Jason Arya Varzaly MBBS Centre for Heart Rhythm Disorders, The … · 2018-08-30 · Centre for...
Transcript of Jason Arya Varzaly MBBS Centre for Heart Rhythm Disorders, The … · 2018-08-30 · Centre for...
1
The Surgical Management of Atrial Fibrillation
Jason Arya Varzaly
MBBS
Centre for Heart Rhythm Disorders, The School of Medicine, University of Adelaide
April, 2017
I
Table of Contents
Chapter 1 Introduction
1.1. Burden of Atrial Fibrillation
1.1.1. Risk Factors of AF
1.1.2. Consequences of AF
1.1.3. Economic Burden of AF
1.2. Mechanisms of Atrial Fibrillation
1.2.1. Multiple Wavelets
1.2.2. Focal Triggers
1.2.3. Anatomical Structures as Sources of Foci
1.2.4. Cardiac Autonomic Nervous System
1.2.5. Endocardial to Epicardial Connections
1.2.6. Rotors
1.3. Management options available
1.3.1. Risk Factor Modification
1.3.2. Anticoagulation
1.3.3. Rate Control
1.3.4. Rhythm Control
1.3.4.1. Catheter Ablation
1.4. Role of surgery
1.5. Available technologies
1.6. Previous Meta-Analyses
1
1
2
3
3
3
4
5
6
6
7
8
8
9
10
11
11
13
16
19
Meta-Analysis and Literature Review
Chapter 2 Surgical Ablation of Atrial Fibrillation Concomitant to
Mitral Valvular Repair or Replacement – Analysis of
Valvular Disease and Intervention, Ablation Method, Safety
and Success in Sinus Rhythm Maintenance
2.1. Introduction
2.2. Methods
2.2.1. Definitions
2.2.2. Statistical Analysis
2.3. Results
2.3.1. Study & Patient Characteristics
21
22
23
23
24
24
II
2.3.2. Mitral and Concurrent Valvulopathy
2.3.3. Surgical Ablation Procedures
2.3.3.1. Approach and Technique
2.3.3.2. Lesion Set
2.3.3.3. Operative Parameters
2.3.4. Post-Procedure Follow-up
2.3.5. Recurrent Arrhythmia
2.3.6. Antiarrhythmic Drug Use
2.3.7. Outcomes
2.3.8. Procedural Complications
2.3.9. Subgroup Analysis
2.3.9.1. Lesion Set
2.3.9.2. Energy Source
2.3.9.3. LAA Exclusion
2.3.9.4. Randomised Controlled Data
2.4. Discussion
2.4.1. Valvular Disease, Surgery and AF
2.4.2. Surgical Ablation Technique
2.4.3. Complications
2.4.4. Clinical Implications
2.4.5. Study Limitations
2.5. Conclusions
2.6. Figure and Tables
25
25
25
26
28
28
28
29
29
30
30
30
31
32
32
32
32
34
34
35
35
36
37
Chapter 3 Surgical Ablation of Atrial Fibrillation with Concurrent
Cardiac Surgery – Assessment of the Safety and Efficacy of
Ablation procedures during any other Cardiac Surgical
Intervention
3.1. Introduction
3.2. Methods
3.2.1. Definitions
3.2.2. Statistical Analysis
3.3. Results
3.3.1. Study & Patient Characteristics
47
48
50
50
51
51
III
3.3.2. Concurrent Cardiac Surgical Procedures
3.3.3. Surgical Ablation Procedures
3.3.3.1. Approach and Technique
3.3.3.2. Lesion Set
3.3.3.3. Operative Parameters
3.3.4. Post Procedure Follow-up
3.3.5. Definitions of Recurrent Arrhythmia
3.3.6. Antiarrhythmic Drug Use
3.3.7. Outcomes
3.3.8. Procedural Complications
3.3.9. Sub-Group Analysis
3.3.9.1. Baseline Characteristics
3.3.9.2. Lesion Set Comparison
3.3.9.3. Energy Source Comparison
3.3.9.4. LAA Exclusion
3.3.9.5. Randomised Control Trial Data
3.4. Discussion
3.4.1. Concomitant Cardiac Surgery and AF
3.4.2. Comparison to Studies of Concomitant Mitral Valve
Procedures
3.4.3. Surgical Ablation Technique
3.4.4. Clinical Implications
3.4.5. Study Limitations
3.5. Conclusions
3.6. Figures and Tables
52
52
52
54
55
56
56
57
57
57
58
58
59
59
60
60
60
61
63
64
64
65
66
67
Chapter 4 Isolated Surgical Ablation for Atrial Fibrillation – The
Methodology, Safety and Success in Sinus Rhythm
Maintenance of Sole Cardiac Surgical Procedures
4.1. Introduction
4.2. Methods
4.2.1. Definition
4.2.2. Statistical Analysis
4.3. Results
80
81
82
82
83
IV
4.3.1. Study & Patient Characteristics
4.3.2. Surgical Ablation Procedures
4.3.2.1. Surgical Access
4.3.2.2. Energy Source
4.3.2.3. Lesion Set
4.3.2.4. LAA Exclusion
4.3.2.5. Procedural Endpoints
4.3.3. Post Procedure Follow-up
4.3.4. Outcomes
4.3.5. Sub-group Analysis
4.3.5.1. Lesion Set
4.3.5.2. Energy Source
4.3.5.3. LAA Exclusion
4.3.5.4. Procedural Complications
4.4. Discussion
4.4.1. Catheter vs Surgical Ablation for Isolated AF
4.4.2. The LAA Dilemma
4.4.3. Clinical Implications: Developing Superior Results
4.4.4. Study Limitations
4.5. Conclusions
4.6. Figures and Tables
83
84
84
84
85
86
86
87
88
89
89
89
90
90
91
92
94
94
95
96
97
Chapter 5 Hybrid Ablation (Surgical and Electrophysiological) of
Atrial Fibrillation – The Methodology, Safety and Success of
Sinus Rhythm Maintenance
5.1. Introduction
5.2. Methods
5.2.1. Definitions
5.2.2. Statistical Analysis
5.3. Results
5.3.1. Study & Patient Characteristics
5.3.2. Surgical Ablation Procedures
5.3.2.1. Surgical Approach
5.3.2.2. Energy Utilised
110
111
112
113
113
113
114
114
115
V
5.3.2.3. LAA Management and Additional
Approaches
5.3.2.4. Endpoints
5.3.3. Catheter-base Electrophysiological Procedures
5.3.3.1. Timing after Surgery
5.3.3.2. Endocardial Procedure Type
5.3.3.3. Procedural Endpoints
5.3.3.4. Procedural Time
5.3.4. Post-Procedural Follow-up
5.3.5. Outcomes
5.3.5.1. Univariate Predictors of Outcome
5.3.5.2. Sequential vs Staged
5.3.5.3. Block Analysis
5.3.5.4. Surgical Approach
5.3.5.5. Surgical Lesion Set
5.3.5.6. Energy Type
5.3.5.7. Device Type
5.3.5.8. Left Atrial Appendage
5.3.5.9. Ganglionated Plexi Ablation
5.3.5.10. AAD Management
5.3.6. Procedural Complications
5.3.6.1. Sub-Group Analysis of Complications
5.4. Discussion
5.4.1. Outcomes and Complications of Hybrid Ablation
5.4.2. Lessons from Early Experiences
5.4.3. Clinical Implications
5.4.4. Study Limitations
5.5. Conclusions
5.6. Figure and Tables
116
116
117
117
117
119
120
120
121
121
121
121
122
122
122
122
123
123
123
124
125
125
126
128
130
130
131
132
Chapter 6 The Contact Force Effect of Bipolar Radio-Frequency
Clamps on Lesion Formation in the Surgical Management of
Atrial Fibrillation
6.1. Introduction
143
VI
6.2. Methods
6.2.1. Pressure Assessment
6.2.1.1. Surgical Equipment
6.2.1.2. Tissue Selection
6.2.1.3. Clamp Pressure Assessment Methodology
6.2.1.4. Contact Force
6.2.1.5. Pressure Film and Analysis
6.2.2. Ablation Lesion Analysis
6.2.2.1. Experimental Preparation
6.2.2.2. Tissue Selection
6.2.2.3. Ablation and Evaluation of Transmurality
6.2.3. Statistical Analysis
6.3. Results
6.3.1. Pressure Analysis
6.3.2. Lesion Analysis
6.3.3. Total RF Time and Impedance
6.3.4. Lesion Depth
6.3.5. Fat Presence and Effect
6.3.6. Transmurality
6.4. Discussion
6.4.1. Major Findings
6.4.2. Importance of Contact Force
6.4.3. Number of Burns
6.4.4. The Interaction of Fat
6.4.5. Impedance as a Measure of Ablation Endpoint
6.4.6. Clinical Implications
6.5. Conclusion
6.6. Figures
144
144
144
145
145
145
146
146
146
146
147
148
149
149
150
151
152
153
153
154
154
154
155
155
156
157
158
159
Chapter 7 Conclusions and Discussion 170
Chapter 8 Future Directions 174
Chapter 9 References
175
VII
ABSTRACT
Atrial fibrillation (AF) is the most common atrial arrhythmia with an increasing prevalence
identified worldwide. Symptomatic episodes have resulted in increasing hospitalisations such
that this is now one of the dominant cardiovascular reasons for hospital admission in Australia.
Beyond this, AF predisposes to heart failure, stroke and increasing cognitive decline. While
medical therapy is available for the treatment of this condition the use of medication has not
always proven sufficient for its management leading to the development of catheter and
surgical ablation therapies. However, while surgical ablation procedures are widely utilised,
the true efficacy of these approaches has not been well understood due to the wide variety of
methodologies, concomitant procedures, technologies utilised and follow-up implemented.
This thesis evaluates the current state of surgical ablation for AF.
Chapter 1 provides an overview of the burden of AF on our community, the
mechanisms of arrhythmia, a summary of medical and catheter based therapies and an in-depth
analysis of how surgical therapies for atrial fibrillation were developed and the current
approaches to surgical ablation with the current technologies available for use. Chapters 2
through 5 examine the efficacy and safety of the surgical ablation of AF when performed
concurrent to mitral valve disease intervention, concomitant cardiac surgery in general, as a
standalone procedure and as part of an emerging technique of hybrid ablation (joint surgical
and cardiology electrophysiology) procedure. This evaluation has been performed through
systematic literature review and meta-analyses identifying that surgical ablations procedures
offer good medium term results in sinus rhythm maintenance (SRM) on and off antiarrhythmic
drugs(AAD) achieving 79.6% SRM concurrent to mitral valve intervention, 79.6% on AAD
and 65.3% off AAD concurrent to general cardiac surgical procedures, 85.3% on and 65.5%
off AAD as a standalone procedure, and 79.4% on AAD and 70.7% off AAD when performed
as a hybrid ablation procedure. The complication rates were low across all series, with the
VIII
lowest complication rates associated with the hybrid procedure. Chapter 6 then presents the
results of a series of experiments analysing the properties of surgical bipolar RF clamps and
how ablation lesions are formed, with the identification that the efficacy of these surgical
ablative tools is subject to the effects of contact force for the delivery of transmural lesions,
increased numbers of ablation increase the likelihood of a transmural lesion irrespective of the
device indicating transmurality, the presence of fat significantly attenuates the ability to deliver
of a transmural lesion and these effects are amplified with increased thickness of the tissue
being ablated. The observations of this thesis thus provide insight into the optimal conditions
under which the surgical ablation of AF may obtain the best results in sinus rhythm
maintenance.
IX
Declaration of Authorship
I certify that this work contains no material which has been accepted for the award of any other
degree or diploma in my name in any university or other tertiary institution 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. In addition, I certify that no part
of this work will, in the future, be used in a submission in my name for any other degree or
diploma in any university or other tertiary institution without the prior approval of the
University of Adelaide and where applicable, any partner institution responsible for the joint
award of this degree.
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.
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 Search and also through web
search engines, unless permission has been granted by the University to restrict access for a
period of time.
Jason Arya Varzaly
27-04-2017
X
ACKNOWLEDGEMENT
I would like to thank Professor Prashanthan Sanders, Mr Michael Worthington and Dr Dennis
Lau for their mentorship, encouragement, direction and friendship throughout my candidature
period. I have been grateful for the opportunity to work under the supervision of such esteemed
clinical scientists and I will carry forward the skills and knowledge I have acquired throughout
this journey onwards through all my future endeavours. I am truly appreciative for all your
support allowing me to explore the world of medical research and make this thesis possible.
My further thanks to the research group of the Centre of Heart Rhythm Disorders who
have been my colleagues and friends along this pathway. Foremost a special thanks to Mr
Darius Chapman whose friendship, teamwork, assistance and ingenuity through so many
experiments, and much hard work, has made so much possible throughout these exciting times
of science and discovery. I would also like to thank Drs Rajiv Mahajan, Anthony Brooks and
Adrian Elliott for their help throughout my candidature in analysis, review and statistics of my
work, even in the tightest of times frames.
I am also grateful for the perpetual support of the Cardiothoracic Surgery Department
of the Royal Adelaide Department for encouraging me to undertake this pathway of research
and supporting me along it over the years.
Further, I would like to thank the assistance of the team at the Preclinical, Imaging and
Research Laboratory of the South Australian Health and Medical Research Institute for
accommodating me throughout my experiments.
I would like to acknowledge the assistance of the Dr Jennie Louise and Ms Suzanne
Edwards of the Biostatistics Department of the University of Adelaide for their statistical
assistance throughout my candidature.
Most importantly I would like to thank my family. The most special of thanks to my
parents Afsaneh and Laird Varzaly whose limitless love, support, encouragement, guidance
and inspiration throughout my life has not only guided me through the toughest of times, but
has given me strength to forge forward and always have the belief to know that anything is
possible. To my sister, Jenifer Varzaly whose love and friendship have been constant in my
life as you have inspired me to work harder through your ongoing achievments, to my brother-
in-law Thomas Mulraney for the friendship and support he has offered for as long as I have
known him and to Amy Reid who continues to encourage and support me in all my life’s
pursuits. Through all your support this thesis was made possible.
XI
PRESENTATIONS TO LEARNED SOCIETIES
1) Awards and Grants
a) Nimmo Prize for Medical Research – Royal Adelaide Hospital – 2016
Awarded for “The Pressure Effect of Surgical Clamps on Ablation for Atrial
Fibrillation”
b) Young Achiever Award – Australia New Zealand Society of Cardio Thoracic
Surgeons (ANZSCTS) – 2015
Awarded for “The Pressure Effect of Bipolar RF Clamps on Ablation for Atrial
Fibrillation”
c) Mentorship and Travel Grant – American Heart Association – 2014
Awarded for “Hybrid approach for ablation of Atrial Fibrillation – a Meta-Analysis
and review of the literature”
Mentor: Prof Ralph Damiano Jr
2) Presentations:
a) The Pressure Effect of Bipolar RF Clamps on Ablation for Atrial Fibrillation
JA Varzaly, D Chapman, DH Lau, F Viana, J Edwards, RG Stuklis, P Sanders and M
Worthington
Presentation at the 2015 Australia New Zealand Society of Cardio Thoracic Surgeons
scientific meeting
b) The Pressure Effect of Bipolar Radio-Frequency Clamps for Treatment of Atrial
Fibrillation
JA Varzaly, D Chapman, DH Lau, F Viana, J Edwards, RG Stuklis, P Sanders and M
Worthington
XII
Presentation at the 2015 European Association of Cardio Thoracic Surgeons annual
meeting
c) Surgical Contact Force Effect in Creating RF Lesions in the Isolated and Hybrid
Ablation of Atrial Fibrillation
JA Varzaly, DH Lau, D Chapman, F Viana, J Edwards, RG Stuklis, M Worthington
and P Sanders
Presentation at Heart Rhythm Society 2016
d) The Efficacy of Joint Catheter and Surgical Ablation for Atrial Fibrillation – a
Review of the Literature and Meta-Analysis
JA Varzaly, DH Lau, D Chapman, F Viana, J Edwards, RG Stuklis, M Worthington
and P Sanders
Presentation at Asia Pacific Heart Rhythm Society 2015
e) Feasibility, Efficacy and Safety of isolated Surgical Ablation for Atrial
Fibrillation: A Systematic Review
JA Varzaly, DH Lau, D Chapman, R Mahajan, F Viana, J Edwards, RG Stuklis, M
Worthington and P Sanders
Presentation at Asia Pacific Heart Rhythm Society 2015
f) Surgical ablation of Isolated atrial fibrillation – a systematic review of the
literature
JA Varzaly, DH Lau, D Chapman, R Mahajan, F Viana, J Edwards, RG Stuklis, M
Worthington and P Sanders
Presentation at Heart Rhythm Society 2015
g) Hybrid approach for ablation of Atrial Fibrillation – a Meta-Analysis and review
of the literature
JA Varzaly, DH Lau, D Chapman, AG Brooks, J Edwards, RG Stuklis, M Worthington
XIII
and P Sanders
Presentation at American Heart Association 2014
h) Surgical ablation of “lone” atrial fibrillation – a systematic review of the
literature
JA Varzaly, A Brooks, J Edwards, RG Stuklis, P Sanders and M Worthington
Presentation at Australia New Zealand Society of Cardio Thoracic Surgeons 2013
i) Hybrid approach for ablation of Atrial Fibrillation – a systematic review of the
literature
JA Varzaly, AG Brooks, DH Lau, J Edwards, RG Stuklis, P Sanders and M
Worthington
Presentation at Australia New Zealand Society of Cardio Thoracic Surgeons 2013
1
Chapter 1
INTRODUCTION
1.1. THE BURDEN OF ATRIAL FIBRILLATION
Atrial Fibrillation (AF) is the most common atrial arrhythmia affecting 0.4% of the general
population with population based studies have estimated the lifetime risk of developing AF is
1 in 4, for both men and women, aged over 40 years of age.(1,2) Further long term analyses of
patients revealed that the incidence of AF has been increasing over time. An observational
study looking at a population over 50 years between the 10 year periods of 1958 to 1967 and
1998 to 2007 demonstrated that the age-adjusted prevalence of AF quadrupled from 20.4 to
96.2 cases per 1000 person-years in men and from 13.7 to 49.4 cases per 1000 person-years in
women.(3) In studies assessing community based trends in AF it has been estimated that by
2050 the number of people with AF will reach 12.1 million in the USA, however, when
corrected for the increasing incidence seen in the last 20 years this estimate rises to 15.9
million.(4) Indeed these estimates in the United States may be dwarfed by predictions in the
Asia-Pacific region, where accurate data still remains elusive.(5)
1.1.1. RISK FACTORS AND AF
The development of AF has been linked to structural heart disease with mitral valvopathy
identified as being associated with the development of AF in 30% to 40% of medically
managed patients in early studies.(6,7) Subsequent studies have confirmed the finding of a
high prevalence of AF in mitral valve disease, but have further identified that after the onset of
AF the incidence of cardiac morbidity and mortality increased.(8)
The growing burden of this disease, out of keeping from currently known risk factors,
has led the search for novel risk factors and new treatment targets for known risk factors.
Established risk factors for AF include increasing age, hypertension, congestive heart failure,
2
myocardial infarction, and valvular heart disease.(9-12) Increasingly recognised new risk
factors include obesity, sleep apnoea, physical inactivity, endurance sports, aortic stiffness and
a genetic predisposition.(13-17)
1.1.2. CONSEQUENCES OF AF
The significance of the increasing burden of AF in the community stems from the identification
that it is associated with risk of thromboembolic stroke and loss of atrial systole with resultant
negative impact on ventricular function.(18-20) Studies analysing the link of AF to
thromboembolic stroke have identified that AF increases the risk of stroke fivefold,
independent of heart failure and coronary artery disease, and this effect was amplified to
seventeen fold in patients with rheumatic heart disease.(19,20)
The association of AF and heart failure has been established with the presence of heart
failure increasing the risk for developing AF 4.5 to 6 fold,(9,10) and associations have been
made to suggest that AF itself can result in heart failure through atrial and ventricular mediated
tachycardia driving cardiomyopathy.(18,21)
From a broader perspective AF has been associated with decreased quality of life and
increased incidence dementia, memory impairment and cognitive impairment, even in the
absence of an overt stroke.(22-26) The cumulative effect of these consequences of AF is an
increased risk of mortality estimated to be increased by 1.5 to 1.9 fold beyond the general
population.(9,10,19-21,27,28)
3
1.1.3. ECONOMIC BURDEN OF AF
Further to the rising prevalence of AF it has significant monetary cost implications as the
number of hospitalisations related to AF increase in Australia(29) and the use catheter ablation
therapies for the control of this arrhythmia have been increasing exponentially.(30) Studies on
the economic burden of AF have reported increasing expenditures being required by health
systems to combat the increasing burden of this disease through general practitioner
consultations, referrals, drug prescriptions and hospital admissions, all of which have been
increasing over time.(29,31) In a report prepared for the National Stroke Foundation by
PricewaterhouseCoopers in June of 2010 it was estimated that the costs annual costs to the
Australian economy were at least $1.25 billion per annum in 2008-2009 owing to medical costs
of $874 million in addition to the long term care for those with this disability and the loss of
productive output to the community.(32)
1.2. MECHANISMS OF ATRIAL FIBRILLATION
The mechanisms of AF have drawn significant focus of study over many years with major
hypotheses about how it is initiated and triggered being documented.
1.2.1. Multiple Wavelet Hypothesis
The initial predominant theory of the mechanism of AF was based on multiple wavelets
of irregular electrical activity occurring simultaneously and propagating throughout the
atria in such wise that multiple wavelets would co-exist, divide around areas of
refractory tissues and result in further wavelet propagation.(33) This initial work based
on computer simulation was then confirmed by further experimental work by mapping
activation wavefronts in pacing induced AF in a canine model.(34) This model
suggested that a critical number of wavelets were required to co-exist for AF to develop,
and this condition was favoured by slowed conduction, shortened refractory periods
4
and increased atrial mass. This phenomenon was subsequently mapped and described
in humans by Cox et al. leading to the early development of surgical procedures for the
control of AF.(35) His landmark work in the field of the surgical management of AF
postulated that surgical lines of conduction block could be “Cut and Sewn” into the
atrial tissue itself to interrupt the propagating wavefront of these wavelets and thus
terminate AF.(36) This work formed the basis for the Maze procedure which
demonstrated the clinical observation that chronic AF could be treated in patients by
compartmentalising the atria into small enough regions that were then unable to sustain
macro re-entrant circuits for the propagation of AF.(36,37)
While the multiple wavelets theory suggests the process of AF is a random and
chaotic one, other studies suggest that AF could be maintained by single or multiple re-
entrant sources within the Left Atrium (LA) alone. In a canine model with increasing
concentrations of acetylcholine; multiple re-entrant circuits converted to a single,
relatively stable circuit that resulted in fibrillation.(38) Based on these findings it was
postulated that such sources of high frequency waves from within the LA may act as
drivers for AF, and this was then supported through further animal studies.(39,40)
Alternatively, it was postulated that these findings could be equally explained by the
multiple anatomic structures that form the LA resulting in greater wavelet fractionation
thus leading to higher frequency waves and AF.(41) Perhaps the most interesting
demonstration of the left atrial source of driving AF was seen in the surgical isolation
of the LA resulting in right atrial sinus rhythm and persisting left atrial AF.(42)
1.2.2. Focal Triggers
The importance of atrial ectopy acting as a trigger for the development of AF was a
significant finding, only eclipsed by the finding that ablation of these areas can result
5
in the termination of AF.(43-45) As this understanding was furthered multiple areas of
sites of atrial ectopy have been identified including the ligament of Marshall, the
coronary sinus, the crista terminalis and the superior vena cava.(43,45,46) However,
the area attracting the most interest for containing the vast majority of atrial ectopic
sites is the pulmonary veins, which had been previously been identified as structures
composed of cardiac tissue, and not simply blood vessels.(43,47)
1.2.3. Anatomical Structures as Sources of Foci
Both the right and left atria contain sites from which originate foci of automaticity
which in turn give rise to atrial tachycardia, and it has been recognised these sites are
in close relationship to anatomical structures such as the crista terminalis, the tricuspid
annulus, the interatrial septum, the coronary sinus, the ligament of Marshall, the
superior vena cava, the right or left atrial appendage, the mitral annulus and the
pulmonary veins.(47-54)
The pulmonary veins have become clinically apparent as one of the major sites of
not only atrial ectopy, but atrial tachycardia and AF as well.(45,50,53,55) This is due
in part to the sleeving of the atrial tissue into the pulmonary vein for ~1-3cm being
thickest most proximally to the LA itself.(56,57) Further, the cardiac myocytes that are
found along these sleeves of pulmonary vein tissue are predisposed to
arrhythmogenesis by discrete ion channel and action potential properties inherent
within them.(58,59) However, despite these properties, the pulmonary vein isolation
(PVI) has primarily proven effective only in paroxysmal AF, but not in persistent forms
of AF.(60-62) In studies utilising mapping technology coupled with high-frequency
amplitude analysis to localise the dominant sites of activation it has been shown that in
paroxysmal AF patients the pulmonary veins contain the highest frequency sites
6
allowing termination of AF in up to 87% of patients through ablation, but in persistent
AF additional sites outside the pulmonary veins were seen.(63,64)
1.2.4. Cardiac Autonomic Nervous System
The atria possess autonomic input from the central nervous system through pre-
ganglionic nerve fibres as well as through the intrinsic cardiac autonomic nervous
system.(65) These intrinsic cardiac autonomic nervous system clusters, termed
ganlionated plexi(GP) are located in epicardial fat pads as well as the ligament of
Marshall and have been linked with the initiation of AF, thus making them a target for
therapeutic intervention.(66) It has been identified that stimulation of these GP
produces repetitive, short bursts of activity in the spatially associated pulmonary veins
initiating AF through parasympathetic and sympathetic activity.(67,68) Thus PVI
ablation may also disrupt the GP nerves reducing the firing of focal sites within the
pulmonary veins.(69)
It has also been observed that GP exist near areas of complex fractionated atrial
electrograms(CFAE) have been identified.(70) Though some studies have indicated
that a stepwise fashion in ablation with the incorporation of CFAE ablation may
decrease the inducibility of AF, other mapping studies have investigated areas of CFAE
for spatial and temporal stability, making it a dynamic target for AF elimination.(71-
74)
1.2.5. Endocardial to Epicardial Connections
Deeper analyses have been performed to examine the substrate of AF to better
understand the sub-optimal outcomes achieved in the more persistent forms of this
arrhythmia. Such analyses have focused on developing a further understanding of the
7
sustaining mechanism of AF on a substrate level. One such body of work is focusing
on the role of both the dissociation and connections between the endocardial and
epicardial surfaces of the atria in contributing to the complexity of the substrate of
AF.(75,76) While initially mapped in animal studies, subsequent human studies have
identified that the breakthrough of fibrillatory waves from the epicardial surface to the
enocardial surface of the atria which may account for the focal activation waves
observed in persistent forms of AF.(77)
1.2.6. Rotors
Animal studies by the Jalife group have consistently visualised the presence of
rotors.(78-82) The phenomenon of electrical rotors has also been described as a
localised, unstable source of AF in mapping studies and has also formed targets for
ablation therapies with results suggestive of offering improved outcomes.(83,84)
Recently there have been several studies that have evaluated the possibility of such a
phenomenon in humans.(85,86) Sanders et al undertook a retrospective study
evaluating the relationship between sites of high dominant frequency and the impact of
ablation.(63) Indeed they were able to demonstrate that during ablation at sites
harbouring high dominant frequency sites (presumed to represent the rotor) there was
slowing and termination of AF. This was further evaluated in a prospective study of
ablation guided by mapping high dominant frequency sites by Atienza and colleagues
but demonstrated only modest results.(87) Narayan and colleagues used basket catheter
mapping to performed phase analysis and demonstrated the presence of rotors in most
individuals with ablation at these sites resulting in significant outcomes with
ablation.(88) In contrast, Haissaguerre et al. used body surface mapping and have
demonstrated the presence of short lived rotors.(89) Similarly, the Kalman group have
8
shown that with epicardial mapping of the atria there was rotors observed in a small
number of patients.(90) Although the existence of such a phenomenon remains debated,
their possible role in the maintenance of AF is intriguing and may have implications on
future therapeutic ventures.
The development of AF requires areas not only the aforementioned areas of focal activity to
act as triggers, but also changes in the atrial tissue to act as substrate for the perpetuation of the
arrhythmia.(91) Atrial re-entry is an important concept in the development of AF as it sets-up
a potential substrate AF, with the atrial wavelength key to the number of re-entry circuits that
a given mass of atrial tissue can sustain.(92) Through this finding it has been identified that
increased LA size reduces the ability for ablation strategies to return the patient to sinus
rhythm.(93) Further, the nature of the anatomic structures within the atria, such as the crista
terminalis, triangle of Koch, coronary sinus and pulmonary veins, have also been implicated in
the development of re-entry.(94-97) There is also an increasing body of evidence focusing on
how the development of atrial fibrosis results in atrial electro-structural remodelling
supporting the potentiation of AF on a substrate level.(98,99)
1.3. MANAGEMENT OF ATRIAL FIBRILLATION
The management of AF has been summarised in the 2016 European Society of Cardiology
Guidelines for the Management of AF developed in collaboration with the European
Association of Cardio-Thoracic Surgery. There are now recognised 4 pillars to the management
of AF comprising; 1) Risk Factor Modification, 2) Anticoagulation, 3) Rate Control and 4)
Rhythm Control.(100)
1.3.1. Risk Factor Modification
The impact of risk factor modification in the management of AF has been a new
development of significant impact as it was identified through such intervention alone
9
the burden and severity of AF can be reduced.(101) Initial work in the animal model
established that risk factors, such as obesity, significantly impact the atria to establish
substrate for its development.(102) Most importantly, the landmark observations were
then made that lifestyle adjustment and aggressive risk factors modification in patients
suffering from AF are not only able to improve outcomes directly, but are also able to
improve outcomes of other rhythm control strategies, such as catheter ablation.(103-
106) The ability of such integrated forms of care for patients with AF has been
demonstrated to have such an effect on the quality of life of patients with AF that it
now is integral to the successful management of the patient with AF.(107,108)
1.3.2. Anticoagulation
Presently, scoring systems are being utilised to determine the type of anticoagulation
most suited to a certain patient as stratified by stroke risk such as CHA2DS2-VASc
which is guiding clinical practice to stratify patients that do not need antiplatelet or
anticoagulant treatment, those that should be considered and those that require
it.(100,109,110) The need for anticoagulation for prevention of thromboembolic
strokes in the setting of AF has been established with the mainstay therapy being the
use of Vitamin K antagonists (warfarin) which has seen a reduction in the risk of stroke
in patients with AF by two-thirds over aspirin alone.(111) However, the use of warfarin
has come at the cost of increased risk of major bleeding events.(112) This has led to the
interest in alternative drugs for use in stroke prevention in patients with AF. These
alternative drugs have been termed novel oral anti-coagulant (NOAC) medications and
are being examined closer for their therapeutic use with the potential for lower rates of
stroke or systemic thromboembolism while reducing the risk of major bleeding
events.(113-116) At present it has been seen that Warfarin or NOACs offer a superior
10
result to aspirin in stroke prevention while not having an increased risk of bleeding to
aspirin.(117,118) Though of note only warfarin has evidence to suggest the safety of
use in patients with rheumatic mitral valve disease or those with mechanical heart
valves.(119) The alternative method to pharmacological management is the mechanical
closure of the left atrial appendage with one such device (Watchman) having been
compared to warfarin therapy and having shown non-inferiority for stroke
prevention.(120,121) The surgical exclusion of the LAA concomitant to cardiac surgery
is an alternative method which has been employed, but only limited data is available at
present with a large randomised trial underway.(122-124)
1.3.3. Rate Control
Several well-constructed studies have looked to assess the differences in outcome
obtained through rate or rhythm control.(125-131) Overall there was no difference in
outcome when comparing the strategies of rate and rhythm control, the rhythm control
strategy was associated with significant adverse events owing to the pharmacological
profiles of the AAD, but analysis of the patients who actually maintained sinus rhythm
shows a survival advantage.(132,133) Consideration of this suggests that if there were
a way to maintain sinus rhythm without the adverse events associated with AAD there
may be a benefit. Of further note, when similar trials have been conducted on patients
who developed AF post cardiac surgery the same relationship of no superior difference
in rhythm to rate control was upheld.(134) Importantly of note in all these studies the
need for anticoagulation was not ameliorated.(127,134)
11
1.3.5. Rhythm Control
The use of AAD compared to placebo has doubled the rate of SRM.(135,136) And
while no significant improvement in mortality or cardiovascular complications has been
demonstrated through the use of AAD, it still provides a pathway for the reduction of
the symptoms of AF.(137) In consideration of this, catheter ablation therapy has been
shown to be superior to AAD in controlling AF, the symptoms of AF and improving
left ventricular function.(138-140)
1.3.5.1. Catheter ablation
Whilst several studies have shown no difference in the outcomes of rate or rhythm
control,(125,134) in practice, several patients remain highly symptomatic and have
required more invasive approaches to rhythm control. Catheter ablation is an
increasingly utilised effectively in highly symptomatic patients. Indeed, several
randomised controlled studies have consistently demonstrated that ablation therapy is
superior to medical therapy alone.(141,142)
The goal of catheter ablation is the treatment of symptomatic AF through the
creation of cardiac lesions which block the propagation of arrhythmic wavefronts of
conduction and alter the atrial substrate to eliminate the triggers of AF.(143,144)
Multiple energy sources are available for the creation such lesions with the most
common technologies being radio-frequency (RF) and cryoablation.(145,146) The use
of RF energy effects the creation of a lesion through the delivery of alternating
electrical current which is dissipated as heat in the cardiac myocytes, which in turn
causes coagulative necrosis that evolves into non-conducting myocardial scar.(147)
Further iterations of RF technology have recognised that lesion size and depth is
proportional to the contact and force of the catheter delivering the
12
technology.(148,149) This in turn has given rise to contact force sensing catheters with
results suggestive of improved efficacy at sinus rhythm maintenance at follow-up of
ablation procedures.(150,151) Cryothermal lesions are delivered through the creation
of an endothermic reaction at the catheter interface which in turn cools the cells,
creating intracellular ice crystals that disrupt the cellular membrane and eventually
result in scar formation.(152,153) Other devices such as ultrasound and laser based
ablation systems are continuing to be investigated with various results.(154,155)
With the identification of the site of triggers for AF commonly being within the
pulmonary veins the early catheter ablation interventions focused on elimination of
these focal triggers themselves.(43,44) However, this was limited not only by initial
difficulties with reproducibly identifying the trigger sites, but also subsequent
complications with pulmonary vein stenosis, thus the target for catheter ablation
evolved to segmental pulmonary vein isolation and then on to wider antral ablation
techniques.(156-158) PVI has been proven an effective treatment in paroxysmal AF,
but with high rates of recurrence seen in persistent forms of AF further ablation
approaches have been investigated to improve outcomes.(60,61) Additional
approaches have included the use of further linear ablation lines of block akin to the
lesions of the Maze procedure,(159,160) the identification and elimination of non-
pulmonary vein activation sites,(161,162) the mapping and elimination of areas of
CFAE(163,164) and the ablation of GP targets.(69,165) These additional interventions
are being investigated for their role in the treatment of persistent AF patients with
reports suggesting improved rates of sinus rhythm maintenance compared to PVI
alone.(166-169) However, in a subsequent large, multi-centre randomised control trial
assessing the addition of the linear ablation and CFAE ablation to PVI in persistent AF
patients has found no improvement in the rate of SR maintenance.(170) With the
13
progressive advances in catheter ablation technique and technology there remain
several inherent limitations to the current catheter approach which include the accurate
evaluation of transmurality, the creation of contiguous lesions, the potential for
collateral injury, and the accepted need for multiple procedures.(151,171-174)
Regardless, the long-term results in SR maintenance of catheter ablation has been an
area of great study with meta-analyses performed in the area. An analysis performed
on the results of the randomised control trials assessing catheter ablation efficacy
identified a 76.8% success rate of freedom of recurrence of tachyarrhythmia.(175)
Further, a meta-analysis not confined to randomised control trials reported a single
procedure success of 53% and multiple procedure success approaching 80% at ≥3
years of follow up.(176)
1.4. ROLE OF SURGERY
The first surgically described procedures for the surgical management of AF date back to 1980
in which Williams et al described the Left Atrial Isolation Procedure in which patients with AF
had the LA surgical electrically isolated from the rest of the heart.(177) While this left the rest
of the heart in sinus rhythm, as it was driven by the SA node, it did not terminate the AF of the
LA, thus the risk of thromboembolism was not ameliorated.(36) Following this work the
Corridor Procedure was then described in 1985 by Guiraudon et al in which a corridor of atrial
tissue was isolated from the remaining atrial tissue to allow the SA node to drive ventricular
contraction via the AV node at regular rate without the chaotic interference of AF.(178) While
this was successful to drive the ventricles, it once more left the atria fibrillating, thus retaining
the thromboembolic risk, and further resulting in haemodynamic compromise from the
asynchronous activity of the atria from the ventricles.(36) The landmark study of Cox et al in
1991 in which he first described his Maze technique for the elimination of AF through the
14
development of lines of conduction block to disrupt the propagation of macro re-entrant circuits
within the atria formed the foundation for the ideal surgical procedure in the management of
AF.(35,36,179,180) The first iteration of this procedure, now known as the Maze I, was based
on the principle of the interruption of all the potential pathways of macro re-entrant circuits
that had been identified through mapping studies, while still maintain the ability of a wave
front of conduction to propagate from the SA node to the AV node.(35,36) While the procedure
was reported to be successful in regards to elimination of AF, the follow-up of the first 32
patients undergoing this procedure identified a frequent association of an attenuated response
of the SA node to exercise as well as LA dysfunction.(181) This finding was thought likely to
be related to an incision directly anterior the location of the SA node and the Maze II procedure
was then formulated to eliminate this incision. However, the alteration of lesion sets proposed
and investigated found the newer Maze II lesion set had further increased technical difficulty,
requiring the SVC to be completely transected in order to allow adequate access for completion
of the lesion set.(181,182) The Maze III lesion set was subsequently described in 1995 in which
the left atrial incision was shifted posteriorly and facilitated reduced technical difficulty and
functional improvements in the procedure.(181,182) The Maze III procedure then formed the
gold standard for the surgical treatment of AF with excellent results reported at 5.4 years
follow-up of 83% SRM off AAD and 97% SRM with AAD. However, despite these results
there was not wide uptake of the procedure and it was postulated this was due to a combination
of perceived technical difficulty, requirement for cardiopulmonary bypass, invasiveness and
associated significant morbidity, particularly that of pacemaker requirement and left atrial
dysfunction.(183-186) In order to maintain the clinical results being obtained in SRM the
adoption of alternative energy course for the creation of lines of conduction block was
investigated. The Maze IV procedure was described in 2004 by Gaynor et al in which a bipolar
RF surgical clamp was utilised for the creation of ablation lesions to replace the traditional cut
15
and sewn lines.(187) This procedure has had a wider uptake with results as high as achieving
84% SRM off AAD and 90% SRM with AAD at 2 years follow-up.(188,189) Though the Maze
procedures were considered invasive in nature with the use a median sternotomy and
cardiopulmonary bypass, they were able to be conducted not only as a standalone procedure,
but also concurrent to the treatment of mitral valve repair and other cardiac procedures to which
AF commonly coexists.(190-193) Further studies on surgical populations of patients have also
demonstrated the strong link between mitral valvular disease and AF with 40% to 60% of
patients presenting for mitral valve surgery suffering from this arrhythmia.(9,194,195) Thus
the ability to treat this arrhythmia at the time of mitral valve surgery identifies a population of
significant need. Importantly, in the setting of valvular heart disease, AF confers a much higher
risk of stroke than patients with valvular heart disease who do not have AF.(196)Following the
establishment of the Maze IV procedure the form of evolution has been down minimising the
invasiveness of the procedure as a median sternotomy had been required. In 2004 Ad et al
described the ability to perform a Maze III lesion set from an anterolateral mini-thoracotomy
instead of the standard median sternotomy with subsequent confirmation of maintaining good
results at 36 months of 92% SRM and 80% SRM off AAD.(197,198) Continuing to further
minimise the invasiveness of surgical procedures for AF a range of thoracoscopic “key hole”
procedures, with or without robotic assistance began to be described.(199-203) The most recent
of such procedures report the reproduction of the Maze lesion set thoracoscopically through
the creation of 5-boxes of ablation, without the need for cardiopulmonary bypass, achieving
SRM off AAD of 95.6% at 13 months.(204-206) The long term results of this approach are still
pending. Finally, a new hybrid procedure is evolving in which both a cardiac
electrophysiologist and cardiac surgeon undertake a joint procedure for treatment of AF.(207)
The early results of this procedure are encouraging as a potential procedure to combine the
strengths of surgical ablation, through the use direct epicardial access allowing direct
16
visualisation of the heart for ablation, and the strengths of catheter ablation, with endocardial
access allowing direct mapping of triggers, conduction and block.(208,209) Most recently, the
Society of Thoracic Surgeons has released an update of their clinical practice guidelines
focused on the surgical ablation of AF providing the most updated recommendations in which
they assigned Class I recommendations to the ablation of AF concurrent to mitral valve
operations without added risk (Level A), concurrent to aortic valve, CABG and a combination
(Level B), however, they only assigned Class II recommendations to stand alone ablation
(Level B) and left atrial appendage management at the time of surgery for AF or other cardiac
surgery to ameliorate thromboembolic risk (Level C).(210)
1.5. AVAILABLE TECHNOLOGIES
Since the evolution of the Maze procedure beyond the cut and sewn lesions to alternative
energy sources an increasing interest in available technologies has arisen. The current
technologies implemented in the surgical field for the creation of such lesions are RF and
Cryothermy, the tissue injury mechanisms of which of have been described earlier.
