transcript of David Haines, M.D., Atrial Fibrillation

http://www.med.virginia.edu/med-ed/cme/5a£ltranscript.html

Atrial Fibrillation: New approaches in Management

David E. Haines, M.D.

 

8/10/1999

KR: Good morning, on behalf of the University of Virginia School of Medicine I would like
to welcome you to this morning's Cutting Edge broadcast entitled Atrial Fibrillation: New
approaches in Management, presented by Dr. David Haines. I am Dr. Karen Rheuban,
Associate Dean for Continuing Medical Education. Dr. Haines Professor of Internal
Medicine of the Division of Cardiology received his M.D. Degree from the University of
Rochester and completed his post-graduate training at the Universities of Vermont and
the University of Virginia. His clinical and research interests are in the arena of arrhythmia
diagnosis and management, pacemakers, and implantable defibrillators, and the use of
catheter ablation therapy in the treatment of arrhythmias. It is my great pleasure to
introduce Dr. Haines.

DH: Thank you Karen, I appreciate having the opportunity to speak with you today. I'd like
to review this morning some of the new areas of work being done in the management of
atrial fibrillation and to give to you the audience some insight into where this very exciting
field is moving.

To begin, atrial fibrillation is a very prevalent arrhythmia. It fully accounts for one-third of
all patient discharges with arrhythmia as the principal diagnosis and by far in a way is the
most common arrhythmia requiring treatment today. In the United States, atrial fibrillation
accounts for 35% of all arrhythmia hospitalizations with an average hospital stay of five
days. The mean cost of hospitalization if $4,800 and this does not include the cost of
outpatient cardioversions, the prolonged length of stay from post-operative atrial
fibrillation, the cost of drugs, side effects monitoring, and the cost of AF induced stroke.
When we are managing patient's with atrial fibrillation two main management goals come
to mind. First, to increase the survival of the patient's and second to improve patient
quality of life. If these two goals are not achieved with any given therapeutic intervention
then one should consider changing to a different strategy.

Now first I would like to review the acute conversion of patient's who present with new
onset atrial fibrillation or recurrent atrial fibrillation. There is a lot of interest in
pharmacological conversion of these patients. This graph shows the proportion of
patients with successful conversion in a meta-analysis of drug versus control studies.
The line of unity is present there and this area indicates that the drug was better than the
control and as you can see, most of the agents including Quinidine, Disopyramide,
Flecainide, Propafenone, Amiodarone, Sotalol, and Ibutilide, or Dofetilide show some
effectiveness in the acute pharmacological conversion of this arrhythmia.

This graph on the left shows acute cardioversion rate with electrical cardioversion and as
you can see it is approximately 80% compared to the controls in this meta-analysis of
about 30%. On the right hand graph, you can see the relative rate of cardioversion
compared to spontaneous cardioversion with various drugs. The dotted yellow line
represents unity and the mean values for the meta-analysis are indicated by the magenta
dots with the range of outcomes in the studies indicated by the arrow-bars. As you can
see, the variety of drugs all have some effectiveness in acute cardioversion of atrial

fibrillation, although Sotalo] did not do as well as placebo and Ibutilide and Dofetilide show
the highest rate of acute cardioversion.

When a patient is acutely cardioverted, however, there are risks associated with this
procedure. The main risk is thromboembolism. This image shows images before and after
electrical cardioversion performed with transesophageal echo monitoring. The chamber
here is the left atrial chamber and the left panel shows normal flow through the atrial chamber pre-cardioversion. However, post-cardioversion we see evidence of spontaneous echo contrast. This indicates stasis of blood flow and this IS due to atrial stunning, a common phenomenon after cardioversion. During this period of time, patients

are at high risk of thrombus formation and potential thromboembolism.

