablation of hepatocellular carcinoma: effect of histologic grade on therapeutic results. AJR Am J Roentgenol. 2006 May;186(5 Suppl):S327-33. 247. Bilchik AJ, Wood TF, Allegra DP. Radiofrequency ablation of unresectable hepatic malignancies: lessons learned. Oncologist. 2001;6(1):24-33. 248. Curley SA, Izzo F, Ellis LM, Nicolas Vauthey J, Vallone P. Radiofrequency ablation of hepatocellular cancer in 110 patients with cirrhosis. Ann Surg. 2000 Sep;232(3):381-91. 249. Lin SM, Lin CC, Chen WT, Chen YC, Hsu CW. Radiofrequency ablation for hepatocellular carcinoma: a prospective comparison of four radiofrequency devices. J Vasc Interv Radiol. 2007 Sep;18(9):1118-25. 250. Giorgio A, Tarantino L, de Stefano G, Scala V, Liorre G, Scarano F, et al. Percutaneous sonographically guided saline-enhanced radiofrequency ablation of hepatocellular carcinoma. AJR Am J Roentgenol. 2003 Aug;181(2):479-84. 251. Lee JM, Han JK, Kim SH, Lee JY, Choi SH, Choi BI. Hepatic bipolar radiofrequency ablation using perfused-cooled electrodes: a comparative study in the ex vivo bovine liver. Br J Radiol. 2004 Nov;77(923):944-9. 252. Kettenbach J, Kostler W, Rucklinger E, Gustorff B, Hupfl M, Wolf F, et al. Percutaneous saline-enhanced radiofrequency ablation of unresectable hepatic tumors: initial experience in 26 patients. AJR Am J Roentgenol. 2003 Jun;180(6):1537-45.

174

253. Bruners P, Pfeffer J, Kazim RM, Gunther RW, Schmitz-Rode T, Mahnken AH. A newly developed perfused umbrella electrode for radiofrequency ablation: an ex vivo evaluation study in bovine liver. Cardiovasc Intervent Radiol. 2007 Sep-Oct;30(5):992-8. 254. Vijh AK. Electrochemical field effects in biological materials: electro-osmotic dewatering of cancerous tissue as the mechanistic proposal for the electrochemical treatment of tumors. J Mater Sci Mater Med. 1999 Jul;10(7):419-23. 255. Dyce K, Sack W, Wensing C. Textbook of Veterinary Anatomy. 3rd ed. Philadelphia: Saunders; 2002. 256. Geoghegan JG, Scheele J. Treatment of colorectal liver metastases. Br J Surg. 1999 Feb;86(2):158-69. 257. Steele G, Jr., Ravikumar TS. Resection of hepatic metastases from colorectal cancer. Biologic perspective. Ann Surg. 1989 Aug;210(2):127-38. 258. Mulier S, Ni Y, Jamart J, Ruers T, Marchal G, Michel L. Local recurrence after hepatic radiofrequency coagulation: multivariate meta-analysis and review of contributing factors. Ann Surg. 2005 Aug;242(2):158-71. 259. Hori T, Nagata K, Hasuike S, Onaga M, Motoda M, Moriuchi A, et al. Risk factors for the local recurrence of hepatocellular carcinoma after a single session of percutaneous radiofrequency ablation. J Gastroenterol. 2003;38(10):977-81. 260. Faes TJ, van der Meij HA, de Munck JC, Heethaar RM. The electric resistivity of human tissues (100 Hz-10 MHz): a meta-analysis of review studies. Physiol Meas. 1999 Nov;20(4):R1-10. 261. Inoue T, Minami Y, Chung H, Hayaishi S, Ueda T, Tatsumi C, et al. Radiofrequency ablation for hepatocellular carcinoma: assistant techniques for difficult cases. Oncology. 2010 Jul;78 Suppl 1:94-101. 262. Song I, Rhim H, Lim HK, Kim YS, Choi D. Percutaneous radiofrequency ablation of hepatocellular carcinoma abutting the diaphragm and gastrointestinal tracts with the use of artificial ascites: safety and technical efficacy in 143 patients. Eur Radiol. 2009 Nov;19(11):2630-40. 263. Rosendal T. Concluding Studies on the Conducting Properties of Human Skin to Alternating Current. Acta Physiologica Scandinavica. 1945;9:39-49. 264. Rosendal T. Studies on the Conducting Properties of the Human Skin to Direct Current. Acta Physiologica Scandinavica. 1943;5:130-51. 265. Corcuff P, Leveque JL. Corneocyte changes after acute UV irradiation and chronic solar exposure. Photodermatol. 1988 Jun;5(3):110-5. 266. Rajadhyaksha M, Gonzalez S, Zavislan JM, Anderson RR, Webb RH. In vivo confocal scanning laser microscopy of human skin II: advances in instrumentation and comparison with histology. J Invest Dermatol. 1999 Sep;113(3):293-303. 267. Mahajan Y. Does combing the scalp reduce scalp electrode impedances? Journal of Neuroscience Methods 2010;188:287-9. 268. Takayasu K, Choi BI, Wang CK, Ikai I, Okita K, Chen RC, et al. First international symposium of current issues for nationwide survey of primary liver cancer in Korea, Taiwan and Japan. Japanese Journal of Clinical Oncology. [Conference Paper]. 2007 Mar;37 (3):233-40. 269. Laeseke PF, Sampson LA, Frey TM, Mukherjee R, Winter TC, 3rd, Lee FT, Jr., et al. Multiple-electrode radiofrequency ablation: comparison with a conventional cluster electrode in an in vivo porcine kidney model. J Vasc Interv Radiol. 2007 Aug;18(8):1005-10. 270. Laeseke PF, Sampson LA, Haemmerich D, Brace CL, Fine JP, Frey TM, et al. Multiple-electrode radiofrequency ablation creates confluent areas of necrosis: in vivo porcine liver results. Radiology. 2006 Oct;241(1):116-24. 271. Lee JM, Han JK, Kim SH, Sohn KL, Lee KH, Ah SK, et al. A comparative experimental study of the in-vitro efficiency of hypertonic saline-enhanced hepatic bipolar and monopolar radiofrequency ablation. Korean J Radiol. 2003 Jul-Sep;4(3):163-9.