RF is available in unipolar and bipolar forms. Bipolar RF clamps are available from
Atricure (West Chester, OH, USA) in the form of 2 curved, parallel jaws that clamp together
in a sliding fashion.(211) They offer 3 such varieties at present named Synergy OLL2/OSL2
at 53mm and 65mm jaw length, EML2/EMR2 at 60mm jaw length each facing alternate curves
of left and right, and EMT1 at 64mm jaw length with an adjustable jaw angle; all of which are
non-irrigated.(212-214) Each jaw of the clamp houses a pair of electrodes running the length
of the jaw through which RF energy is delivered to create an RF lesion.(212) All Atricure
bipolar clamp devices are non-irrigated, the jaws are rigid and are suitable for epicardial use or
a combination of endo/epicardial use with a jaw on each surface of the atrial tissue.(212-214)
Medtronic (Minneapolis, MN, USA) offers the alternative bipolar RF clamps available on the
17
market with Cardioblate LP/BP2 and the Gemini, both of which clamps closed in a scissor type
fashion from a proximal hinge.(215-218) These devices house an RF electrode on each jaw
underneath an irrigated pad and these devices are also suitable for epicardial use or a
combination of endo/epicardial use with a jaw on each surface of the atrial tissue. The LP/BP2
are designed for open access procedures with adjustable necks and adjustable articulation of
the jaw clamp with the neck to allow for increased flexibility of angle.(216,217) The most
recent of these devices allows for the jaws themselves to be malleable and shaped by the
surgeon to deliver the best fit for the patient specific anatomy. The Gemini system offers a
thoracoscopic alternative of this device with a longer neck for improved access.(218) Both
Atricure and Medtronic have their own RF generators for use with their products, with visual
displays indicating energy delivery and ablation status.(219,220) The Medtronic device is
specifically reported to supply “Audible tones and touch screen messages notify user of
ablation status, transmurality, and other critical information”.(220)
A series of ablation pens are available through both Atricure (West Chester, OH, USA)
and Medtronic (Minneapolis, MN, USA) and are suitable for epicardial or endocardial use. The
Atricure devices are primarily bipolar in nature with 2 electrodes on the same surface of the
pen (in a rail fashion) generating RF between these 2 electrodes.(211) Three such pens are
available in 3 forms; Coolrail Linear at 30mm electrode length, Isolator Linear at 20mm
electrode length and Isolator Transpolar Pen with short tip electrodes.(221-223) The Medtronic
devices are 2 unipolar type pens; the Cardioblate irrigated pen and the Cardioblate MAPS pen
device.(224,225) The Atricure Isolator Linear, Transpolar Pen and the Medtronic MAPS all
have the ability to stimulate, pace and sense, in addition to ablation.(222,223,225)
The most recent RF device companies have recently been acquired by Atricure (West
Chester, OH, USA) and are now featured within their product line; the COBRA devices and
the EPi-Sense Coagulation Device, both of which are designed for epicardial use, utilise suction
18
technology to maintain tissue contact and are internally cooled as opposed to externally
irrigated.(211) The COBRA range of devices (previously produced by Estech, San Ramon,
CA, USA) are designed for thoracoscopic access utilising a series of longitudinal electrodes
housed in a soft casing designed to be placed epicardially on the atrial surface of the heart. The
devices come in 50mm length and 150mm length, with the most recent iterations of these
devices have upgraded from purely unipolar energy delivery to a combination of poles able to
be activate in order to achieve a transmural lesion.(226) The EPi-Sense devices are a series of
unipolar devices designed for minimally invasive access via laparoscopy to gain access to the
pericardium through the diaphragm. These devices house a series of 4 electrodes across 3 cm
array of the distal end of the device.(227)
Surgical Cryothermy devices are also produced by Atricure (West Chester, OH, USA)
and Medtronic (Minneapolis, MN, USA). The Atricure Cryo devices are 3 in number (CRYO2,
CRYO3 and CRYOF) all in the form of a single, malleable shaft probes which utilise Nitrous
Oxide as their cooling agent.(228) They represent a series of probes that have been developed
progressively and require less force to shape with the newest model of the CRYOF.(228) The
Medtronic Cryo devices include 2 single malleable shaft devices (7cm and 10cm lengths
available) and 1 clamp device, all of which utilise Argon gas as their cooling
agent.(215,229,230)
Other forms of ablation devices have been trialled with Micro-Wave technology,(231-
234) Laser ablation(235-237) and High-intensity focused ultrasound.(238-240) However, most
of these devices have not been able to create consistently reproducible lesions and are not
currently commercially on offer for clinical use.(153,241)
It is worth considering some of the already understood limitations of RF and Cryo
energy in creating lesions of conduction block. Firstly, all aforementioned technologies have
the ability to cause collateral damage; reports of damage to the oesophagus in animal
19
studies,(242,243) with unipolar RF reports of injury documented in human trials.(244) Even
though Cryo technology is reported to be preservative of the fibrous skeleton of the heart,
experimental studies have demonstrated variable amounts of hyperplasia of coronary arteries
exposed to this energy source.(153,245-248) Secondly, no device has been documented to
demonstrate 100% transmurality; Cryo lesions have varied in efficacy of animal studies from
25% to 100% transmurality of tested lesions with results negatively impacted by beating heart
procedures caused by blood acting as a heat sink on the endocardial surface;(242,249) Unipolar
RF lesions have also seen low reports of transmural lesions ranging from 7% to 92% efficacy
owing it’s wide variability to the effect of epicardial fat, tissue thickness, use on a beating heart
model and the development of char on the electrode;(250-252) Bipolar RF clamps have been
shown to be the most reliable and quickest at achieving transmurality, even in the beating heart
model, but studies cannot confirm consistent 100% achievement of transmurality.(153,253-
255) Third, ablation times reported as necessary for achievement of conduction block have a
wide variance from as quick as 5-10 seconds with a bipolar RF clamp to 3 minutes of
application for Cryo.(153,253)
1.6. PREVIOUS META-ANALYSES
Multiple meta-analyses are currently published on the topic of surgical ablation of atrial
fibrillation; however, a large proportion of the published studies included within these meta-
analyses were of small sample size and not well controlled series.(256-261) This has led to
clouding of the data in this field
To rationalise the current state of the field of the surgical ablation of atrial fibrillation
this thesis will undertake a high-quality analysis of the literature. To ensure the articles selected
are representative by only including studies containing at least 100 patients. Further, through
assessment of previous published analyses within this area it has become apparent that distinct
20
groups of surgical patients and procedures are contained within the broad context of surgical
ablation. The first group to be addressed in Chapter 2 are patients solely undergoing mitral
valve surgery concurrent to surgical ablation, as these patients represent a specific cohort of
surgical patients with cardiac valvular disease that affects the atrial substrate. In Chapter 3 we
analyse studies in which patients presenting for cardiac surgery, with concurrent AF, undergoes
ablation – this is a unique cohort that is separate from the isolated pathology of mitral disease
as it includes patients with AF potentially unrelated to any cardiac valvular pathology. Further,
this group is of particular interest as many patients present for cardiac surgery without mitral
valve pathology, but with AF nonetheless, and thus the decision of how to best manage such
general cardiac surgical patients is a separate and pertinent issue. In Chapter 4 we address
patients with isolated AF in the absence of any other cardiac surgical pathology as not only is
the substrate for AF potentially different, but these patients would not undergo surgery for any
other reason and thus the threshold for undertaking ablation may be different from concurrent
cardiac surgical ablation procedures. The final meta-analysis of Chapter 5 is an assessment of
hybrid ablation for AF in which both a cardiac surgical procedure and cardiac
electrophysiological procedure are performed on a single patient to deliver the best results.
21
Chapter 2
Surgical ablation of atrial fibrillation concomitant to mitral valvular repair or
replacement – analysis of valvular disease and intervention, ablation method, safety and
success in sinus rhythm maintenance
2.1. INTRODUCTION
Atrial Fibrillation (AF) is the most common atrial arrhythmia with a 1 in 4 lifetime risk of
developing AF for both men and women aged over 40 years.(1,2) Significant risk factors for
development of AF include increasing age, hypertension, congestive heart failure, myocardial
infarction and valvular heart disease.(9,10) In particular, mitral valvular disease has a strong
association with AF and it has been identified in 40% to 60% of patients presenting for mitral
valve surgery.(9,194,195) Importantly, in the setting of valvular heart disease, AF confers a
much higher risk of stroke.(196)
The Cox-Maze procedure has defined the gold standard for the surgical management of
AF in which lines of conduction block were cut and sewn into the atria achieving sinus rhythm
maintenance (SRM) at 5.4 years follow-up, of 83% off anti-arrhythmic drugs (AAD) and 97%
with AAD.(186) The Cox-Maze IV procedure is the latest iteration of this in which lines of
conduction block were created using RF, achieving 84% SRM off AAD and 90% SRM with
AAD at 2 years follow-up.(188,189) This procedure is commonly undertaken in conjunction
with other cardiac surgical procedures to treat the AF concurrently; however, a relative paucity
of level 1 evidence exists and as such high-level recommendations for when to employ this
strategy has been lacking in the HRS/EHRA/ECAS Expert Consensus Statement on Catheter
and Surgical Ablation for AF.(262)
Here, we undertook a systematic review of the literature and meta-analysis to evaluate
the efficacy and safety of surgical ablation, concurrent to mitral valve repair or replacement, at
22
maintaining sinus rhythm in patients with AF. We hypothesised that AF ablation concurrent to
mitral valve surgery is a safe and effective intervention.
2.2. METHODS
This study was registered with the PROSPERO International prospective register of systematic
reviews website (Registration Number CRD42017058123). The MOOSE Guidelines for Meta-
Analyses and Systematic Reviews of Observational Studies were utilised as a guide for this
study. A systematic review of electronic databases including Pubmed, Embase and Cochrane
Database of Systematic Reviews was carried out using search terms of “[(Atrial
Fibrillation[MeSH Terms] OR Atrial Fibrillation OR AF[Ti] OR Cox OR Maze OR Mini-
maze) AND (Unipolar [All fields] OR Bipolar[All fields] OR Radiofrequency OR Ablation
technique*[All Fields] OR Cryoabla*[All Fields] OR Cryomaze OR Cryo-maze OR
Cryosurgery[MeSH Terms] OR Cryosurg*[All Fields]) AND (Surgery[All Fields] OR
Surgical procedures, operative[MeSH Terms])]”. The search was conducted from January 1991
to March 2016. The results were imported into Endnote reference database for further review.
These searches were supplemented by reviewing the bibliographies of published studies and
reviews. Studies were accepted if they conducted a surgical ablation procedure for the
treatment of AF and reported outcomes on SRM. Studies that did not perform ablation
concurrent to mitral valve intervention or that had primarily other concurrent cardiac surgical
procedures (coronary artery bypass grafting, etc) and sample size <100 patients were excluded.
Citations related to animal studies and those published in journals in languages other than
English were not included. After exclusions, the publications were analysed for the following
outcomes: SRM on and off anti-arrhythmic drugs (AAD), block success if assessed, mortality,
pericardial effusion/tamponade, significant bleeding, atrio-esophageal fistula, conversion to
open procedure if minimally invasive, stroke, length of stay, permanent pacemaker (PPM)
23
requirement and 30-day mortality. Figure 2.1 shows breakdown of the search, number of
studies retrieved and review process of citations. At the completion of the review process 16
studies remained for analysis.
2.2.1. Definitions
For the purposes of this study, surgical ablation procedures concurrent to mitral valve surgery
were defined as a surgical procedure, of any incision approach, in which a series of lesions
were placed in the heart (Cut and Sew/Radiofrequency/Cryothermy/Micro-Wave) for the
management of AF, with concomitant mitral valve repair or replacement. Additional, cardiac
valve procedures, coronary artery bypass or any other cardiac surgery procedure are
permissible if performed concurrent to mitral valve intervention. AF ablation procedures were
broadly categorised into Bi-Atrial Maze, Left Atrial Maze and Pulmonary Vein Isolation.
Surgical exclusion of the LAA appendage is part of the Maze procedure and has been included
as part of ablation procedures.
The classification of AF was made according to HRS/EHRA/ECAS Expert Consensus
Statement on Catheter and Surgical Ablation for AF, with any study patients that were
described as in permanent AF reclassified to long-standing persistent (LSP) AF.(262)
2.2.2. Statistical Analysis
Meta-analysis was performed using Stata/SE version 14. Means were weighted and standard
deviation of continuous variables or proportion and Exact binomial lower and upper confidence
limits of categorical variables were pooled across studies and analyzed using a random effects
meta-analysis model. This type of model was chosen to partially compensate for any inherent
heterogeneity. Linear regressions were performed to investigate associations between
outcomes of sinus rhythm maintenance and complications in each paper versus ablation
24
method, lesions set utilized, exclusion of LAA and use of RCT, weighting for number of
patients who completed follow-up. Assumptions of a linear model were upheld throughout.
Statistical significance was assigned to P value <0.05.
2.3. RESULTS
2.3.1. Study & Patient Characteristics
The 16 included studies (Table 2.1) were published between 2002 and 2015 and consisted of
13 case series studies (11 retrospective and 2 prospective) and 3 randomized control trials.
(193,263-277) Fourteen of these studies were single center and 2 were multi-center studies.
Overall, the total cohort contained 3313 patients with a mean sample size of 207 patients
(Range: 103-340). The composition of patients was 14.9% paroxysmal, 19.4% persistent and
59.1% long standing persistent. One study did not delineate the AF type of their patients leaving
6.6% of the cohort unspecified.(273) The average AF duration was 45.5 months (Range: 30.3-
108), though 2 studies reported median length of AF(193,268) and 3 studies did not report a
pre-operative length of AF.(264,270,274) The mean patient age was 58.4 years (Range: 45-
69.7), with a mean left atrial (LA) diameter of 5.6cm (Range: 4.6-6.1) and mean left ventricular
ejection fraction (LVEF) 56.2% (Range: 47-61.3). Though, 1 study did not report LA
diameter(193) and 3 studies did not report on LVEF.(266,269,273) Beyond these factors there
was incomplete reporting of previous ablation, pre-operative anti-arrhythmic drug (AAD) use,
warfarin use, or other risk factors for AF development (Hypertension, COPD, diabetes, BMI,
etc). The mean follow-up duration across the studies was 27 months (Range: 10.1-46.1), though
of note 1 study provided follow-up length in median.(270)
25
2.3.2. Mitral and Concurrent Valvulopathy
The various forms of mitral valve pathology were grouped into rheumatic, degenerative or
others (infective, ischemic, cardiomyopathy, congenital, mixed, etc). Four studies did not
report on the valvular pathology of their patients.(266,269,272,273) Of those studies that
reported the valve pathology, rheumatic valvular disease accounted for 1748 (72%),
degenerative valvular disease 561 (23%) and others in 125 (5%). Analysis of surgical
procedure undertaken identified the rate of mitral valve repair was 31.6% and the rate of
replacement was 56.7%, with 11.7% undefined due to 1 study not providing a breakdown of
mitral valve procedure.(273) Concomitant valvular procedures included aortic valve
intervention in 687 (20.7%) patients and tricuspid valve intervention in 1247 (37.6%) patients.
Concomitant non-valvular procedures were performed in 430 (13%) of patients of which
coronary artery bypass grafting was the most common procedure.
2.3.3. Surgical Ablation Procedures
2.3.3.1. Approach and Technique
The surgical access, lesion set, energy source and operative characteristic are presented in the
Table 2.2. In mixed ablation cohorts (where some patients had Bi-Atrial Maze and others PVI)
the minimum basic lesions that all patients had performed were documented in the “Included
Lesions” section of the Table 2.2. A generalised diagrammatic representation of lesions
employed is displayed in the Figure 2.2. Surgical access was obtained by sternotomy in all
studies except the studies by Gillinov et al (which utilised either sternotomy or
thoracotomy)(193) and the study by Jeanmart et al (which utilised thoracotomy alone).(269)
The methodology of creating a lesion varied among the studies and was usually created by a
combination of approaches. This point becomes particularly of note as the studies included
26
performed their lesions through a combination of endocardial and epicardial energy delivery
or only endocardial – in either case an atriotomy is required resulting in an effective “Cut and
Sew” (C&S) lesion being placed, even if not with intent of this to be a lesion to terminate
arrhythmia. In general, there were 3 studies that used a combination of C&S, radio-frequency
(RF) and cryothermy (Cryo),(193,268,270) 8 studies that used a combination of C&S and
RF,(264,265,267,269,272,274-276) 4 studies that utilised a combination of C&S and
Cryo,(263,266,273,277) and 1 study that utilised a combination of C&S, Cryo and Micro-Wave
(MW).(271)
The RF devices used to create lesions were Mono-Polar RF in 5
studies(267,269,270,272,276) with the range of devices being Cardioblate (Medtronic,
Minneapolis, MN, USA),(269,270,272,274) ThermaLine (Boston Scientific, Marlborough,
MA, USA),(267) Cobra (Boston Scientific, Marlborough, MA, USA – at the time study was
published),(267) a custom made probe,(276) and a combination of the above utilised over the
length of the study.(267) Bi-Polar RF devices were utilised in 6
studies(264,265,268,270,274,275) which consisted of Synergy (Atricure, West Chester, OH,
USA),(264,265,268,274,275) and Cardioblate Bipolar (Medtronic, Minneapolis, MN,
USA).(270) The cryo devices utilised varied over the studies consisting of either Frigitronics
(Cooper Surgical, Shelton, CN, USA),(263,266,268,271,273) SurgiFrost (CryoCath
Technology, Pointe-Claire, Canda),(277) a 140 Cryo Unit (Spembly Medical, Andover,
UK).(270) The micro-wave device utilised was FLEX (Afx Inc, Fremont, CA, USA).(271)
There was one study that did not specify energy source.(193)
2.3.3.2. Lesion Set
There was also variability of lesion sets used throughout all studies. These have been broadly
classified as follows for reporting: Bi-Atrial Maze (BAM); Left Atrial Maze (LAM); and
27
Pulmonary Vein Isolation (PVI) (Figure 2.2). A BAM lesion set was used in 11
studies(193,263-266,268,270,271,273,275,276) with 3 of those studies not exclusively using
BAM on their patients and having a number of patients with either LAM or PVI lesion
sets.(193,268,276) In the mixed cohort study by Gillinov et al. a combination of PVI (n=31;
RF:28, 3:Cryo), PVI + Left Atrial Connecting Lesion (n=80; Both RF and cryo all), CM (n=41;
Cut and sew + cryo: CMII in 6 and CMIII in 35) was utilised in their patients with no difference
in ablation cohorts freedom from AF.(268) In the next mixed cohort study of Gillinov et al a
combination of PVI (n=67) and BAM (n=66) was utilised with no difference in freedom from
AF in either group.(193) In the mixed cohort study of Wang et al the breakdown was LAM
(n=70) and BAM (n=70) with no difference in terms of AF recurrence between the
groups.(276) Three studies used a LAM lesion set exclusively(269,272,274) and 2 studies used
PVI with the study of Geidel et al an additional line across the posterior LA joining the PV
lesions.(267,277)
The Left Atrial Appendage (LAA) was excluded in all but 2 studies(269,273) with
exclusion methods consisting of oversewing/suture exclusion, (264,265,267,270,275)
excision,(263) either method allowed in 2 studies,(266,274) and method not
specified.(193,268,271-273,276) Additional intervention undertaken related to arrhythmia
control included left atrial reduction in 2 studies,(270,275) and division of the Ligament of
Marshall in 3 studies.(264,265,276)
The practice of Direct Current Cardioversion (DCC) was variable across the studies
with it utilized in 10 studies(263,265-267,269,272-276) and not specified in the remaining
studies.(193,264,268,270,271) In those studies that utilized DCC; 1 study reported immediate
use,(263) 8 studies utilized DCC on a proportion of their patients during the hospital stay as
required,(266,267,269,272-276) and 1 study reported use in the first 3 to 6 months post-op as
required.(265)
28
2.3.3.3. Operative Parameters
The operative time indices of cardio-pulmonary bypass (CPB) time and aortic cross clamp time
were reported in 11 studies with the meta-analysed CPB time being 138 minutes (Median: 148,
95% CI: 115-162) and average cross clamp time of 96 minutes (95% CI: 78-114). One study
reported total operative time of 192±27 minutes.(267)
2.3.4. Post Procedure Follow-up
The mean follow-up duration across the studies was 27 months (95% CI: 18.5-35.4). Table 2.3
details the follow-up frequency and methodology. All studies reported follow-up as a mean,
except 1 study which reported as median.(270) The 2 primary methods of follow-up rhythm
assessment were ECG and Holter monitoring. ECG was utilised in all studies except the study
by Gillinov et al.(193) Seven studies reported specific time periods of follow-up ranging from
3 monthly to 1 year.(263,265,266,271,273,275,276) Ten studies utilised a form of Holter
monitor or continuous telemetry for their follow-up;(193,263,266,267,270,272,275,276) the
length of monitoring was a 3 day Holter monitor in 1 study,(193) a 2 day holter in 1 study,(275)
1 day in 4 studies,(267,270,272,276) and not specified in the remaining 2 studies.(263,266).
Further, in 3 studies the use of Holter monitoring was only employed if AF was suspected, and
not routinely.(266,272,275)
2.3.5. Recurrent Arrhythmia
There was insufficient reporting on the results of specific AF subgroups
(Paroxysmal/Persistent/Long Standing Persistent) at follow-up to comment on sub-group
outcomes. The way the studies defined a recurrence of AF was also variable; 3 studies used a
documented of AF/Atrial Flutter/Atrial Tachycardia lasting >30 seconds,(193,264,265,276) 1
29
study also used documented AF/Atrial Flutter/Atrial Tachycardia though did not define length
of episode required,(268) 1 study defined as any recurrent AF/Atrial Flutter/Atrial Tachycardia
after a 3 month blanking period,(271) 2 studies defined maintenance of SR, instead of AF, as
a supraventricular rhythm with P-waves present on ECG,(272,273) and the remaining studies
did not provide a specific definition.(263,266,267,269,271,274,275,277)
2.3.6. Antiarrhythmic Drug Use
The post-operative use of AAD was also variable in practice; 11 studies utilised amiodarone
as their primary choice,(263-265,268,270-272,274-277) 3 studies did not provide any
statement on AAD use,(193,268,269) 1 study stated Amiodarone and Sotalol were never
used,(273) and 1 study stated AAD was used only if post-operative arrhythmia developed, but
did not comment on the one selected.(266) Of those that utilised Amiodarone post-operatively;
5 commenced an intra-venous form initially followed by an oral regime for 3 to 6 months,(263-
265,267,274) 4 only used the oral form of amiodarone for 3 to 6 months,(270,272,275,276)
and 2 studies did not commence it routinely, but used it as their first line option if any recurrent
of AF developed.(271)
2.3.7. Outcomes
The mean SRM across the studies was 79.9% (95% CI: 78.4-81.3; Figure 2.3). Only 1 study
reported the use of AAD at the time of follow-up, but did not specify the number of patients in
SR and on AAD.(193) Warfarin use at final follow-up was only described in 3 studies with a
total rate 49.3%.(267,269,272) The need for re-intervention by catheter ablation in the follow-
up period was only reported in 1 study and was required for 2 out of 133 patients.(193)
30
2.3.9. Procedural Complications
The complications recorded from across the studies is displayed in the Table 2.4 and Forest
Plots of complications in Figure 2.4. The weighted mean complication rate across the studies
was 6.9% (95% CI: 5.8-8.0;), which composed the following complications (Figure 2.5):
mortality of 0.92% (95% CI: 0.5-1.4;); CVA/TIA of 0.31% (95% CI: 0-0.6;);
bleeding/tamponade requiring intervention or return to the operating theatre of 0.24% (95%
CI: -0.1-0.5;); pulmonary embolus of 0.02% (95% CI: -0.2-0.3); and requirement for permanent
pacemaker of 1.2% (95% CI: 0.7-1.6;). There were no reported cases of atrio-esophageal
fistula.
Further analysis of mortality across the studies found the most common reported cause
to by multi-organ dysfunction in 7 patients, followed by heart failure in 4 patients, myocardial
infarction in 3 patients, cerebral haemorrhage in 2 patients, CVA/neurological cause in 1
patient, infection/sepsis in 1 patient and pulmonary embolus in 1 patient. One study only
provided a mortality cause breakdown of “cardiac” vs “non-cardiac”,(267) another study
provided their mortality caused grouped together with a comparator cohort so was not suitable
for further analysis(275) and 4 studies did not list causes of death.(193,268,271,277)
The meta-analysed mean length of stay was 10.7 (95% CI: 4.0-17.4) days with only 3
studies reporting a mean length of stay.(193,267,269)
2.3.10. Subgroup Analysis
2.3.10.1. Lesion Set
The categorised lesion sets for analysis comprised 1881 (62%) patients in the BAM group, 486
(16%) in the LAM group, 271 (9%) patients in the PVI group and 398 (13%) patients in the
combination group with patients not restricted to 1 specific lesion set group. Lesion set was
shown to be a significant univariate predictor of SRM across the included studies (p<0.001).
31
The SRM of the BAM group was 84% (95% CI: 78.7-88.9), of the LAM group was 64% (95%
CI: 54.0-74.0), of the PVI group 75% (95% CI: 61.8-88.7) and of the Combination group 66%
(95% CI: 54.5-76.7) (Figure 2.6A). Comparison of these groups found the BAM group superior
to the LAM (p<0.001) and superior to the combination group (p<0.01), but no statistical
difference to the PVI group (p=0.25). There were no significant differences between LAM,
combination and PVI groups.
Complication rates related to lesion set are shown in Figure 2.6B. Interestingly, both
the BAM group and the LAM group were statistically superior to the PVI group in terms of
reduced mortality (BAM vs PVI: 1.5% [95% CI: 1.1-1.9] vs 3.0% [95% CI: 1.8-4.1], p=0.02;
LAM vs PVI: 1.0% [95% CI: 0.2-1.9] vs 3.0% [95% CI: 1.8-4.1], p<0.01). Further sub-group
analysis demonstrated a significantly lower rate of PPM placement when comparing BAM to
LAM and BAM to Combination groups (BAM vs LAM: 2.2% [95% CI: 0.0-4.7] vs 7.8% [95%
CI: 2.4-12.8], p=0.048; BAM vs Comb: 2.2% [95% CI: 0.0-4.7] vs 11% [95% CI: 5.6-16.6],
p<0.01). In total complication rate analysis, the BAM group showed a statistical significant
superiority to the Combination group (6.6% [95% CI: 2.6-10.6] vs 16.8% [95% CI: 8.1-25.5],
p=0.04). There were no other statistical differences between any of the groups regarding
mortality, CVA, bleeding, PPM or total complication rates in the remaining sub-group analyses
between ablation groups.
2.3.10.2. Energy Source
The categorised lesion sets for analysis comprised C&S/RF/Cryo in 477 (16%) patients,
C&S/RF in 1464 (48%), C&S/Cryo in 765 (25%) patients and C&S/MW/Cryo in 330 (11%)
patients. The rate of SRM achieved with each energy source was 79.9% (95% CI: 66.7-93.1)
with C&S/RF/Cryo, 72.8% (95% CI: 65.2-80.3) with C&S/RF, 80.7% (95% CI: 70.2-91.1)
with C&S/Cryo in 765 (25%) patients and 87.6% (95% CI: 71.7-103.5) C&S/MW/Cryo in 330
32
(11%) patients. There was no statistically significant difference between any combination of
energy source used and SRM, mortality rate or total complication rates.
2.3.10.3. LAA Exclusion
The number of patients who underwent LAA exclusion were 2717 (89.5%) and the number
who did not were 319. There was no statistically significant difference between the groups in
terms of SRM, but a statistically significant difference between mortality rate was identified
with a reduction shown when the LAA was not excluded (LAA Excluded vs NOT: 1.7±0.3%
vs 0.3±0.1%, p=0.02). No other differences in complication rates were statistically significant
between these groups.
2.3.11. Randomized-Controlled Data
Analysis was performed to compare the Randomised Control Trials (RCT) to the Case Series
of the total cohort. The RCTs demonstrated a SRM of 70.8% (95% CI: 57.7-83.9) compared to
the Case Series cohort of 79% (95% CI: 72.8-85.1) which was not statistically significant.
Further, there was no statistical difference in the mortality or complication rates of either group.
2.4. DISCUSSION
Atrial fibrillation occurring in the context of valvular heart disease is common. Surgical
ablation for the treatment of AF at the time of concurrent mitral valve surgery is an opportune
therapeutic procedure. However, the level of evidence for such an approach and safety of
concurrent surgery are issues that are frequently raised. This systematic review and meta-
analysis on the efficacy and safety of these procedures demonstrates the following information:
1. The ablation of AF concurrent to mitral valve surgery is an effective procedure for
maintaining SR with the BAM lesion set appearing to offer a superior result compared
to procedures confined to the left atrium.
33
2. The overall total complication rate was low at 6.9%. This number was included the
PPM requirement rate of 1.2%.
3. The mean follow-up across this series was 27 months with a SRM of 79.6% achieved
displaying good medium term results for ablation concurrent to mitral valve surgery.
4. The exclusion of the LAA in this population does not appear to reduce stroke risk or
offer a statistically significant difference in rhythm control. Interestingly, the exclusion
of the LAA may confer a high mortality rate, so consideration should be given when
including this intervention.
Although this review has highlighted the considerable variability in the methods of follow up
and reporting of the data in these studies, a substantial proportion of the population had
persistent AF at baseline which was not present at follow up. In addition, this group represent
the most advanced substrate in which there has been limited success with catheter ablation
techniques. In this context, the current data provides evidence for the feasibility of intervention
in this group to rhythm control.(278)
2.4.1. Valvular Disease, Surgery and AF
The incidence of AF concurrent to mitral valve disease is significant and the association of this
with increased risk of stroke has been demonstrated.(9,194-196) As such, in symptomatic
patients, if there is an opportunity to reduce the risk of AF it should be strongly considered.
Previous large analyses conducted utilising the Society of Thoracic Surgeons database with a
cohort on 58,370 isolated mitral valve operations described a median CPB time of 112 minutes,
a median aortic cross-clamp time of 82 minutes with major complications of permanent stroke
was 1.31% and operative mortality rate of 2.45%.(279) The length of these median indices
measured in this meta-analysis are indeed longer by more than 30 minutes for CPB time and
20 minutes for aortic cross-clamp time. However, importantly, the complications rates of CVA
34
and mortality are both significantly reduced in this analysis when compared to previously
published data on the natural history, which may suggest a protective effect for the patient in
returning them to SR post mitral valve surgery.
The most recent guidelines produced by the STS on the surgical management of AF
provided their highest recommendation (Class A, Level 1) to the concurrent ablation of AF to
mitral valve surgery with no increase in morbidity or mortality.(210) The findings of this meta-
analysis are consistent with these guidelines, although, it must be recognised that there still
remains a paucity of level 1 evidence in this area.
2.4.2. Surgical Ablation Technique
The optimal technique for the surgical ablation of AF is a complex consideration. This analysis
has demonstrated no statistically significant difference in lesion energy source selection in
obtaining SRM at follow-up. The goal must remain to obtain a transmural lesion, regardless of
energy source. The optimal lesion set to apply appears to favour BAM in this analysis with a
statistically significant superior result obtained in SRM when compared to studies only
performing LAM ablation or studies that included a multiple of lesion sets, and a statistically
significant superior result in mortality rate compared to PVI. The question of optimal ablation
technique is one of careful consideration and would warrant future focused studies to answer
this question.
