Europace 2003 5(s1):S20-S29; doi:10.1016/j.eupc.2004.07.004
© 2003 by European Society of Cardiology
Surgical treatment of atrial fibrillation: a review
James L. Cox*
Division of Cardiothoracic Surgery, Washington University School of Medicine St. Louis, MO, USA
Manuscript submitted 25 May 2004. Accepted after revision 24 June 2004.
*13523 Rosewood Lane, Naples, FL, USA. Tel.: +1 239 598 1622; fax: +1 239 598 4090. E-mail address: jamescoxmd{at}aol.com
 |
Abstract
|
|---|
The optimal treatment of atrial fibrillation (AF) depends upon
a proper understanding of the electrophysiological basis of
its clinical manifestations. Whether AF is
continuous (persistent)
or
intermittent (paroxysmal) depends on variable underlying
electrophysiology and determines the choice of interventional
treatment. The initiation of
intermittent AF requires a "trigger",
often, though not always, located near the orifices of pulmonary
veins. In contrast,
continuous AF does not need a "trigger"
to be re-induced repetitively and does not depend on the pulmonary
veins or other abnormal automatic foci for its induction or
perpetuation. Simple pulmonary vein encircling confines the
trigger to the pulmonary veins and, if expertly performed, will
cure the majority of patients with
intermittent AF. On the other
hand,
continuous AF requires a Maze procedure or variant thereof
to eliminate atrial macro-reentry while allowing sinus rhythm
to activate the entire atrial myocardium and preserve atrial
transport function. This article reviews the development of
the surgical Maze procedure and its implications for the treatment
of AF by percutaneous intracardiac or epicardial, minimally
invasive techniques. High-intensity focussed ultrasound, a new
energy source generating frictional heat, appears promising
in the creation of focussed transmural lesions, while preserving
the integrity of coronary arterial walls.
Key Words: atrial fibrillation, antiarrhythmic surgery, maze operation, high intensity focussed ultrasound, surgical treatment of atrial fibrillation
|
Introduction
|
|---|
The optimal treatment of atrial fibrillation (AF) depends upon
a proper understanding of the electrophysiological basis of
each of its clinical manifestations. It is fortuitous that in
the case of AF, both its clinical manifestation and its proper
interventional treatment are dictated by its underlying electrophysiology.
AF presents clinically in one of two ways. The patient either
has AF
all of the time or not. If AF is present all of the time,
it is classified as
continuous (persistent). If it is not present
all of the time, it is classified as
intermittent (paroxysmal).
This simple classification correctly implies that there is a fundamental pathophysiological difference in the AF of a patient who presents with intermittent episodes versus one who presents with a continuous arrhythmia, and that all other differences are irrelevant from a perspective of interventional management. Indeed, the clinical presentations are different specifically because the underlying electrophysiology is different. This correlation explains why their interventional treatment must also be different.
|
Intermittent atrial fibrillation
|
|---|
It is intuitive that
intermittent AF is initiated repetitively,
and thus requires a "trigger" to induce each individual episode.
In 1998, Haïssaguerre and colleagues suggested that this trigger
is located in the orifices of the pulmonary veins in approximately
90% of patients [1]
, although more recent studies suggest a
considerably lower percentage [2]
. Regardless of the true prevalence
of pulmonary vein triggers, the premature event originating
within the pulmonary vein propagates into the left atrium and
induces the formation of multiple
macro-reentrant circuits involving
both atria (
Fig. 1). The multiple circuits cause the atria to
quiver, a state known as AF (
Fig. 2) [3]
.
View larger version (26K):
[in this window]
[in a new window]
|
Figure 1 Induction of AF by a premature atrial complex (PAC) originating in the orifice of a pulmonary vein. These triggers may be caused by micro-reentrant circuits that occur in the cellular transitional zone between pulmonary vein endothelium and left atrial endocardium. However, these triggers are most likely automatic rather than reentrant, since recurrent atrial tachycardias are known to arise from automatic foci in other parts of the atria [11]. Reproduced with permission.
|
|
View larger version (75K):
[in this window]
[in a new window]
|
Figure 2 Once induced, AF is characterized by the presence of multiple macro-reentrant circuits in the atria (posterior view). This electrophysiological basis of AF is unrelated to (1) how, or from where, it was induced, (2) whether it is intermittent or continuous, and (3) its duration. Reproduced with permission.
|
|
Once induced, AF persists until the macro-reentrant circuits
terminate spontaneously or are terminated by drugs or other
interventions. If AF terminates spontaneously, normal sinus
rhythm naturally returns (
Fig. 3). Each subsequent episode of
AF then requires another premature event to initiate the episode,
the trigger again being the pulmonary veins in the majority
of cases.