Transesophageal electrocardiography has been used to identify high risk patients and
help prevent stroke associated with this intervention. Listed here are several studies who
used TEE guided cardioversion and as you can see, the prevalence of thrombi on these
studies range from 2.5% to 23% depending on the population studied. Those patients did
not undergo cardioversion. The patients who had low risk TEE echos underwent
cardioversion and as you can see the risks of emboli in these series was very low. It is

. very important to continue anticoagulation in the three to four weeks after electrical
cardioversion. The reason for this is that atrial stunning persists despite return of normal
electrical activity. This graph shows the peak a wave velocity, a measure of left
contractility, after cardioversion, immediately after, one day after, one week, one month,
and then finally three months after cardioversion. The green bar represents patients with
atrial fibrillation of brief duration, the blue bar those of medium duration, and the light bar
those patients who had long duration atrial fibrillation greater than a month prior to
cardioversion. As you can see, in those patients with long duration atrial fibrillation, the
return of mechanical activity did not occur until one month after cardioversion. 80 in that
window of time, up to one month, they are still at risk for thrombus formation and
potentially thromboembolism.

Now when one considers the various therapeutic options in management of chronic or
recurrent atrial fibrillation, these come to mind. First rate control and anticoagulation using
AV nodal blockers and Coumadin or Aspirin, or possibly AV junctional ablation and
Coumadin or Aspirin. One can use suppressive anti-arrhythmic drugs, primarily Class 1
and Class 3 anti-arrhythmic drugs. There has been much interest in the use of devices to
prevent recurrent atrial fibrillation particularly atrial synchronous pacing in patients with
sinus node disease or the new implantable atrial defibrillator that is currently in
development. And finally, there is preliminary evidence that atrial fibrillation recurrence
may be prevented with curative ablative therapy. Either with a focal technique or a linear
technique.

Now to briefly review the strategy of rate control and anticoagulation for atrial fibrillation. A
number of agents can be used to control the ventricular response to the AF including
Beta blockers, Calcium channel blockers, Digitalis compounds, or finally Ailjunctional
ablation. We consider A V junctional ablation to be yet one more strategy for rate control
in drug refractory patients. The choices for anticoagulation include Warfarin or Aspirin.

A point worth reviewing is that Digoxin, probably the most widely prescribed drug for rate
control is also probably our least effective agent. The reason for this is that Digoxin has
its effect on A V nodal conduction slowing through a centrally mediated vagotonic effect.

               With exercise or physiological stress, the body's response is vagolysis and hence the                       

effects of Digoxin are dissipated rapidly in these settings. This graph shows the use of
Digoxin in atrial fibrillation and shows the rest and exercise ventricular rates. The left bar
is the rest heart rate and the right bar is exercise heart rate. As one can see, when no
Digoxin is used, there is an elevated heart rate at rest, as well as an elevated heart rate
with exercise. However, if one increases to a relatively high dose of Digoxin, we do
appreciate slowing of the rest heart rate and yet the exercise heart rate does not change.
So Digoxin, while effective in many patients, is not our first drug of choice. We do use A V
nodal modification or ablation regularly in drug refractory patients. Here is a radiograph
showing the positioning of catheters in the RAO and LAO position. The ablation catheter
is positioned in the region of the A V node and radiofrequency current is delivered to this
catheter to achieve partial or complete AV nodal block. Of course, the patients are
usually left with a junctional escape rhythm that is too slow to allow for normal healthy
activity. So this procedure is almost always accompanied with implantation of permanent
pacemaker. Having performed this procedure, most patients benefit greatly from it. This
graph shows the quality of life after AV junctional ablation for paroxysmal atrial fibrillation.
There are two measures of psychological well-being and both measures show significant
symptomatic improvement for patients undergoing this procedure. So, this is considered
to be a good strategy for rate control.

Now, the concern with any rate control method including A V junctional ablation is that
patients continue to be at risk for thromboembolism. Those patients who are highest risk
for thromboembolic events include those with a history of hypertension, those with prior
stroke or TIA, or other systemic embolic event, those with diabetes, heart failure, or age
greater than 65 years.

Here is the CVA risk in the left graph rate per year in various subgroups of patients. The
left hand group are those without thromboembolic risk factors. They have a relatively low
risk but still in this case approaching 4% per year. However, in patients with
thromboembolic risk factors, the risk increases significantly especially for the older age
group. If one compares the risk of bleeding, CNS bleed, with Aspirin versus Warfarin
therapy. As expected, those receiving Warfarin have a somewhat higher risk but this is
only in the range of 0.5% per year. Hence, the trade-off for high-risk patients is clearly in
favor of Warfarin versus Aspirin to prevent CVA. In addition, it must be noted that there is
a small but real risk of acute CVA in patients undergoing cardioversion.