2.4.3. Complications
Catheter ablation of AF is a safe and effective treatment for AF typically employed in the
absence of structural heart disease. A recent systematic literature review on the complications
of catheter ablation extracted data of 83,236 patients undergoing catheter ablation found a 2.9%
total complication rate, a mortality rate of 0.06%, CVA/TIA rate of 0.6%, pericardial
35
tamponade rate of 1.0% and phrenic nerve injury of 0.4%. Though it comes with the additional
complications of atria-oesophageal fistula of 0.08% and valve damage of 0.2%.(280) When
these are compared to the complication rates of this analysis, it is likely that the surgical cohort
represents a sicker cohort of patients to require surgery in the first place, and this is then coupled
with a larger procedure to treat their heart disease in totality (mitral valvulopathy in addition to
AF).
2.4.4. Clinical Implications
Patients presenting for mitral valve intervention with concurrent AF should be considered for
the addition of a concurrent ablation procedure. The rate of SRM achievable is significant, and
regardless of the energy source preference of the surgeon, a bi-atrial lesion set will offer the
best results without a significant increase in patient morbidity. Previous studies analysing the
30-day mortality rate of general patients undergoing mitral valve repair or replacement
identified a range rate of 2.2% to 10.4%, with more recent studies finding lower rates of
mortality.(281-284) In consideration of this, no increase in mortality rate is seen with the
addition of an ablation procedure.
2.4.5. Study Limitations
The composition of the included studies identifies a relative paucity of level 1 evidence
identified within this area. Further, there exists significant heterogeneity in terms of surgical
procedure and follow-up method and intervals among these included studies, and this should
be considered in interpreting the results. There is also variability in the quality of studies with
discrepancies in recording of baseline variables, consistency of ablation method and follow-
up. These variations serve to not only highlight the features of the highest quality studies, but
also the need for standardisation of reporting within the field. Additionally, as with any meta-
36
analysis there may be confounding variables, measured and unmeasured, which may impact
the results of this analysis. Finally, we are unable to provide more extensive analysis on long
term results of patients off AADs and in AF subtypes given the lack of patient level data
available. Nonetheless, these identified limitations serve to highlight key areas of attention in
future studies.
2.5. CONCLUSIONS
Surgical ablation procedures for the management of AF carried out concurrent to mitral valve
intervention have good results in terms of sinus rhythm maintenance at medium-term follow-
up with no additive mortality effect. The best results have been seen with more extensive lesion
sets (bi-atrial) regardless of energy source utilised. However, it must be noted that there is
significant variability in study follow up length and method. Further prospective studies with
structured follow-up could confirm these results and help delineate procedural success for
specific patient AF subtype populations presenting for mitral valve surgery.
37
2.6. FIGURES AND TABLES
Figure 2.1: Search Strategy
38
Figure 2.2 – Lesion Sets
Depicts the common lesions performed for AF ablation in the given studies. The Blue and
Green represent PVI and Box lesions respectively (considered together), the addition of Red
represents a Left Atrial Maze lesion set (PVI/Box + Left Atrial Appendage Line + Exclusion
+ Mitral Line), with the further addition of Purple to the afore mentioned representing a Bi-
Atrial Lesion Set (Superior Vena Cava Line + Inferior Vena Cava Line + Right Atrial
Appendage Line + Tricuspid Line).
39
Figure 2.3 – Sinus Rhythm Maintenance Forest Plot
40
Figure 2.4 –Complication Forest Plots
41
Figure 2.5 – Mean Complication Rates
The mean complication rates of mortality, CVA, RTOT and PPM requirements are shown
here with error bars representing the 95% CI. Tabulated from meta-analysis forest plot
findings.
42
Figure 2.5 – Lesion Set Comparisons
A) The comparative SRM rates achieved based on lesion set with statistically significant
differences highlighted. B) Comparative complication rates based on lesion set.
43
44
45
46
47
Chapter 3
Surgical Ablation of Atrial Fibrillation with Concurrent Cardiac Surgery – Assessment
of the Safety and Efficacy of Ablation procedures during any other Cardiac Surgical
Intervention
3.1. INTRODUCTION
Atrial Fibrillation (AF) is the most common atrial arrhythmia effecting the general population
with the major risk factors for development being the common risk factors of those patients
presenting for cardiac surgery of advanced age, hypertension, congestive cardiac failure,
myocardial infarction and cardiac valulopathy.(1,2,9,10) Studies analysing the outcome of
patients undergoing cardiac surgery in the form of Coronary Artery Bypass Grafting (CABG)
and heart valve disease have identified a negative impact on in hospital complications and long
term outcomes when preoperative AF is present.(285-289) Thus in those patients with a
mixture of structural heart disease and coronary artery disease concomitant to AF the
elimination of AF may offer a protective effect against in hospital complications and long term
patient outcomes.
Historically, the Cox-Maze procedure has provided the best results for the surgical
elimination of AF achieving sinus rhythm maintenance results of 89% sinus rhythm
maintenance (SRM) with the use of antiarrhythmic drugs (AAD) and 78% SRM without AAD
at 5.4 years follow-up.(188,189) This procedure is commonly undertaken in conjunction with
potentially any other cardiac surgical procedures to treat AF concurrently; however, there has
been a relative paucity of level 1 evidence and as such high-level recommendations for whether
to employ this strategy concurrent to other cardiac surgical procedures has been lacking in the
HRS/EHRA/ECAS Expert Consensus Statement on Catheter and Surgical Ablation for
AF.(262)
48
The majority of literature published on the surgical ablation of AF has not been
restrictive of inclusion criteria for other cardiac procedures carried out concomitantly,
consequently, the largest series are formed of mixed cohorts of patients that have undergone
surgical ablation concurrent to either coronary artery bypass grafting, valvular intervention of
any cardiac valve or a mixture of these procedures.(190,191) This presents a highly
heterogenous cohort of patients undergoing surgical ablation of AF that is of interest as it is
representative of the common patients presenting for cardiac surgery that may also be identified
to have AF incidentally. We have elected to analyse this cohort as a separate group,
independent of surgical ablation carried out concurrent to purely mitral valve intervention or
surgical ablation as a standalone procedure as this cohort not only represents a population with
a highly variable risk profile, but this cohort also does not reflect one sole mechanism for the
development of AF, as what may be seen in the ablation of AF concurrent to mitral valve
intervention.(290) Rather, it presents a cohort of patients requiring cardiac surgery that
incidentally have AF and now are provided with an opportunity for interventional treatment
while their primary cardiac pathology is being treated.
Here, we undertook a systematic review of the literature and meta-analysis to evaluate
the efficacy and safety of surgical ablation, concurrent to any cardiac surgical procedure, but
excluding studies of purely mitral valve surgical procedures or standalone cardiac surgical
ablation procedures. We hypothesised that AF ablation concurrent to other cardiac surgical
procedures is a safe and effective intervention.
3.2. METHODS
This study was registered with the PROSPERO International prospective register of systematic
reviews website (Registration Number CRD42017059107). The MOOSE Guidelines for Meta-
Analyses and Systematic Reviews of Observational Studies were utilised as a guide for this
49
study. A systematic review of electronic databases including Pubmed, Embase and Cochrane
Database of Systematic Reviews was carried out using search terms of “[(Atrial
Fibrillation[MeSH Terms] OR Atrial Fibrillation OR AF[Ti] OR Cox OR Maze OR Mini-
maze) AND (Unipolar [All fields] OR Bipolar[All fields] OR Radiofrequency OR Ablation
technique*[All Fields] OR Cryoabla*[All Fields] OR Cryomaze OR Cryo-maze OR
Cryosurgery[MeSH Terms] OR Cryosurg*[All Fields]) AND (Surgery[All Fields] OR
Surgical procedures, operative[MeSH Terms])]”. The search was conducted from January 1991
to March 2017. The results were imported into Endnote reference database for further review.
These searches were supplemented by reviewing the bibliographies of published studies and
reviews. Studies were accepted if they conducted a surgical ablation procedure for the
treatment of AF and reported outcomes on sinus rhythm maintenance (SRM). Studies that
performed ablation concurrent to only mitral valve intervention were excluded as this was felt
to represent a separate population cohort of AF. The following studies were excluded: (i)
sample size <100 patients; did not consistently report on rhythm outcome; highly
heterogeneous intervention arms with multiple ablation strategies in the one paper; those with
an absence of data on lesion set utilisation; animal studies; and those published in journals in
languages other than English. If citations had multiple ablation arms for analysis that were
>100 patients each they were divided and both kept for analysis. If the individual arms were
statistically different and <100 patients they were excluded and if they were not statistically
different they were combined and included as a single cohort for analysis. If multiples citations
were identified from the same author and/or institute with patients captured from the same
period the most complete data set article was accepted and the other articles were discarded. If
studies contained multiple ablation strategies the predominate or most basic lesion set was
chosen for study categorisation. Selected publications were analysed for the following
outcomes: SRM on and off anti-arrhythmic drugs (AAD), block success if assessed, mortality,
50
pericardial effusion/tamponade, significant bleeding, atrio-oesophageal (AO) fistula,
conversion to open procedure if minimally invasive, stroke, length of stay, PPM requirement
and 30-day mortality. Figure 3.1 shows the number of studies retrieved and review process of
citations. At the completion of the review process 29 studies remained which supplied 31 data
sets for analysis.
3.2.1. Definitions
For the purposes of this study, surgical ablation procedures concurrent to cardiac surgical
procedures were defined as any adult cardiac surgical procedure, of any incision approach, in
which a series of lesions were placed in the heart (Cut and
Sew/Radiofrequency/Cryothermy/Micro-Wave/Combination) for the management of AF.
Additional procedures recorded included any specific cardiac valve procedures, coronary artery
bypass or any other cardiac surgery procedure. Lone procedures for AF were excluded if they
represented the entire cohort, but were permissible if they represented only a subsection of the
study cohort that is otherwise composed of patients have undergone other cardiac surgical
procedures. AF ablation procedures were broadly categorised into Bi-Atrial Maze, Left Atrial
Maze and Pulmonary Vein Isolation. Surgical exclusion of the LAA appendage is part of the
Maze procedure and has been included as part of ablation procedures.
The classification of AF was made according to HRS/EHRA/ECAS Expert Consensus
Statement on Catheter and Surgical Ablation for AF, with any study patients that were
described as being in permanent AF reclassified to long-standing persistent.(262)
3.2.2. Statistical Analysis
Meta-analysis was performed using R (Version 3.3.0). Mean and standard deviation of
continuous variables or proportion and Exact binomial lower and upper confidence limits of
51
categorical variables were pooled across studies and analysed using a random effects meta-
analysis model. This type of model compensated for any inherent heterogeneity. Mixed model
meta-regression techniques were performed to investigate associations between outcomes of
sinus rhythm maintenance and complications in each paper versus ablation method, lesions set
utilised, exclusion of LAA and use of RCT, weighting for number of patients who completed
follow-up. Assumptions of a linear model were upheld throughout. Statistical significance was
assigned to P value <0.05.
3.3. RESULTS
3.3.1. Study & Patient Characteristics
The 31 included studies supplied 33 cohorts for analysis (Table 3.1). These studies were
published between 2002 and 2015 that consisted of 25 case series studies (15 retrospective and
10 prospective), 3 cohort comparator studies and 2 randomized control trials.(188,190-
192,291-317) Twenty-five of these studies were single center and 4 were multicenter studies.
Overall, the total cohort contained 6791 patients with a mean sample size of 206
patients (Range: 100-576). The pooled cohort consisted of 27% paroxysmal, 20% persistent,
45% longstanding persistent AF. Two studies did not delineate the AF subtype of their patients
leaving 8% of patients in the total cohort unspecified.(305,315) The average AF duration was
46.3 months (Range: 12-98), though 4 studies reported length of AF in
median(190,296,299,312) and 5 studies did not report a length of pre-op
AF.(295,301,303,308,311)
The mean patient age was 64.2 years (Range: 38-73.1), with a mean left atrial (LA)
diameter of 5.3 cm (Range: 4.3-7.3) and mean left ventricular ejection fraction (LVEF) 54.9%
52
(Range: 48-60.1). Of note, 1 study reported age in median,(312) 7 studies did not report mean
LA diameter(295,297,301,303,304,311,312) and 13 studies did not report
LVEF.(192,292,295,297,303,304,307-313). The incidence of pre-operative catheter ablation,
pre-operative use of anti-arrhythmic drugs (AAD) and warfarin were inconsistently reported.
Baseline risk factors for development of AF were not consistently reported across the studies
(Table 3.2).
The mean follow-up duration across the studies was 22.1 months (Range: 6-48), with 2
studies reporting a median duration.(294,296)
3.3.2. Concurrent cardiac surgical procedures
The reporting of concurrent cardiac surgical procedures was not uniform across the studies as
overlap of procedures was not routinely documented. The most common concurrent procedure
performed in addition to AF ablation was mitral valve repair or replacement in 3625 patients
(53.4%), followed by tricuspid valve procedures in 1125 patients (16.6%), coronary artery
bypass grafting (CABG) in 1197 patients (17.6%), aortic valve procedures in 975 patients
(14.4%) and other non-valve or CABG procedures in 517 patients (7.6%) (Table 3.3). Further,
the reported incidence of combined CABG and valvular procedures was 902 patients (13.3%),
isolated valve procedures in 3100 patients (45.7%) and isolated AF ablation in 595 patients
(8.8%).
3.3.3. Surgical Ablation Procedures
3.3.3.1. Approach and Technique
The surgical access, lesion set, energy source and other operative characteristics are presented
in Table 3.4. In mixed ablation cohorts (where some patients had Bi-Atrial Maze and others
Pulmonary Vein Isolation) the minimum basic lesions that all patients had performed were
53
documented in the “Included Lesions” section of the Table 3.4. A generalised diagrammatic
representation of lesions employed is displayed in Figure 2.2. Across the studies surgical access
was obtained through sternotomy. Though of note the study by Ad et al reported thoracotomy
access in 219 patients,(190) the study by Damiano et al had 10% of patients with access via
thoracotomy(188), Guden et al reported a minimally invasive procedure in 43 patients without
further specification,(301) the study by Henn et all reported 44 patients undergoing a mini-
thoracotomy(191) and the study by Yanagawa et al reported thoracotomy access in 40
patients.(317) The methodology of creating lesions of conduction block varied across the
studies with the majority utilising an atriotomy to gain access to the inside of the heart, which
is effectively a “Cut and Sew” (C&S) lesion. Beyond this most studies then utilised a separate
device to create the remaining lesion set. As such, all energy sources utilised in a study are
tabulated with the primary method of creating most lesions shown underlined in Table 3.4. In
general terms, the studies could be categorised as primarily Radio-Frequency(RF) for 34% of
the patients,(192,291,292,294,295,298,301,302,305,308,310,313,315) Cryothermy in 15% of
the patients,(293,296,303,306,307,316,317) RF or Microwave in 3% of the patients,(300) C&S
in 12% of the patients(312) and a combination of C&S,RF and Cryo in 36% of the
patients.(188,190,191,299,304,309,311,314)
Devices used to created lesions in studies varied; Bipolar RF devices used included the
Isolator Synergy devices (Atricure, West Chester, OH, USA) in 8 studies,(188,190,295,297-
299,304,313,314) and the Cardioblate devices (Medtronic, Minneapolis, MN, USA) in 8
studies.(188,190,192,292,293,295,311,315) Monopolar RF devices utilised included
Thermaline (Boston Scientific, San Jose, CA, USA) in 3 studies,(291,302,310) Cobra (Boston
Scientific, San Jose, CA, USA) in 2 studies,(291,298) Cardioblate Monpolar Device
(Medtronic, Minneapolis, MN, USA) in 4 studies(188,192,292,300) and a custom made RF
device (Osypka GmbH, Grenzach-Wyhlen, Germany) in 3 studies(292,305,308). A non-stated
54
RF device was utilised in 5 studies.(191,291,294,301,309) The cryo devices used across the
studies included the ATS system (Medtronic, Minneapolis, MN, USA) in 6
studies,(190,293,296,303,316,317) Frigitronics (Cooper Surgical, Shelton, CN, USA) in 1
study,(299) 142 Cyro Unit (Spembly Medical, Andover, Hampshire, UK) in 1 study,(306)
Cryo1 (Atricure, West Chester, OH, USA) in 1 study(190) and cryo device not specified in 4
studies.(191,295,307,314) Micro-wave was utilised in 1 study with the Flex 4 device (Guidant
Corp, Santa Clara, CA, USA).(300)
In closer study analysis the study by Budera et al used cryo with 97% of the patients
and RF with the remaining 3% and was classified as a cryo study(293) and the study by
Grubitzsch et al utilised RF energy in 102 patients and MW in 110 patients.(300) Of note, the
study of Doty et al was broken into 2 cohorts of C&S BAM and RF BAM for inclusion.(295)
3.3.3.2. Lesion Set
Variability of lesion sets also existed across the studies and were broadly categorised into Bi-
Atrial Maze (BAM), Left Atrial Maze (LAM) and Pulmonary Vein Isolation (PVI) for analysis.
The common lesion sets are diagrammatically represented in Figure 2.2.
Primary lesion set utilisation consisted of a BAM lesion set in 15 studies comprising
59% of the total cohort,(188,190-192,292,295-297,304,306,308,312,315-317) the LAM lesion
set in 10 studies comprising 21% of the total
cohort,(291,293,294,300,301,303,307,311,312,315) the PVI lesion set in 5 studies comprising
19% of the total cohort(298,299,302,305,313) and 1 study had a cohort of patients who
underwent any of the 3 basic lesion sets to comprise 2% of the total cohort.(309)
Of note the study by Deneke at al utilised either the LAM in 66 patients or the BAM in
64 patients with no difference in outcome between the groups,(294) the RF cohort by Doty et
al had a BAM in 125 patients and a PVI in 32 patients,(295) the study by Gaynor et al had 3
55
variants of a BAM consisting of a Cox-Maze I in 33 patients, Cox-Maze II in 16 patients, Cox-
Maze III in 197 patients and a Cox-Maze IV in 30 patients,(297) Guden et al reported a BAM
in 48 patients and a LAM in 57 patients,(301) Martinez-Comendador et al reported a LAM in
89 patients and BAM in 66 patients(303) and Sternik et al reported a LAM in 142 patients and
BAM in 50 patients.(311) In addition, the study by Wang et al was broken into 2 cohorts for
analysis of LAM and BAM.(315)
Additional ablation lines beyond the basic categorising lesion set included Ligament of
Marshall ablation in 2 studies,(291,315) PVI joining line in 3 studies,(298,302,305) Mitral line
in 1 study(310) and RA isthmus line with or without floor or roof lines in 1 study.(299)
The Left Atrial Appendage (LAA) was excluded in all but 2 studies.(300,305) The
method was by resection in 12 studies,(188,292-295,297,304,309-311,315,317) oversewn in 8
studies,(291,296,298,303,306-308,316), either oversewn or resected in 3 studies,(299,301,315)
stapled in 1 study,(313) either oversewn, resected or stapled in 1 study(192) and not stated in
4 studies.(190,191,302,314)
The use of Direct Current Cardioversion (DCC) was variable among studies with it
being used post-op if necessary in 11 studies,(292,298,300,303,305,307-309,311,312,314) as
part of the procedure either routinely or if required in 4 studies,(192,302,309,314) if there was
an early recurrence without time frame being listed in 1 study,(315) not performed in 1
study(301) and not stated in the remaining 16 studies.(188,190,191,291,293-
297,299,304,306,310,313,316,317)
3.3.3.3. Operative Parameters
The operative time indices reported included cardio-pulmonary bypass (CPB) time, aortic cross
clamp time and total procedure time. The mean CPB time was 141 minutes (Range: 113-188),
the mean aortic cross clamp time was 88 minutes (Range: 69-119) and the mean total
56
procedural time was 220 minutes (Range: 167-294). Of note, 8 studies did not provide any time
indices for analysis.(190,295,301,307-309,313,316)
3.3.4. Post Procedure Follow-up
The mean follow-up duration across the studies was 22.1 months (Range: 6-48) with Table 3.5
detailing follow-up frequency and methodology. All studies reported follow-up duration as a
mean with only 2 studies reporting as a median.(294,296) The primary methods of follow-up
across all studies was ECG and Holter monitor; 7 day Holter monitor was the highest level of
monitoring utilised in 3 studies,(296,313,316) a 1d Holter (or Holter length not specified) was
utilised in 15 studies,(188,190-192,291,293,294,301-303,305,306,311,315,317) 3 studies
utilised a Holter monitor if possible or symptomatic, but not routinely,(297,307,314) but
otherwise those studies reverted to ECG follow-up which then comprised 11
studies.(292,295,297-299,307-310,312,314)
There was not routine delineation of follow-up to specific groups of paroxysmal,
persistent or long standing persistent to comment on the outcomes of these sub-groups.
3.3.5. Definitions of Recurrent Arrhythmia
The adopted definition for recurrence of arrhythmia was variable among the studies with the
most common definition being the recurrence of AF, atrial flutter or atrial tachycardia lasting
longer than 30 seconds on follow-up monitoring in 11 studies,(188,190-
192,293,294,296,300,313,316,317) 1 study defined as any occurrence of AF or atrial flutter on
ECG without specifying time period or length,(299) 1 study defined as any occurrence of AF
or atrial flutter after 6 months,(297) 1 study conversely defined SR as the presence of p-
waves(305) and the remaining 17 studies did not provide a definition.(291,292,295,298,301-
304,306-312,314,315)
57
3.3.6. Antiarrhythmic Drug Use
The post-operative use of AAD had varying practice among the studies; 8 studies routinely
used Amiodarone or Sotalol post-operatively and continued for a period after
surgery,(291,293,298,301,304,312-315) 10 studies utilised Amiodarone or Sotalol post-
operatively if there was a recurrence of AF,(292,296,303,305,308-311,314,316) 4 studies listed
the use of a Class I or III AAD without specifying beyond,(188,191,192,299) 1 study listed the
use of either the patients pre-operative beta blocker or commencement of amiodarone or
sotalol,(300) 1 study listed the use of AAD as not routine(306) and the remaining 5 studies did
not specify.(190,294,295,297,307)
3.3.7. Outcomes
The mean SRM across all included studies was 79.6% (95% CI: 76.2-82.5) and the mean SRM
off AAD was 65.3% (95% CI: 57.4-72.5) (Figure 3.2 and 3.3). Of note, 9 studies did not supply
data on the number of patients in SRM without AAD.(295,298,301-303,305,307,312,315)
3.3.8. Procedural Complications
The complications recorded across the studies is presented in Table 3.6, the complication rates
Forest Plots in Figure 3.4 and comparative rates in Figure 3.5. The mean total complication
rate was 15.1% (95% CI: 11.7-19.3), which comprised mean mortality rate of 4.3% (95% CI:
3.4-5.4), mean CVA rate of 1.8% (95% CI: 1.3-2.5), mean rate of return to the operating theatre
(RTOT) for bleeding of 4.3% (95% CI: 3-6.1), mean rate of post-operative permanent pace
maker (PPM) requirement of 5.4% (95% CI: 3.8-7.5), and mean rate of atrio-esophageal(AO)
fistula of 0.5% (95% CI: 0.3-0.9). There was no incidence reported of phrenic nerve injury. Of
note, the study by Ad et al did not report number of complications.(190)
58
Analysis of mortality found 191 deaths reported across all studies with breakdown
analysis of cause demonstrating cardiac to be most common with 35 cases (18%), followed by
multi-organ dysfunction with 22, sepsis in 10 cases, respiratory failure in 5 cases, CVA in 2
cases, AO fistula in 2 cases, PE in 2 cases, renal failure in 1 case and gastro-intestinal
complication in 1 case. The majority were unknown, not stated or not reported with 109 cases.
The mean length of stay was 11.2 days (Range: 6-14) with 9 studies reporting mean
length of stay(191,192,291,293,300,302,303,306,316) and 7 studies reporting median length
of stay.(188,296,297,304,311,314,317)
3.3.9. Sub-Group Analysis
3.3.9.1. Baseline Characteristics
Univariate analysis was performed across the studies examining their relationship to SRM with
age and paroxysmal AF status being statistically significant. For every 1 year increase in
baseline patient age there was a 0.6% decline in SRM (p=0.02) and for every 1% in increase
in the percentage of patients in a given study having paroxysmal AF there was a 0.2% increase
in SRM (p=0.02). The pre-operative duration of AF, LA diameter, percentage of mitral valve
intervention, year of study publication and follow-up duration were not shown to be statistically
significant predictors of SRM.
Further univariate analysis was then performed to determine predictors of
complications and it was identified that every 1% increase in the proportion of patients
undergoing mitral valve surgery was associated with a 0.2% decline in complication rate
(p=0.02). In other words, patients not undergoing mitral valve surgery in this cohort were more
likely to experience a complication. Analysis of age, AF duration, follow-up duration,
percentage of paroxysmal AF patients, LA diameter, publication year or CPB time were not
predictive of increased complication rate.
59
3.3.9.2. Lesion Set Comparison
The categorised lesion sets for analysis comprised BAM in 3996 patients (59% of total cohort),
LAM in 1400 patients (21% of total cohort), PVI in 1275 patients (19% of total cohort) and a
combination study of 120 patients (1% of total cohort). Comparison of lesion sets did not
identify any significant difference in terms of SRM with or without AAD achieved at final
follow-up (Figure 3.6). Further, there was no significant difference in terms of any individual
complication or total complication rates.
3.3.9.3. Energy Source Comparison
The categorised energy sources for comparison comprised RF as the primary modality in 2293
patients (34% of total cohort), cryo in 1030 patients (15% of total cohort), RF/MW in 212
patients (3% of total cohort), C&S in 831 patients (12% of total cohort) and a combination of
C&S, RF and Cryo in 2425 patients (36% of total cohort).
The effect of energy source on SRM on and off AAD is shown in Figure 3.7, with
analysis of SRM demonstrating that the studies utilising a combination of ablation modalities
in their operations or those utilising C&S to be superior to the use of RF(Combo: 85.6% [95%
CI: 80.3-89.7] vs RF: 75.4% [95% CI: 70.7-79.6], p=0.007; C&S: 87.3% [95% CI: 73.8-94.4]
vs RF: 75.4% [95% CI: 70.7-79.6], p=0.03), and the combination approach was also shown to
be superior to the use of Cryo(Combo: 85.6% [95% CI: 80.3-89.7] vs Cryo: 76.5% [95% CI:
68.5-83.1], p=0.04). However, no statistically significant difference was found in SRM off
AAD between any approach, nor in any of the rates of complication.
60
3.3.9.5. LAA Exclusion
The LAA was excluded in 6345 patients (93%) and not excluded in the remaining 446 patients
(7%). Comparison revealed no significant difference in SRM with or without AAD, and no
significant difference in rate of CVA or any other complications. However, the rate of mortality
was lower in the LAA exclusion group (LAA Excl: 4% [95% CI: 3.2-5] vs LAA Not Excl:
10.3% [95% CI: 7.2-14.5], p=0.001) and the overall rate of complications was lower in the
LAA exclusion group(LAA Excl: 14.2% [95% CI: 11-18.2] vs LAA Not Excl: 33.3% [95%
CI: 10.2-68.9]. p=0.04).
3.3.10. Randomised Control Trial Data
Analysis was performed to compare the Randomised Control Trials to the Case Series studies
of the total cohort. This revealed no statistically significant difference between the groups in
terms of SRM or total complications.
3.4. DISCUSSION
Atrial fibrillation is commonly present in patients presenting for cardiac surgery and this
presents a unique opportunity for the potential treatment of their arrhythmia. The treating
physician must then consider if the efficacy and safety of an additional procedure for the
management of AF is justified to outweigh the negative effects of on-going AF for the patient.
This systematic review and meta-analysis demonstrates the following information:
1. The ablation of AF concurrent to general cardiac surgical procedures is an effective
treatment for maintaining SR with mean follow-up across the series of 22.1 months
demonstrating SRM of 79.6% (95% CI: 76.2-82.5) and SRM off AAD of 65.3% (95%
CI: 57.4-72.5).
61
2. No statistically significant difference exists in results obtained with BAM, LAM or PVI
lesion set, however, the use of C&S for creating lesions or a combination of modalities
is associated with superior outcomes in SRM on AAD.
3. Younger patient age and paroxysmal AF are associated with superior results in terms
of SRM.
4. The overall total complication rate was low at 15.1% (95% CI:11.7-19.3) with mortality
rate 4.3% (95% CI:3.4-5.4), CVA rate of 1.8% (95% CI:1.3-2.5), RTOT of 4.3% (95%
CI:3-6.1), PPM of 5.4% (95% CI:3.8-7.5) and AO fistula of 0.5% (95% CI:0.3-0.9).
5. Ablations which incorporate exclusion of the LAA and those performed concurrent to
mitral valve intervention offer reduced risk of complications.
6. There is notable variability in follow-up methodology, with ECG and 1 day Holter
monitoring being the most utilised methods.
3.4.1. Concomitant Cardiac Surgery and AF
It has previously been established that the presence of preoperative AF in cardiac surgery
patients is associated with worse long term outcomes.(285-289) Thus, in patients presenting
for cardiac surgery, a concomitant intervention capable of returning the patient to sinus rhythm
may offer long term benefits to patients if it does not significantly increase the risk of the
planned surgery. The Society of Thoracic Surgeons reported in 2016 the results of CABG,
AVR, MVR, AVR + CABG and MVR + CABG reported over 201,553 operations throughout
the 2014 calendar year with in-hospital mortality ranging across these procedures from 1.7%
to 9.2%, permanent stroke ranging from 1.1% to 3.8% and reoperation ranging from 2.3%
to7.5% with the higher complication rates seen in CABG+MVR procedures.(290) While this
analysis does not represent a matched cohort of patients for comparison the mean complication
rates of mortality 4.3% (95% CI: 3.4-5.4), CVA/TIA 1.8% (95% CI: 1.3-2.5) and RTOT 4.3%
62
(95% CI: 3-6.1) identified here are not representative of a higher complication rate. As such,
the addition of an ablation procedure to common cardiac surgical procedures may improve
longer term results for patients through maintaining SRM while not incurring additive risk.
However, this inference would need further study to be evaluated.
Interestingly, a similar sub-group of patients has been evaluated in the recent guidelines
of the STS focused on the practice of the surgical ablation of AF in which they evaluated the
outcomes of ablation providing recommendations on ablation concurrent to AVR, CABG or
the combination AVR+CABG.(210) They provided a Class I, Level B recommendation in this
setting for the addition of concurrent ablation in consideration that it did not increase mortality
or morbidity. These same guidelines evaluated separately the ablation of AF concurrent to
mitral valve surgery arriving at Class I, Level A recommendations for the addition of ablation
in also seeing no additional mortality or major morbidity risk.(210) Both findings are in
keeping with results produced by this meta-analysis looking at an effective cohort that spans
both of these guideline recommendations. However, there is a difference in the level of
evidence identified in this meta-analysis with a paucity of level 1 evidence and those suggested
in the guidelines. Further, the guidelines were only able to assign Class II, Level C
recommendations to intervention performed on the LAA at the time of ablation, which has been
identified as a protective factor in this meta-analysis reducing the risk of mortality.(210) This
may warrant further analysis in future prospective trials such as are currently underway.(122)
3.4.2. Comparison to Studies of Concomitant Mitral Valve Procedures
This cohort provides an interesting comparator group to those analysed in Chapter 2 that was
comprised of studies with 100% mitral valve intervention. The cohort in this Chapter by
comparison is a sundry group of patients which have been undergone ablation concurrent to a
63
variety of other interventions; however, this still offers a population of interest as many patients
presenting to the cardiac surgeon with AF will not only have a single pathology in the mitral
valve for treatment, but potentially multiples pathologies in the form of other valves and
ischaemic heart disease. Thus, the surgeon is faced with a far more complex task of providing
a risk profile for a wider combination cohort of patients that may benefit from the elimination
of AF. Comparison of the results of this Chapter to Chapter 2 have revealed the following
information:
1. There is no significant difference in the SRM obtained in either cohort (p=NS)
2. The total number of complications trended towards significance, but did not reach
significant (6.9% [95% CI: 5.8-8.0] vs 15.1% [95% CI: 11.7-19.3], p=0.058)
3. The mortality rate in patients undergoing ablation concurrent to mitral valve
intervention was significantly lower than those in this mixed cohort of patients (0.92%
[95% CI: 0.5-1.4] vs 4.3% [95% CI: 3.4-5.4], p=0.001).
The STS database report described a notable increase of mortality when CABG is added to
mitral valve surgery with the in-hospital mortality going from 4.2% to 9.2% with mitral valve
replacement and from 1.0% to 4.3% in the mitral valve repair cohort.(290) This comparative
analysis has identified an increase in mortality factor in keeping with the mitral valve repair
cohort of the STS report, though of interest the repair rate in the mitral valve repair rate in
Chapter 2 was only 31.6% with a replacement rate of 56.7% and 11.7% undefined. Thus, the
significant increase in mortality and trend towards increased complications identified in this
series with procedures of increasing complexity is in keeping with larger database cohorts and
the addition of AF ablation does not indicate an increase in complication rate compared to the
reports of large cross sectional studies.(290)
64
3.4.4. Surgical Ablation Technique
This meta-analysis has identified no statistically significant difference in lesion set but suggests
a trend towards BAM being most efficacious. The ablation methodologies of C&S and
combination approaches were both found to offer superior results in SRM compared to RF and
the combination approach was found to be superior to Cryo.
The most balanced approach would be to offer ablation for any patient with a history
of AF – particularly the younger patients with paroxysmal AF and mitral valve disease. The
ideal ablation lesion set is not apparent from this analysis, however, a more extensive lesion
set, such as with a BAM trends towards offering the best results. The most ideal ablation
methodology suggested from this analysis would be C&S or a combination approach, but the
ultimate goal of any approach is the creation of contiguous lines of conduction block and this
should be the guiding principle of the proceduralist. The LAA should be excluded as part of
the procedure.
3.4.5. Clinical Implications
The addition of AF ablation to any cardiac surgical procedure should be considered in patients
presenting with a history of AF. While the complication rates are not insignificant in this series,
when compared to the STS data of a similar mixed cohort of patients the rates of the major
complications of mortality, stroke and returning to theatre for bleeding fall within the range
reported.(290) Utilising this as a comparison cohort represents the most comprehensive
published cohort of patients undergoing cardiac surgery and thus a robust comparator. As such,
this represents a cohort of patients of inherently higher risk presenting for cardiac surgery with
incidental AF that have the opportunity to undergo concurrent ablation in the context that it has
been identified that pre-operative AF when left untreated is associated with higher
complication rates and worse long term outcomes.(285-289) Encouragingly though the rate of
65
SRM achieved at mean follow-up 22 months was 79.6% representing a large reduction in the
burden of AF of this population, which has the potential translate to improved long term
survival in patients presenting with AF. However, such an assertion would need to be
confirmed by large prospective studies.
3.4.6. Study Limitations
The composition of included studies in this analysis represents a highly heterogenous cohort
of patients that are not specific to valvular intervention or coronary artery bypass intervention,
as such, they likely represent multiple mechanisms behind the development of AF within the
any given study. There is also variability in the quality of studies in this area with discrepancies
in recording of baseline variables, consistency of ablation method, lesion set and follow-up.
These variations serve to not only highlight the features present within the highest quality
studies, but also the need for standardisation of reporting within the field. Further, the included
studies identify a relative paucity of level 1 evidence within the literature of this area.
Additionally, as with any meta-analysis there may be confounding variables, measured and
unmeasured, which may impact the results of this analysis. Further, significant heterogeneity
exists in the literature in terms of surgical ablation procedure, follow-up length and
methodology, and this should be considered in interpreting the results. To manage the
significant literature heterogeneity with RCTs with <100 patients only those that did not have
statistically significant outcome measures were retained – this however does not retain
statistically significant control arms for comparison analysis. Finally, we are unable to provide
more extensive analysis on long term results of patient subtypes of AF given the lack of patient
level data available. Nonetheless, these identified limitations serve to highlight key areas of
attention for future studies.
66
3.5. CONCLUSIONS
Surgical ablation procedures for the management of AF carried out concomitant to other
cardiac surgical procedures are effective at maintaining SR on and off AAD at medium term
follow-up with relatively low complication rates compared to mixed cohorts of cardiac surgical
patients not undergoing ablation. The best results in SRM are obtained in younger patients with
paroxysmal AF and lower complication rates are seen when performed concomitant to mitral
valve intervention and with the exclusion of the LAA. However, it should be noted that this is
a highly heterogeneous cohort of patients at baseline with a wide variability in study follow-up
method and length. Further prospective studies with structured delineation of the wide variety
of cardiac surgical procedures that may be performed concurrent to ablation with a structured
follow-up methodology could confirm these results and help delineate the most ideal cardiac
surgical populations for the concomitant ablation of AF.