View larger version (49K):
[in this window]
[in a new window]
|
Figure 3 Intermittent AF. The patient is in normal sinus rhythm (NSR) some of the time. A premature atrial complex (PAC) induces AF until atrial macro-reentry can no longer sustain itself and ends spontaneously. Each episode of AF requires a PAC to trigger its induction. Reproduced with permission.
|
|
At least 10% of patients with intermittent AF have a triggering
mechanism that does
not involve the pulmonary veins (
Fig. 4)
[1,
2]
. AF can be induced episodically by the Wolff–Parkinson–White
syndrome, atrioventricular (AV) node reentry, or the deterioration
of atrial flutter [3,
4]
, although, in most cases, non-pulmonary
vein triggers, like their counterparts in the pulmonary veins,
are automatic foci located elsewhere in the atria.
View larger version (52K):
[in this window]
[in a new window]
|
Figure 4 The trigger of intermittent AF was located in the pulmonary veins in 90% of patients and outside the pulmonary vein area in 10% of patients with normal or small-sized atria [1]. Reproduced with permission.
|
|
|
Continuous atrial fibrillation
|
|---|
If normal sinus rhythm is not re-established either spontaneously
or by drugs, AF is no longer episodic and, once induced, the
atria will continue indefinitely to fibrillate. Once AF is
continuous,
it no longer needs a "trigger" to be re-induced repetitively
since the atria are fibrillating all of the time. Thus, continuous
AF does not depend on the pulmonary veins for its induction
or perpetuation.
AF becomes continuous because the macro-reentrant circuits in the atrial myocardium are capable of sustaining themselves. Allessie has attributed this to "atrial remodelling" [5]. This electrical remodelling may or may not be associated with anatomic remodelling, i.e., enlargement, hypertrophy or stretching of the atria. The essence of this concept, as Allessie points out, is that "atrial fibrillation begets atrial fibrillation". In other words, the more a patient develops atrial fibrillation, the more that patient is likely to re-develop it.
|
Interventional therapy
|
|---|
Simple pulmonary vein encircling confines the trigger to the
pulmonary veins and, if expertly performed, will cure the majority
of patients with intermittent AF (
Fig. 5). Unfortunately, it
is unsuccessful as an isolation procedure in a significant number
of patients with intermittent AF because the trigger mechanism
is not in the pulmonary veins. In addition, since continuous
AF is not dependent upon the pulmonary veins for its maintenance,
simple pulmonary vein encircling is not a rational treatment
goal in patients with continuous AF (
Fig. 6).
View larger version (39K):
[in this window]
[in a new window]
|
Figure 5 Pulmonary vein isolation for intermittent AF. Since the putative trigger is located within one or more pulmonary veins in 90% of patients with intermittent AF, simple pulmonary vein encirclement, if performed correctly, should be curative in 90% of cases. In 10% of patients with intermittent AF, simple pulmonary vein isolation will not be successful since the trigger is away from the pulmonary veins. Reproduced with permission.
|
|
View larger version (50K):
[in this window]
[in a new window]
|
Figure 6 Failure of pulmonary vein isolation for continuous AF. Since the maintenance of continuous AF is not dependent upon triggers located in the pulmonary veins, simple pulmonary vein isolation is not a legitimate treatment. However, if a large amount of left atrium is ablated during prolonged catheter attempts to achieve "pulmonary vein isolation", it may eliminate continuous AF. This procedure is not a "pulmonary vein isolation" but an "ablation of the left atrium". Reproduced with permission.
|
|
The only way to ablate atrial macro-reentry, and at the same
time, leave the resultant sinus rhythm capable of activating
the entire atrial myocardium and assure atrial transport function
postoperatively is to perform the Maze procedure or a variant
thereof. The Maze procedure is a classic ablation procedure
which eliminates the macro-reentry responsible for AF. Unfortunately,
since it is a major open-heart operation, it is too complex
to be applied to all patients with AF and therefore, vigorous
efforts are being made to develop a simplified version of the
Maze procedure that can be applied more widely while retaining
its efficacy.
|
Atrial lesions necessary to cure atrial fibrillation
|
|---|
All surgical interventions for cardiac arrhythmias can be classified
as isolation or ablation procedures. Isolation procedures do
not terminate arrhythmias but confine them and/or their trigger
mechanisms to a desired region of the heart to minimise their
adverse effects. Ablation procedures eliminate arrhythmias by
destroying their trigger mechanism, or by altering or removing
the substrate allowing their induction or sustenance.