Despite the clear benefit of Warfarin in preventing thromboembolic events, especially in
those patients with risk factors for thromboembolism, the prevalence of Warfarin use is
still low. This survey of Warfarin use in the United States showed that from the years
1989 - 1993 as the results of large multi-centered randomized trials were being published,
the use of Warfarin in patients with chronic atrial fibrillation increased significantly.
However, from 1993 - 1996, the trend has been flat and under 40% of patients who are
probable candidates for Warfarin therapy are actually receiving this drug.

Now when one considers strategies for suppressing atrial fibrillation, there are a number
of drugs that are effective. The Class I drugs include Quinidine, Procainamide, Norpace,
Flecainide, and Propafenone. Class III drugs include Amiodarone, Sotalol, and two drugs
that are not yet released yet in the United States but will be within the next six months,
Dofetilide and Azimilide. All these drugs have some effectiveness in suppression of atrial
fibrillation and all anti-arrhythmic drugs have some potential risks with their use. This
graph shows the relapse rate of atrial fibrillation relative to the spontaneous relapse rates

               with various agents, various drugs. This is a meta-analysis looking at drugs including Quinidine, Flecainide, Propafenone, Amiodarone and Sotalol. The spontaneous relapse
rate is about 40% and this is the relative risk. So a level of I indicates similar to control.  
As you can see, all of the drugs that are tested have about a 50% efficacy at reducing the relapse rate of atrial fibrillation. So they all have some effectiveness but no drug is 100% effective. Dofetilide is one of the new drugs. It is a Class III agent and it does show reasonable efficacy in preventing atrial fibrillation. This is a dose ranging study of Dofetilide showing three increasing doses of this drug versus placebo and Sotalol. You can see that the medium and high dose Dofetilide were better than Sotalol in this study. A second study, the SAFIRE-D trial showed that Dofetilide was significantly better than placebo in terms of maintaining normal sinus rhythm in patients with atrial fibrillation. The other new drug is a drug Azimilide and this trial shows a placebo versus Azimilide trial in preventing recurrent atrial fibrillation. And again, the drug showed significant improvement in maintenance of sinus rhythm. However, all anti-arrhythmic drugs do have some potential for toxicity. In this trial from Coplan, we see that Quinidine was effective in maintaining normal sinus rhythm versus control, Control in the left bar, Quinidine in the green bar on the right. However, in this meta-analysis we identified a three-fold higher mortality in the Quinidine treated group indicating a potential for pro-arrhythmia with the use of Quinidine. This echos the results of the CAST trial in which Encainide and Flecainide were tested in patients post-myocardial infarction and there was a three-fold higher mortality in the drug treated group. The DIAMOND trials have not shown increased mortality with Dofetilide and large trials with Amiodarone have not shown excess mortality with this drug. So this does appear to either be a class specific or drug specific finding. However, we need to approach the use of all of these drugs with some caution. Here is another meta-analysis looking at the potential risk of drug therapy in AF, specifically looking at the risks of arrhythmic deaths. The top analysis shows all AF patients and you can see again a 2.6 increased risk of dying with this drug. Importantly, if you look at patients without congestive heart failure, that risk is 0 or the increased appears to be negligible. So the general rule of thumb is particularly for Class I drugs in patients with heart failure and structural heart disease these drugs should probably be avoided. Again the Class III drugs, Amiodarone, Sotalol, and now Dofetilide, Azimilide appear to be safer to use in patients with structural heart disease.

So the final question, what is better? Rate control and anticoagulation or suppressive
drug therapy. There are good sides and bad sides with both of these strategies and
currently underway in the United States is the AFFIRM trial which will specifically address
the risks and benefits of these two competing traditional strategies. We are approaching
completion of enrollment in that trial and results of that should be available in the next
couple of years certainly.

Moving onto device therapy for atrial fibrillation. This can be divided into the use of
permanent pacing with either AAI pacemakers or DDD pacemakers. Again, atrial
synchronous pacing is important in atrial fibrillation prevention. The other devices that are
being tested are implantable devices including the implantable atrioverter or implantable
atrial and ventricular defibrillators. That is a single device that has the capability of
cardioverting both arrhythmias.