67
3.6. FIGURES AND TABLES
Figure 3.1
68
Figure 3.2 – Sinus Rhythm Maintenance Forest Plot
Forest plot for sinus rhythm maintenance achieved across the studies
69
Figure 3.3 – Sinus Rhythm Maintenance without AAD Forest Plot
Forest plot for sinus rhythm maintenance achieved without anti-arrhythmic drugs
70
Figure 3.4 – Complication Rates Forest Plots
71
Figure 3.5 – Mean Complication Rates
Mean complication rates from meta-analysis with error bars representing 95% CI
72
Figure 3.6 – Lesion Set Comparison
Comparison of lesion sets utilised across the studies
73
Figure 3.7 – Energy Source Comparison
Comparison of energy sources for ablation
74
75
76
77
78
79
80
Chapter 4
Isolated Surgical Ablation for Atrial Fibrillation – The Methodology, Safety and
Success in Sinus Rhythm Maintenance of Sole Cardiac Surgical Procedures
4.1. INTRODUCTION
The management of symptomatic atrial fibrillation (AF) remains a question of ongoing
importance with increasing hospitalizations placing significant burden on the healthcare
system.(29) Whilst several studies have shown no difference in the outcomes of rate or rhythm
control,(125,134) in practice, several patients remain highly symptomatic and have required
approaches to rhythm control. Catheter ablation is an increasingly utilised effectively in highly
symptomatic patients. Indeed, several randomised controlled studies have consistently
demonstrated that ablation therapy is superior to medical therapy alone.(141,142) However,
there remain several inherent limitations to the current catheter approach which include the
accurate evaluation of transmurality, the creation of contiguous lesions, the potential for
collateral injury, and the accepted need for multiple procedures.(151,171-174) In this setting,
there is interest in developing more definitive techniques. A meta-analysis of catheter ablation
procedures for AF has demonstrated that at long-term follow-up of more than three years,
single procedural success has been estimated at 53% and with multiple ablation procedures
(average 1.5), this figure is approaching 80% (176).
The gold standard for the surgical management of AF has been the Maze procedure
since first being described in 1991 by Cox et al.(36) This procedure has evolved since its
inception with the latest iterations of this procedure delivering 84% freedom from AF off AAD
and 90% freedom from AF with AAD at 2 years follow-up (188,189). Most commonly, this
procedure is undertaken in conjunction with other cardiac surgical procedures; however, the
experience with this procedure as a stand-alone surgical ablation for AF is limited (262).
81
Here we review the current literature to provide the current state of play with the use of
stand-alone AF ablation by undertaking a systematic literature review to evaluate the efficacy
and safety of the isolated surgical approach at maintaining sinus rhythm in patients with AF.
4.2. METHODS
A systematic review of electronic databases including Pubmed, Embase and Cochrane
Database of Systematic Reviews was carried out using search terms of “[(Atrial
Fibrillation[MeSH Terms] OR Atrial Fibrillation OR AF[Ti] OR Cox OR Maze OR Mini-
maze) AND (Unipolar [All fields] OR Bipolar[All fields] OR Radiofrequency OR Ablation
technique*[All Fields] OR Cryoabla*[All Fields] OR Cryomaze OR Cryo-maze OR
Cryosurgery[MeSH Terms] OR Cryosurg*[All Fields]) AND (Surgery[All Fields] OR
Surgical procedures, operative[MeSH Terms])]”. The search was conducted from January 1991
to April 2017. The results were imported into Endnote reference database for further review.
These searches were supplemented by reviewing the bibliographies or published studies and
reviews. Studies were accepted if they conducted a surgical ablation procedure for the
treatment of AF and reported outcomes on sinus rhythm maintenance (SRM). Studies that
performed other concurrent cardiac surgical procedures (valve repair or replacement, coronary
artery bypass grafting, etc) and sample size <100 patients were excluded. In addition, studies
that reported techniques of hybrid ablation were excluded (see definition below). Citations
related to animal studies and those published in journals in languages other than English were
not included. Selected publications were analysed for the following outcomes: SRM on and off
anti-arrhythmic drugs (AAD), block success if assessed, mortality, pericardial
effusion/tamponade, significant bleeding, atrio-oesophageal (AO) fistula, conversion to open
procedure if minimally invasive, stroke, length of stay, PPM requirement and 30-day mortality.
This study was registered with the PROSPERO international prospective register of
82
systematic reviews website (Registration Number CRD42017058255). The MOOSE
Guidelines for Meta-Analyses and Systematic Reviews of Observational Studies were utilised
as a guide for this study.
4.2.1. Definition
For the purposes of this study, isolated surgical ablation procedure was defined as a surgical
procedure, of any incision approach, for the sole management of AF, without concomitant
cardiac valve, coronary artery bypass or any other cardiac surgery procedure. Surgical
exclusion of the LAA appendage is part of the Maze procedure and has been included as part
of ablation procedures. Hybrid or convergent procedures in which an Electrophysiologist
undertaking catheter mapping and ablation study have also been excluded.
4.2.2. Statistical Analysis
Meta-analysis was performed using Stata/SE version 14. Means were weighted and standard
deviation of continuous variables or proportion and Exact binomial lower and upper confidence
limits of categorical variables were pooled across studies and analyzed using a random effects
meta-analysis model. This type of model was chosen to partially compensate for any inherent
heterogeneity. Linear regressions were performed to investigate associations between
outcomes of sinus rhythm maintenance and complications in each paper versus ablation
method, lesions set utilized, exclusion of LAA and use of RCT, weighting for number of
patients who completed follow-up. Assumptions of a linear model were upheld throughout.
Statistical significance was assigned to P value <0.05. All confidence intervals reported
represented 95% Confidence Intervals (CI).
83
4.3. RESULTS
4.3.1. Study & Patient Characteristics
Figure 4.1 shows the number of studies retrieved and review process of citations. At the
completion of the review process 9 studies remained for analysis. The 9 included studies (Table
4.1) were all non-randomised, case series published between 2003 and
2013.(186,198,202,206,318-322) The overall cohort size was 1082 patients with an average
study sample size of 120 patients. Three studies were multi-centre(318-320) with the remaining
being single centre, and two studies provided a comparator group for analysis (Concomitant
cardiac surgical procedures in addition to a Maze).(186,322)
Overall, the patient cohort comprised 32.3% paroxysmal, 22.7% persistent and 35.4%
long-standing persistent, with 9.6% unspecified. It should be noted that one study did not
present a breakdown of AF subtype.(321) The mean AF duration was 5.7 years (Range: 4.9-
9.6), although 2 studies reported median AF duration,(198,320) and 2 studies did not report a
pre-operative AF duration.(319,321) The mean patient age was 60.4 years (Range: 51.3-67.5),
the mean LA diameter was 4.7cm (Range: 4.3-5.2) and the mean left ventricular ejection
fraction was 53.6% (Range: 50-55). However, three studies did not report any
echocardiographic patient parameters.(186,202,321)
There was incomplete to no reporting of AF risk factors across the studies (Table 4.2).
Two studies reported the incidence of hypertension and diabetes,(318,321) two studies reported
NYHA class,(318,322) one study reported the incidence of hypercholesterolemia, COPD,
smoking and CVA/TIA(318) and one study reported the mean CHADS2 score of their cohort
and BMI.(320)
The mean follow-up duration was 17.4 months (Range: 6.8-64.8) utilising ECG (1),
Holter (2) or a combination of both (6) at variable time points. Of note, 3 of the studies provided
84
time interval follow-up periods instead of mean and in these cases the longest follow-up period
with >50% of the cohort was selected as the follow-up period.(198,206,322)
4.3.2. Surgical Ablation Procedures
4.3.2.1. Surgical Access
The surgical access, lesion set, energy source and operative characteristic are presented in the
Table 4.3. Procedural descriptions were provided in all studies. Surgical access was by
sternotomy in 2 of the studies,(186,322) mini-thoracotomy in 5 studies(198,318-321) and
Video Assisted Thoracoscopy (VAT) in 2 studies.(202,206)
4.3.2.2. Energy Source
Energy source varied amongst the studies with 5 studies utilising Radio-Frequency (RF) to
create the lesion set,(206,318-321) “Cut and Sew” in 1 study,(186) Cryo in 1 study,(198)
Micro-Wave (MW) in 1 study(202) and a combination of modalities (RF/Cut and Sew/Cryo)
in 1 study.(322) Of further note, all the RF devices used were bipolar in nature (they developed
a lesion through heating tissue between 2 electrodes) through either a bipolar clamp (in which
2 opposing jaws housed the electrodes of the device) or a bipolar rail (in which the 2 electrodes
were mounted on a device at a fixed distance and the device applied epicardially). Within the
studies that utilised RF, 2 studies used a bipolar clamp to create the lesions,(318,319) 1 study
used only a bipolar rail device(321) and 2 studies utilised a combination of clamp and rail
devices.(206,320) The RF Devices were manufactured by Medtronic (Minneapolis, MN) in 1
study,(318) Atricure (Cincinnati, OH) in 3 studies(206,320,322) and Estech (San Ramon, CA)
in 1 study.(321) The Cryo devices were produced by Medtronic (Minneapolis, MN) in 1
study(198) and Cooper Surgical (Trumbull, CT) in 1 study.(322) The MW device was
manufactured by Guidant (Indianapolis, IN).(202)
85
4.3.2.3. Lesion Set
The two major base lesion sets performed in the studies were either a Maze type lesion set or
Pulmonary Vein Isolation (PVI)/Box lesion set (Isolation of posterior left atrium including
pulmonary veins. A maze type lesion set was defined as any study that defined their procedure
as a “Maze” – in these studies bi-atrial lesions were performed with some variability in the
specific lesions placed. General examples of the lesion sets are displayed diagrammatically in
the Figure 2.2. In total there were 4 studies that utilised a Maze lesion set(186,198,206,322)
and 5 that utilised a PVI/Box lesion set.(202,318-321)
In consideration of the Maze classified procedures, the study by Ad et al described their
lesion set using a diagram – this displayed a box lesion encircling all the pulmonary veins, a
Left Atrial Appendage (LAA) line, a Mitral Valve Line (MVL), a line to the Superior Vena
Cava (SVC), a line to the Inferior Vena Cava (IVC), a line to the Tricuspid Valve Line (TVL)
and one towards the Right Atrial Appendage (RAA).(198) The study by Prasad et al did not
specifically described the procedural lines placed, beyond describing it as the “full Cox Maze
III” with the majority of these procedures performed by Dr James Cox himself.(186) The study
by Saint et al referenced a further study by Gaynor et al for procedural description – this
description interpreted from a diagram entailed bilateral PVI, floor line joining the inferior
aspect of the PVI, LAA line, MVL, SVC line, IVC line, TVL and RAA line.(187,322) The
study by Sirak et al utilised a unique “5 box” lesion set in order to recreate much of the same
lines as the traditional Maze procedure in thoracoscopic fashion – these boxes incorporated
bilateral PVI, roof line, floor line, coronary sinus line, anterior mitral trigone line, LAA line
and SVC line.(206)
In consideration of the PVI/Box lesion studies, additional lines may have been placed
in the PVI/Box studies and consisted of: Ligament of Marshall ablation in 2 studies;(318,319)
86
Roof and floor lines in 1 study;(320) and the addition of a line to the LAA, right atrial
appendage and superior and inferior vena caval lines in 1 study.(202)
4.3.2.4. LAA Exclusion
The LAA was excluded in all but 1 study,(321) with the method being stapler excision in 3
studies,(202,318,319) clipped in 1 study,(206) oversewn in 1 study,(198) a combination of
stapling, excision and clipping in 1 study(320) and 2 studies did not state the method of
exclusion.(186,322) In the study by Edgerton et al the LAA was excluded in 89% of cases, in
the study by Kasirajan et al it was excluded in 95% of cases and in the study by Pruitt et al it
was excluded in 85% of cases.(202,319,320) The operative time, cross clamp time and
cardiopulmonary bypass time across the studies were inconsistently reported.
4.3.2.5. Procedural Endpoints
Three studies reported procedural duration with a mean of 172.2 minutes (Range:82-253)
(202,318,321) and two studies reported the use of cardiopulmonary bypass and cross clamp
times with respective means of 149 minutes (Range:135-163) and 66.5 minutes (Range:40-93).
The endpoint of the surgical procedure was bidirectional block across lesions in 3
studies(202,206,318), exit block in 2 studies,(321,322) entrance block combined with no vagal
response on pacing stimulation in 1 study(319) and not stated in 3 studies.(186,198,320) The
study by Pruitt et al with the endpoint of bidirectional block described a low rate of success in
achieving this endpoint, but did not provide a figure for the incidence of success.(202) Three
studies reported the use of Direct Current Cardioversion (DCC); with the study by Beyer et al
utilising DCC at the end of the procedure if the patient was still in AF (47/100 patients).(318)
In the study by Kasirajan et al it was utilised (8/118 patients) either immediately post-
operatively or 3 months later.(320) In the study by Saint et al DCC was performed on patients
87
in AF at the time of surgery as part of the procedure.(322) Only the study by Beyer et al reported
the number of patients who achieved Sinus Rhythm (SR) by the completion of the procedure,
which was 86/100.(318)
4.3.3. Post-Procedure Follow-up
The mean follow-up duration was 17.4 months (Range: 6.8-64.8). Table 4.4 details follow-up
frequency and methodology. In studies where specific time points of follow-up were provided
instead of mean, the latest follow-up with >50% of the cohort was selected for analysis. In the
study by Edgerton et al the follow-up period was presented in days and this was converted into
months for statistical analysis in this review.(319) In the study by Edgerton et al the patients
received either a 1 day Holter monitor or long-term monitoring (defined as a 14 to 21-day event
monitor) at 6 months follow-up, but the number of patients in each category was not
specified.(319) In the study by Pruitt et al no specific time follow–up period was documented
but the majority of patients were between 12 and 36 months at last follow-up. In the study by
Kasirajan et al the specific time points of planned follow-up was not stated.(320) The follow-
up methodology was performed utilising ECG (1), Holter (2) or a combination of both (6). The
use of ECG was documented in 7 studies,(186,198,202,318,319,321,322) 1 day Holter monitor
in 7 studies,(198,202,318-322) 7 day Holter monitor in 3 studies,(198,202,206) Long Term
Monitoring (defined as the use of either a 14 to 21 or 30 day event monitor) in 2
studies(319,320) and Permanent Pacemaker (PPM) interrogation in 4 studies.(206,318-320)
Closer analysis of the studies utilising Holter monitor identified that either a 1 day or 7 day
Holter was performed at 6,12 and 24 months in the study by Ad et al,(198) a 1 day Holter at
the 6 month mark in the study by Edgerton et al,(319) a 1 day Holter monitor at 6,12 and 24
months in the study by Kasirajan et al,(320) a 1 day Holter monitor at 3 months, 6 months and
6 monthly thereafter in the study by Nasso et al,(321) a 1 day or 7 day Holter monitor in the
88
study by Pruitt et al,(202) a 1 day Holter monitor in the study by Saint et al,(322) and 7 day
Holter monitoring at 3,6,13 and 24 months in the study by Sirak et al.(206) Only 2 studies
provided information on success rate of maintaining SR in the different classifications of AF
(Paroxysmal/Persistent/Long Standing Persistent).(318,319)
Definition of recurrent arrhythmia was not uniform across the studies with 4 studies
utilising an atrial arrhythmia of >30 seconds duration,(198,206,320,322) 1 study stated to be
using the “HRS/EHRA/ECAS Expert Consensus Statement on Catheter and Surgical Ablation
of AF”(319) and the remaining 4 studies did not specifically define recurrent
arrhythmia.(186,202,318,321)
4.3.4. Outcomes
Overall SRM at final follow-up from this meta-analysis with and without AAD respectively
was 83.3% (95% CI: 75.2-91.5) and 65.5% (95% CI: 50.6-80.3; Figures 4.2 and 4.3). Analysis
of the 2 studies which provided success in AF types supplies a cohort of 214 patients with
88.9% SRM of 99 paroxysmal patients, 75.4% SRM of 61 persistent patients and 63% SRM
of 54 long-standing persistent. Within the study by Pruitt et al 10 patients went on to have a
formal Maze procedure after their initial procedure PVI and these patients were excluded from
the follow-up analysis.(202) Further, in the study by Kasirajan et al 6 patients went on to have
a catheter ablation procedure in the follow-up period to treat AF or atrial tachycardia within
the 3 month blanking period and they were considered failure of treatment in the articles
analysis.(320) Analysis of post-operative AAD usage found that it was not utilised uniformly
across the studies and the AAD utilised was also varied. Regarding AAD choice; the study by
Edgerton et al documented use of Amiodarone, Sotalol, Flecainamide or Digoxin at
discharge,(319) the study by Nasso et al utilised Amiodarone or Flecainamide at
discharge,(321) the study by Pruitt et al utilised Amiodarone(202) and the study by Saint et al
89
utilised Class I/III AADs.(322) Kasirajan et al aiming to discontinue AAD use by 6 months
with ideally 2 months free of AAD prior to the 6 month follow-up.(320) Nasso et al used AAD
in all and would discontinue if the patient was in SR at both 3 and 6 month follow-up
visits.(321) Saint et al used AAD in all and would discontinue it at 2 months post-procedure if
the patient was in SR.(322) The remaining 6 studies did not specify an AAD usage
protocol.(186,198,202,206,318,319)
4.3.5. Sub-Group Analysis
4.3.5.1. Lesion Set
The 2 primary base lesion sets employed across the studies were a Maze type lesion set or a
PVI/Box lesion set. The Maze type lesion set was associated with a higher rate of SRM both
off and on AAD than PVI/Box lesion sets reaching 79.2% (95% CI: 64.7-93.6) vs 54.1% (95%
CI: 42.2-67.1; p<0.01) and 93.8% (95% CI: 81.2-106.3) vs 73.8% (95% CI: 62.6-85.1; p=0.02)
respectively (Figure 4.4).
4.3.5.2. Energy Source
The energy source utilised to create lesions of cardiac conduction block had a wide variance
across the studies and comparison between these in shown in Figure 4.5A. The use of Cut and
Sew, RF or the use of a combination of modalities were superior to the use of MW for achieving
SRM (p<0.05) and the use of Cut and Sew, RF, Cryo or a combination of modalities were
superior to the use of MW for achieving SRM with AAD (p<0.001). The use of Cut and Sew
to create lines of conduction block was associated with a 79.6% (95% CI: 71.7-87.5) rate of
SRM off AAD and 95.9% (95% CI: 85.5-106.3) with AAD. The use of RF to create lesions
was associated with SRM of 67.1% (95% CI: 54.7-79.5) and 84.5% (95% CI: 79-90) with
AAD. The use of Cryo to create lesions was associated with SRM of 65.2% (95% CI: 57.3-73)
90
and 92.4% (95% CI: 82-104.8) with AAD. The use of MW energy to create lesions was
associated with SRM of 30.7% (95% CI: 22.8-38.6) and 42.1% (95% CI: 31.6-52.5) with AAD.
The use of a combination of all of these modalities in a procedure achieved a SRM of 76.2%
(95% CI: 68.4-84.1) and 91.1% (95% CI:80.7-101.5) with AAD.
Further analysis of RF devices conducted to compare rail and clamp devices (Figure
4.5B) shows that the joint use of clamp and rail devices in a single procedure was superior to
clamp alone (p<0.001) and clamp alone was superior to the use of rail alone (p<0.001) in SRM.
There was no statistical difference between RF modalities in SRM with AAD use. The SRM
of joint clamp and rail use was associated with SR off AAD of 87.6% (95% CI: 82.1-93.2) and
87.6% (95% CI: 80.3-95) with AAD, the use of clamp devices alone was associated with SR
off AAD of 60% (95% CI: 54.4-65.6) and 79% (95% CI: 71.7-86.4) with AAD, and the use of
rail devices alone was associated with SR off AAD of 40.2% (95% CI: 6.6-73.8) and 89.2%
(95% CI: 78.8-99.6) with AAD.
4.3.5.3. LAA Exclusion
Studies that excluded the LAA were compared to the study which did not for analysis of effect
on rhythm and incidence of stroke (Figure 4.6). There was no statistical difference in rhythm
outcomes on or off AAD but there was a statistically significant reduction in stroke risk with
exclusion of the LAA (0.7% [95% CI: 0.4-1.1] vs 2.0% [95% CI: 0.9-3], p=0.028).
4.3.6. Procedural Complications
The procedural complications reported in the studies are outlined in Table 4.5 and complication
rates Forest Plots in Figure 4.7. The overall complication rate was 7.4% (95% CI: 0.8-14) with
a 0.5% (95% CI: 0-2) mortality rate, 0.9% (95% CI: 0-2) CVA rate, 1% (95% CI: 0-3.4) rate
of conversion to an open procedure from a minimally invasive one, 0.6% (95% CI: 0-3.1) rate
91
of bleeding requiring return to operating theatre, 0.3% (95% CI: 0-3) rate of pleural or
pericardial effusion, a 3.2% (95% CI: 0-9.2) rate of PPM requirement, 0.2% (95% CI: 0-1) of
post-operative pulmonary embolus and a 1.4% (CI:0-8) rate of phrenic nerve injury. There
were no reports of atrio-oesophageal fistula in any of the studies. The mean length of hospital
stay was 4.3 days (CI:1.9-6.6), though of note 2 studies reported median instead of mean length
of stay (186,320) and 1 study did not report length of stay (321). The 0.5% mortality rate
consisted of a total of 5 deaths from the time of procedure to within the first 30 days following.
The causes consisted of 1 case of an avulsed LAA, 1 from an unknown cause (inconclusive
autopsy), 1 case of multi-organ failure from delayed cardiac tamponade, 1 case of acute
respiratory failure secondary to amiodarone toxicity and cause not specified in 1 case
(186,319,322). No statistically significant difference existed in complication rates between
procedures though a trend towards significance existed with PPM requirement (Figure 4.8).
4.4. DISCUSSION
Isolated surgical procedures for the treatment of AF are aimed to restore patients to SR in the
absence of structural heart disease. This systematic review and meta-analysis on the efficacy
and safety of these procedures demonstrates the following information:
1. Good medium to long term results in SRM are achievable, particularly with use of
AAD;
2. Higher success rates are seen with earlier stages of disease (paroxysmal > persistent >
long-lasting persistent AF cohorts);
3. Utilisation of a more extensive lesion set delivers superior results in SRM with or
without AAD;
4. Lesion sets developed through “Cut and Sew”, RF or a combination of modalities in a
single procedure are the most effective;
92
5. There is a difference in RF modalities in terms of efficacy in achieving SRM without
AAD. In this analysis using a combination of RF clamp and rail devices was most
effective, though this may reflect a more extensive lesion set employed. The use of RF
clamp alone was superior to rail alone in achieving SRM off AAD;
6. Exclusion of LAA during procedure does not provide a difference in rhythm control,
but it is protective of future CVA event;
7. An overall complication rate of 7.4% (95% CI: 0.9-14) has been noted and is driven
largely by a significant PPM requirement of 3.2% (95% CI: 0-9.2).
Importantly, this review is based on non-randomised, case series with only 2 studies providing
a comparator cohort for analysis. Of note, this analysis has highlighted a significant lack of
level 1, randomised control trials in this area. Nevertheless, the current review demonstrating
the safety and efficacy of the approach suggests that such an approach could be considered in
selected individuals and warrants further evaluation.
4.4.1. Catheter vs Surgical Ablation for Isolated AF
Patients with symptomatic AF that have failed medical therapy should be considered for
ablation to treat their arrhythmia.(262) Catheter ablation techniques have rapidly and continue
to evolve. Indeed, studies have consistently demonstrated the superiority of catheter ablation
over medical therapy for rhythm control. Not only has the evidence base for the use of catheter
ablation has expanded but there have now been several randomized controlled studies that have
evaluated a variety of techniques.(323-325) Nevertheless, there remain several inherent
limitations to the current catheter approach which include the accurate evaluation of
transmurality, the creation of contiguous lesions, the potential for collateral injury, procedural
93
duration, and the accepted need for multiple procedures. It is in this context that one could
consider the optimal method for ablation of AF and the potential to use a surgical technique.
Compared to a contemporary meta-analysis on catheter ablation of AF, which has
identified a single procedure SRM success of 53% and multiple procedure success of ~80% at
≥3 years of follow up of (176), this systematic review and meta-analysis has identified a
procedural success of 65% without AAD and 83% with AAD, which offers results superior to
a single catheter procedure and similar to multiple catheter procedures with the addition of
AAD use. However, this is delivered at the cost of higher procedural complication rate,
primarily that of increased PPM requirement.
The FAST Trial sought to answer this question through a randomised clinical trial
comparing catheter ablation and minimally invasive surgical ablation across 2 centres over a
12-month period. The enrolled patient cohort comprised 124 patients with AAD refractory AF
with left atrial dilatation and hypertension (42 patients [33%]) or failed previous catheter
ablation (82 patients [67%]). This trial included patients of both paroxysmal and persistent AF
with 63 patients randomised to catheter ablation and 61 patients to surgical ablation. At final
analysis it had identified superior results with surgical ablation in terms of SRM (65.6% vs
36.5%), but was once again at the cost of higher complication rate (34.4% vs 15.9%).(326)
There are however several questions that remain using such a technique with the best
results being achieved using a more extensive procedure, the need for an open-chest procedure,
the in hospital and recovery time for the individual and the longer-term maintenance of these
results. Nevertheless, this review suggests that further evaluation of the use of a surgical
approach is warranted.
94
4.4.3. The LAA Dilemma
The role of the LAA in arrhythmia and stroke risk is one of on-going scrutiny and interest.
There are increasing modalities of how to handle the LAA (excision, ligation, stapler excision,
clip) the most apt term to describe the fundamental intervention is LAA exclusion as in all
cases it is removed from the circulating blood volume. The LAAOS trial identified that not all
methods were equal in eliminating the LAA with stapling being superior to over-sewing, but it
was not able to determine a protective role in stroke prevention through its elimination.(123)
Further, recent studies have suggested that incomplete exclusion of the LAA is a predictor of
stroke or systemic embolization.(327) Within the most recent guidelines produced by the STS
examining the issue of the LAA at the time of surgery it was felt a reasonable intervention to
perform for the reduction in thromboembolic events; however, this recommendation only
reached Class IIA, Level C evidence.(210) The results of our analysis are consistent with these
aforementioned studies and guidelines, with primary method of exclusion being stapling,
suggesting a protective effect against CVA and as such should be considered at the time of
surgery.
4.4.4. Clinical Implications: Developing Superior Results
The most recent guidelines produced by the Society of Thoracic Surgeons focusing on the
clinical practice of surgical ablation for AF identified the standalone surgical procedure as a
reasonable intervention to undertake for symptomatic AF refractory to AAD or catheter
ablation, assigning Class IIA, Level B evidence.(210) Interestingly, they suggested the use of
the Cox Maze procedure over PVI as reasonable, assigning the same level of evidence.(210) In
consideration of these recommendations and the results identified in this meta-analysis the
most reasonable approach would be to only undertake such procedures where appropriate
experience and expertise are available to achieve the most efficacious and safe results.
95
This review has identified three important technical considerations for the clinician
embarking upon this treatment strategy for their patient. First, undertaking a more extensive
lesion set, such as a Maze procedure, is more likely to deliver increased rates of success in long
term maintenance of SR and therefore if this should always be considered when planning
surgery. Second, these results suggest that not all methods of creating lines of conduction block
are of equal efficacy and the incorporation of Cut and Sew or RF seems to deliver the best
results. Assessment of conduction block at either the time of surgery or post procedurally with
an electrophysiologist may help overcome this technology limitation. Third, exclusion of the
LAA is associated with protection against CVA and as such should routinely be excluded at
the time of operation. In these considerations, the isolated surgical procedure for treatment of
AF should be considered as part of the armamentarium for clinicians treating medically
refractory, symptomatic AF.
4.4.5. Study Limitations
There exists significant heterogeneity in terms of surgical procedure and follow-up method and
intervals among these included studies, and this should be considered in interpreting the results.
There is variability in the quality of studies with discrepancies in recording of baseline
variables, patient characteristics, consistency of ablation method and follow-up. These
variations serve to highlight the features of the highest quality studies and also the need for
standardisation of reporting within the field. Further, the composition of these studies is all
non-randomised, patient cohort studies. Additionally, as with any meta-analysis there may be
confounding variables, measured and unmeasured, which may impact the results of this
analysis. Finally, we are unable to provide more extensive analysis on patient AF subtype given
the lack of patient level data available. Nonetheless, these identified limitations serve to
highlight key areas of attention in future studies.
96
4.5. CONCLUSION
Stand-alone surgical procedures carried out for the treatment of AF have good results in sinus
rhythm maintenance with moderate results in freedom from AAD at medium term follow up.
The greatest results have been seen with more invasive procedures which typically have a more
aggressive ablation or electrical isolation strategy. However, it must be recognised that there is
significant variability in study follow up length and method due to the generally retrospective
nature of these studies. Further prospective randomised studies, with structured follow up are
required to confirm these results and help delineate specific procedure success for specific
patient populations based on AF type.
97
4.6. FIGURES AND TABLES
Figure 4.1 –Search Strategy
98
Figure 4.2 –Sinus Rhythm Maintenance Forest Plot
99
Figure 4.3 – Sinus Rhythm Maintenance without AAD Forest Plot
100
Figure 4.4 – Lesion Set Comparison
The lesion set comparison highlights the difference in SR on and off AAD of Maze vs
PVI/Box. Error bars represent 95% confidence interval.
101
Figure 4.5 – Energy Sources Comparison for Sinus Rhythm Maintenance
A) The rates of SR on and off AAD is shown for different energy sources. No statistically
significant difference was detected between lesions sources. B) Comparison made between
the different forms of RF ablation device types. Only statistically significant differences are
shown – the SR off AAD achieved between all comparisons. Error bars represent 95%
confidence interval.
102
Figure 4.6 – LAA Exclusion Comparison
The LAA excluded comparison highlights the difference in SR on and off AAD with CVA
rate. Only significant p-values are depicted with error bars representing 95% confidence
interval.
103
Figure 4.7 – Complications Rates Forest Plots
104
Figure 4.7 – Complications Comparison
105
106
107
108
109
110
Chapter 5
Hybrid Ablation (Surgical and Electrophysiological) of Atrial Fibrillation – The
Methodology, Safety and Success of Sinus Rhythm Maintenance
5.1. INTRODUCTION
Symptomatic atrial fibrillation (AF) remains a significant public health burden with increasing
hospitalizations.(29) In symptomatic patients, ablation has consistently been demonstrated to
be superior to medical therapy.(139,141)
Although multiple catheter-based procedures may be required for sinus rhythm
maintenance, a recent meta-analysis has demonstrated that this is an effective and durable long-
term therapeutic option.(176) Specifically, at long-term follow-up of more than three years,
single procedural success has been estimated at 53% and with multiple ablation procedures
(average 1.5), this figure is approaching 80%.(176) In comparison, surgical Cox- Maze
procedures have also documented sinus rhythm maintenance results of up to 96% with the Cox-
Maze III and 91% for the Cox-Maze IV, with respective results off anti-arrhythmic drugs
(AAD) of up to 83% and 78% respectively at up to 5.4 years follow-up.(186,188,189,322)
These apparent superior results may be due to the ability to obtain direct or videoscopic
visualization, better stabilization of the cardiac tissue substrate at the time of surgery and
confirmation of lesion transmurality, particularly when the lesions are cut and sewn in.
However, the greater invasiveness of the Cox Maze procedures is also accompanied by greater
peri-operative morbidity and potential major complications including death, bleeding and
stroke.(186,189)
More recently, a new hybrid approach for AF comprising initial surgical epicardial
ablation with concurrent or sequential endocardial catheter based ablation has emerged. This
strategy seeks to combine the strengths and minimise the drawbacks of the individual
111
approaches by balancing invasiveness and duration of ablation procedure with improved
delivery of ablative lesions sets followed by endocardial electrophysiological evaluation and
additional consolidative ablation if required. Initial studies have shown promising results of
sinus rhythm maintenance.(208) Here, we undertook a systematic literature review to evaluate
the efficacy and safety of this approach at maintaining sinus rhythm in patients with AF.
5.2. METHODS
This study was registered with the PROSPERO International prospective register of systematic
reviews website (Registration Number CRD42017059106). The MOOSE Guidelines for Meta-
Analyses and Systematic Reviews of Observational Studies were utilised as a guide for this
study. A systematic review of electronic databases including Pubmed, Embase and Cochrane
Database of Systematic Reviews was carried out using search terms of “[(Atrial
Fibrillation[MeSH Terms] OR Atrial Fibrillation OR AF[Ti] OR Cox OR Maze OR Mini-
maze) AND (Unipolar [All fields] OR Bipolar[All fields] OR Radiofrequency OR Ablation
technique*[All Fields] OR Cryoabla*[All Fields] OR Cryomaze OR Cryo-maze OR
Cryosurgery[MeSH Terms] OR Cryosurg*[All Fields]) AND (Surgery[All Fields] OR
Surgical procedures, operative[MeSH Terms])]” and “[(Atrial Fibrillation[MeSH Terms] OR
Atrial Fibrillation OR AF[Ti] OR Cox OR Maze OR Mini-maze) AND (Unipolar [All fields]
OR Bipolar[All fields] OR Radiofrequency OR Ablation technique*[All Fields] OR
Cryoabla*[All Fields] OR Cryomaze OR Cryo-maze OR Cryosurgery[MeSH Terms] OR
Cryosurg*[All Fields]) AND (Surgery[All Fields] OR Surgical procedures, operative[MeSH
Terms])]”. The search was conducted from January 1991 to March 2017. The results were
imported into Endnote reference database for further review. These searches were
supplemented by reviewing the bibliographies of published studies and reviews. Studies were
accepted if they conducted a surgical ablation procedure for the treatment of AF and reported
112
outcomes on SRM. Studies were accepted if they conducted a hybrid (joint surgical and
electrophysiology procedure) ablation procedure for the treatment of AF and reported
outcomes on sinus rhythm maintenance (SRM). Studies with different definitions of hybrid
(catheter ablation + AAD use, cryo + radiofrequency, two types of surgical ablation, etc) and
sample size <15 were excluded. Citations related to animal studies and those published in
journals in languages other than English were not included. After exclusions, the publications
were analysed for the following outcomes: SRM on and off anti-arrhythmic drugs (AAD),
block success if assessed, mortality, pericardial effusion/tamponade, significant bleeding,
atrio-esophageal (AO) fistula, conversion to open procedure if minimally invasive, stroke,
length of stay, permanent pacemaker (PPM) requirement and 30-day mortality. Figure 5.1
shows breakdown of the search, number of studies retrieved and review process of citations.
At the completion of the review process 22 studies remained for analysis.
5.2.1. Definitions
For the purposes of this study hybrid surgical ablation were defined as a combined surgical and
electrophysiological endocardial catheter-based approach for the sole management of AF.
Typically, a surgically conducted ablation is followed by catheter-based study either during the
same procedure or in the coming months. The aim is to confirm the integrity of the surgically
placed ablation lines with endocardial mapping, consolidate these ablation lesions sets as
required and perform additional substrate-based ablation. Surgical incision may be of
sternotomy, thoracotomy, video-assisted thoracoscopy (VAT) or sub-xiphoid/laparoscopic
trans-diaphragmatic approach. Surgical lesions were defined as any series of lesions placed on
the heart (Cut and Sew/Radiofrequency/Cryothermy/Micro-Wave) for the management of AF.
Additional, cardiac valve procedures, coronary artery bypass or any other cardiac surgery
procedure are permissible if performed concurrent to AF ablation. AF ablation procedures were
113
broadly categorised into Maze, Box and Pulmonary Vein Isolation. Surgical exclusion of the
LAA appendage has been included as part of ablation procedures for analysis.
The classification of AF was made according to HRS/EHRA/ECAS Expert Consensus
Statement on Catheter and Surgical Ablation for AF, with any study patients that were
described as being in permanent AF reclassified to long-standing persistent (LSP) AF.(262)
5.2.2. Statistical Analysis
Data was collected by the investigators and all statistical calculations were checked and further
analysis performed by an independent statistician. Meta-analysis was performed using R
version 3.2. Mean and standard deviation of continuous variables and Exact binomial lower
and upper confidence limits of categorical variables were pooled across studies and analysed
using a random effects meta-analysis model. Random effects meta-analysis of single
proportions was performed using the Freeman-Turkey Double arcsine transformation. Mixed
model meta-regression models were implemented to investigate associations between
outcomes of sinus rhythm maintenance and complications in each paper versus ablation
method, lesions set utilised, exclusion of LAA and use of RCT, weighting for number of
patients who completed follow-up. Assumptions of a linear model were upheld throughout.