We have learned empirically that several lesions created in the original surgical Maze-III procedure may not be essential to ablate AF. The Maze procedure was designed to preclude the development of macro-reentry anywhere and everywhere in the atria. There is growing evidence, however, that, although continuous AF is sustained by multiple macro-reentrant circuits, only a limited number of left atrial sites are capable of sustaining such circuits [6]. This is the theory behind the so-called "focal" AF that seems to respond to lesions critically created away from the pulmonary veins. It also supports the empirical observation of the elimination of continuous AF despite forgoing several lesions originally included in the Maze procedure (Fig. 7).
View larger version (57K):
[in this window]
[in a new window]
|
Figure 7 The five left atrial lesions of the standard Maze-III surgical procedure for AF. Reproduced with permission.
|
|
I have never been convinced that the septal lesion was essential
to the efficacy of the Maze procedure and performed it primarily
to gain better exposure of the inside of the left atrium. Reports
of recent interventional and surgical therapy for AF suggest
that the septal lesion is not necessary. In addition, the lesion
extending out into the left atrial appendage was included in
the Maze procedure in order to prevent the theoretically possible
reentry around the base of the appendage. That reentrant circuit
has never been documented and may not be a clinical entity.
Therefore, that lesion may also be unnecessary.
If neither the septal nor the left atrial appendage lesions are critical to ablate AF, the essential left atrial lesions would be limited to the incision encircling the pulmonary veins and the lesion across the "isthmus", between the inferior pulmonary veins and the mitral annulus. Experience with the Maze and other surgical antiarrhythmic procedures has amply confirmed that the left atrial isthmus lesion is extremely important in eliminating reentry responsible for AF. Indeed, this lesion, along with its companion cryolesion in the coronary sinus, is the "Achilles Heel" of the Maze procedure, in that every failure in our own series was associated with persistent conduction across this isthmus (Fig. 8) [7].
View larger version (49K):
[in this window]
[in a new window]
|
Figure 8 Left panel: The lesion across the left atrial isthmus is essential to prevent the postoperative development of a macro-reentrant tachyarrhythmia ("slow left atrial flutter") around the pulmonary vein lesions. However, this atriotomy alone does not preclude the development of this circuit and must be accompanied by the creation of conduction block in the coronary sinus. Right panel: The simplest technique for creating conduction block through the coronary sinus wall is by cryoablation. The cryolesion must be placed in the precise plane of the isthmus atriotomy to be effective. Reproduced with permission.
|
|
Besides the simplification of left atrial lesions, experimental
observations by Allessie regarding differences in left versus
right atrial refractory periods suggest that the right atrial
lesions of the Maze procedure may be replaced by a simpler approach
[8]
. The duration of the local refractory period determines
the
minimum size of an atrial macro-reentrant circuit [9]
. Since
refractory periods are shorter in the left atrium, it can sustain
the small macro-reentrant circuits that are characteristic of
AF, whereas, in the right atrium, the longer refractory periods
are probably not capable of sustaining AF, unless the atrium
is pathologically enlarged. Thus, the treatment of AF can be
focussed on the left side, knowing that AF will not recur if
macro-reentry can be prevented by lesions critically placed
in the left atrium. In the right atrium, capable of sustaining
only atrial flutter, a lesion in the isthmus between coronary
sinus and tricuspid valve will disable a critical limb of the
macro-reentrant circuit sustaining the vast majority of atrial
flutters (
Fig. 9) [10]
. In summary, most patients with AF of
either type could probably be cured by a "Mini-Maze Procedure"
which includes (1) an incision encircling the pulmonary veins,
(2) a left atrial isthmus and companion coronary sinus lesions,
and (3) a right atrial isthmus lesion (
Fig. 10).