If one looks at the risks of recurrent symptomatic atrial fibrillation using patients as their

                .-------,          own controls. One can see a significant benefit from implantation of the pacemaker
particularly with sinus node dysfunction. This study published by Dr. Saksena shows that
the baseline number of AF episodes per week in this population was three and after

                implantation of a pacemaker was significantly reduced. This issue was addressed                           

prospectively in the pacemaker selection in the elderly trial that was published last year in
the New England Journal of Medicine. This was a prospective randomized trial of VVI R
versus DOOR pacing modes in patients over 65 years old. It was an equal randomization
between the two arms and cross-over was allowed if patients had significant symptoms.
The primary endpoint was quality of life and secondary end points included death, CVA,
atrial fibrillation or pacemaker syndrome. This is a summary of the results in all patients,
those with sinus node disease, and those with A V block as the indication for pacemaker
implantation. The red bars are those patients receiving VVIR pacing and the green bar is
those with dual chamber pacing. As you can see in those patients with sinus node
dysfunction, dual chamber pacing significantly reduced risk of most of the endpoints
including the risk of atrial fibrillation. So it appears that in patients with sinus node
disease, there is something about the bradycardia that precipitates onset of atrial
fibrillation and use of pacing in these patients can sometimes have a dramatic effect at
reducing atrial fibrillation.

The Pac-A-Tach trail was another prospective trial to address this issue. This was a
prospective randomization of pacing modes in 198 patients with sinus node dysfunction
and a pretty equal randomization with a mean follow-up of 23 months. Now in this trial, in
fact, the prevalence of atrial tachycardia including atrial fibrillation was actually higher in
the dual chamber pacing arm which was not expected. However, the key factor with this
is that there was a huge cross-over from single chamber to dual chamber pacing and that
probably accounts for the unexpected results in this trial.

Another large trial that has been presented and yet has not been published yet is the
CTOPP trial which is a Canadian trial of permanent pacing. Again, this was a large trial
where patients were randomized between VVI and Physiological pacing, that is AAI or
ODD pacing and you can see that the characteristic of the two groups of patients were
fairly similar and approximately one-third of patients had sinus node dysfunction as their
indication for permanent pacemaker implant. The risk of atrial fibrillation at four years was
significantly reduced with the physiological pacing mode. So we have several trials that
now seem to support the use of dual chamber or atrial synchronous pacing as a strategy
to decrease the risk of atrial fibrillation.

The implantable atrial defibrillator is an interesting concept but clearly still in its
investigational phases. Pictured here is one of the Metrix defibrillator devices with the
electrodes that go into the coronary sinus, one electrode that goes into the right atrium
and then an electrode going into the ventricle to allow synchronization of shocks. These
devices automatically detect atrial fibrillation occurrence and will deliver either a patient
directed shock, a physician directed shock, or can be put in automatic mode and deliver
an automatic shock when a patient has atrial fibrillation.

Here is a composite graph from two different studies showing efficacy of atrial
cardioverters at termination of atrial fibrillation. This represents the atrial defibrillation
threshold, that is the level of energy that is required to successfully convert a patient from
atrial fibrillation back to sinus rhythm. These were just two different series of patients. The
first group was more highly selected and hence had lower defibrillation thresholds. The
second group was a little bit less selected and you can see have slightly higher energy

»=>:                  thresholds. Now the dotted line here and these two confidence intervals represent the
pain threshold for atrial defibrillation. An importantly, one can see particularly in the less

selected patient group that their average defibrillation threshold was about equivalent to C'

                the pain threshold. So probably half of these patients .were receiving shocks that they                   

considered to be intolerably painful. This has proven to be the main limitation of this
technique, particularly in patients with frequent atrial fibrillation recurrence, that is they
would rather have the atrial fibrillation than to have multiple shocks delivered to them.