Statistical significance was assigned to P value <0.05.
5.3. RESULTS
5.3.1. Study & Patient Characteristics
The 22 included studies (Table 5.1) were published between 2010 and 2017 and were all single
centre, non-randomised studies with 17 case studies (7 prospective, 6 retrospective, 4 not
stated) and 5 case series studies with a comparator group (3 prospective, 1 retrospective, 1 not
stated).(207,208,328-347) The studies that had comparator groups comprised Cut and sew
114
Cox-Maze (342), lone catheter ablation (331,338,343) and isolated surgical thoracoscopic
ablation (341). The total cohort contained 925 patients with a mean cohort size of 42 patients
(Range: 15-101) comprised of 103 paroxysmal (11%), 351 persistent (38%) and 471 long
standing persistent (51%) in terms of AF type. The mean length of AF was 5.1 years (Range:
2.42-8) with 3 studies reporting a median length. (337,342,346)
The mean patients age was 60.5 years (Range: 53.7-63.8), with mean LAD of 4.91cm
(Range: 4.3-5.3) and mean LVEF of 54.8% (Range: 47-59.3). Of note 3 studies did not report
LAD,(331,338,342) 1 study reported median LAD(345) and 3 studies did not report
LVEF(336,337,341) with 1 study providing median LVEF.(345)
The rate of AF related comorbidities reported across the total cohort included 59.3%
hypertension (Range: 0-83), 12.1% diabetes (Range: 0-31%), average BMI of 29 (Range: 26.8-
32.8) and 18.5% having had a prior failed catheter ablation (Range: 0-76). Of note 4 studies
did not report on rate of hypertension,(333,336,344,345) 7 studies did not report on rate of
diabetes,(207,333,336,339,340,344,345) 5 studies did not report average
BMI(208,333,336,343,345) with 2 reporting number of patients in the obese range,(328,334)
and only 9 studies reporting the rate of prior catheter ablation.(207,329,330,334,337,338,340-
342,347)
The mean follow-up duration achieved across the studies was 19 months (Range: 6-
128.4) Of note, 5 of the studies provided only time interval follow-up periods and in these cases
the longest follow-up period with >50% of the cohort included was selected as the follow-up
period.(331,334-336,340)
115
5.3.3. Surgical Ablation Procedures
5.3.3.1. Surgical Approach
The surgical access site, surgical lesion set, energy source and other basic operative
characteristics are detailed in Table 5.2 and Figure 5.2. Both surgical access and ablation lesion
sets varied amongst the studies with one study necessitating interpretation of lesion set from
their published diagram.(335) Surgical access was via unilateral or bilateral VATS
(videoscopic assisted thoracoscopic surgery) in 13 studies, (207,208,328-330,337-341,343-
345) sub-xiphoid/trans-diaphragmatic laparoscopy in 6 studies, (331,334-336,346,347)
sternotomy in 2 studies (332,333) and mini-thoracotomy.(342) The fundamental ablation lesion
sets included pulmonary vein isolation (PVI) for 11 studies, (207,333-337,339,341-343,346)
box isolation that encompasses isolation of all pulmonary veins and the posterior left atrium in
10 studies (208,328-331,338,340,344,345,347) or a Maze procedure in 1 study.(332)
Additional lesions included roof line,(207,333-337,339,341,343,346) floor line,
(334,337,339,341) mitral isthmus line, (207,330,333,334,336,337,339,341,343) ganglionated
plexi ablation,(208,329,335,339-343,345-347) ligament of Marshall
ablation,(329,331,332,334-337,339,343,345-347) between the right inferior pulmonary vein
and inferior vena cava,(207,334-336,346) superior vena cava,(207,330,343,345) inferior vena
cava,(207,345) bicaval line,(207,330) between the right inferior pulmonary vein and Thebesian
valve,(336) oblique sinus,(336) Waterson’s groove line(334,336) and complete posterior wall
ablation.(347)
5.3.3.2. Energy Utilized
Twenty studies utilized Radio-Frequency (RF) for surgical ablation energy
source(207,208,328-331,334-347) while two used cryothermy (332,333). Devices for surgical
ablation varied among the studies which utilised RF; 3 utilizing the monopolar Cobra Adhere
116
XL device(Estech, San Ramon, CA, USA),(208,328,340), 1 utilising the bipolar Cobra Fusion
150(Estech, San Ramon, CA, USA)(344), 9 the Atricure Synergy bipolar ablation system
clamp with or without bipolar ablation rail device (Coolrail)(Atricure, West Chester, OH,
USA),(207,329,330,337-339,341,343,345) 6 the nContact ablation system (nContact Surgical,
Morrisville, NC, USA),(331,334-336,346-348) and 1 primarily the Medtronic Gemini-X
ablation system (10% of the cases with the Atricure system)(Medtronic, Minneapolis, MN,
USA).(342) Of the 2 studies utilising cryo; 1 used the Frigitronics cryothermy system
(Frigitronics, Cooper Surgical, Shelton, Connecticut),(333) and 1 the CryoFlex Medtonic
system (Medtronic, Minneapolis, MN, USA).(332) Figure 5.2 shows the RF devices used in
these studies.
In consideration of energy applications 11 studies reported on amount of energy
delivered per lesion (9 RF studies and 1 cryo). Regarding the RF studies; 2 studies reported 4
to 6 applications of RF,(207,330) 2 studies reported >5 applications,(329,339) 3 studies
reported 3,(337,341,342) 1 study reported 3-7 applications,(343) 1 study reported >2
applications(208) and 1 study reported >1 minute time of application.(345) The cryo study
which provided data on application length reported 2 minutes of cryo time per lesion.(332)
5.3.3.3. LAA Management and Additional Approaches
The LAA was surgically managed in 11 of the studies; it was stapled off in 5
studies,(207,337,338,342,343) clipped off in 3 studies,(329,330,339) a combination of stapled
or clipped in 2 studies(341,345) and method not stated in 1 study.(332) Among the studies that
surgically excluded the LAA the number of patients within each study that had this intervention
ranged from 27% to 100%.
Additional procedures included; removal of the fat along Waterson’s groove to
facilitate disruption of the ganglionic plexi In the study of Muneretto et al,(208) the addition
117
of concomitant surgery consisting of valve replacement in 19 patients, CABG in 7 patients and
combination of these procedures of 8 patients in the study of Eisenberger et al(332) and 1
patient underwent a concurrent minimally invasive mitral valve repair in the study by Kiser et
al.(336) In the study by Pison et al, the lesions were placed in a stepwise fashion with additional
lines only added when further testing with isoprenaline was able to demonstrate inducible
sustained AF (207). The commonly employed lesions are diagrammed in Figure 5.2.
5.3.3.4. Endpoints
The end point of the surgical procedure was conduction block in 12 studies
(208,328,329,334,335,337,339,341-345) with 5 of those being endocardial block confirmation
by the electrophysiologist,(208,328,334,343,344) completion of the planned lesion set in 9
studies(207,330-333,338,340,346,347) and visual confirmation of contiguity of ablation lines
in 1 study.(336) The mean procedure time was 218.2 minutes (Range: 85-450.1) with 10 studies
reporting total procedural time,(207,328,330,331,335,337,338,340,341,343) 6 studies
reporting their surgical procedure time(208,329,336,339,344,347) and 6 studies not reporting
total or surgical procedural time.(332-334,342,345,346)
5.3.4. Catheter-based Electrophysiological Procedures
5.3.4.1. Timing after Surgery
Nine studies undertook immediate follow on to endocardial catheter-based ablation in a hybrid
theatre,(207,330,331,334,336-338,340,341) while nine studies utilized a staged approach with
a variable interval period ranging from 4 to 167 days before the endocardial
procedure,(208,328,329,332,333,339,342-344) and 4 studies utilising a combination of both
sequential and staged approaches (Table 5.3).(335,345-347) Of note, Lee et al only performed
catheter-based procedure in patients with recurrent AF more than 3 months after the initial
118
surgical procedure.(342) In this study, only 23 out of the cohort of 25 underwent further
electrophysiological study and ablation, with 2 patients declining further intervention given
symptomatic improvement (342). Further, the study of Gersak et al had 5 patients that declined
the endocardial ablation stage of the procedure (335).
5.3.4.2. Endocardial Procedure Type
Seventeen studies utilised RF energy for catheter ablation; 6 studies reported use of irrigated
Thermocool [Biosense Webster, CA, USA],(329,332,335,339,345,346) 2 reported use
Thermocool with no comment on being irrigated,(340,341) 2 reported utilising irrigated
Navistar Thermocool [Biosense Webster, CA, USA],(330,347) 2 Thermocool Navistar with no
comment on being irrigated,(328,333) 2 reported use of SmartTouch [Biosense Webster, CA,
USA],(338,344) 1 a combination of either irrigated Navistar Thermocool or non-
cooled/irrigated Blazer II [Boston Scientific, MA, USA],(334) 1 a combination of an irrigated
RF catheter from either Biosense or Boston, but without specification of catheter model (343)
and 1 study used a combination of RF Thermocool (no comment on irrigated) and cryo-balloon
technology(Arctic Front, Arctic Front, Cryocath, Montreal, Quebec)(207). Five studies did not
state the device employed for catheter procedure.(208,331,336,337,342)
Fourteen studies reported the use of mapping with 12 studies using the Carto system (Biosense
Webster, CA, USA),(208,328-330,332,333,336,338,339,344,345,347) 2 studies utilised either
Carto or Ensite NavX (St Jude Medical, Austin, TX, USA) (334,343), and 8 studies did not
specify a mapping method.(207,331,335,337,340-342,346)
Catheter ablation strategy differed among the studies apart from all studies ensuring the
pulmonary veins were isolated (Table 5.3). Eleven studies added a cavo-tricuspid isthmus line
as required,(207,329,330,334,336,338-341,344,345) 6 studies added a mitral line as
required,(329,330,340,342,345,346) 4 studies added a coronary sinus line as
119
required,(331,334,343,347) 3 studies added a roof line,(335,342,346) 4 studies performed an
additional ablation line of the SVC with or without IVC(335,341,343,346) and 1 study
performed GP ablation.(344) In addition, 6 studies procedures carried out Complex
Fractionated Atrial Electrogram (CFAE) ablation and this was utilised on a case by case basis
if SR was not able to be obtained.(330,331,334,338,343,345) Four studies did not perform
additional lines of ablation other than consolidating the surgically placed lines as required.
(208,328,332,333,337) Two studies did not provide a specific written description of catheter
ablation lines, necessitating interpretation from diagrams in the respective studies (335,348).
5.3.4.3. Procedural Endpoints
At the beginning of the procedure, 16 studies reported checking for conduction block with
variable results in gap free lesions (Range: 0-91.1%).(207,208,328-333,338-345) Post
procedural block confirmation was described in all studies with 11 studies reporting
specifically to what extent they were successful.(328-330,332-335,339-341,344) The
remaining studies listed achieving block as the end point of the procedure, but did not specify
how successful they were.(207,208,331,336-338,342,343,345-347) Lee et al only tested for
conduction block if the patients were still in AF, and this occurred in seven (29%) patients with
a further 2 being offered catheter ablation, but declined (342). Further testing with AF induction
was performed with 4 studies using pacing manoeuvres (208,328,339,342) and 4 others using
isoproterenol challenge (207,336,341,343). In the study by Kiser et al, if SR not achieved at
end of EP procedure DC cardioversion was performed and mapping of triggers using Isuprel
was performed (336). In the study by Mahapatra et al at the end of ablation procedure
isoproterenol was administered at increasing quantities every 2 minutes or until MAP
<60mmHg. If atrial flutter induced it was mapped and ablated. If the patient returned to SR at
this point PV isolation and block confirmed. Then isoproterenol was administered again and
120
further targeted ablation of triggers carried out. If the patient was then in AF a multipolar
catheter was used to check PV isolation and CFAE performed. After 30 minutes if SR was still
not obtained DC cardioversion was performed (343). In the studies by Pison et al and La Meir
et al, if after the completion of PV isolation, the patient reverted to SR, coronary sinus pacing
was carried out in an attempt to induce AF. If AF could not be induced with pacing,
isoproterenol was infused and linear lesions applied if AF became inducible with this
(207,341). Only 5 studies reported rhythm status of their patients at the end of the procedure
(207,328,334,336,344).
5.3.4.4. Procedural Time
The procedural time for the catheter-based component was not uniformly reported with 13
studies not reporting any procedural time indices.(207,328,333-335,337,339-342,345-347)
Seven studies reported time of fluoroscopy use with mean of 18.3 minutes (Range: 7.3-
36.2)(329-332,338,343,344) and 5 studies providing length of procedure with mean 112.3
minutes (Range: 18-216).(208,329,332,336,344) The study by Gaita et al reported that no
fluoroscopy was used (333).
5.3.5. Post-Procedural Follow-up
Follow-up methodology consisted primarily of the use of implantable loop recorders for 5
studies,(208,328,335,345,346) Holter monitoring (1 or 7 day) in 16 studies(207,329-334,336-
341,343,344,347) and 12 lead ECG in 1 study.(342) However, the follow-up frequency was
variable as detailed in Table 5.4. The overall mean follow-up duration for all studies was 19
months (Range: 6-128.4) after initial surgical procedure with follow-up completed by 93.3%
of the total cohort. Definition of recurrent arrhythmia varied across studies with 13 defining
AF recurrence as any atrial tachyarrhthmia lasting >30 seconds,(207,329-332,334,337-
121
339,341,343,345,347) 3 studies as the total burden of AF on a monthly basis being greater than
0.5%(208,328,346) and not defined in 6 studies.(333,335,336,340,342,344)
5.3.6. Outcomes
Overall Sinus Rhythm Maintenance(SRM) at final follow-up from this meta-analysis was
79.4% (95% CI: 72.4-85.7) and 70.7%(95% CI: 62.2-78.7) without AAD (Figure 5.3 and 5.4).
Of note SRM off AAD was not reported in 2 studies.(328,334)
5.3.6.1. Univariate Predictors of Outcome
Univariate analysis was conducted across the variables of age, AF type, LA diameter, AF
duration and follow-up length with none shown to be statistically significant in predicting
SRM.
5.3.6.2. Sequential vs Staged
There was no significant difference in SRM with or without AAD in comparison of the studies
utilising a staged procedure, sequential procedure or a combination of both approaches (SR
with AAD: Staged 85.3% [95% CI: 77.1-92.1], Sequential 75.5% [95% CI: 61.6-87.2] and
Combined 74.6% [95% CI: 59.6-87.2]) (SR without AAD: Staged 75.9% [95% CI: 64.4-85.9],
Sequential 70.4% [95% CI: 51.9-86.2] and Combined 60.4% [95% CI: 51.1-69.3]).
5.3.6.3. Block Analysis
Analysis was performed comparing those studies that confirmed 100% block by the end of the
EP study to those that achieved <100% block with no statistically significant difference
identified. (SRM with AAD: Block 100%: 87.6% [95% CI: 78.4-94.7], Block<100%: 73.1%
[95% CI: 60.2-84.4]) (SRM without AAD: Block 100%: 81.5% [95% CI: 72.9-88.8],
Block<100%: 62% [95% CI: 39.4-82.3]) (Figure 5.5).
122
5.3.6.4. Surgical Approach
Comparison of surgical approach comparing laparoscopic trans-diaphragmatic approach, mini-
thoracotomy, sternotomy and VATS showed no significant difference in SR with or without
AAD.
5.3.6.5. Surgical Lesion Set
No significant difference was found between Box, PVI or Maze in terms of SR with or without
AAD (SR with AAD: Box: 77.9% [95% CI: 66.2-87.8], PVI: 79.3% [95% CI: 69.7-87.6],
Maze: 91.4% [95% CI: 79.4-98.9]) (SR without AAD: Box: 63.3% [95% CI: 50.4-75.2], PVI:
76% [95% CI: 63.6-86.6], Maze: 85.7% [95% CI: 71.9-95.7]).
5.3.6.6. Energy Type
No significant differences existed between Bipolar RF, Unipolar RF or Cryo in terms of SR
with or without AAD (SR with AAD: Bipolar RF: 84% [95% CI: 75-91.4], Unipolar RF: 72.9%
[95% CI: 58.9-85], Cryo: 77.9% [95% CI: 42.7-99.2] and Combo: 86.5% [95% CI: 73.6-95.8])
(SR without AAD: Bipolar RF: 78.1% [95% CI: 67.8-87], Unipolar RF: 56.8% [95% CI: 40.5-
72.5], Cryo: 68.6% [95% CI: 29.4-96.9] and Combo: 82.1% [95% CI: 57.3-97.9]).
5.3.6.7. Device Type
No significant difference was identified between devices in terms of SR with AAD; however,
the use of the Atricure Synergy system was superior to the nContact system in SRM without
AAD (Atricure: 82% [95% CI: 72.5-90] vs nContact: 68.6% [95% CI: 29.4-96.9],p=0.01)
(Figure 5.5).
123
Further, analysis of the catheters utilised for ablation found no statistically significant
difference in outcome between irrigated, non-irrigated and contact force sensing catheters in
terms of SRM on or off AAD.
5.3.6.8. Left Atrial Appendage
No significant difference in SR with AAD existed between the groups who excluded the LAA
as opposed to those who did not, but a superior result in SR without AAD was found with the
LAA exclusion group (79.5% [95% CI: 71.2-86.8] vs 55.8% [95% CI: 41.4-69.8], p<0.001)
(Figure 5.5).
5.3.6.9. Ganglionated Plexi Ablation
No significant difference in SR with or without AAD was identified between the studies that
ablated the ganglionated plexi and those that did not.
5.3.7. AAD management
Analysis of post-operative AAD usage found 6 studies initiated an AAD post-operatively,
without stating which specific drug, and left AAD direction to the discretion of the referring
physician,(208,333,334,336,340,341) 6 studies utilised amiodarone as their post-operative
AAD of choice with varying protocols,(335,339,343,344,346,347) 4 studies initiated the
patients pre-operative AAD as soon as possible,(207,330,331,338) 2 studies discharged
patients without any AAD post-operatively,(329,332) 1 study continued whichever AAD the
patient was on throughout the admission(337) and 3 studies did not state an AAD
protocol.(328,342,345)
124
5.3.9. Procedural Complications
Table 5.5 outlines the various procedural complications from the 22 included studies with
complication rates forest plots in Figure 5.6. Ten studies reported mean post-operative length
of stay providing a total mean of 5 days (Range: 3.4-7.6).(207,208,328-330,334,339,342-344)
Of note 5 studies reported a median length of stay(331,337,340,341,345) and 7 studies did not
report on hospital length of stay.(332,333,335,336,339,346,347) The mean total complication
rate across the studies was 6.5% (95% CI: 3.4-10.2) comprising of a mean mortality rate of
0.2% (95% CI: 0-0.9), mean stroke rate of 0.3% (95% CI: 0-1.1), mean return to OT for
bleeding (RTOT) of 1.6% (95% CI: 0.6-3.0), post op PPM rate of ~0% (95% CI: 0-0.5),
conversion to sternotomy rate of 0.3% (95% CI: 0-1.1), mean incidence of AO fistula of ~0%
(95% CI: 0-0.5) and phrenic nerve injury either temporary or permanent of 0.3% (95% CI: 0-
1.1). Mortality was considered in the first 30 days of either surgical or catheter procedure in
which time frame there were 8 deaths; 3 were secondary to AO fistula, 1 secondary to bleeding
and cardiac tamponade, 1 sudden cardiac death, 1 secondary to CVA and the remaining 2 not
determined. One of the cases of death recorded as secondary to AO fistula resulted in the death
at 33 days, but this was still recorded in the mortality rate of this analysis due to the time frame
remaining pertinent to the complication rate.(335) In total, there was 4 AO fistulae reported
across the series with 3 mortalities resultant and 1 being repaired surgically, though this patient
died 8 months later from a cerebrovascular event. Of note all AO fistula occurred in studies
that utilised mono-polar RF. There was not routine reporting of post-operative pain in any study
though the study by Kumar et al reported 1 patient representing with recurrent chest pain after
a few weeks(338) and the study by Pison et al reported 1 patient representing with pain from a
VATS working port.(207) The study by Gehi et al reported a significant improvement in the
Canadian Cardiovascular Society Severity of AF [CCS-SAF] as being significantly improved
125
to their final follow-up of 12 months,(334) but no other study reported on this or resumption
of normal activity.
5.3.9.1. Sub-Group Analysis of Complications
Comparison across the studies did not reveal any statistically significant predictors of
major complications.
5.4. DISCUSSION
Hybrid (joint surgical and catheter-based) ablation for AF is an emerging technique purporting
to improve rhythm control outcomes in patients with this debilitating arrhythmia. This
systematic review and meta-analysis on the standalone hybrid ablation approach in an overall
cohort of 925 (89% non-paroxysmal AF) subjects demonstrates the following new information:
1. Favourable overall mean sinus rhythm maintenance of 79% and 71% with and without
AAD at mean follow-up of 19 months;
2. Ensuring lesion completeness at the end of the endocardial procedure with electrical
testing for bi-directional block is indicative of better outcomes though this observation
did not reach statistical significance;
3. No difference in rhythm outcome on whether the catheter ablation was conducted
immediately following or staged for a later date;
4. The use of the Atricure Synergy bipolar RF ablation system was associated with
superior outcomes in terms of SRM off AAD when compared to the use of the nContact
unipolar RF ablation system;
5. The best results in terms of SR without AAD are achieved when the LAA is excluded;
6. Comparable risk profile to that described with the initial use of catheter ablation alone
with a complication rate of approximately 6.5%;
126
Whilst in its infancy, the approach to potentially improve the longer-term durability of ablation
results and reduce the need for multiple procedures, is of potential importance. A recent study
based on a US administrative data set has suggested that a 1% reduction in repeat procedures
is associated with a $30M/year saving to the health care system.(349) The current data
demonstrates the safety and efficacy of the approach when compared to catheter ablation alone
and suggests that the technique warrants further evaluation.
5.4.1. Outcomes and Complications of Hybrid Ablation
The mean overall sinus rhythm maintenance with AAD of 79% and without AAD of 71% at
mean 19-month follow-up with the standalone hybrid ablation approach in largely non-
paroxysmal AF patients (89% of overall cohort) appears to be favourable in comparison to
other established approaches although the heterogeneity of the included studies must also be
taken into consideration. Recent meta-analysis of the outcome of catheter ablation approach
has demonstrated sinus rhythm maintenance at 2-year follow-up to be 60% after single ablation
procedure and 80% following multiple procedures.(176) Further, single procedural catheter
ablation outcome for those with non-paroxysmal AF was only 42% at 2-year follow-up (176).
The success rates for standalone surgical Cox-Maze procedure for AF were reported to range
from 79 to 99% from earlier series although the follow-up methods were highly variable with
most studies relying on rhythm status or symptoms at last follow-up visit and not utilizing
extended monitoring.(312) However, contemporary reports of the standalone Cox-Maze
approach with stricter follow-up regimens have demonstrated sinus rhythm maintenance off
AAD of 76 to 82% at 1 to 2 year follow-up from patient cohorts with 50 to 70% having non-
paroxysmal form of AF.(188,189,322)
127
In comparison to 2 catheter ablation procedures the hybrid procedure offers the
advantage of a surgical epicardial ablation placing broad, continuous lesions to isolate the
pulmonary veins in a relatively quick time frame. This can be done, in addition to placing a
number of additional lines and LAA exclusion, in between 80 to 200 minutes. Therefore, a
hybrid ablation approach provides a surgically placed broad, epicardial lesion substrate at the
beginning of the EP procedure, which allows mapping to focus on block confirmation and
targeted endocardial ablation to ensure lesion contiguity and electrical isolation. Additionally,
this combined approach can see reduced procedure times, with the study by Muneretto et al
completing the surgical procedure with a mean time of 80±7minutes, mean EP procedural time
of 18±2 minutes and an additional 25±4 minutes if additional ablation was required beyond
confirming block.(208) Such times suggest faster procedures are possible with this approach
compared to a multiple catheter ablations. Therefore, the hybrid ablation approach appears to
be superior to single catheter ablation procedure and comparable to multiple catheter ablation
procedures and surgical Cox-Maze procedure.
The overall rate of complications with the standalone hybrid approach from this review
was 6.5%. That no specific major complication had a statistically significant association with
energy source, device, lesion set or procedure type is most likely reflective of the overall low
number of complications causing the comparisons to be underpowered. However, it is worth
noting that all incidence of AO fistula occurred with use of nContact unipolar RF device, which
suggests an area for further study. In consideration of comparison of the complication rates of
the hybrid procedure to other interventions for AF recent systematic review on complications
from catheter-based AF ablation procedures indicated a decline over recent years of major
complications to 2.6%.(280) Data from the worldwide surveys have also demonstrated a
decline in major complication rates from 6% between 1995-2012 to 4.5% between 2003-
2006.(350,351) In addition, a recently published retrospective cohort analysis from the
128
Californian Medicare beneficiaries undergoing catheter ablation for AF between 2007-2009
has shown a major procedural complication rate of 4.1%.(352) The major complication rate for
standalone surgical Cox-Maze procedure for AF has been reported at 7.1% although the rate is
much lower with the latest iteration of the Cox-Maze IV procedure at 2 to 4%.(189,322)
Therefore, the hybrid approach appears to be comparable to other established treatment
modalities with similar major complication rates. Interestingly, both the hybrid and Cox-Maze
approaches have been shown to result in a significant proportion of patients needing permanent
pacing post-procedurally at 1.5% and 7-9% respectively, a complication not usually seen with
catheter-based ablation procedures.(189,322) Further, patients undergoing the hybrid approach
seem to require a shorter hospital stay of 5 days as compared to the Cox-Maze procedure at 7-
8 days,(189,322) although this remains substantially longer than in the current catheter based
approaches. The hybrid ablation approach appears to provide the ‘best of both worlds’ with
less invasiveness than the Cox-Maze procedures and seemingly comparable outcomes to
catheter-based approach that often requires multiple procedures.
5.4.2. Lessons from Early Experiences
The move to a hybrid procedure from a surgical standpoint represents the potential for a less
invasive surgical procedure, than that of the tradition Cox-Maze procedure, while potentially
offering equivalent results. The traditional Cox-Maze procedure is typically done through a
median sternotomy and requires cardiopulmonary bypass to complete, thus increasing the
invasiveness of the procedure through incision coupled with the sequelae of a cardiopulmonary
bypass run. While advances in surgical techniques have been progressing the evolution of the
Cox-Maze procedure to be completed via mini-thoracotomy or equivalent lesions through
VATS approach,(198,206) the hybrid procedure adds an additional dimension through
assessment of the endocardium.
129
This review identified three important technical considerations with this novel
standalone hybrid ablation technique. First, ensuring bi-directional conduction block as the
procedural endpoint following endocardial ablation is indicative of a superior rate of sinus
rhythm maintenance off AAD of 81.5% when 100% block was achieved as to 66.5% in those
with less than 100% confirmation of conduction block at 19-month follow-up. While this
observation did not reach statistical significance, it suggests a strong trend towards significance
which may be realised in the observations of future prospective studies. This observation is not
surprising, as experience from catheter-based ablation procedures has shown that recovery of
pulmonary vein conduction was associated with early recurrence of atrial arrhythmia
(353,354). The hybrid approach allows for the seeming advantage of direct visualization of the
lesion sets to achieve more durable ablation. Given our findings above, efforts should be made
to ensure the endpoint of complete isolation is obtained, to justify the greater invasiveness of
the hybrid procedure. Second, this review has not been able to identify a statistically significant
difference in outcome when the hybrid procedure is performed sequentially in the same setting
as opposed to the staged approach with follow on catheter based ablation after an interval of
weeks to months. There are potential advantages offered in each procedure type. Sequential
procedures offer the potential for immediate identification of lesion gaps which can be treated
at that time and thereby provide the ability for substrate remodelling to take place without
delay. Staged procedures offer the potential for complete lesion formation and regression of
oedema to aid identification of lesion gaps once any oedema has regressed, and further, such
procedures do not necessitate a hybrid theatre to be available. Further work is needed to
delineate the differences in the two approaches. Third, the use of the Atricure bipolar RF system
offered superior results in SRM off AAD at final follow-up, being statistically significant when
compared to the nContact unipolar RF system. Further of note to this point is that in this series
all atrio-esophageal fistulae occurred with the use of the nContact unipolar RF devices
130
suggesting that the use of bipolar RF devices may be protective against this serious
complication.
5.4.3. Clinical Implications
The standalone hybrid ablation approach for AF appears promising with this systemic review
and meta-analysis demonstrating favourable outcomes without significant additional
procedural related complications. This novel technique may fill the gap in our current
armamentarium of ablating patients with more persistent form of AF given that the optimal
catheter ablation approach for this subgroup of patients remains unclear (355). The
invasiveness and complexity of the Cox-Maze III procedure might have limited its uptake
despite excellent sinus rhythm maintenance results (186). In contrast, the hybrid ablation
procedure is less invasive, does not require cardiopulmonary bypass and allows for shorter
hospitalization. As with all novel techniques, there remains room for improvement with further
reduction in complication rates following greater experience, improvement in ablation devices
and ablation techniques. This hybrid ablation approach can be potentially extended to AF
patients undergoing concomitant valvular or coronary artery bypass graft surgery in place of a
traditional Cox-Maze procedure, to reduce cardiopulmonary bypass time and associated
complications. Recently, novel risk factors for AF have been identified and aggressive
targeting of risk factors for AF has been shown to reduce AF burden and recurrence following
catheter ablation (15,101,356,357). Incorporation of these strategies should not be neglected in
patients undergoing invasive ablation procedures.
5.4.4. Study Limitations
We are unable to detail the specific hybrid ablation outcomes according to types of AF
(paroxysmal vs. non-paroxysmal) given the lack of available patient level data. We recognize
131
the limitations of the individual included studies may affect the robustness of this review and
meta-analysis. First, the average number of subjects in most studies is low. Second, all the
available studies on hybrid ablation were non-randomized, single-centre cohort series. Third,
there is significant heterogeneity of the hybrid ablation approach with variable epicardial and
endocardial lesion sets, different ablation devices used and surgical access employed. Fourth,
as with any meta-analysis there may be confounding variables, measured and unmeasured,
which may impact the results of this analysis. Fifth, there is also variability in the quality of
studies with variations in recording of baseline variables, consistency of both epicardial and
endocardial ablation method and follow-up. These variations serve to not only highlight the
features of the highest quality studies, but also the need for standardisation of reporting within
this field to determine the true efficacy of this evolving technique. Last, there is a lack of
reporting standard in the included studies with variable follow-up methodology and time
points. Nevertheless, these limitations serve to highlight areas of focus for future studies.
5.5. CONCLUSIONS
The standalone hybrid ablation procedure for AF shows favourable outcome without increased
complication rates as compared to more established catheter-based or surgical approach. Early
results suggest a potential role for this novel strategy in selected AF patients given the
suboptimal results of catheter-based approach in those with more persistent form of the
arrhythmia. Further studies are needed to determine the optimal hybrid ablation technique and
longer-term outcome results.
132
5.6. FIGURES AND TABLES
Figure 5.1
133
Figure 5.2 – Surgical Lesions and Devices
(Left) Figure depicting the commonly placed surgical lines of Pulmonary Vein Isolation
(Blue), Box lesion (Green), Roof Line (Purple), Mitral Line (Yellow), Right inferior
pulmonary vein to Right Atrium Line (Orange), Left Atrial Appendage Exclusion (Red).
(Right) The Radio-Frequency Devices utilised in the studies: A) Medtronic Gemini, B)
Atricure Synergy, C) nContant system, D) Estech Cobra
134
Figure 5.3 – Sinus Rhythm Maintenance Forest Plot
Sinus rhythm maintenance achieved across studies in terms of mean and 95% Confidence
Interval
135
Figure 5.4 – Sinus Rhythm Maintenance without AAD Forest Plot
Sinus rhythm maintenance without Anti-Arrhythmic Drugs (AAD) achieved across studies in
terms of mean and 95% Confidence Interval.
136
Figure 5.5 – Sinus Rhythm Subgroup Analysis
Procedural success in obtaining sinus rhythm maintenance at final follow-up related to device
utilised, LAA exclusion and block confirmation at the end of the EP procedure.
137
Figure 5.6 – Complication Rates Forest Plots
138
139
140
141
142
143
Chapter 6
The Contact Force Effect of Bipolar Radio-Frequency Clamps on Lesion Formation in
the Surgical Management of Atrial Fibrillation
6.1. INTRODUCTION
The surgical treatment of atrial fibrillation (AF) has grown in popularity since the development
of the Cox-Maze procedure in 1987 where lines were cut and sewn in the atria to create scars
that inhibit electrical conduction.(36) As technology advanced the technique was modified to
utilise modern energy sources such as cryothermy and radiofrequency (RF) to create lesions of
conduction block, by developing myocardial cell damage which evolves into non-conducting
myocardial scar, with greater efficiency and ease.(187,192) The mainstay tool of the
contemporary Cox-Maze IV procedure is the bipolar RF clamp which delivers RF energy
through opposing jaws of a clamp.(153) This technique has been shown to be effective for the
treatment of AF; however, the cut-and-sew technique has been shown to be superior to RF
ablation, despite advances in ablation technology.(358)
Recent studies of RF catheter technology have shown that RF lesion size is related to
contact force.(148,359-361) Further studies have also demonstrated that higher contact force
is associated with superior rhythm control at follow-up.(362,363) Additionally, studies
analysing the lesions formed during surgical ablation procedures for AF have found the
presence of gaps in ablation lines made with RF.(364,365) This in turn has led to the
development of hybrid procedures for the treatment of AF to optimise results by following
surgical ablation by an electrophysiology study to assess contiguity of ablation lines and
complete them as necessary delivering excellent results.(207,341)
In consideration of these findings, we hypothesized that the jaws of the bipolar RF
clamps used in AF surgery may yield varying force profiles over different thickness of tissue,
144
which may in turn effect subsequent RF lesions. An analysis of two commonly used bipolar
RF clamps with different systems of jaw apposition (hinged and parallel) was undertaken to
determine contact and force profiles of these clamps, and the impact on the likelihood of
achieving transmurality during ablation.
6.2. METHODS
6.2.1. Pressure Assessment
6.2.1.1. Surgical Equipment
There are presently 2 types of bipolar RF clamps with jaws that possess different articulation
designs and therefore close in 2 different manners. The “hinged” have jaws that approximate
from a hinge at the proximal portion of the jaw, and the “parallel” have jaws that approximate
along a straight axis. Two different styles of bipolar RF clamps were identified with these
different jaw apposition mechanisms (Hinged: Cardioblate LP, Medtronic, Minneapolis, MN,
USA and Parallel: Synergy, Atricure, West Chester, OH, USA; Figure 6.1). The hinged clamp
is 70mm in length and at the fulcrum of the jaw is 67.5mm wide which tapers to 3.5mm wide
after 1cm and remains that width to the tip. The parallel jaw is 60mm in length and 4.0mm
wide along the length. Both clamps possess locking mechanisms to keep the jaws closed once
applied and both clamps were fully closed as per their design when utilised.
It is important to note that the hinged clamp is an irrigated clamp whereas the parallel
clamp is non-irrigated. As both clamps were subjected to the same battery of tests it is unlikely
that this feature in isolation should have a significant interaction that would affect the validity
of comparing them in for contact force. Irrigation technology in RF lesions may be protective
against the build-up of char or debris on the probe itself but it does not seem to overcome the
effect of decreased contact.
145
6.2.1.2. Tissue Selection
Bovine skeletal muscle was used as biological tissue model. Initial testing was performed on
atrial tissue extracts but due to the uneven surface of such tissue and the focus of analysis being
pressure assessment skeletal muscle was selected as an analogue that was able to have
properties of thickness controlled along the entire length of tissue sample. Thickness was
controlled in muscle preparations to have three samples for testing at each thickness from 1mm
to 8mm.
6.2.1.3. Clamp Pressure Assessment Methodology
Ablation clamps were loaded with the muscle sample and a pressure sensitive film sample, and
the clamp jaws were then closed in a standard fashion until the locking mechanism engaged.
Each sample was clamped for 40 seconds, then released and the pressure film removed for
analysis. This procedure was repeated three times for each millimetre tissue thickness between
1mm and 8mm. New tissue samples and pressure films are used for each test.