View larger version (46K):
[in this window]
[in a new window]
|
Figure 9 Left panel: In the absence of right atrial enlargement, the right atrium will only sustain macro-reentrant circuits substantially longer than in the left atrium since the right atrial refractory period is longer. The lesions of the standard Maze procedure interrupt this reentrant circuit at two sites. Right panel: The usual macro-reentrant circuit in the right atrium passes through the "right atrial isthmus", between the os of the coronary sinus and the tricuspid valve annulus. The isthmus may also include tissue between the coronary sinus os and the inferior vena cava orifice. A lesion placed across this right atrial isthmus interrupts this common macro-reentrant circuit responsible for most cases of typical atrial flutter. Reproduced with permission.
|
|
|
Characteristics of the ideal operation for atrial fibrillation
|
|---|
Besides establishing this lesion pattern as an intelligent simplification
of the original Maze procedure, several years of clinical experience
and vigorous industrial support have also defined the characteristics
of the ideal technique by which this pattern should be applied.
First, the procedure should be epicardial because of the greater
safety of delivering the energy from epicardium to endocardium,
instead of from the endocardium, where over-penetration may
cause catastrophic complications (
Fig. 11). Second, the energy
source should be capable of penetrating fat as well as atrial
muscle to eliminate the need to dissect away the epicardial
fat prior to its application. Third, the procedure should be
capable of ablating intermittent AF, continuous AF and atrial
flutter. Fourth, cardiopulmonary bypass must be avoided. Fifth,
the procedure should be endoscopic or minimally invasive, and
be completed within <60 min. Sixth, in absence of complications,
patients should be able to leave the hospital no later than
on the first postoperative day. In summary, the ideal AF procedure
would be performed via a minimally invasive incision, endoscopically
or robotically, off cardiopulmonary bypass, in <1 h, with
hospital discharge planned for the next morning.
View larger version (11K):
[in this window]
[in a new window]
|
Figure 11 Left panel: Energy delivered from the endocardium of the left atrium is associated with several uncertainties and risks. Energy sources applied in this manner do not allow monitoring of lesion depth. Excessive penetration of the left atrial wall can cause injury to surrounding structures, such as the oesophagus (esophagus). It is also impossible to determine when an endocardial lesion created by energy of any type is not transmural, therefore when it will fail to cause conduction block. Right panel: Transmural lesions created by energy sources directed from the epicardium are only in contact with the atrial blood pool. Therefore, epicardial ablation is safer than endocardial ablation. Unfortunately, transmural lesions are difficult or impossible to create from the epicardial surface with certain energy sources because of the competing effects of the atrial blood pool. For example, the blood pool serves as a "heat-sink" with cryothermia, and as a "cooling-sink" when radiofrequency and microwave sources are applied to the epicardium. Reproduced with permission.
|
|
|
High-intensity focussed ultrasound (HIFU) – the ideal energy source?
|
|---|
This ideal AF procedure demands the development of new energy
sources capable of ablating tissue in critical areas, such as
the left atrial isthmus, more rapidly, uniformly and safely
than any energy source currently available. No energy source,
including cryothermy, uni- or bipolar radiofrequency, irrigated
radiofrequency, microwave, or laser energy, is currently capable
of creating the left atrial isthmus lesion from the epicardial
surface because of the unavoidable penetration through the circumflex
coronary artery to reach the left atrial wall near the posterior
mitral annulus (
Fig. 12). Therefore, the Mini-Maze procedure
cannot be performed by an epicardial approach with any presently
available energy source.
View larger version (18K):
[in this window]
[in a new window]
|
Figure 12 Cross-section of the posterior left heart showing the left posterior AV groove and its most important contents, the coronary sinus and the circumflex coronary artery. No currently available energy source is capable of penetrating through the AV groove safely to create the left atrial isthmus lesion from the epicardial surface. PV = pulmonary vein; LA = left atrial; LV = left ventricular; AV = atrioventricular; RF = radiofrequency; CS = cryosurgery; MW = microwave. Reproduced with permission.
|
|
However, high-intensity focussed ultrasound (HIFU), a new energy
source currently in clinical trials in Europe, seems to overcome
most, if not all the limitations identified thus far (
Fig. 13).
HIFU is capable of ablating efficiently atrial myocardium as
well as epicardial fat. It creates tissue injury by causing
cells to oscillate or "rub against each other", generating frictional
heat that destroys the cells. The absence of injury to the coronary
arteries, perhaps the most important and unique characteristic
of HIFU, is explained by the distance between circulating red
blood cells, which prevents them from generating the heat of
friction. Since the blood remains at body temperature, it protects
the integrity of the arterial wall. This allows the ablation
of myocardium and fat on both sides of a coronary artery without
causing any injury to the vessel itself. Thus, HIFU can be used
to create lesions across the left atrial isthmus from the epicardium
without injury to the circumflex coronary artery, which no other
energy source can accomplish (
Fig. 14). HIFU also very rapidly
creates uniform, contiguous transmural lines of tissue ablation,
and is amenable to minimally invasive surgery without cardiopulmonary
bypass.