Here is another interesting observation from the early experience with atrial cardioverter
implants. This is a complex slide but the horizontal bars represent the durations of
periods of sinus rhythm and the dots represent durations of atrial fibrillation episodes in
hours. So this is in days and these are in hours. The hypothesis was that, with an atrial
defibrillator, if you immediately cardioverted patients, over time they would have less atrial
remodeling and would have fewer and fewer episodes of atrial fibrillation, and have
shorter duration of Afib and longer intervals between the Afib episodes. And in fact in
these four patients that are shown, over time you can see that the pattern of Afib didn't
improve. The episodes of arrhythmia occurred just as frequently, and the interval

between AF episodes did not lengthen. Therefore, use of an implantable atrial defibrillator
should be considered as a viable treatment option but probably one that will be relatively
limited in its applicability.

Finally, we move to an area that I have a great deal of personal and research interest in
and that is ablative therapy for atrial fibrillation. That is using a catheter or catheters to
actually eliminate the source of the atrial fibrillation and hopefully cure the patient from
this potentially disabling arrhythmia. The two techniques that are currently being tested at
our institution and others include: the focal ablation technique, primarily with patients with
paroxysmal atrial fibrillation in particular those with lone atrial fibrillation, and then the
linear atrial ablation technique, more suited for patients with persistent atrial fibrillation
and structural heart disease.

Now the rationale for focal ablation of paroxysmal AF is that regions of local atrial activity,
either automatic, triggered or microreentry, may exist and produce rapid focal atrial
tachycardias. In turn these rapid focal atrial tachycardias can disorganize and have the
appearance identical to reentrant atrial fibrillation, or can induce typical reentrant atrial
fibrillation. Finally, focal ablation at the site of atrial premature depolarizations and atrial
tachycardia origin may prevent the initiation of clinical paroxysmal atrial fibrillation. Here is
an example of recordings taken from a patient who had very frequent symptomatic atrial
fibrillation. We have leads I, AVL, AVF, and V1 and then we have recordings from the left
superior pulmonary vein, the right superior pulmonary vein and another recording from
the right superior pulmonary vein. We have sinus rhythm at the initiation of the tracing
and then at this point, we have onset of atrial fibrillation.

What is apparent and is highlighted by the arrow is that in the right superior pulmonary
vein there is a focal firing of electrical activity that precedes any other electrical activity in
either the right or the left atrium and it is our hypothesis in this particular patient that that
may in fact be the source of their atrial fibrillation. Later in the study, atrial fibrillation starts
up and we are unable to terminate the atrial fibrillation even with a direct current electrical
shock. You can see surface leads I, AVF, and V1 and then a number of leads measured
from the right and left atrium. Here you can see the disorganized activity of atrial
fibrillation and despite this DC shock disorganized activity continues and the patient is
entirely refractory to cardioversion. We mapped out that focus in the right superior

-------.    pulmonary vein and delivered radiofrequency electrical current to that focus. This is a
continuous tracing showing ongoing atrial fibrillation. This offset here is the initiation of the
radiofrequency pulse and as you can see, atrial fibrillation continues, continues and then
at this point atrial fibrillation stops and we have return of normal sinus P-waves and

normal sinus rhythm. So that this patient, whom we could not even cardiovert out of atrial
fibrillation, by delivering a pinpoint radiofrequency burn in the right superior pulmonary
vein, we eliminated the atrial fibrillation. That patient is cured of this arrhythmia.

Two of the pioneers in this area are Michelle Haissaguerre from Bordeaux, France, and
Shih-Ann Chen from Taiwan. This is Dr. Haissaguerre's initial published series of 45
patients who had documented PAF on a frequent basis and they performed activation
mapping of the atrial ectopic site of origin. Interestingly, you can see the distribution of
these foci is almost exclusively in the pulmonary veins. This schematic representation
shows right atrium and left atrium, and you can see that only a scattering handful
occurred in the right atrium. Essentially, all of these had their origin in the pulmonary
veins. We are working hard now to understand why the pulmonary veins are such fertile
territory for the initiation of these abnormal impulses, but clearly these are the areas
where we target first. In their series, they had successful catheter ablation in 62% of their
patients overall which compares very favorably to any drug that one could use that has
perhaps a 50% success in suppressing atrial fibrillation. This is a series from Shih-Ann
Chen in Taiwan with a similar schematic of the right atrium and left atrium, and again you
see almost all of their sites of origin were pulmonary veins. In his series, mostly superior
pulmonary veins. In their series of 111 patients with 160 foci, they had acute success in
all but five foci. They had a high recurrence rate of 23% over seven months and they did
have complications including two transient ischemic attacks, two patients with
tamponnade, and 42% of patients that were studied that showed some focal stenosis of
the pulmonary vein which is a new complication that has arisen in this era of pursuing
focal ablation. There was no mortality. In order to improve the efficacy of these
techniques and decrease the risks of pulmonary vein stenosis and decrease procedure
time, new innovative catheters for pulmonary vein ablation are being introduced. This is
one such catheter that uses an ultrasound energy source and a compliant balloon to fill
the vein and then circumferentially heat the vein with ultrasonic energy.