6.2.1.4. Contact Force
Catheter ablation technology utilising RF creates lesions to eliminate the triggers of AF and
block the propagation of AF wave fronts.(145) To optimise the creation of such lesions catheter
technology evolved to incorporate cooled-tip catheters, irrigated catheters, adjustable power
delivery and contact force sensing catheters.
It has been described that the contact force during catheter ablation for AF correlates
with clinical outcomes and when lesions are placed with an average contact force of >20g
superior rhythm control results are obtained.(362) Catheter contact force is traditionally
expressed in grams, but as this figure is simply a weight, a conversion was required to calculate
actual contact force in g/mm2 (Calculation: 7Fr Catheter diameter is 2.333mm. Assuming a half
146
spherical shape the surface area is 2πr2. This gram pressure is then divided by this area
calculation). Figure 6.2 displays the contact force of a 7Fr Catheter in g/mm2 of 10g, 20g, 30g,
40g, 50g, 60g and 70g as well as the mean contact force of the hinged and parallel clamps at
each 1cm increment from tissue thickness 1mm-8mm. A colour chart has been applied to these
tables such that weights of ≤10g are displayed in white to denote areas with contact force which
would be associated with rhythm control failure in catheter studies, 20g to 70g in shades of
blue to denote areas of expected superior rhythm control and >70g in red to denote pressure
that would be considered exceptionally high in catheter studies.
6.2.1.5. Pressure film and analysis
A pressure sensitive film (4LW Prescale, Fujifilm) was used to assess the pressure profile of
the clamps on the biological tissue model samples. The film develops an image with colour
intensity correlating to pressure. These film images were then digitally scanned and the length
of the image was divided into 10mm strips for pressure calculation of each portion of the jaw.
The method of pressure film use and analysis is displayed in Figure 6.3.
6.2.2. Ablation Lesion Analysis
6.2.2.1. Experimental Preparation
A total of 6 Merino Cross Wethers were studied. All procedures were conducted in accordance
with the guidelines outlined in the “Position of the American Heart Association on Research
Animal Use” adopted on November 11, 1984, by the American Heart Association. Approval
for the performance of the study was provided by the Animal Ethics Committees of the South
Australian Health and Medical Research Institute and the University of Adelaide, Adelaide,
Australia
147
6.2.2.2. Tissue Selection
Assessment was initially performed on ventricular muscle to determine its suitability for use
with determining a pressure effect with ablation; however, it was found that it was not uniform
over the length of a given clamp due to papillary muscles and trabeculae carneae, and only 1
sample per muscle thickness could be obtained in each sheep. As the experimental question to
be answered was regarding a pressure effect a uniform muscle was required for analysis. Cuts
of skeletal muscle were thus selected over cardiac muscle as they could be controlled for
thickness, are uniform over the length (no trabeculations) and, are therefore, ideal for pressure
testing experiments which require an even surface to minimise confounding effects. As cell
types of skeletal and cardiac muscle are both striated in nature the effect RF has on each should
be the same, and further, each clamp was submitted to the same test design and therefore
remains valid in this experimental model. Ovine pectoral muscle was thus selected for
experimentation as it provides a broad, flat sheet of muscle with area for multiple, similar
thickness tissue samples to be utilised.
6.2.2.3. Ablation and Evaluation of Transmurality
Intact ovine pectoral muscle flap was used as in-vivo tissue model for ablation assessment.
Right and left pectoral muscle flaps were exposed surgically on 6 anaesthetised animals. Each
flap was created to allow for 6 parallel ablation lesion with a margin of ~20mm between lesion.
Thus, each animal provided 12 lesions and 1 clamp style was used per animal, providing a total
of 36 ablation lesions per clamp for study. The ablation clamp was positioned with the proximal
edge of the clamp against the muscle flap edge prior to engaging the locking mechanism
(Figure 6.4). RF energy was applied according to manufacturer guidelines. A single burn was
defined as delivery of RF energy to the tissue until the energy source generator indicated that
a lesion was created (displayed as ‘transmural’ indicator light/label on generator display as
148
determined by the manufacturer). In the case of the hinged clamp, irrigation was tested prior to
the delivery of all burns energy in accordance with manufacturer guidelines. Following each
lesion, the ablation clamp jaws were wiped clean of char and blood prior to further ablation.
To determine the effect of multiple applications of RF energy on lesion formation, either one,
two or three applications of RF was applied to a single area of muscle tissue. This was repeated
such that for each muscle flap, there were two samples of one burn, two samples of two burns
and two samples of three burns of RF energy. Each clamp was measured and marked with 1cm
lines along the clamp such that when the jaw was closed the tissue itself was able to be marked
at each 1cm segment correlating correctly to the conditions of the clamp when the lesion was
created. Following lesion creation, whole muscle flaps were excised and sectioned in 10mm
increments (along the aforementioned marks) for histological analysis.
Tissue samples were soaked in 1% Triphenyl-Tetrazolium Chloride (TTC) for 30
minutes and were then placed into Formaldehyde for tissue fixation. Lesion assessment was
then carried out digitally. TTC is a vital stain such that it stains viable tissue red and non-viable
tissue is left white, allowing for visual assessment. A camera was used to acquire images of the
tissue samples at a resolution of 3264x1840 pixels. All measurements on digital images were
calibrated for size by including a ruler in the image and all measurements scaled accordingly.
Digital images were analysed using commercially available tools in Photoshop CC2014
(Adobe Systems, San Jose, CA, USA). For each 10mm section-image, analysis was performed
to quantify; tissue thickness, presence of overlying fat, thickness of fat, lesion depth and
transmurality. All images were verified by 2 investigators.
6.2.3. Statistical Analysis
All continuous variables reported as mean ± SD. Statistical analyses were performed using
SAS v9.3 (Cary, NC, US) and Stata v13 (College Station, TX: StataCorp). Differences
149
between clamp styles were investigated using linear regression (for continuous outcomes) or
logistic regression (for binary or categorical outcomes). An interaction term (clamp style by
distance) was included to test for modification of the effect of clamp style (hinged vs parallel)
by distance (proximal vs distal), or vice versa; separate estimates were derived for (a) the
difference between proximal and distal for each clamp style, and (b) the difference between
hinged and parallel clamp for each distance. Generalised Estimating Equations (or random
effects where required) were used to account for clustering due to sheep and side (left/right).
In the case of burn thickness, the linear regression model also included an offset term for total
lesion thickness, while percentage burn thickness was investigated via several methods
(standard linear regression, Tobit regression, and ordinal logistic regression with outcome
grouped into 6 categories) due to a high proportion of values at the upper bound of 100%. In
the case of pressure, log transformation was applied due to a skewed distribution, with
estimates back-transformed to the original scale. Additional models were also fitted to
investigate the effect of thickness (and potential interaction with distance) for each clamp style.
Transmurality, as a binary outcome, was investigated using logistic regression; additional
models were fitted to investigate the effect of epicardial fat (considered as both a continuous
and a binary outcome) on transmurality for each clamp type. Statistical significance was taken
as P<0.05.
6.3. RESULTS
6.3.1. Pressure Analysis
Three muscle samples were tested for each 1mm thickness yielding 24 pressure profile
samples per clamp for pressure assessment.
The hinged clamp is 70mm in length and the parallel clamp is 60mm in length (Figure
6.1). The cross-sectional area of each segment of the hinged clamp is 35mm2 with the most
150
proximal segment (0-10mm) having a slightly larger area of 67.5mm2. The cross-sectional area
of each segment of the parallel clamp is 40mm2.
The mean pressure exerted (g/mm2) of every 10mm segment of each clamp style (from
proximal to distal) at increasing tissue thickness (from 1-8mm) is plotted for analysis in Figure
6.5. The range of mean pressures exerted on the tissue by the hinged clamp was 0.12 to
25.05g/mm2 and by the parallel clamp was 1.96 to 10.89g/mm2.
Analysis was then conducted to compare the proximal and distal portions of each jaw
(The initial 2cm of each jaw compared to the distal 2cm of each jaw). Due to the skewed
distribution of data the statistical analysis was conducted on median value. Figure 6.6 depicts
the pressure comparisons of each clamp comparing the proximal 2cm and distal 2cm. The
parallel clamp demonstrates no statistical difference in pressure exerted between the proximal
and distal clamp segments for all tissue thickness tested except 4mm. The hinged clamp
demonstrates a statistically significant pressure drop off between proximal and distal clamp
segments for all tissue thicknesses tested. If all tissue thicknesses between 1mm and 8mm are
analysed together the hinged clamp demonstrates a statistically significant pressure drop off
(p<0.0001) and the parallel clamp does not demonstrate and statistically significant pressure
change (p=0.6).
6.3.2. Lesion Analysis
Data was extracted and analysed from every 1cm segment of muscle tissue along the burn. An
example of tissue segments for 1 burn, 2 burns and 3 burns for each clamp style is presented in
Figure 6.7. The mean thickness of muscle was 13.5mm.
151
6.3.2.2. Total RF Time and Impedance
The time in seconds of RF energy delivered for each burn was recorded for both clamp styles.
In lesions created using 2 and 3 RF energy delivery cycles, the individual RF times were
recorded and summed to provide total RF time for analysis. Statistical analysis was then
conducted to see if a correlation existed between total burn time of each lesion and
transmurality. With the parallel clamp for every 1 second increase in burn time the odds of
creating a transmural lesion increased by 7% (p<0.001); however, no statistically significant
association between burn time and transmurality existed for the hinged clamp (p=0.6).
Impedance results differed between the 2 clamp styles. The hinged clamp console
possessed a digital display which provided a specific figure of impedance and as such the value
at the commencement of RF delivery as well as the value and the completion of RF delivery
could be recorded. At the completion of each burn the device would display transmural. The
parallel clamp did not provide a digital figure read out, but only a graph which tracked
impedance until it plateaued below a set level, at which point the device would indicate
transmurality. As such only the peak level interpreted from the graph was recorded. Due to the
different nature of impedance recordings direct comparison between clamp styles is not
possible, but analysis of the association of impedance figures with transmurality of each clamp
was conducted.
For the parallel clamp, the peak or maximum impedance was associated with
transmurality, such that for every unit of conductance increase the odds of transmurality
increased by 9.5% (p=0.0002). However, this association was not statistically significant for
peak conductance reached on the 2nd or 3rd RF deliveries.
For the hinged clamp no statistically significant association between the transmurality
and the initial impedance, the final impedance or the magnitude of change between the initial
and final impedance existed, except with the start impedance of the 2nd RF delivery, which
152
found that for every unit increase in start impedance the odds of transmurality decreased by
2% (p=0.018).
6.3.2.3. Lesion Depth
Lesion depth was measured at 10mm increments along the length of the burn for each clamp
style at 1, 2 and 3 burns. The lesion depth was then divided by the measured tissue thickness
at the specific burn increment (to standardise burn depth for each data point) and is then
expressed as a percentage. Figure 6.8 demonstrates the mean lesion depth percentage for both
hinged and parallel clamps with application of 1, 2 and 3 burns. The mean burn depth for 1
burn of energy delivery for the parallel clamp was 62.9% (95% CI: 49.0-76.9%) and in the
hinged clamp was 40.6% (95% CI: 27.9-53.4%; p=0.02). The mean burn depth for 2 burns of
energy delivery for the parallel clamp was 77.2% (95% CI: 62.6-91.8%) and in the hinged
clamp was 52.3% (95% CI: 39.2- 65.5%; p=0.01). The mean burn depth for 3 burns of energy
delivery for the parallel clamp was 99.6% (95% CI: 84.3-114.9%) and in the hinged clamp was
74.0% (95% CI: 60.1-87.9%; p=0.02).
Analysis was then undertaken to compare burn depth at the proximal and distal
segments of each clamp (Figure 6.9). Burn depth was analysed using a linear regression model
with an offset for total lesion thickness; an interaction term between clamp style and segment
was included to derive separate estimates of (a) difference between clamp style for each
segment, and (b) difference between segments for each clamp style. For the hinged clamp the
proximal 2cm segment estimated mean burn depth was 10.2mm (95% CI: 9.1–11.4mm) and
the distal 2cm estimated mean burn depth was 8.0mm (95% CI: 7.1–9.0mm). For the parallel
clamp the proximal 2cm segment had an estimated mean burn depth of 11.7mm (95% CI: 10.2-
13.2mm) and in the distal 2cm 10.2mm (95% CI: 9.1–11.2mm).
153
Tests of the differences between these estimated burn depths showed that the difference
between proximal and distal segments in the hinged clamp was statistically significant
(p=0.002), whereas the difference between proximal and distal segments in the parallel clamp
was not statistically significant (p=0.07). Secondly, the proximal segment comparison across
clamp style revealed no statistically significant difference (p=0.1) whereas comparison of the
distal segments of each clamp identified a statistically significant difference in burn depth
(p=0.003).
6.3.3. Fat Presence and Effect
Analysis was performed to assess odds of transmurality in relation to the presence and
thickness of overlying fat between the clamp and the muscle (Figure 6.10). It was found that
for every 1mm increase in fat tissue thickness the odds of being transmural decreased with both
clamps; parallel clamp by 13% (p=0.002), hinged clamp by 15% (p=0.001) and overall by 11%
(p=0.006). Further, in analysis of larger amounts of fat in between the clamp and the tissue it
was identified that in the presence of 15mm of fat the odds of transmurality with a parallel
clamp decreases from 0.85 to 0.41 representing a 51.6% decrease and with a hinged clamp
decreases from 0.59 to 0.1 representing a 82.5% decrease.
6.3.4. Transmurality
The overall odds ratio of 1, 2 or 3 burns being transmural with either clamp is depicted on a
forest plot for analysis (Figure 6.11). This graph demonstrates that the odds of transmurality
are greater with a parallel clamp for 1, 2 or 3 burns (p=0.03, p=0.003, p=0.002, respectively).
154
DISCUSSION
6.3.5. Major Findings
This study provides new information characterizing the ablation profile of bipolar RF surgical
ablation clamps. It demonstrates that:
1) The hinged clamp loses tissue contact force at the tips of the jaw (away from the hinge
of jaw closure);
2) Increasing the number of burns delivered, by either bipolar clamp device, increases the
likelihood of true transmurality - regardless of indications provided by the devices;
3) When ablating tissue with overlying fat the likelihood of transmurality is significantly
reduced; and
4) Reduced contact force reduces lesion depth and likelihood of transmurality.
These findings have significant implications to achieving transmural and contiguous lesions,
both important elements for the success of ablation.
6.3.6. Importance of Contact Force
It has been demonstrated that increased contact force results in increased lesion depth as
exemplified in the comparison of proximal and distal segments of lesion depth of each clamp.
The parallel clamp did not demonstrate a statistically significant difference between the
proximal and distal segments in terms of lesion depth, but it trended towards a decrease lesion
depth. Further inspection of the clamp revealed that is constructed from a more inherent flexible
material which likely renders a small decrease in contact force at the distal tip, which may
explain the down trend noticed. This suggests that, regardless of which clamp is utilised for AF
surgery, when ablating across thicker structures such as the pulmonary veins, ablating from 2
directions should negate this effect (ie. position the clamp around the pulmonary veins from
both the superior and inferior directions for separate burn delivery).
155
As can be readily seen with the hinged jaw type, regardless of tissue thickness, there is
significantly lower pressure distally in the clamp when compared to a parallel jaw type which
delivers a more uniform pressure across all tissue thicknesses. It has previously been reported
that the posterior left atrial wall thickness is 4.1±0.7mm with a range of 2.5mm to 5.3mm.(366)
If we consider this when planning epicardial pulmonary vein isolation the thickness of tissue
that would be ablated could range from 5mm to 10.6mm in thickness (with epicardial fat in
addition to the measurement) and therefore, maintaining contact force with such ablations is of
paramount importance to deliver a truly transmural lesion.
The use of a parallel clamp as associated with increased lesion depth and increased
likelihood of delivering a transmural lesion with 1, 2 or 3 burns.
6.3.7. Number of Burns
For either clamp style an increasing number of burns is associated with a higher likelihood of
transmurality. Determining the number of burns required clinically to develop a transmural
lesion is of the utmost importance in successful surgery for the ablation of AF. While up to 3
burns did not deliver transmurality all the time, the parallel clamp was very close to achieving
this. This study only went up to 3 burns over a given area so anything beyond this number of
burns would only be extrapolation, but with the parallel clamp it would seem likely that with 4
or more burns over a single area transmurality along the length of the lesion should be obtained.
6.3.8. The Interaction of Fat
The presence of fat intervening between the clamp and muscle tissue is a common occurrence
in the clinical setting and, as such, it is important to understand the effect on the RF lesion
created when this takes place. This analysis has revealed significant effect on lesion
transmurality from even 1mm of intervening fat in a bipolar clamp and this effect is important
156
to consider. Firstly, when utilising a bipolar clamp to generate RF lesions selecting an area with
minimal fat or clearing the intervening epicardial fat becomes of great importance to guarantee
a transmural lesion. Second, when utilising non-clamp style device (such a bipolar rail device
or unipolar device) to generate an RF lesion the area of selection and the ability to clear any
overlying fat becomes of paramount importance as not only is the contact force difficult to
maintain uniformly with the tissue, but the entire delivery of RF energy must pass through the
attenuating effect of the intervening fat to then create a transmural lesion. Finally, it is still
unclear if simply increasing the number of burns delivered over an area of fat overcomes the
decreased odds of transmurality or if fat clearance is indeed required.
6.3.9. Impedance as a Measure of Ablation Endpoint
Increasing amounts of RF burn time are associated with increased likelihood of transmurality
with the parallel clamp, however, not for the hinged clamp. The loss of this association with
the hinged clamp may be more reflective of the physical design suffering from loss of distal
pressure. As with decreased contact the RF energy is unable to bridge the physical gap between
the jaw and the tissue resulting in decreased lesion depth and this is a barrier which is simply
not able to be overcome by burning longer.
Analysis of impedance is complex as the algorithm utilised by either device in
calculating transmurality has not been provided. It would be expected that as the RF energy
excites the molecules within the cell and the resultant energetic movement of molecules would
generate heat and thereby causing thermal damage and lysis of the cell. As such, with increased
cellular death the intracellular ions would be released and allow current to flow more freely
across the lesion area. Thus, with as a lesion approaches transmurality the impedance should
drop.
157
Interestingly, for the most part impedance did not have a statistically significant
association with transmurality outside of 2 instances. First, increased peak impedance with the
first RF delivery of a lesion with the parallel clamp was associated with increased likelihood
of transmurality. Second, increased initial impedance with the second RF delivery of a lesion
with the hinged clamp was associated with a decreased likelihood of transmurality.
Importantly, there is no obvious association of impedance changes with transmurality. While
we do not know the algorithm used for transmurality determination with either device, it should
also be noted that with up to 3 burns with the device displaying ‘transmural’ the histological
staining of these burns lesions demonstrates that this label is not accurate. This supports the
hypothesis that impedance may not be completely accurate in transmurality determination. The
impedance measures in the hinged clamp are likely hindered by the design as when thick tissue
is clamped there is loss of tissue contact distally in between the jaws. As such, the RF electrode
would be interfacing with air instead of tissue and this would invariably have a higher
impedance which would negatively impact any algorithm in determining transmurality. In
addition, in the clinical setting only relying on impedance changes when delivering RF energy
may not be the best guide to determine lesion transmurality. Therefore, this is an area which
would benefit from further analysis in futures studies.
6.3.10. Clinical Implications
The ability to generate consistently transmural lesions is a key feature of the appropriate tool
for the surgical ablation of AF. This study aids the clinician in providing information of how
surgical tools function to create lesions and should be utilised in the selection and use of bipolar
RF clamps.
Obesity is frequently linked to the development of AF and is associated with increased
pericardial fat. As the presence of fat reduces the efficacy of bipolar clamps to deliver a
158
transmural lesion it is highly important that epicardial fat located at the site of a burn is cleared
prior to delivering energy to increase the likelihood of transmurality.
The use of impedance alone may not be the best measure of transmurality. Utilisation
of conduction testing at the time of procedure may be a useful adjunct for transmurality
assessment.
6.4. CONCLUSION
Parallel and hinged design bipolar RF clamps have different pressure profiles and resultant RF
lesions. The use of a parallel clamp as associated with increased lesion depth and increased
likelihood of delivering a transmural lesion with 1, 2 or 3 burns. The presence of fat
significantly decreases the likelihood of obtaining transmurality in bipolar RF clamps. Ablation
lesions along thick tissues, such as may be experienced with epicardial pulmonary vein
isolation, may not be transmural along the entire length of the burn despite the device indicating
that the lesion is transmural. These findings have implications for all surgical ablation
procedures carried out with bipolar RF clamps and should be considered when selecting
ablation device, number of burns, location and clearance of epicardial fat.
159
6.5. FIGURES
Figure 6.1 – Bipolar Ablation Clamps and Corresponding Footprints
The Parallel (Left) and Hinged (Right) closure style bipolar RF clamps are shown with their
corresponding foot prints below. The hinged clamp has a proximal increased width for the
first 10mm.
160
Figure 6.2 – Pressure Tables
Pressure tables of Hinged Jaw, Parallel Jaw and 7 Fr Catheter for comparison. Pressure of
10g or less is represented in white shading, 20g to 70g in deepening shades of blue and
greater than 70g is represented in red.
161
Figure 6.3 – Methods - Pressure Analysis
Methodology sequence of Pressure Analysis: A) Muscle is clamped between clamp jaws with
pressure sensitive film, B) Pressure Film is scanned as a raw image, C) Film image converted
to greyscale for intensity standardisation, D) Image is divided into 1cm segments for
segmented analysis and E) The image is then digitally sampled against the pressure film
reference table for analysis.
162
Figure 6.4 – Methods - Experimental Ablation Diagram
The clamp is engaged over the pectoral muscle and ablation delivered. The superior burn line
of the diagram represents where a single burn was delivered with incision marks placed at the
superior edge to denote the segments. The clamp is shown over the next segment down where
2 burns will be applied to the same area. The next burn in the pattern will be 3, and this series
will repeat itself once more.
163
Figure 6.5 – Clamp Pressure Profiles
Clamp Pressure Profiling with A representing the Hinged Clamp pressure profile and B
representing the Parallel Clamp pressure profile. The x-axis of each graph displays the
distance from proximal to distal along the clamp, the y-axis displays the thickness of muscle
on which the clamp was applied and the z-axis represents the pressure exerted.
164
Figure 6.6 – Proximal and Distal Clamp Segment Pressure Comparisons
A comparison of exerted pressures of the Proximal (First 2cm) and Distal (Last 2cm) portions
of each clamp at increasing tissue thickness (1mm to 8mm). Error bars indicate 95%
confidence interval [The Upper Confidence Intervals of 7mm and 8mm for the Hinged Clamp
are 39.58mm and 48.38mm respectively] (* indicates a non-significant p-value)
165
Figure 6.7 – Burn Lesion Samples and
Examples of lesion specimens from 1 burn, 2 burns and 3 burns for each clamp style are
shown. The lesions have been sectioned at 1cm increments and stained with Triphenyl-
Tretrazolium Chloride (TTC) which stains viable tissue red and is not taken up by dead cells.
166
Figure 6.8 Mean Burn Lesion Depth
Mean burn depth standardised to tissue thickness for 1 burn, 2 burns and 3 burns of energy
delivery across all samples. Comparator p-values are presented for the difference between
parallel and hinged clamps.
167
Figure 6.9 – Burn Lesion Analysis
Burn depth analysis of the proximal and distal segments of each clamp in comparison. Linear
regression model estimates of burn depth (mm) are plotted. The p-values in black denote the
represented burn segments with comparison across clamp type. The coloured p-values are
matched with clamp style showing comparison between segments of the clamp. Error bars
represent a 95% confidence interval.
168
Figure 6.10 – Odds of Transmurality in Relation to Depth of Fat
Each data point represents a single burn segment for each clamp style expressed in terms of
odds of obtaining transmurality and the depth of fat at that point of the burn. Odds based off
of measurement of overlying fat for every segment of ablated tissue.
169
Figure 6.11 – Odds of Transmurality Forest Plot
Forest plot depicting the odds of transmurality. Increased likelihood of obtaining
transmurality with a hinged clamp is represented on the left side of the graph and increased
likelihood with a parallel clamp on the right. The estimate point for 1 burn, 2 burns and 3
burns lies to the right with highly significant p-values and the 95% CI completely to the right
of the graph.
170
Chapter 7
Conclusion and Discussion
This thesis presents current perspectives on the nature of the surgical management of AF. In
the introductory chapter an overview of the impact that AF is currently having on the general
population and the burden this has placed on patients and health care systems is presented.
Mechanisms for how this arrhythmia develops are then addressed followed by the paradigm
for medical management presently available. The development of the therapy of catheter
ablation is then addressed incorporating the rational, technique and results that are currently
being achieved. Finally, the history of the surgical ablation of AF is analysed, culminating in
the present state of this surgical procedure with an analysis of the current technologies available
for carrying out this technique. While surgical treatment of AF has evolved, this has largely
been in small series and with variable experiences.
In this context, Chapters 2 through 5 then perform an in-depth analysis of the current
literature on the surgical ablation of AF through meta-analyses covering four distinct
populations of surgical ablation procedures; 1) those performed concomitant to mitral valve
repair or replacement, 2) those performed concurrent to mixed cardiac surgical procedures, 3)
those performed as a standalone procedure or 4) those performed as part of a joint procedure
with a cardiac electrophysiologist in the hybrid ablation for AF. These chapters summarise the
efficacy of the surgical ablation of AF as follows:
• When carried out concurrent to mitral valve repair or replacement the surgical ablation
of AF achieves SRM of 79.6% at mean follow-up of 27 months. The rhythm follow-up
was carried out primarily with ECG with only half of the studies utilising Holter
monitoring. The best results are obtained with a Bi-Atrial Maze procedure in this
cohort, with a low total complication rate of 6.9% driven by a post-op PPM requirement
of 1.2%. The mortality rate was low at 0.9% and the stroke rate of 0.3%.
171
• When carried out concomitant to general cardiac surgical procedures, such as coronary
artery bypass grafting or any valvular intervention, the surgical ablation of AF achieves
SRM of 79.6% and SRM off AAD of 65.3% at mean follow-up of 22 months. The
rhythm follow-up consisted primarily of ECG with a further two thirds of the studies
utilising Holter monitoring for follow-up. No lesion set proved statistically superior,
but the best results are obtained in younger patients with paroxysmal AF. The overall
total complication rate was low for this cohort of higher complexity concomitant
procedures at 15.1% with mortality rate of 4.3%, CVA rate of 1.8%, PPM rate of 5.4%
and atria-oesophageal fistula rate of 0.5%. When these procedures were carried out
concurrent to mitral valve intervention and they excluded the left atrial appendage the
complication rate was lowest.
• When carried out as a standalone procedure the surgical ablation of AF achieves SRM
of 83.3% and SRM off AAD of 65.5% at mean follow-up of 17 months. The primary
method of follow-up was the use of single day Holter monitors with half of the studies
utilising a longer form of assessment with seven day Holter monitor. The best results
are achieved with a Bi-Atrial Maze procedure and the creation of this lesion set through
a Cut and Sew technique, the use of RF or the incorporation of multiple ablative
modalities were the most successful. The overall complication rate was low at 7.4%,
which was largely driven by a PPM rate of 3.2%. The exclusion of the left atrial
appendage corresponded with a lower risk of stroke.
• When performed jointly with a cardiac electrophysiologist the hybrid ablation
procedure achieves SRM of 79.4% and 70.7% off AAD at mean follow-up of 19
months. The primary method of rhythm assessment was a seven day Holter monitor
with a further one third utilising loop recorder technology. The best results were
achieved through the surgical use of a bipolar RF clamp and the exclusion of the left
172
atrial appendage. Ensuring lesion completeness at the end of the endocardial procedure
with electrical testing for block was indicative of better outcomes. The overall
complication rate was low at 6.5% being driven by post-operative bleeding requiring
return to the operating theatre in 1.6% of patients.
As the surgical ablation of AF has evolved the use of alternative ablative strategies to cut and
sewn lesions it has seen the rise in many technologies to achieve the delivery of a safe and
effective transmural lesion, but even the most efficacious devices as seen in bipolar RF clamps
have not been able to achieve 100% reliably transmural lesions.(153) In analysis of this point
we undertook a series of experiments to develop an increased understanding of RF lesion
formation with surgical bipolar RF clamps. In particular, are they subject to effects of contact
force as has been identified in catheter ablation.(151) Chapter 6 thus elucidates how different
mechanisms of clamp have different pressure profiles which result in a variable degrees of
contact force being transmitted through the clamp to the tissue which is to be ablated, an effect
which is amplified at greater tissue thickness. This hypothesis thus confirmed was taken
through to animal experimentation where it was then demonstrated how contact force directly
correlates with the ability of surgical clamps to deliver a transmural lesion along the entire
length of the clamp, regardless of the device indicating to that the lesion is transmural. The use
of a clamp with parallel jaws that close evenly on tissue is associated with a higher likelihood
of delivering a transmural lesion than those with a hinged jaw. Through the delivery of
increasing numbers of burns the likelihood of transmurality increases, however, the presence
of intervening fat between the clamp and the ablated tissue is associated with a significant
decrease in the ability to deliver a transmural lesion, an effect that increases with every
millimetre of fat intervening. These results thus describe potentially why bipolar RF clamps
for the surgical ablation of AF have not attained 100% reliable efficacy, and most importantly,
these results are instructive to the surgeon using these devices for the surgical ablation of AF
173
as to how to ensure the best results. Firstly, the area of the heart planned for ablation must be
cleared of any intervening epicardial fat to optimise lesion creation efficacy and likelihood of
transmurality. Secondly, clamp selection as it relates to the mechanism of jaw closure is of
paramount importance for the delivery of a truly transmural lesion across the length of the jaw.
Thirdly, increasing the number of burns delivered to an area increases the likelihood of
obtaining a transmural lesion, irrespective of the device indicating transmurality to the
operator.
In summary, this thesis provides not only an overview for the current state of the
surgical ablation of AF, but also offers new understandings on how to optimise the efficacy of
these ablative strategies for the elimination of AF.
174
Chapter 8
Future Direction
The analyses conducted over the course of this thesis have provided important understandings
into the optimal management strategies for the surgical ablation of AF. However, many
questions remain to be answered in this arena of study. Firstly, a common thread across all the
meta-analyses we conducted in this area revealed a paucity of level 1 randomised control trial
evidence with a large variability in both ablation technique and follow-up methodology. To
understand what the best lesion set is, to confirm the most efficacious way to apply this lesion
set and to determine what the ultimate efficacy of the surgical ablation of AF is a large,
prospective RCT will need to be conducted. Secondly, an evolving body of evidence is
emerging in which superior results in the elimination of AF are being obtained through the
aggressive control of risk factors in patients with AF.(106) In the vast majority of studies
analysing the results of surgical ablation of AF the risk factors of the patients have not even
been documented for consideration, so if the surgical arena mimics that of catheter ablation an
increased level of surgical ablation efficacy may be able to obtained with the same risk factor
management. This is certainly an area that will warrant further study in future. Thirdly, as the
current ablation technology available for the surgical management of AF is imperfect at this
present time as, we do not possess an ablation source capable of delivering 100% reliably
transmural lesions beyond cut and sew. Until such a device is developed the remaining
questions of optimal lesion set, technique of delivery and impact of concomitant procedures
will be confounded by the overarching issue of uncertainty that a true lesion of conduction
block has been created to abolish AF. In relation to this point the role of assessing conduction
block in cardiac surgical ablation procedures also requires further investigation as to the impact
this provides on long term results in sinus rhythm maintenance. The testing of block in the
surgical arena also has implications in technique, as when ablation is performed concomitant
175
to other cardiac surgical procedures cardiopulmonary bypass is necessitated and cardioplegia
would acutely inhibit the ability to determine electrical block. Finally, as the field of cardiac
electrophysiology advances achieving strong results in sinus rhythm maintenance through
catheter ablation, the results of the surgical ablation of AF must be compared to this to
determine the ideal ablative therapies for patients with AF. The goal must be to deliver the best
long term results in sinus rhythm maintenance and lowest complication rates across the range
of patients with AF; from those undergoing concomitant cardiac surgery to those with a lone
arrhythmia. Through such analysis in future we will be able to form the ideal treatment
paradigm for patients suffering from AF in the modern era of medicine.
176
Chapter 9
References
1. Fuster V, Ryden LE, Asinger RW et al. ACC/AHA/ESC guidelines for the management of patients with atrial fibrillation: executive summary. A Report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients With Atrial Fibrillation): developed in Collaboration With the North American Society of Pacing and Electrophysiology. J Am Coll Cardiol 2001;38:1231-66.
2. Lloyd-Jones DM, Wang TJ, Leip EP et al. Lifetime risk for development of atrial fibrillation: the Framingham Heart Study. Circulation 2004;110:1042-6.
3. Schnabel RB, Yin X, Gona P et al. 50 year trends in atrial fibrillation prevalence, incidence, risk factors, and mortality in the Framingham Heart Study: a cohort study. Lancet 2015;386:154-62.
4. Miyasaka Y, Barnes ME, Gersh BJ et al. Secular trends in incidence of atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications on the projections for future prevalence. Circulation 2006;114:119-25.
5. Tse HF, Wang YJ, Ahmed Ai-Abdullah M et al. Stroke prevention in atrial fibrillation--an Asian stroke perspective. Heart Rhythm 2013;10:1082-8.
6. Diker E, Aydogdu S, Ozdemir M et al. Prevalence and predictors of atrial fibrillation in rheumatic valvular heart disease. Am J Cardiol 1996;77:96-8.
7. Rowe JC, Bland EF, Sprague HB, White PD. The course of mitral stenosis without surgery: ten- and twenty-year perspectives. Ann Intern Med 1960;52:741-9.
8. Grigioni F, Avierinos JF, Ling LH et al. Atrial fibrillation complicating the course of degenerative mitral regurgitation: determinants and long-term outcome. J Am Coll Cardiol 2002;40:84-92.
9. Benjamin EJ, Levy D, Vaziri SM, D'Agostino RB, Belanger AJ, Wolf PA. Independent risk factors for atrial fibrillation in a population-based cohort. The Framingham Heart Study. JAMA : the journal of the American Medical Association 1994;271:840-4.
10. Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrillation: the Framingham study. The New England journal of medicine 1982;306:1018-22.
11. Wang TJ, Parise H, Levy D et al. Obesity and the risk of new-onset atrial fibrillation. JAMA : the journal of the American Medical Association 2004;292:2471-7.
12. Go AS, Hylek EM, Phillips KA et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA : the journal of the American Medical Association 2001;285:2370-5.
13. Vizzardi E, Sciatti E, Bonadei I, D'Aloia A, Curnis A, Metra M. Obstructive sleep apnoea-hypopnoea and arrhythmias: new updates. Journal of cardiovascular medicine (Hagerstown, Md) 2014.
14. Huxley RR, Misialek JR, Agarwal SK et al. Physical activity, obesity, weight change, and risk of atrial fibrillation: the Atherosclerosis Risk in Communities study. Circulation Arrhythmia and electrophysiology 2014;7:620-5.
15. Lau DH, Middeldorp ME, Brooks AG et al. Aortic stiffness in lone atrial fibrillation: a novel risk factor for arrhythmia recurrence. PloS one 2013;8:e76776.
16. Olshansky B, Sullivan R. Increased prevalence of atrial fibrillation in the endurance athlete: potential mechanisms and sport specificity. The Physician and sportsmedicine 2014;42:45-51.
177
17. Roberts JD, Gollob MH. A contemporary review on the genetic basis of atrial fibrillation. Methodist DeBakey cardiovascular journal 2014;10:18-24.
18. Grogan M, Smith HC, Gersh BJ, Wood DL. Left ventricular dysfunction due to atrial fibrillation in patients initially believed to have idiopathic dilated cardiomyopathy. Am J Cardiol 1992;69:1570-3.
19. Wolf PA, Dawber TR, Thomas HE, Jr., Kannel WB. Epidemiologic assessment of chronic atrial fibrillation and risk of stroke: the Framingham study. Neurology 1978;28:973-7.
20. Wolf PA, Kannel WB, McGee DL, Meeks SL, Bharucha NE, McNamara PM. Duration of atrial fibrillation and imminence of stroke: the Framingham study. Stroke 1983;14:664-7.
21. Lin C, Edwards C, Armstrong GP et al. Prevalence and prognostic significance of left ventricular dysfunction in patients presenting acutely with atrial fibrillation. Clinical Medicine Insights Cardiology 2010;4:23-9.
22. Ott A, Breteler MM, de Bruyne MC, van Harskamp F, Grobbee DE, Hofman A. Atrial fibrillation and dementia in a population-based study. The Rotterdam Study. Stroke 1997;28:316-21.