View larger version (51K):
[in this window]
[in a new window]
|
Figure 13 High-intensity focussed ultrasound is delivered via a series of 1-cm concave transducers focusing the ultrasound beam at a depth of 1 cm. Reproduced with permission.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Figure 14 Cross-section of the posterior left heart showing the left posterior AV groove and its most important contents, the coronary sinus and the circumflex coronary artery. High-intensity focussed ultrasound is the only known energy source capable of penetrating through the AV groove safely to create the left atrial isthmus lesion from the epicardial surface. PV = pulmonary vein; LA = left atrial; LV = left ventricular; AV = atrioventricular; HIFU = high-intensity focussed ultrasound. Reproduced with permission.
|
|
|
Closing comments
|
|---|
Despite the promise of developing a simpler operation for AF
that would be applicable in virtually all patients, it is noteworthy
that many of the more recent concepts, particularly those relating
to the
induction of AF, are based on limited scientific data,
on very few clinical cases, or on observations made in highly
selected patients. The electrophysiology of AF is unforgiving
and, despite our fondest wishes, the truth regarding AF and
its treatment cannot be altered to fit preconceived notions.
It is predictable that a significant number of patients with
AF will continue to require the more extensive and complex surgical
procedures. Nevertheless, these newer simplified surgical approaches
offer the possibility of curing patients who are currently doomed
to live with their disease.
In summary, up to 90% of intermittent ("paroxysmal" or "persistent") atrial fibrillation can be cured by a properly performed isolation of the pulmonary veins. In order to attain a similar cure rate in patients with continuous ("permanent") atrial fibrillation, however, it is essential to add additional lesions in the left atrium, specifically the left atriotomy from the pulmonary vein encircling incision down to the mitral valve annulus posteriorly and ablation of coronary sinus conduction at the same point. Since this additional lesion is now becoming easier to perform, there is little reason not to add it in patients with intermittent atrial fibrillation as well.
|
References
|
|---|
[1] Haïssaguerre M, Jaïs P, Shah DC, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998; 339: 659–666.
[2] Schmitt C, Ndrepepa G, Weber S, et al. Biatrial multisite mapping of atrial premature complexes triggering onset of atrial fibrillation. Am J Cardiol 2002; 89: 1381–1387.
[3] 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. J Thorac Cardiovasc Surg 1991; 101: 406–426.
[4] Canavan TE, Schuessler RB, Boineau JP, Corr PB, Cain ME, Cox JL. Computerized global electrophysiological mapping of the atrium in patients with Wolff–Parkinson–White syndrome. Ann Thorac Surg 1988; 46: 223–231.
[5] Allessie MA. Atrial fibrillation-induced electrical remodeling in humans: what is the next step? Cardiovasc Res 1999; 44: 10–12.
[6] Harada A, Konishi T, Fukata M, Higuchi K, Sugimoto T, Sasaki K. Intraoperative map guided operation for atrial fibrillation due to mitral valve disease. Ann Thorac Surg 2000; 69: 446–450.
[7] Cox JL and Ad N. The importance of cryoablation of the coronary sinus during the Maze procedure. Semin Thorac Cardiovasc Surg 2000; 12: 20–24.
[8] Lammers WJ, Schalij MJ, Kirchhof CJ, Allessie MA. Quantification of spatial inhomogeneity in conduction and initiation of reentrant atrial arrhythmias. Am J Physiol 1990; 259: H1254–H1263.
[9] Cox JL, Boineau JP, Schuessler RB, Kater KM, Lappas DG. Five-year experience with the maze procedure for atrial fibrillation. Ann Thorac Surg 1994; 56: 814–824.
[10] Nakagawa H, Lazzara R, Khastgir T, et al. Role of the tricuspid annulus and the eustachian valve/ridge in atrial flutter. Relevance to catheter ablation of the septal isthmus and a new technique for rapid identification of ablation success. Circulation 1996; 94: 407–424.
[11] Gallagher JJ, Cox JL, German LD, Kasell JH. Nonpharmacologic treatment of supraventricular tachycardia. In Josephson ME and Wellens HJ (Eds.). Tachycardias: mechanisms, diagnosis, treatment 1984; 1st ed. Philadelphia Lea & Febiger pp. 271–285.
CiteULike 
Connotea 
Del.icio.us What's this?
Top
Abstract
Introduction