Now in patients who are not the typical focal atrial fibrillation patient, what options are
available? Well we are pursing simultaneously investigation into linear atrial ablation for
these patients. Now the rationale behind this is that a critical number of atrial fibrillation
wavelets, that is regions of small swirling reentry, are required for atrial fibrillation
propagation. Each wavelet requires a minimum geographical area which is a function of
its wavelength or its size. If new anatomical barriers are created that reduce the number
of simultaneously propagating wavelets, then atrial fibrillation cannot be sustained. That is
if we segment the heart into smaller pieces, the spinning wheels of electrical activity that
maintain reentrant atrial fibrillation will run into barricades and extinguish. To achieve this
a variety of innovative catheter designs have been tested. We are using a loop ablation
catheter that has multiple coil electrodes and temperature monitors. Here is a radiograph
showing introduction of the loop catheter into the left atrium. This is an LAO view. This
catheter is a mapping catheter in the right atrium. Here is a catheter in the coronary sinus
showing the posterior border of the heart and here is the loop catheter that has been
deployed in the left atrium filling the chamber and allowing us to create a long linear lesion
to segment the left atrial chamber. Here is an example of a patient who underwent linear
atrial ablation. The top three leads show surface leads 1, AVF, and V1 and then the
bottom leads are multiple leads from the right atrial chamber showing the disorganized
atrial fibrillation rhythm. We initiate radiofrequency ablation and this was after creation of
a number of linear lesions and we have some organization of the rhythm as you can see
more discreet activity, but the radiofrequency energy is turned on. Part way through the
energy delivery, we suddenly see a shift from faster disorganized rhythm to more discreet

flutter-like activity indicating that the wavelets are becoming more organized. Then finally
with continued RF energy delivery, we have termination of the arrhythmia and restoration
of normal sinus rhythm. This patient was cured from his atrial fibrillation.

Now the initial results of atrial fibrillation ablation from our Center and others around the
world performing this technique show more modest success than that achieved with the
focal ablation. It is important to note that in the worldwide experience, we are still looking
at very small numbers of patients and the cure rate ranges from 10% - probably 46% was
the highest cure rate. However, we find that these patients sometimes with a combination
of previously ineffective suppressive drugs, plus the ablation procedure can have
suppression of their atrial fibrillation somewhere between 50 - 75% of the time.
Complication rate is higher because this is a much more extensive procedure.

The risks of the linear atrial ablation include thromboembolism, bleeding, proarrhythmia,
pulmonary vein stenosis, phrenic nerve injury, loss of atrial transport function, and
radiation exposure.

Now the proposed benefits of ablation therapy for atrial fibrillation include decrease in
symptoms of atrial fibrillation. Balancing that include the potential risks or proarrhythmia
from linear ablation, complications such as pulmonary vein stenosis, and we need to
keep in mind that A V junctional ablation should be the gold standard against which we
compare symptom reduction because we know that standard approach is very effective.
We hope that these approaches will decrease the long-term risks of stroke. However,
manipulating catheters and ablating in the left atrium exposes the patients to some risk of
acute stroke. In addition, with extensive atrial ablation, they may suffer loss of atrial
transport function and it may increase the risk of stroke long-term. We would hope that
this approach would improve survival, but the excess mortality from AF is very low to
begin with and again with extensive catheter procedures, there may be some procedure
related risk of death. So the jury is still out with regard to curative ablation therapy for
atrial fibrillation.