23. Knecht S, Oelschlager C, Duning T et al. Atrial fibrillation in stroke-free patients is associated with memory impairment and hippocampal atrophy. Eur Heart J 2008;29:2125-32.
24. Ball J, Carrington MJ, Stewart S. Mild cognitive impairment in high-risk patients with chronic atrial fibrillation: a forgotten component of clinical management? Heart 2013;99:542-7.
25. Marzona I, O'Donnell M, Teo K et al. Increased risk of cognitive and functional decline in patients with atrial fibrillation: results of the ONTARGET and TRANSCEND studies. CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne 2012;184:E329-36.
26. Thrall G, Lane D, Carroll D, Lip GY. Quality of life in patients with atrial fibrillation: a systematic review. The American journal of medicine 2006;119:448.e1-19.
27. Stewart S, Hart CL, Hole DJ, McMurray JJ. A population-based study of the long-term risks associated with atrial fibrillation: 20-year follow-up of the Renfrew/Paisley study. The American journal of medicine 2002;113:359-64.
28. Wang TJ, Larson MG, Levy D et al. Temporal relations of atrial fibrillation and congestive heart failure and their joint influence on mortality: the Framingham Heart Study. Circulation 2003;107:2920-5.
29. Wong CX, Brooks AG, Leong DP, Roberts-Thomson KC, Sanders P. The increasing burden of atrial fibrillation compared with heart failure and myocardial infarction: a 15-year study of all hospitalizations in Australia. Arch Intern Med 2012;172:739-41.
30. Kalman JM, Sanders P, Brieger DB et al. National Heart Foundation of Australia consensus statement on catheter ablation as a therapy for atrial fibrillation. The Medical journal of Australia 2013;198:27-8.
31. Stewart S, Murphy NF, Walker A, McGuire A, McMurray JJ. Cost of an emerging epidemic: an economic analysis of atrial fibrillation in the UK. Heart 2004;90:286-92.
32. PricewaterhouseCoopers. The Economic Costs of Atrial Fibrillation in Australia. 2010. 33. Moe GK, Rheinboldt WC, Abildskov JA. A COMPUTER MODEL OF ATRIAL FIBRILLATION. Am
Heart J 1964;67:200-20. 34. Allessie MA LW, Bonke F, Hollen J. Experimental evaluation of Moe's multiple wavelet
hypothesis of atrial fibrillation. Cardiac arrhythmias 1985. 35. Cox JL, Canavan TE, Schuessler RB et al. The surgical treatment of atrial fibrillation. II.
Intraoperative electrophysiologic mapping and description of the electrophysiologic basis of atrial flutter and atrial fibrillation. The Journal of thoracic and cardiovascular surgery 1991;101:406-26.
36. Cox JL, Schuessler RB, D'Agostino HJ, Jr. et al. The surgical treatment of atrial fibrillation. III. Development of a definitive surgical procedure. The Journal of thoracic and cardiovascular surgery 1991;101:569-83.
178
37. Cox JL, Boineau JP, Schuessler RB, Kater KM, Lappas DG. Five-year experience with the maze procedure for atrial fibrillation. The Annals of thoracic surgery 1993;56:814-823; discussion 823-4.
38. Schuessler RB, Grayson TM, Bromberg BI, Cox JL, Boineau JP. Cholinergically mediated tachyarrhythmias induced by a single extrastimulus in the isolated canine right atrium. Circulation research 1992;71:1254-67.
39. Mansour M, Mandapati R, Berenfeld O, Chen J, Samie FH, Jalife J. Left-to-right gradient of atrial frequencies during acute atrial fibrillation in the isolated sheep heart. Circulation 2001;103:2631-6.
40. Morillo CA, Klein GJ, Jones DL, Guiraudon CM. Chronic rapid atrial pacing. Structural, functional, and electrophysiological characteristics of a new model of sustained atrial fibrillation. Circulation 1995;91:1588-95.
41. Jalife J, Berenfeld O, Mansour M. Mother rotors and fibrillatory conduction: a mechanism of atrial fibrillation. Cardiovascular research 2002;54:204-16.
42. Nitta T, Lee R, Watanabe H et al. Radial approach: a new concept in surgical treatment for atrial fibrillation. II. Electrophysiologic effects and atrial contribution to ventricular filling. The Annals of thoracic surgery 1999;67:36-50.
43. Haissaguerre M, Jais P, Shah DC et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. The New England journal of medicine 1998;339:659-66.
44. Haissaguerre M, Marcus FI, Fischer B, Clementy J. Radiofrequency catheter ablation in unusual mechanisms of atrial fibrillation: report of three cases. J Cardiovasc Electrophysiol 1994;5:743-51.
45. Jais P, Haissaguerre M, Shah DC et al. A focal source of atrial fibrillation treated by discrete radiofrequency ablation. Circulation 1997;95:572-6.
46. Hwang C, Wu TJ, Doshi RN, Peter CT, Chen PS. Vein of marshall cannulation for the analysis of electrical activity in patients with focal atrial fibrillation. Circulation 2000;101:1503-5.
47. Nathan H, Eliakim M. The junction between the left atrium and the pulmonary veins. An anatomic study of human hearts. Circulation 1966;34:412-22.
48. Chen CC, Tai CT, Chiang CE et al. Atrial tachycardias originating from the atrial septum: electrophysiologic characteristics and radiofrequency ablation. J Cardiovasc Electrophysiol 2000;11:744-9.
49. Hoffmann E, Reithmann C, Nimmermann P et al. Clinical experience with electroanatomic mapping of ectopic atrial tachycardia. Pacing and clinical electrophysiology : PACE 2002;25:49-56.
50. Kay GN, Chong F, Epstein AE, Dailey SM, Plumb VJ. Radiofrequency ablation for treatment of primary atrial tachycardias. J Am Coll Cardiol 1993;21:901-9.
51. Morton JB, Sanders P, Das A, Vohra JK, Sparks PB, Kalman JM. Focal atrial tachycardia arising from the tricuspid annulus: electrophysiologic and electrocardiographic characteristics. J Cardiovasc Electrophysiol 2001;12:653-9.
52. Natale A, Breeding L, Tomassoni G et al. Ablation of right and left ectopic atrial tachycardias using a three-dimensional nonfluoroscopic mapping system. Am J Cardiol 1998;82:989-92.
53. Tracy CM, Swartz JF, Fletcher RD et al. Radiofrequency catheter ablation of ectopic atrial tachycardia using paced activation sequence mapping. J Am Coll Cardiol 1993;21:910-7.
54. Volkmer M, Antz M, Hebe J, Kuck KH. Focal atrial tachycardia originating from the musculature of the coronary sinus. J Cardiovasc Electrophysiol 2002;13:68-71.
55. Chen SA, Hsieh MH, Tai CT et al. Initiation of atrial fibrillation by ectopic beats originating from the pulmonary veins: electrophysiological characteristics, pharmacological responses, and effects of radiofrequency ablation. Circulation 1999;100:1879-86.
179
56. Ho SY, Sanchez-Quintana D, Cabrera JA, Anderson RH. Anatomy of the left atrium: implications for radiofrequency ablation of atrial fibrillation. J Cardiovasc Electrophysiol 1999;10:1525-33.
57. Weiss C, Gocht A, Willems S, Hoffmann M, Risius T, Meinertz T. Impact of the distribution and structure of myocardium in the pulmonary veins for radiofrequency ablation of atrial fibrillation. Pacing and clinical electrophysiology : PACE 2002;25:1352-6.
58. Chen YC, Pan NH, Cheng CC, Higa S, Chen YJ, Chen SA. Heterogeneous expression of potassium currents and pacemaker currents potentially regulates arrhythmogenesis of pulmonary vein cardiomyocytes. J Cardiovasc Electrophysiol 2009;20:1039-45.
59. Ehrlich JR, Cha TJ, Zhang L et al. Cellular electrophysiology of canine pulmonary vein cardiomyocytes: action potential and ionic current properties. The Journal of physiology 2003;551:801-13.
60. Nilsson B, Chen X, Pehrson S, Kober L, Hilden J, Svendsen JH. Recurrence of pulmonary vein conduction and atrial fibrillation after pulmonary vein isolation for atrial fibrillation: a randomized trial of the ostial versus the extraostial ablation strategy. Am Heart J 2006;152:537.e1-8.
61. Oral H, Knight BP, Tada H et al. Pulmonary vein isolation for paroxysmal and persistent atrial fibrillation. Circulation 2002;105:1077-81.
62. Voskoboinik A, Moskovitch JT, Harel N, Sanders P, Kistler PM, Kalman JM. Revisiting pulmonary vein isolation alone for persistent atrial fibrillation: A systematic review and meta-analysis. Heart Rhythm 2017;14:661-667.
63. Sanders P, Berenfeld O, Hocini M et al. Spectral analysis identifies sites of high-frequency activity maintaining atrial fibrillation in humans. Circulation 2005;112:789-97.
64. Lin YJ, Tsao HM, Chang SL et al. Role of high dominant frequency sites in nonparoxysmal atrial fibrillation patients: insights from high-density frequency and fractionation mapping. Heart Rhythm 2010;7:1255-62.
65. Pauza DH, Skripka V, Pauziene N, Stropus R. Morphology, distribution, and variability of the epicardiac neural ganglionated subplexuses in the human heart. The Anatomical record 2000;259:353-82.
66. Sakamoto S, Fujii M, Watanabe Y et al. Exploration of theoretical ganglionated plexi ablation technique in atrial fibrillation surgery. The Annals of thoracic surgery 2014;98:1598-604.
67. Patterson E, Po SS, Scherlag BJ, Lazzara R. Triggered firing in pulmonary veins initiated by in vitro autonomic nerve stimulation. Heart Rhythm 2005;2:624-31.
68. Scherlag BJ, Yamanashi W, Patel U, Lazzara R, Jackman WM. Autonomically induced conversion of pulmonary vein focal firing into atrial fibrillation. J Am Coll Cardiol 2005;45:1878-86.
69. Pokushalov E, Romanov A, Artyomenko S, Turov A, Shirokova N, Katritsis DG. Left atrial ablation at the anatomic areas of ganglionated plexi for paroxysmal atrial fibrillation. Pacing and clinical electrophysiology : PACE 2010;33:1231-8.
70. Scherlag BJ, Hou YL, Lin J et al. An acute model for atrial fibrillation arising from a peripheral atrial site: evidence for primary and secondary triggers. J Cardiovasc Electrophysiol 2008;19:519-27.
71. Lau DH, Maesen B, Zeemering S, Verheule S, Crijns HJ, Schotten U. Stability of complex fractionated atrial electrograms: a systematic review. J Cardiovasc Electrophysiol 2012;23:980-7.
72. O'Neill MD, Jais P, Takahashi Y et al. The stepwise ablation approach for chronic atrial fibrillation--evidence for a cumulative effect. Journal of interventional cardiac electrophysiology : an international journal of arrhythmias and pacing 2006;16:153-67.
73. Stiles MK, Brooks AG, Kuklik P et al. High-density mapping of atrial fibrillation in humans: relationship between high-frequency activation and electrogram fractionation. J Cardiovasc Electrophysiol 2008;19:1245-53.
180
74. Verma A, Wulffhart Z, Beardsall M, Whaley B, Hill C, Khaykin Y. Spatial and temporal stability of complex fractionated electrograms in patients with persistent atrial fibrillation over longer time periods: relationship to local electrogram cycle length. Heart Rhythm 2008;5:1127-33.
75. Verheule S, Tuyls E, van Hunnik A, Kuiper M, Schotten U, Allessie M. Fibrillatory conduction in the atrial free walls of goats in persistent and permanent atrial fibrillation. Circulation Arrhythmia and electrophysiology 2010;3:590-9.
76. Eckstein J, Zeemering S, Linz D et al. Transmural conduction is the predominant mechanism of breakthrough during atrial fibrillation: evidence from simultaneous endo-epicardial high-density activation mapping. Circulation Arrhythmia and electrophysiology 2013;6:334-41.
77. de Groot NM, Houben RP, Smeets JL et al. Electropathological substrate of longstanding persistent atrial fibrillation in patients with structural heart disease: epicardial breakthrough. Circulation 2010;122:1674-82.
78. Chen J, Mandapati R, Berenfeld O, Skanes AC, Gray RA, Jalife J. Dynamics of wavelets and their role in atrial fibrillation in the isolated sheep heart. Cardiovascular research 2000;48:220-32.
79. Filgueiras-Rama D, Price NF, Martins RP et al. Long-term frequency gradients during persistent atrial fibrillation in sheep are associated with stable sources in the left atrium. Circulation Arrhythmia and electrophysiology 2012;5:1160-7.
80. Kalifa J, Tanaka K, Zaitsev AV et al. Mechanisms of wave fractionation at boundaries of high-frequency excitation in the posterior left atrium of the isolated sheep heart during atrial fibrillation. Circulation 2006;113:626-33.
81. Mandapati R, Skanes A, Chen J, Berenfeld O, Jalife J. Stable microreentrant sources as a mechanism of atrial fibrillation in the isolated sheep heart. Circulation 2000;101:194-9.
82. Skanes AC, Mandapati R, Berenfeld O, Davidenko JM, Jalife J. Spatiotemporal periodicity during atrial fibrillation in the isolated sheep heart. Circulation 1998;98:1236-48.
83. Haissaguerre M, Hocini M, Denis A et al. Driver domains in persistent atrial fibrillation. Circulation 2014;130:530-8.
84. Narayan SM, Krummen DE, Shivkumar K, Clopton P, Rappel WJ, Miller JM. Treatment of atrial fibrillation by the ablation of localized sources: CONFIRM (Conventional Ablation for Atrial Fibrillation With or Without Focal Impulse and Rotor Modulation) trial. J Am Coll Cardiol 2012;60:628-36.
85. Rodrigo M, Guillem MS, Climent AM et al. Body surface localization of left and right atrial high-frequency rotors in atrial fibrillation patients: a clinical-computational study. Heart Rhythm 2014;11:1584-91.
86. Jones AR, Krummen DE, Narayan SM. Non-invasive identification of stable rotors and focal sources for human atrial fibrillation: mechanistic classification of atrial fibrillation from the electrocardiogram. Europace 2013;15:1249-58.
87. Atienza F, Almendral J, Ormaetxe JM et al. Comparison of radiofrequency catheter ablation of drivers and circumferential pulmonary vein isolation in atrial fibrillation: a noninferiority randomized multicenter RADAR-AF trial. J Am Coll Cardiol 2014;64:2455-67.
88. Narayan SM, Krummen DE, Rappel WJ. Clinical mapping approach to diagnose electrical rotors and focal impulse sources for human atrial fibrillation. J Cardiovasc Electrophysiol 2012;23:447-54.
89. Haissaguerre M, Hocini M, Shah AJ et al. Noninvasive panoramic mapping of human atrial fibrillation mechanisms: a feasibility report. J Cardiovasc Electrophysiol 2013;24:711-7.
90. Lee G, Kumar S, Teh A et al. Epicardial wave mapping in human long-lasting persistent atrial fibrillation: transient rotational circuits, complex wavefronts, and disorganized activity. Eur Heart J 2014;35:86-97.
91. Allessie MA, Boyden PA, Camm AJ et al. Pathophysiology and prevention of atrial fibrillation. Circulation 2001;103:769-77.
181
92. Rensma PL, Allessie MA, Lammers WJ, Bonke FI, Schalij MJ. Length of excitation wave and susceptibility to reentrant atrial arrhythmias in normal conscious dogs. Circulation research 1988;62:395-410.
93. Henry WL, Morganroth J, Pearlman AS et al. Relation between echocardiographically determined left atrial size and atrial fibrillation. Circulation 1976;53:273-9.
94. Olgin JE, Kalman JM, Fitzpatrick AP, Lesh MD. Role of right atrial endocardial structures as barriers to conduction during human type I atrial flutter. Activation and entrainment mapping guided by intracardiac echocardiography. Circulation 1995;92:1839-48.
95. Jais P, Shah DC, Haissaguerre M et al. Mapping and ablation of left atrial flutters. Circulation 2000;101:2928-34.
96. Papageorgiou P, Monahan K, Boyle NG et al. Site-dependent intra-atrial conduction delay. Relationship to initiation of atrial fibrillation. Circulation 1996;94:384-9.
97. Sun Y, Arruda M, Otomo K et al. Coronary sinus-ventricular accessory connections producing posteroseptal and left posterior accessory pathways: incidence and electrophysiological identification. Circulation 2002;106:1362-7.
98. Burstein B, Nattel S. Atrial fibrosis: mechanisms and clinical relevance in atrial fibrillation. J Am Coll Cardiol 2008;51:802-9.
99. Abed HS, Samuel CS, Lau DH et al. Obesity results in progressive atrial structural and electrical remodeling: implications for atrial fibrillation. Heart Rhythm 2013;10:90-100.
100. Kirchhof P, Benussi S, Kotecha D et al. 2016 ESC Guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Europace 2016;18:1609-1678.
101. Abed HS, Wittert GA, Leong DP et al. Effect of weight reduction and cardiometabolic risk factor management on symptom burden and severity in patients with atrial fibrillation: a randomized clinical trial. JAMA : the journal of the American Medical Association 2013;310:2050-60.
102. Mahajan R, Lau DH, Brooks AG et al. Electrophysiological, Electroanatomical, and Structural Remodeling of the Atria as Consequences of Sustained Obesity. J Am Coll Cardiol 2015;66:1-11.
103. Gallagher C, Hendriks JM, Mahajan R et al. Lifestyle management to prevent and treat atrial fibrillation. Expert review of cardiovascular therapy 2016;14:799-809.
104. Malmo V, Nes BM, Amundsen BH et al. Aerobic Interval Training Reduces the Burden of Atrial Fibrillation in the Short Term: A Randomized Trial. Circulation 2016;133:466-73.
105. Pathak RK, Middeldorp ME, Meredith M et al. Long-Term Effect of Goal-Directed Weight Management in an Atrial Fibrillation Cohort: A Long-Term Follow-Up Study (LEGACY). J Am Coll Cardiol 2015;65:2159-69.
106. Pathak RK, Middeldorp ME, Lau DH et al. Aggressive risk factor reduction study for atrial fibrillation and implications for the outcome of ablation: the ARREST-AF cohort study. J Am Coll Cardiol 2014;64:2222-31.
107. Carter L, Gardner M, Magee K et al. An Integrated Management Approach to Atrial Fibrillation. Journal of the American Heart Association 2016;5.
108. Hendriks JM, de Wit R, Crijns HJ et al. Nurse-led care vs. usual care for patients with atrial fibrillation: results of a randomized trial of integrated chronic care vs. routine clinical care in ambulatory patients with atrial fibrillation. Eur Heart J 2012;33:2692-9.
109. Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest 2010;137:263-72.
110. Lip GY, Al-Khatib SM, Cosio FG et al. Contemporary management of atrial fibrillation: what can clinical registries tell us about stroke prevention and current therapeutic approaches? Journal of the American Heart Association 2014;3.
111. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study. Lancet 1994;343:687-91.
182
112. Oral anticoagulation in patients after cerebral ischemia of arterial origin and risk of intracranial hemorrhage. Stroke 2003;34:e45-6.
113. Connolly S, Pogue J, Hart R et al. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet 2006;367:1903-12.
114. Granger CB, Alexander JH, McMurray JJ et al. Apixaban versus warfarin in patients with atrial fibrillation. The New England journal of medicine 2011;365:981-92.
115. Lopes RD, Al-Khatib SM, Wallentin L et al. Efficacy and safety of apixaban compared with warfarin according to patient risk of stroke and of bleeding in atrial fibrillation: a secondary analysis of a randomised controlled trial. Lancet 2012;380:1749-58.
116. Lip GY, Lim HS. Atrial fibrillation and stroke prevention. Lancet neurology 2007;6:981-93. 117. Mant J, Hobbs FD, Fletcher K et al. Warfarin versus aspirin for stroke prevention in an elderly
community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet 2007;370:493-503.
118. Diener HC, Eikelboom J, Connolly SJ et al. Apixaban versus aspirin in patients with atrial fibrillation and previous stroke or transient ischaemic attack: a predefined subgroup analysis from AVERROES, a randomised trial. Lancet neurology 2012;11:225-31.
119. Eikelboom JW, Connolly SJ, Brueckmann M et al. Dabigatran versus warfarin in patients with mechanical heart valves. The New England journal of medicine 2013;369:1206-14.
120. Holmes DR, Reddy VY, Turi ZG et al. Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation: a randomised non-inferiority trial. Lancet 2009;374:534-42.
121. Reddy VY, Sievert H, Halperin J et al. Percutaneous left atrial appendage closure vs warfarin for atrial fibrillation: a randomized clinical trial. JAMA : the journal of the American Medical Association 2014;312:1988-98.
122. Whitlock R, Healey J, Vincent J et al. Rationale and design of the Left Atrial Appendage Occlusion Study (LAAOS) III. Annals of cardiothoracic surgery 2014;3:45-54.
123. Healey JS, Crystal E, Lamy A et al. Left Atrial Appendage Occlusion Study (LAAOS): results of a randomized controlled pilot study of left atrial appendage occlusion during coronary bypass surgery in patients at risk for stroke. Am Heart J 2005;150:288-93.
124. Whitlock RP, Vincent J, Blackall MH et al. Left Atrial Appendage Occlusion Study II (LAAOS II). The Canadian journal of cardiology 2013;29:1443-7.
125. Olshansky B, Rosenfeld LE, Warner AL et al. The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study: approaches to control rate in atrial fibrillation. J Am Coll Cardiol 2004;43:1201-8.
126. Saksena S, Slee A, Waldo AL et al. Cardiovascular outcomes in the AFFIRM Trial (Atrial Fibrillation Follow-Up Investigation of Rhythm Management). An assessment of individual antiarrhythmic drug therapies compared with rate control with propensity score-matched analyses. J Am Coll Cardiol 2011;58:1975-85.
127. Wyse DG, Waldo AL, DiMarco JP et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. The New England journal of medicine 2002;347:1825-33.
128. Carlsson J, Miketic S, Windeler J et al. Randomized trial of rate-control versus rhythm-control in persistent atrial fibrillation: the Strategies of Treatment of Atrial Fibrillation (STAF) study. J Am Coll Cardiol 2003;41:1690-6.
129. Hohnloser SH, Kuck KH, Lilienthal J. Rhythm or rate control in atrial fibrillation--Pharmacological Intervention in Atrial Fibrillation (PIAF): a randomised trial. Lancet 2000;356:1789-94.
130. Ogawa S, Yamashita T, Yamazaki T et al. Optimal treatment strategy for patients with paroxysmal atrial fibrillation: J-RHYTHM Study. Circulation journal : official journal of the Japanese Circulation Society 2009;73:242-8.
183
131. Shelton RJ, Clark AL, Goode K et al. A randomised, controlled study of rate versus rhythm control in patients with chronic atrial fibrillation and heart failure: (CAFE-II Study). Heart 2009;95:924-30.
132. Caldeira D, David C, Sampaio C. Rate versus rhythm control in atrial fibrillation and clinical outcomes: updated systematic review and meta-analysis of randomized controlled trials. Archives of cardiovascular diseases 2012;105:226-38.
133. de Denus S, Sanoski CA, Carlsson J, Opolski G, Spinler SA. Rate vs rhythm control in patients with atrial fibrillation: a meta-analysis. Arch Intern Med 2005;165:258-62.
134. Gillinov AM, Bagiella E, Moskowitz AJ et al. Rate Control versus Rhythm Control for Atrial Fibrillation after Cardiac Surgery. The New England journal of medicine 2016;374:1911-21.
135. Lafuente-Lafuente C, Longas-Tejero MA, Bergmann JF, Belmin J. Antiarrhythmics for maintaining sinus rhythm after cardioversion of atrial fibrillation. Cochrane database of systematic reviews (Online) 2012:Cd005049.
136. Roy D, Talajic M, Nattel S et al. Rhythm control versus rate control for atrial fibrillation and heart failure. The New England journal of medicine 2008;358:2667-77.
137. Al-Khatib SM, Allen LaPointe NM, Chatterjee R et al. Rate- and rhythm-control therapies in patients with atrial fibrillation: a systematic review. Ann Intern Med 2014;160:760-73.
138. Cosedis Nielsen J, Mortensen LS, Hansen PS. Radiofrequency ablation as initial therapy in paroxysmal atrial fibrillation. The New England journal of medicine 2013;368:478-9.
139. Wilber DJ, Pappone C, Neuzil P et al. Comparison of antiarrhythmic drug therapy and radiofrequency catheter ablation in patients with paroxysmal atrial fibrillation: a randomized controlled trial. JAMA : the journal of the American Medical Association 2010;303:333-40.
140. Anselmino M, Matta M, D'Ascenzo F et al. Catheter ablation of atrial fibrillation in patients with left ventricular systolic dysfunction: a systematic review and meta-analysis. Circulation Arrhythmia and electrophysiology 2014;7:1011-8.
141. Jais P, Cauchemez B, Macle L et al. Catheter ablation versus antiarrhythmic drugs for atrial fibrillation: the A4 study. Circulation 2008;118:2498-505.
142. Wilber DJ, Pappone C, Neuzil P et al. Comparison of antiarrhythmic drug therapy and radiofrequency catheter ablation in patients with paroxysmal atrial fibrillation: a randomized controlled trial. JAMA : the journal of the American Medical Association 2010;303:333-40.
143. Kobza R, Kottkamp H, Dorszewski A et al. Stable secondary arrhythmias late after intraoperative radiofrequency ablation of atrial fibrillation: incidence, mechanism, and treatment. J Cardiovasc Electrophysiol 2004;15:1246-9.
144. Ouyang F, Ernst S, Chun J et al. Electrophysiological findings during ablation of persistent atrial fibrillation with electroanatomic mapping and double Lasso catheter technique. Circulation 2005;112:3038-48.
145. Calkins H, Kuck KH, Cappato R et al. 2012 HRS/EHRA/ECAS Expert Consensus Statement on Catheter and Surgical Ablation of Atrial Fibrillation: recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design. Europace 2012;14:528-606.
146. Kuck KH, Brugada J, Furnkranz A et al. Cryoballoon or Radiofrequency Ablation for Paroxysmal Atrial Fibrillation. The New England journal of medicine 2016;374:2235-45.
147. Nath S, DiMarco JP, Haines DE. Basic aspects of radiofrequency catheter ablation. J Cardiovasc Electrophysiol 1994;5:863-76.
148. Wong MC, Edwards G, Spence SJ et al. Characterization of catheter-tissue contact force during epicardial radiofrequency ablation in an ovine model. Circulation Arrhythmia and electrophysiology 2013;6:1222-8.
149. Holmes D, Fish JM, Byrd IA et al. Contact sensing provides a highly accurate means to titrate radiofrequency ablation lesion depth. J Cardiovasc Electrophysiol 2011;22:684-90.
184
150. Natale A, Reddy VY, Monir G et al. Paroxysmal AF catheter ablation with a contact force sensing catheter: results of the prospective, multicenter SMART-AF trial. J Am Coll Cardiol 2014;64:647-56.
151. Reddy VY, Dukkipati SR, Neuzil P et al. Randomized, Controlled Trial of the Safety and Effectiveness of a Contact Force-Sensing Irrigated Catheter for Ablation of Paroxysmal Atrial Fibrillation: Results of the TactiCath Contact Force Ablation Catheter Study for Atrial Fibrillation (TOCCASTAR) Study. Circulation 2015;132:907-15.
152. Holman WL, Ikeshita M, Douglas JM, Jr., Smith PK, Lofland GK, Cox JL. Ventricular cryosurgery: short-term effects on intramural electrophysiology. The Annals of thoracic surgery 1983;35:386-93.
153. Melby SJ, Schuessler RB, Damiano RJ, Jr. Ablation technology for the surgical treatment of atrial fibrillation. ASAIO journal (American Society for Artificial Internal Organs : 1992) 2013;59:461-8.
154. Meininger GR, Calkins H, Lickfett L et al. Initial experience with a novel focused ultrasound ablation system for ring ablation outside the pulmonary vein. Journal of interventional cardiac electrophysiology : an international journal of arrhythmias and pacing 2003;8:141-8.
155. Doll N, Suwalski P, Aupperle H et al. Endocardial laser ablation for the treatment of atrial fibrillation in an acute sheep model. J Card Surg 2008;23:198-203.
156. Robbins IM, Colvin EV, Doyle TP et al. Pulmonary vein stenosis after catheter ablation of atrial fibrillation. Circulation 1998;98:1769-75.
157. Haissaguerre M, Jais P, Shah DC et al. Electrophysiological end point for catheter ablation of atrial fibrillation initiated from multiple pulmonary venous foci. Circulation 2000;101:1409-17.
158. Arentz T, Weber R, Burkle G et al. Small or large isolation areas around the pulmonary veins for the treatment of atrial fibrillation? Results from a prospective randomized study. Circulation 2007;115:3057-63.
159. Ernst S, Ouyang F, Lober F, Antz M, Kuck KH. Catheter-induced linear lesions in the left atrium in patients with atrial fibrillation: an electroanatomic study. J Am Coll Cardiol 2003;42:1271-82.
160. Jais P, Hocini M, Hsu LF et al. Technique and results of linear ablation at the mitral isthmus. Circulation 2004;110:2996-3002.
161. Lin WS, Tai CT, Hsieh MH et al. Catheter ablation of paroxysmal atrial fibrillation initiated by non-pulmonary vein ectopy. Circulation 2003;107:3176-83.
162. Tsai CF, Tai CT, Hsieh MH et al. Initiation of atrial fibrillation by ectopic beats originating from the superior vena cava: electrophysiological characteristics and results of radiofrequency ablation. Circulation 2000;102:67-74.
163. Nademanee K, McKenzie J, Kosar E et al. A new approach for catheter ablation of atrial fibrillation: mapping of the electrophysiologic substrate. J Am Coll Cardiol 2004;43:2044-53.
164. Oral H, Chugh A, Good E et al. Radiofrequency catheter ablation of chronic atrial fibrillation guided by complex electrograms. Circulation 2007;115:2606-12.
165. Scherlag BJ, Nakagawa H, Jackman WM et al. Electrical stimulation to identify neural elements on the heart: their role in atrial fibrillation. Journal of interventional cardiac electrophysiology : an international journal of arrhythmias and pacing 2005;13 Suppl 1:37-42.
166. Hocini M, Jais P, Sanders P et al. Techniques, evaluation, and consequences of linear block at the left atrial roof in paroxysmal atrial fibrillation: a prospective randomized study. Circulation 2005;112:3688-96.
167. Sauer WH, Alonso C, Zado E et al. Atrioventricular nodal reentrant tachycardia in patients referred for atrial fibrillation ablation: response to ablation that incorporates slow-pathway modification. Circulation 2006;114:191-5.
185
168. Haissaguerre M, Hocini M, Sanders P et al. Catheter ablation of long-lasting persistent atrial fibrillation: clinical outcome and mechanisms of subsequent arrhythmias. J Cardiovasc Electrophysiol 2005;16:1138-47.
169. Takahashi Y, O'Neill MD, Hocini M et al. Effects of stepwise ablation of chronic atrial fibrillation on atrial electrical and mechanical properties. J Am Coll Cardiol 2007;49:1306-14.
170. Verma A, Jiang CY, Betts TR et al. Approaches to catheter ablation for persistent atrial fibrillation. The New England journal of medicine 2015;372:1812-22.
171. Arujuna A, Karim R, Caulfield D et al. Acute pulmonary vein isolation is achieved by a combination of reversible and irreversible atrial injury after catheter ablation: evidence from magnetic resonance imaging. Circulation Arrhythmia and electrophysiology 2012;5:691-700.
172. Ouyang F, Tilz R, Chun J et al. Long-term results of catheter ablation in paroxysmal atrial fibrillation: lessons from a 5-year follow-up. Circulation 2010;122:2368-77.
173. Tzou WS, Marchlinski FE, Zado ES et al. Long-term outcome after successful catheter ablation of atrial fibrillation. Circulation Arrhythmia and electrophysiology 2010;3:237-42.
174. Weerasooriya R, Khairy P, Litalien J et al. Catheter ablation for atrial fibrillation: are results maintained at 5 years of follow-up? J Am Coll Cardiol 2011;57:160-6.
175. Bonanno C, Paccanaro M, La Vecchia L, Ometto R, Fontanelli A. Efficacy and safety of catheter ablation versus antiarrhythmic drugs for atrial fibrillation: a meta-analysis of randomized trials. Journal of cardiovascular medicine (Hagerstown, Md) 2010;11:408-18.
176. Ganesan AN, Shipp NJ, Brooks AG et al. Long-term outcomes of catheter ablation of atrial fibrillation: a systematic review and meta-analysis. Journal of the American Heart Association 2013;2:e004549.
177. Williams JM, Ungerleider RM, Lofland GK, Cox JL. Left atrial isolation: new technique for the treatment of supraventricular arrhythmias. The Journal of thoracic and cardiovascular surgery 1980;80:373-80.
178. Guiraudon G, Campbell CS, Jones DL, McLellan JL, MacDonald JL. Combined sino-atrial node atrio-ventricular node isolation: a surgical alternative to His bundle ablation in patients with atrial fibrillation [Abstract]. Circulation 1985;72.
179. Cox JL. The surgical treatment of atrial fibrillation. IV. Surgical technique. The Journal of thoracic and cardiovascular surgery 1991;101:584-92.
180. Cox JL, Boineau JP, Schuessler RB et al. Operations for atrial fibrillation. Clin Cardiol 1991;14:827-34.
181. Cox JL, Boineau JP, Schuessler RB, Jaquiss RD, Lappas DG. Modification of the maze procedure for atrial flutter and atrial fibrillation. I. Rationale and surgical results. The Journal of thoracic and cardiovascular surgery 1995;110:473-84.
182. Cox JL, Jaquiss RD, Schuessler RB, Boineau JP. Modification of the maze procedure for atrial flutter and atrial fibrillation. II. Surgical technique of the maze III procedure. The Journal of thoracic and cardiovascular surgery 1995;110:485-95.
183. Lonnerholm S, Blomstrom P, Nilsson L, Blomstrom-Lundqvist C. Atrial size and transport function after the Maze III procedure for paroxysmal atrial fibrillation. The Annals of thoracic surgery 2002;73:107-11.
184. Schaff HV, Dearani JA, Daly RC, Orszulak TA, Danielson GK. Cox-Maze procedure for atrial fibrillation: Mayo Clinic experience. Semin Thorac Cardiovasc Surg 2000;12:30-7.
185. Feinberg MS, Waggoner AD, Kater KM, Cox JL, Lindsay BD, Perez JE. Restoration of atrial function after the maze procedure for patients with atrial fibrillation. Assessment by Doppler echocardiography. Circulation 1994;90:Ii285-92.
186. Prasad SM, Maniar HS, Camillo CJ et al. The Cox maze III procedure for atrial fibrillation: long-term efficacy in patients undergoing lone versus concomitant procedures. The Journal of thoracic and cardiovascular surgery 2003;126:1822-8.
186
187. Gaynor SL, Diodato MD, Prasad SM et al. A prospective, single-center clinical trial of a modified Cox maze procedure with bipolar radiofrequency ablation. The Journal of thoracic and cardiovascular surgery 2004;128:535-42.
188. Damiano RJ, Jr., Schwartz FH, Bailey MS et al. The Cox maze IV procedure: predictors of late recurrence. The Journal of thoracic and cardiovascular surgery 2011;141:113-21.
189. Weimar T, Schena S, Bailey MS et al. The cox-maze procedure for lone atrial fibrillation: a single-center experience over 2 decades. Circulation Arrhythmia and electrophysiology 2012;5:8-14.
190. Ad N, Holmes SD, Shuman DJ, Pritchard G. Impact of Atrial Fibrillation Duration on the Success of First-Time Concomitant Cox Maze Procedures. The Annals of thoracic surgery 2015;100:1613-8; discussion 1618-9.
191. Henn MC, Lancaster TS, Miller JR et al. Late outcomes after the Cox maze IV procedure for atrial fibrillation. The Journal of thoracic and cardiovascular surgery 2015;150:1168-76, 1178.e1-2.
192. Damiano RJ, Jr., Badhwar V, Acker MA et al. The CURE-AF trial: a prospective, multicenter trial of irrigated radiofrequency ablation for the treatment of persistent atrial fibrillation during concomitant cardiac surgery. Heart Rhythm 2014;11:39-45.
193. Gillinov AM, Gelijns AC, Parides MK et al. Surgical ablation of atrial fibrillation during mitral-valve surgery. The New England journal of medicine 2015;372:1399-409.
194. Brodell GK, Cosgrove D, Schiavone W, Underwood DA, Loop FD. Cardiac rhythm and conduction disturbances in patients undergoing mitral valve surgery. Cleveland Clinic journal of medicine 1991;58:397-9.