175

272. Hansler J, Neureiter D, Wasserburger M, Janka R, Bernatik T, Schneider T, et al. Percutaneous US-guided radiofrequency ablation with perfused needle applicators: improved survival with the VX2 tumor model in rabbits. Radiology. 2004 Jan;230(1):169-74. 273. Leveillee RJ, Hoey MF. Radiofrequency interstitial tissue ablation: wet electrode. J Endourol. 2003 Oct;17(8):563-77. 274. Goldberg SN, Ahmed M, Gazelle GS, Kruskal JB, Huertas JC, Halpern EF, et al. Radio-frequency thermal ablation with NaCl solution injection: effect of electrical conductivity on tissue heating and coagulation-phantom and porcine liver study. Radiology. 2001 Apr;219(1):157-65. 275. Ahmed M, Lobo SM, Weinstein J, Kruskal JB, Gazelle GS, Halpern EF, et al. Improved coagulation with saline solution pretreatment during radiofrequency tumor ablation in a canine model. J Vasc Interv Radiol. 2002 Jul;13(7):717-24. 276. Fernandes ML, Lin CC, Lin CJ, Chen WT, Lin SM. Prospective Study of a 'Popping' Sound during Percutaneous Radiofrequency Ablation for Hepatocellular Carcinoma. Journal of Vascular and Interventional Radiology. 2010 February;21 (2):237-44. 277. Seiler J, Roberts-Thomson KC, Raymond JM, Vest J, Delacretaz E, Stevenson WG. Steam pops during irrigated radiofrequency ablation: feasibility of impedance monitoring for prevention. Heart Rhythm. 2008 Oct;5(10):1411-6. 278. Hiraoka A, Michitaka K, Horiike N, Hidaka S, Uehara T, Ichikawa S, et al. Radiofrequency ablation therapy for hepatocellular carcinoma in elderly patients. J Gastroenterol Hepatol. 2010 Feb;25(2):403-7. 279. Peng ZW, Chen MS, Liang HH, Gao HJ, Zhang YJ, Li JQ, et al. A case-control study comparing percutaneous radiofrequency ablation alone or combined with transcatheter arterial chemoembolization for hepatocellular carcinoma. European Journal of Surgical Oncology. [Comparative Study

Research Support, Non-U.S. Gov't]. 2010 Mar;36(3):257-63.

Documents

Improving the Safety and Efficacy of Bimodal Electric Tissue Ablation · 2012. 11. 7. · size of ablations in RP-BEA was significantly smaller compared to RFA and BETA (9.1mm vs