The spectrum for atrial fibrillation, we are learning more. On one side our patients with
normal hearts, normal atrial size, that have paroxysmal atrial fibrillation and may in fact be
the ones that are purely focal, the lone Afib patients. On the other side of the spectrum,
we have patients with diseased hearts significantly enlarged atria, chronic atrial fibrillation.
Clearly, they have a reentrant mechanism. However, in the middle maybe a transition,
maybe people who start with a focal abnormality and if you catch them early may respond
to focal ablation techniques, but if you wait long enough they may make the transition to
an enlarged atrium and no longer be candidates for focal ablation. So the indications for
catheter ablation include symptomatic Afib patients refractory to medical therapy and
those who have a strong desire for drug and device-free lifestyle but the risks must be
balanced against anticipated improvement in quality of life.

So to conclude, there are several subsets of patients that have AF, multiple treatment
options for AF are viable, and the selection of therapeutic modalities need to be
individualized to the patient with regard to the patient anatomy and likely AF mechanism,
the symptom severity, and their acceptance of a different therapeutic modalities.

Thank you.

KR: Thank you Dr. Haines for your most informative presentation. This would be a

wonderful opportunity to entertain questions from the audience. I do have several
questions that we have received thus far. The first one is at what point do you decide it is
not worth trying to cardiovert and just leave the patient in chronic atrial fibrillation?

DH: It is my personal strategy in the management of atrial fibrillation patients to attempt
one cardioversion at some point in the patient's course. Even if they have had
long-standing, chronic atrial fibrillation. The general rule of thumb is once a patient has
been in atrial fibrillation for two years or more, it is relatively unlikely that one is going to
be able to successfully cardiovert them and maintain long-term sinus rhythm. 80 that
those people with longer duration of atrial fibrillation are probably not the ones that you
would repeatedly cardiovert. But I think at least one cardioversion is always indicated. I
think the benefits of maintaining sinus rhythm are clear. Patients who have intermittent
atrial fibrillation less than one episode every year or two can sometimes be managed with
intermittent cardioversion alone. This is an outpatient procedure. It is a safe procedure. If
it is done within the first 48 hours of Afib onset no anticoagulation is required and patients
do well with it. However, longer-standing atrial fibrillation it's imperative that the patient
receive anticoagulation before cardioversion, as well as a month after cardioversion. But
still I think that everybody should get at least one cardioversion. If they have recurred in
Afib and recurred despite the use of good suppressive anti-arrhythmic drug therapy then I
think it is reasonable to withdraw drugs and treat them with rate control and
anticoagulation long-term.

KR: Another question, how do you decide which anti-arrhythmic drug to use for an
individual patient, especially first line?

DH Well, that is an important issue. Again, there are concerns about the use of Class I
drugs, that is Quinidine, Procainamide, Disopyramide, Flecainide, Propafenone in
patients particularly with ischemic heart disease and prior infarction but probably we could
generalize and say that patients with structural heart disease, that is depressed LV
function may be at increased risk with the use of those drugs. 80 as a rule, I avoid those
drugs in patients who have structural heart disease. However, in patients with structurally
normal hearts or near normal hearts, let's say the patient with hypertension but only mild
LV hypertrophy, those patients do very well with the Type I drugs and I use Propafenone
and Flecainide frequently in those patients with good results. The Class III drugs have
been limited to Sotalol and Amiodarone but we will have a lot of new options coming
down the line with Dofetilide and Azimilide. Sotalol I think is a good drug to use. I think
there is some proarrhythmic risk of the drug and we as a rule hospitalize patients,
particularly those with depressed LV function during the initiation of that drug therapy
because of the small but real risk of proarrhythmia. I think Amiodarone is an excellent
drug. It of course, does have problems with long-term toxicity and we need to monitor
lung function, hepatic function, thyroid function, and caution patients against direct sun
exposure. 80 it is not an easy drug to use but it is a very well tolerated drug from the
EMIAT and CAMIAT trials of Amiodarone post-MI we know that patients with diseased
hearts do not have excess mortality with the use of this drug so it appears to be a safe
drug and that again is generally my first choice off the shelf for patients with structural
heart disease. I think that Dofetilide and Azimilide will also be reasonably safe and
effective in patients with structural heart disease when those drugs become available.

KR: Okay, who is the optimal candidate for curative ablation?

DH: Right now we are being highly selective in patients to whom we offer those