195. Chua YL, Schaff HV, Orszulak TA, Morris JJ. Outcome of mitral valve repair in patients with preoperative atrial fibrillation. Should the maze procedure be combined with mitral valvuloplasty? The Journal of thoracic and cardiovascular surgery 1994;107:408-15.
196. Petty GW, Khandheria BK, Whisnant JP, Sicks JD, O'Fallon WM, Wiebers DO. Predictors of cerebrovascular events and death among patients with valvular heart disease: A population-based study. Stroke 2000;31:2628-35.
197. Ad N, Cox JL. The Maze procedure for the treatment of atrial fibrillation: a minimally invasive approach. J Card Surg 2004;19:196-200.
198. Ad N, Henry L, Friehling T, Wish M, Holmes SD. Minimally invasive stand-alone Cox-maze procedure for patients with nonparoxysmal atrial fibrillation. The Annals of thoracic surgery 2013;96:792-8; discussion 798-9.
199. Balkhy HH, Chapman PD, Arnsdorf SE. Minimally invasive atrial fibrillation ablation combined with a new technique for thoracoscopic stapling of the left atrial appendage: case report. The heart surgery forum 2004;7:353-5.
200. Gillinov AM, Mihaljevic T. Thoracoscopic epicardial radiofrequency ablation for atrial fibrillation: commentary. Heart 2009;95:1110-1.
201. Pruitt JC, Lazzara RR, Dworkin GH, Badhwar V, Kuma C, Ebra G. Totally endoscopic ablation of lone atrial fibrillation: initial clinical experience. The Annals of thoracic surgery 2006;81:1325-30; discussion 1330-1.
202. Pruitt JC, Lazzara RR, Ebra G. Minimally invasive surgical ablation of atrial fibrillation: the thoracoscopic box lesion approach. Journal of interventional cardiac electrophysiology : an international journal of arrhythmias and pacing 2007;20:83-7.
203. Yilmaz A, Geuzebroek GS, Van Putte BP et al. Completely thoracoscopic pulmonary vein isolation with ganglionic plexus ablation and left atrial appendage amputation for treatment of atrial fibrillation. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2010;38:356-60.
204. Sirak J, Jones D, Schwartzman D. The five-box thoracoscopic maze procedure. The Annals of thoracic surgery 2010;90:986-9.
187
205. Sirak J, Jones D, Sun B, Sai-Sudhakar C, Crestanello J, Firstenberg M. Toward a definitive, totally thoracoscopic procedure for atrial fibrillation. The Annals of thoracic surgery 2008;86:1960-4.
206. Sirak JH, Schwartzman D. Interim results of the 5-box thoracoscopic maze procedure. The Annals of thoracic surgery 2012;94:1880-4.
207. Pison L, La Meir M, van Opstal J, Blaauw Y, Maessen J, Crijns HJ. Hybrid thoracoscopic surgical and transvenous catheter ablation of atrial fibrillation. J Am Coll Cardiol 2012;60:54-61.
208. Muneretto C, Bisleri G, Bontempi L, Curnis A. Durable staged hybrid ablation with thoracoscopic and percutaneous approach for treatment of long-standing atrial fibrillation: a 30-month assessment with continuous monitoring. The Journal of thoracic and cardiovascular surgery 2012;144:1460-5; discussion 1465.
209. Pison L, Gelsomino S, Luca F et al. Effectiveness and safety of simultaneous hybrid thoracoscopic and endocardial catheter ablation of lone atrial fibrillation. Annals of cardiothoracic surgery 2014;3:38-44.
210. Badhwar V, Rankin JS, Damiano RJ, Jr. et al. The Society of Thoracic Surgeons 2017 Clinical Practice Guidelines for the Surgical Treatment of Atrial Fibrillation. The Annals of thoracic surgery 2017;103:329-341.
211. Atricure. RF Ablation, Pacing and Sensing. 2017. 212. Atricure. Isolator Synergy Clamps. 2017. 213. Atricure. Isolator Synergy Clamps - EML2,EMR2. 2017. 214. Atricure. Isolator Synergy Access Clamp. 2017. 215. Medtronic. ABLATION PRODUCTS – SURGICAL. 2017. 216. Medtronic. Cardioblate BP2. 2017. 217. Medtronic. Cardioblate LP. 2017. 218. Medtronic. Cardioblate Gemini. 2017. 219. Atricure. Ablation Sensing Unit & Switch Matrix. 2017. 220. Medtronic. Cardioblate 68000. 2017. 221. Atricure. Coolrail Linear Pen. 2017. 222. Atricure. Isolator Linear Pen. 2017. 223. Atricure. Isolator Transpolar Pen. 2017. 224. Medtronic. Cardioblate Surgical Pen. 2017. 225. Medtronic. Cardioblate MAPS. 2017. 226. Atricure. COBRA Fusion 150 and 50 Ablation System. 2017. 227. Atricure. EPi-Sense Coagulation Device. 2017. 228. Atricure. Cryo. 2017. 229. Medtronic. Cardioblate CryoFlex. 2017. 230. Medtronic. Cardioblate CryoFlex Clamp. 2017. 231. Lee SK, Choo SJ, Kim KS, Lee JW. Epicardial microwave application in chronic atrial fibrillation
surgery. Journal of Korean medical science 2005;20:727-31. 232. MacDonald DRW, Maruthappu M, Nagendran M. How effective is microwave ablation for
atrial fibrillation during concomitant cardiac surgery? Interactive Cardiovascular and Thoracic Surgery 2012;15:122-127.
233. Manasse E, Medici D, Ghiselli S, Ornaghi D, Gallotti R. Left main coronary arterial lesion after microwave epicardial ablation. The Annals of thoracic surgery 2003;76:276-7.
234. Molloy TA. Midterm clinical experience with microwave surgical ablation of atrial fibrillation. The Annals of thoracic surgery 2005;79:2115-8.
235. Bokeriia LA, Borisov KV, Revishvili A et al. [Surgical treatment of supraventricular tachyarrhythmias using laser]. Grudnaia i serdechno-sosudistaia khirurgiia / Ministerstvo zdravookhraneniia SSSR [i] Vsesoiuznoe nauchnoe obshchestvo khirurgov 1992:23-30.
188
236. Dorschler K, Muller G. The role of laser in cardiac surgery. The Thoracic and cardiovascular surgeon 1999;47 Suppl 3:385-7.
237. Schade A, Krug J, Szollosi AG, El Tarahony M, Deneke T. Pulmonary vein isolation with a novel endoscopic ablation system using laser energy. Expert review of cardiovascular therapy 2012;10:995-1000.
238. Metzner A, Chun KR, Neven K et al. Long-term clinical outcome following pulmonary vein isolation with high-intensity focused ultrasound balloon catheters in patients with paroxysmal atrial fibrillation. Europace 2010;12:188-93.
239. Mitnovetski S, Almeida AA, Goldstein J, Pick AW, Smith JA. Epicardial high-intensity focused ultrasound cardiac ablation for surgical treatment of atrial fibrillation. Heart, lung & circulation 2009;18:28-31.
240. Neven K, Metzner A, Schmidt B, Ouyang F, Kuck KH. Two-year clinical follow-up after pulmonary vein isolation using high-intensity focused ultrasound (HIFU) and an esophageal temperature-guided safety algorithm. Heart Rhythm 2012;9:407-13.
241. Melby SJ, Zierer A, Bailey MS et al. A new era in the surgical treatment of atrial fibrillation: the impact of ablation technology and lesion set on procedural efficacy. Ann Surg 2006;244:583-92.
242. Doll N, Kornherr P, Aupperle H et al. Epicardial treatment of atrial fibrillation using cryoablation in an acute off-pump sheep model. The Thoracic and cardiovascular surgeon 2003;51:267-73.
243. Aupperle H, Doll N, Walther T et al. Ablation of atrial fibrillation and esophageal injury: effects of energy source and ablation technique. The Journal of thoracic and cardiovascular surgery 2005;130:1549-54.
244. Gillinov AM, Pettersson G, Rice TW. Esophageal injury during radiofrequency ablation for atrial fibrillation. The Journal of thoracic and cardiovascular surgery 2001;122:1239-40.
245. Holman WL, Ikeshita M, Ungerleider RM, Smith PK, Ideker RE, Cox JL. Cryosurgery for cardiac arrhythmias: acute and chronic effects on coronary arteries. Am J Cardiol 1983;51:149-55.
246. Iida S, Misaki T, Iwa T. The histological effects of cryocoagulation on the myocardium and coronary arteries. The Japanese journal of surgery 1989;19:319-25.
247. Mikat EM, Hackel DB, Harrison L, Gallagher JJ, Wallace AG. Reaction of the myocardium and coronary arteries to cryosurgery. Laboratory investigation; a journal of technical methods and pathology 1977;37:632-41.
248. Misaki T, Allwork SP, Bentall HH. Longterm effects of cryosurgery in the sheep heart. Cardiovascular research 1983;17:61-9.
249. Guiraudon GM, Jones DL, Skanes AC et al. En bloc exclusion of the pulmonary vein region in the pig using off pump, beating, intra-cardiac surgery: a pilot study of minimally invasive surgery for atrial fibrillation. The Annals of thoracic surgery 2005;80:1417-23.
250. Bugge E, Nicholson IA, Thomas SP. Comparison of bipolar and unipolar radiofrequency ablation in an in vivo experimental model. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2005;28:76-80; discussion 80-2.
251. Miyagi Y, Ishii Y, Nitta T, Ochi M, Shimizu K. Electrophysiological and histological assessment of transmurality after epicardial ablation using unipolar radiofrequency energy. J Card Surg 2009;24:34-40.
252. Thomas SP, Guy DJ, Boyd AC, Eipper VE, Ross DL, Chard RB. Comparison of epicardial and endocardial linear ablation using handheld probes. The Annals of thoracic surgery 2003;75:543-8.
253. Gaynor SL, Ishii Y, Diodato MD et al. Successful performance of Cox-Maze procedure on beating heart using bipolar radiofrequency ablation: a feasibility study in animals. The Annals of thoracic surgery 2004;78:1671-7.
189
254. Melby SJ, Gaynor SL, Lubahn JG et al. Efficacy and safety of right and left atrial ablations on the beating heart with irrigated bipolar radiofrequency energy: a long-term animal study. The Journal of thoracic and cardiovascular surgery 2006;132:853-60.
255. Voeller RK, Zierer A, Lall SC, Sakamoto S, Schuessler RB, Damiano RJ, Jr. Efficacy of a novel bipolar radiofrequency ablation device on the beating heart for atrial fibrillation ablation: a long-term porcine study. The Journal of thoracic and cardiovascular surgery 2010;140:203-8.
256. Huffman MD, Karmali KN, Berendsen MA et al. Concomitant atrial fibrillation surgery for people undergoing cardiac surgery. Cochrane database of systematic reviews (Online) 2016:CD011814.
257. Je HG, Shuman DJ, Ad N. A systematic review of minimally invasive surgical treatment for atrial fibrillation: a comparison of the Cox-Maze procedure, beating-heart epicardial ablation, and the hybrid procedure on safety and efficacy. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2015;48:531-40; discussion 540-1.
258. Phan K, Xie A, La Meir M, Black D, Yan TD. Surgical ablation for treatment of atrial fibrillation in cardiac surgery: a cumulative meta-analysis of randomised controlled trials. Heart 2014;100:722-30.
259. Krul SP, Driessen AH, Zwinderman AH et al. Navigating the mini-maze: systematic review of the first results and progress of minimally-invasive surgery in the treatment of atrial fibrillation. Int J Cardiol 2013;166:132-40.
260. Cheng DC, Ad N, Martin J et al. Surgical ablation for atrial fibrillation in cardiac surgery: a meta-analysis and systematic review. Innovations (Philadelphia, Pa) 2010;5:84-96.
261. Kong MH, Lopes RD, Piccini JP, Hasselblad V, Bahnson TD, Al-Khatib SM. Surgical Maze procedure as a treatment for atrial fibrillation: a meta-analysis of randomized controlled trials. Cardiovascular therapeutics 2010;28:311-26.
262. Calkins H, Kuck KH, Cappato R et al. 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design: a report of the Heart Rhythm Society (HRS) Task Force on Catheter and Surgical Ablation of Atrial Fibrillation. Developed in partnership with the European Heart Rhythm Association (EHRA), a registered branch of the European Society of Cardiology (ESC) and the European Cardiac Arrhythmia Society (ECAS); and in collaboration with the American College of Cardiology (ACC), American Heart Association (AHA), the Asia Pacific Heart Rhythm Society (APHRS), and the Society of Thoracic Surgeons (STS). Endorsed by the governing bodies of the American College of Cardiology Foundation, the American Heart Association, the European Cardiac Arrhythmia Society, the European Heart Rhythm Association, the Society of Thoracic Surgeons, the Asia Pacific Heart Rhythm Society, and the Heart Rhythm Society. Heart Rhythm 2012;9:632-696 e21.
263. Baek MJ, Na CY, Oh SS et al. Surgical treatment of chronic atrial fibrillation combined with rheumatic mitral valve disease: Effects of the cryo-maze procedure and predictors for late recurrence. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2006;30:728-36.
264. Chen L, Xiao Y, Ma R et al. Bipolar radiofrequency ablation is useful for treating atrial fibrillation combined with heart valve diseases. BMC surgery 2014;14:32.
265. Dong L, Fu B, Teng X, Yuan HS, Zhao SL, Ren L. Clinical analysis of concomitant valve replacement and bipolar radiofrequency ablation in 191 patients. The Journal of thoracic and cardiovascular surgery 2013;145:1013-7.
266. Funatsu T, Kobayashi J, Nakajima H, Iba Y, Shimahara Y, Yagihara T. Long-term results and reliability of cryothermic ablation based maze procedure for atrial fibrillation concomitant with mitral valve surgery. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2009;36:267-71; discussion 271.
190
267. Geidel S, Lass M, Ostermeyer J. Seven-year experience with ablation of permanent atrial fibrillation concomitant to mitral valve surgery in 152 patients. The heart surgery forum 2008;11:E175-80.
268. Gillinov AM, Bakaeen F, McCarthy PM et al. Surgery for paroxysmal atrial fibrillation in the setting of mitral valve disease: a role for pulmonary vein isolation? The Annals of thoracic surgery 2006;81:19-26; discussion 27-8.
269. Jeanmart H, Casselman F, Beelen R et al. Modified maze during endoscopic mitral valve surgery: the OLV Clinic experience. The Annals of thoracic surgery 2006;82:1765-9.
270. Kasemsarn C, Lerdsomboon P, Sungkahaphong V, Chotivatanapong T. Left atrial reduction in modified maze procedure with concomitant mitral surgery. Asian cardiovascular & thoracic annals 2014;22:421-9.
271. Kim JB, Cho WC, Jung SH, Chung CH, Choo SJ, Lee JW. Alternative energy sources for surgical treatment of atrial fibrillation in patients undergoing mitral valve surgery: microwave ablation vs cryoablation. Journal of Korean medical science 2010;25:1467-72.
272. Maltais S, Forcillo J, Bouchard D et al. Long-term results following concomitant radiofrequency modified maze ablation for atrial fibrillation. J Card Surg 2010;25:608-13.
273. Nakajima H, Kobayashi J, Bando K et al. The effect of cryo-maze procedure on early and intermediate term outcome in mitral valve disease: case matched study. Circulation 2002;106:I46-I50.
274. Nardi P, Mve Mvondo C, Scafuri A et al. Left atrial radiofrequency ablation associated with valve surgery: midterm outcomes. The Thoracic and cardiovascular surgeon 2013;61:392-7.
275. Wang W, Buehler D, Martland AM, Feng XD, Wang YJ. Left atrial wall tension directly affects the restoration of sinus rhythm after Maze procedure. European Journal of Cardio-thoracic Surgery 2011;40:77-82.
276. Wang X, Wang X, Song Y, Hu S, Wang W. Efficiency of radiofrequency ablation for surgical treatment of chronic atrial fibrillation in rheumatic valvular disease. Int J Cardiol 2014;174:497-502.
277. Rahmanian PB, Filsoufi F, Salzberg S, Coppolino A, Castillo JG, Adams DH. Surgical treatment of atrial fibrillation using cryothermy in patients undergoing mitral valve surgery. Interact Cardiovasc Thorac Surg 2008;7:990-5.
278. John B, Stiles MK, Kuklik P et al. Electrical remodelling of the left and right atria due to rheumatic mitral stenosis. Eur Heart J 2008;29:2234-43.
279. Gammie JS, Sheng S, Griffith BP et al. Trends in mitral valve surgery in the United States: results from the Society of Thoracic Surgeons Adult Cardiac Surgery Database. The Annals of thoracic surgery 2009;87:1431-7; discussion 1437-9.
280. Gupta A, Perera T, Ganesan A et al. Complications of catheter ablation of atrial fibrillation: a systematic review. Circulation Arrhythmia and electrophysiology 2013;6:1082-8.
281. Albeyoglu SC, Filizcan U, Sargin M et al. Determinants of hospital mortality after repeat mitral valve surgery for rheumatic mitral valve disease. The Thoracic and cardiovascular surgeon 2006;54:244-9.
282. Kaartama T, Heikkinen L, Vento A. An evaluation of mitral valve procedures using the European system for cardiac operative risk evaluation. Scandinavian journal of surgery : SJS : official organ for the Finnish Surgical Society and the Scandinavian Surgical Society 2008;97:254-8.
283. Stahle E, Kvidal P, Nystrom SO, Bergstrom R. Long-term relative survival after primary heart valve replacement. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 1997;11:81-91.
284. Vassileva CM, Mishkel G, McNeely C et al. Long-term survival of patients undergoing mitral valve repair and replacement: a longitudinal analysis of Medicare fee-for-service beneficiaries. Circulation 2013;127:1870-6.
191
285. Anghel D, Anghel R, Corciova F, Enache M, Tinica G. Preoperative arrhythmias such as atrial fibrillation: cardiovascular surgery risk factor. BioMed research international 2014;2014:584918.
286. Ngaage DL, Schaff HV, Mullany CJ et al. Does preoperative atrial fibrillation influence early and late outcomes of coronary artery bypass grafting? The Journal of thoracic and cardiovascular surgery 2007;133:182-9.
287. Quader MA, McCarthy PM, Gillinov AM et al. Does preoperative atrial fibrillation reduce survival after coronary artery bypass grafting? The Annals of thoracic surgery 2004;77:1514-22; discussion 1522-4.
288. Rogers CA, Angelini GD, Culliford LA, Capoun R, Ascione R. Coronary surgery in patients with preexisting chronic atrial fibrillation: early and midterm clinical outcome. The Annals of thoracic surgery 2006;81:1676-82.
289. Russell EA, Walsh WF, Tran L et al. The burden and implications of preoperative atrial fibrillation in Australian heart valve surgery patients. Int J Cardiol 2017;227:100-105.
290. Jacobs JP, Shahian DM, Prager RL et al. The Society of Thoracic Surgeons National Database 2016 Annual Report. The Annals of thoracic surgery 2016;102:1790-1797.
291. Benussi S, Nascimbene S, Agricola E et al. Surgical ablation of atrial fibrillation using the epicardial radiofrequency approach: mid-term results and risk analysis. The Annals of thoracic surgery 2002;74:1050-6; discussion 1057.
292. Beukema WP, Sie HT, Misier AR, Delnoy PP, Wellens HJ, Elvan A. Intermediate to long-term results of radiofrequency modified Maze procedure as an adjunct to open-heart surgery. The Annals of thoracic surgery 2008;86:1409-14.
293. Budera P, Straka Z, Osmancik P et al. Comparison of cardiac surgery with left atrial surgical ablation vs. cardiac surgery without atrial ablation in patients with coronary and/or valvular heart disease plus atrial fibrillation: Final results of the PRAGUE-12 randomized multicentre study. European Heart Journal 2012;33:2644-2652.
294. Deneke T, Khargi K, Voss D et al. Long-term sinus rhythm stability after intraoperative ablation of permanent atrial fibrillation. Pacing and clinical electrophysiology : PACE 2009;32:653-9.
295. Doty JR, Doty DB, Jones KW et al. Comparison of standard Maze III and radiofrequency Maze operations for treatment of atrial fibrillation. The Journal of thoracic and cardiovascular surgery 2007;133:1037-44.
296. Gammie JS, Didolkar P, Krowsoski LS et al. Intermediate-term outcomes of surgical atrial fibrillation correction with the CryoMaze procedure. The Annals of thoracic surgery 2009;87:1452-8; discussion 1458-9.
297. Gaynor SL, Schuessler RB, Bailey MS et al. Surgical treatment of atrial fibrillation: predictors of late recurrence. The Journal of thoracic and cardiovascular surgery 2005;129:104-11.
298. Geidel S, Krause K, Boczor S et al. Ablation surgery in patients with persistent atrial fibrillation: an 8-year clinical experience. The Journal of thoracic and cardiovascular surgery 2011;141:377-82.
299. Gillinov AM, McCarthy PM, Blackstone EH et al. Surgical ablation of atrial fibrillation with bipolar radiofrequency as the primary modality. The Journal of thoracic and cardiovascular surgery 2005;129:1322-9.
300. Grubitzsch H, Dushe S, Beholz S, Dohmen PM, Konertz W. Surgical Ablation of Atrial Fibrillation in Patients With Congestive Heart Failure. Journal of Cardiac Failure 2007;13:509-516.
301. Guden M, Akpinar B, Caynak B et al. Left versus bi-atrial intraoperative saline-irrigated radiofrequency modified maze procedure for atrial fibrillation. Cardiac electrophysiology review 2003;7:252-8.
302. Mantovan R, Raviele A, Buja G et al. Left atrial radiofrequency ablation during cardiac surgery in patients with atrial fibrillation. J Cardiovasc Electrophysiol 2003;14:1289-95.
192
303. Martinez-Comendador J, Castano M, Mosquera I et al. Cryoablation of atrial fibrillation in cardiac surgery: Outcomes and myocardial injury biomarkers. Journal of Cardiothoracic and Vascular Anesthesia 2011;25:1030-1035.
304. Melby SJ, Kaiser SP, Bailey MS et al. Surgical treatment of atrial fibrillation with bipolar radiofrequency ablation: Mid-term results in one hundred consecutive patients. Journal of Cardiovascular Surgery 2006;47:705-710.
305. Mohr FW, Fabricius AM, Falk V et al. Curative treatment of atrial fibrillation with intraoperative radiofrequency ablation: short-term and midterm results. The Journal of thoracic and cardiovascular surgery 2002;123:919-27.
306. Nakamura T, Izutani H, Sawa Y. Mid-term outcomes of the modified Cox-maze procedure for elderly patients: a risk analysis for failure. Interact Cardiovasc Thorac Surg 2011;12:924-8.
307. Sakamoto Y, Takakura H, Onoguchi K et al. Cryosurgical left-sided maze procedure in patients with valvular heart disease: medium-term results. Annals of thoracic and cardiovascular surgery : official journal of the Association of Thoracic and Cardiovascular Surgeons of Asia 2011;17:148-52.
308. Sie HT, Beukema WP, Elvan A, Ramdat Misier AR. Long-term results of irrigated radiofrequency modified maze procedure in 200 patients with concomitant cardiac surgery: six years experience. The Annals of thoracic surgery 2004;77:512-6; discussion 516-7.
309. Srivastava V, Kumar S, Javali S et al. Efficacy of three different ablative procedures to treat atrial fibrillation in patients with valvular heart disease: a randomised trial. Heart, lung & circulation 2008;17:232-40.
310. Starck C, Botha CA, Roser D, Paula J, Rein JG, Hemmer W. Results of a modified left atrial maze procedure as a combined procedure. The Thoracic and cardiovascular surgeon 2003;51:147-53.
311. Sternik L, Luria D, Glikson M, Malachy A, First M, Raanani E. Efficacy of surgical ablation of atrial fibrillation in patients with rheumatic heart disease. The Annals of thoracic surgery 2010;89:1437-42.
312. Stulak JM, Sundt TM, 3rd, Dearani JA, Daly RC, Orsulak TA, Schaff HV. Ten-year experience with the Cox-maze procedure for atrial fibrillation: how do we define success? The Annals of thoracic surgery 2007;83:1319-24.
313. Veasey RA, Segal OR, Large JK et al. The efficacy of intraoperative atrial radiofrequency ablation for atrial fibrillation during concomitant cardiac surgery-the Surgical Atrial Fibrillation Suppression (SAFS) Study. Journal of interventional cardiac electrophysiology : an international journal of arrhythmias and pacing 2011;32:29-35.
314. Voeller RK, Bailey MS, Zierer A et al. Isolating the entire posterior left atrium improves surgical outcomes after the Cox maze procedure. The Journal of thoracic and cardiovascular surgery 2008;135:870-7.
315. Wang J, Meng X, Li H, Cui Y, Han J, Xu C. Prospective randomized comparison of left atrial and biatrial radiofrequency ablation in the treatment of atrial fibrillation. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2009;35:116-22.
316. Watkins AC, Young CA, Ghoreishi M et al. Prospective assessment of the CryoMaze procedure with continuous outpatient telemetry in 136 patients. The Annals of thoracic surgery 2014;97:1191-8; discussion 1198.
317. Yanagawa B, Holmes SD, Henry L, Hunt S, Ad N. Outcome of concomitant cox-maze III procedure using an argon-based cryosurgical system: A single-center experience with 250 patients. Annals of Thoracic Surgery 2013;95:1633-1639.
318. Beyer E, Lee R, Lam BK. Point: Minimally invasive bipolar radiofrequency ablation of lone atrial fibrillation: early multicenter results. The Journal of thoracic and cardiovascular surgery 2009;137:521-6.
193
319. Edgerton JR, McClelland JH, Duke D et al. Minimally invasive surgical ablation of atrial fibrillation: Six-month results. Journal of Thoracic and Cardiovascular Surgery 2009;138:109-114.
320. Kasirajan V, Spradlin EA, Mormando TE et al. Minimally invasive surgery using bipolar radiofrequency energy is effective treatment for refractory atrial fibrillation. The Annals of thoracic surgery 2012;93:1456-61.
321. Nasso G, Bonifazi R, Del Prete A et al. Long-term results of ablation for isolated atrial fibrillation through a right minithoracotomy: toward a rational revision of treatment protocols. The Journal of thoracic and cardiovascular surgery 2011;142:e41-6.
322. Saint LL, Bailey MS, Prasad S et al. Cox-Maze IV results for patients with lone atrial fibrillation versus concomitant mitral disease. The Annals of thoracic surgery 2012;93:789-94; discussion 794-5.
323. Verma A, Jiang CY, Betts TR et al. Approaches to catheter ablation for persistent atrial fibrillation. The New England journal of medicine 2015;372:1812-22.
324. Verma A, Mantovan R, Macle L et al. Substrate and Trigger Ablation for Reduction of Atrial Fibrillation (STAR AF): a randomized, multicentre, international trial. Eur Heart J 2010;31:1344-56.
325. Verma A, Sanders P, Champagne J et al. Selective complex fractionated atrial electrograms targeting for atrial fibrillation study (SELECT AF): a multicenter, randomized trial. Circulation Arrhythmia and electrophysiology 2014;7:55-62.
326. Boersma LV, Castella M, van Boven W et al. Atrial fibrillation catheter ablation versus surgical ablation treatment (FAST): a 2-center randomized clinical trial. Circulation 2012;125:23-30.
327. Aryana A, Singh SK, Singh SM et al. Association between incomplete surgical ligation of left atrial appendage and stroke and systemic embolization. Heart Rhythm 2015;12:1431-7.
328. Bisleri G, Rosati F, Bontempi L, Curnis A, Muneretto C. Hybrid approach for the treatment of long-standing persistent atrial fibrillation: electrophysiological findings and clinical results. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2013.
329. Bulava A, Mokracek A, Hanis J, Kurfirst V, Eisenberger M, Pesl L. Sequential hybrid procedure for persistent atrial fibrillation. Journal of the American Heart Association 2015;4:e001754.
330. de Asmundis C, Chierchia GB, Mugnai G et al. Midterm clinical outcomes of concomitant thoracoscopic epicardial and transcatheter endocardial ablation for persistent and long-standing persistent atrial fibrillation: a single-centre experience. Europace 2017;19:58-65.
331. Edgerton Z, Perini AP, Horton R et al. Hybrid Procedure (Endo/Epicardial) versus Standard Manual Ablation in Patients Undergoing Ablation of Longstanding Persistent Atrial Fibrillation: Results from a Single Center. J Cardiovasc Electrophysiol 2016;27:524-30.
332. Eisenberger M, Bulava A, Mokracek A, Hanis J, Kurfirst V, Dusek L. Sequential Hybrid Surgical CryoMaze and Transvenous Catheter Ablation of Atrial Fibrillation. Pacing and clinical electrophysiology : PACE 2015.
333. Gaita F, Ebrille E, Scaglione M et al. Very long-term results of surgical and transcatheter ablation of long-standing persistent atrial fibrillation. The Annals of thoracic surgery 2013;96:1273-8.
334. Gehi AK, Mounsey JP, Pursell I et al. Hybrid epicardial-endocardial ablation using a pericardioscopic technique for the treatment of atrial fibrillation. Heart Rhythm 2013;10:22-8.
335. Gersak B, Pernat A, Robic B, Sinkovec M. Low rate of atrial fibrillation recurrence verified by implantable loop recorder monitoring following a convergent epicardial and endocardial ablation of atrial fibrillation. Journal of Cardiovascular Electrophysiology 2012;23:1059-1066.
336. Kiser AC, Landers M, Horton R, Hume A, Natale A, Gersak B. The convergent procedure: a multidisciplinary atrial fibrillation treatment. The heart surgery forum 2010;13:E317-21.
194
337. Krul SP, Driessen AH, van Boven WJ et al. Thoracoscopic video-assisted pulmonary vein antrum isolation, ganglionated plexus ablation, and periprocedural confirmation of ablation lesions: first results of a hybrid surgical-electrophysiological approach for atrial fibrillation. Circulation Arrhythmia and electrophysiology 2011;4:262-70.
338. Kumar N, Pison L, Lozekoot P et al. The symbiosis of contact force catheter use for hybrid ablation for atrial fibrillation. Netherlands heart journal : monthly journal of the Netherlands Society of Cardiology and the Netherlands Heart Foundation 2015;23:438-46.
339. Kurfirst V, Mokracek A, Bulava A, Canadyova J, Hanis J, Pesl L. Two-staged hybrid treatment of persistent atrial fibrillation: short-term single-centre results. Interact Cardiovasc Thorac Surg 2014;18:451-6.
340. La Meir M, Gelsomino S, Lorusso R et al. The hybrid approach for the surgical treatment of lone atrial fibrillation: one-year results employing a monopolar radiofrequency source. Journal of cardiothoracic surgery 2012;7:71.
341. La Meir M, Gelsomino S, Luca F et al. Minimally invasive surgical treatment of lone atrial fibrillation: early results of hybrid versus standard minimally invasive approach employing radiofrequency sources. Int J Cardiol 2013;167:1469-75.
342. Lee R, McCarthy PM, Passman RS et al. Surgical treatment for isolated atrial fibrillation: minimally invasive vs. classic cut and sew maze. Innovations (Philadelphia, Pa) 2011;6:373-7.
343. Mahapatra S, LaPar DJ, Kamath S et al. Initial experience of sequential surgical epicardial-catheter endocardial ablation for persistent and long-standing persistent atrial fibrillation with long-term follow-up. The Annals of thoracic surgery 2011;91:1890-8.
344. Osmancik P, Budera P, Zdarska J, Herman D, Petr R, Straka Z. Electrophysiological findings after surgical thoracoscopic atrial fibrillation ablation. Heart Rhythm 2016;13:1246-52.
345. Richardson TD, Shoemaker MB, Whalen SP, Hoff SJ, Ellis CR. Staged versus Simultaneous Thoracoscopic Hybrid Ablation for Persistent Atrial Fibrillation Does Not Affect Time to Recurrence of Atrial Arrhythmia. J Cardiovasc Electrophysiol 2016;27:428-34.
346. Toplisek J, Pernat A, Ruzic N et al. Improvement of Atrial and Ventricular Remodeling with Low Atrial Fibrillation Burden after Hybrid Ablation of Persistent Atrial Fibrillation. Pacing and clinical electrophysiology : PACE 2016;39:216-24.
347. Zembala M, Filipiak K, Kowalski O et al. Staged hybrid ablation for persistent and longstanding persistent atrial fibrillation effectively restores sinus rhythm in long-term observation. Archives of medical science : AMS 2017;13:109-117.
348. Zembala M, Filipiak K, Kowalski O et al. Minimally invasive hybrid ablation procedure for the treatment of persistent atrial fibrillation: one year results. Kardiol Pol 2012;70:819-28.
349. Mansour M, Karst E, Heist EK et al. The Impact of First Procedure Success Rate on the Economics of Atrial Fibrillation Ablation. JACC: Clinical Electrophysiology 2017;3:129-138.
350. Cappato R, Calkins H, Chen SA et al. Up-dated Worldwide Survey on the Methods, Efficacy and Safety of Catheter Ablation for Human Atrial Fibrillation. Circ Arrhythm Electrophysiol 2009.
351. Cappato R, Calkins H, Chen SA et al. Worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circulation 2005;111:1100-5.
352. Piccini JP, Sinner MF, Greiner MA et al. Outcomes of Medicare beneficiaries undergoing catheter ablation for atrial fibrillation. Circulation 2012;126:2200-7.
353. Ouyang F, Antz M, Ernst S et al. Recovered pulmonary vein conduction as a dominant factor for recurrent atrial tachyarrhythmias after complete circular isolation of the pulmonary veins: lessons from double Lasso technique. Circulation 2005;111:127-135.
354. Lemola K, Hall B, Cheung P et al. Mechanisms of recurrent atrial fibrillation after pulmonary vein isolation by segmental ostial ablation. Heart Rhythm 2004;1:197-202.
355. Brooks AG, Stiles MK, Laborderie J et al. Outcomes of long-standing persistent atrial fibrillation ablation: a systematic review. Heart Rhythm 2010;7:835-46.
195
356. Wong CX, Abed HS, Molaee P et al. Pericardial fat is associated with atrial fibrillation severity and ablation outcome. Journal of the American College of Cardiology 2011;57:1745-51.
357. Pathak RK, Middeldorp ME, Lau DH et al. Aggressive Risk factor REduction STudy for Atrial Fibrillation (ARREST-AF Cohort Study): implications for the outcome of ablation. J Am Coll Cardiol 2014;In Press.
358. Stulak JM, Dearani JA, Sundt TM, 3rd et al. Superiority of cut-and-sew technique for the Cox maze procedure: comparison with radiofrequency ablation. The Journal of thoracic and cardiovascular surgery 2007;133:1022-7.
359. Shah DC, Lambert H, Nakagawa H, Langenkamp A, Aeby N, Leo G. Area under the real-time contact force curve (force-time integral) predicts radiofrequency lesion size in an in vitro contractile model. J Cardiovasc Electrophysiol 2010;21:1038-43.
360. Thiagalingam A, D'Avila A, Foley L et al. Importance of catheter contact force during irrigated radiofrequency ablation: evaluation in a porcine ex vivo model using a force-sensing catheter. J Cardiovasc Electrophysiol 2010;21:806-11.
361. Yokoyama K, Nakagawa H, Shah DC et al. Novel contact force sensor incorporated in irrigated radiofrequency ablation catheter predicts lesion size and incidence of steam pop and thrombus. Circulation Arrhythmia and electrophysiology 2008;1:354-62.
362. Reddy VY, Shah D, Kautzner J et al. The relationship between contact force and clinical outcome during radiofrequency catheter ablation of atrial fibrillation in the TOCCATA study. Heart Rhythm 2012;9:1789-95.
363. Kautzner J, Neuzil P, Lambert H et al. EFFICAS II: optimization of catheter contact force improves outcome of pulmonary vein isolation for paroxysmal atrial fibrillation. Europace 2015;17:1229-35.
364. Kasirajan V, Sayeed S, Filler E, Knarik A, Koneru JN, Ellenbogen KA. Histopathology of Bipolar Radiofrequency Ablation in the Human Atrium. The Annals of thoracic surgery 2016;101:638-43.
365. Takahashi K, Miyauchi Y, Hayashi M et al. Mechanisms of postoperative atrial tachycardia following biatrial surgical ablation of atrial fibrillation in relation to the surgical lesion sets. Heart Rhythm 2016;13:1059-65.
366. Ho SY, Cabrera JA, Sanchez-Quintana D. Left atrial anatomy revisited. Circulation Arrhythmia and electrophysiology 2012;5:220-8.