ABSTRACTS

CLEVELAND, OHIO. Atrial fibrillation and atrial flutter often co-exist in the same patient with one readily converting to another. Studies have shown that a successful pulmonary vein antrum isolation (PVAI – Natale method) often is effective in controlling both afib and aflutter in patients who have not undergone previous cardiac surgery. However, it is not known whether patients who have had cardiac surgery are more likely to develop flutter after a PVAI than are those with no previous cardiac surgery (controls).

Researchers at the Cleveland Clinic analyzed data from 1125 patients who had undergone a PVAI. The majority (1062 patients) had no history of cardiac surgery, while the remaining 63 had undergone such surgery. The afib recurrence rate among surgery patients (after about 1.5 years of follow-up) was 21% (success rate of 79%) versus 19% (success rate of 81%) for controls; in other words, no significant difference. The rate of recurrence of atrial flutter, however, was much higher (33%) in surgery patients than in controls (4%). All 21 aflutter patients in the surgery group underwent an ablation procedure for atrial flutter with an immediate success rate of 86% and a long-term success rate of 76%. The immediate success rate for controls was 100% with a long-term rate of 95%.

The researchers point out that incision lines, cannulation sites, scars, etc. may make surgery patients more prone to develop aflutter. They also point out that among patients not exposed to previous cardiac surgery the PVAI cured pre-existing afib in 94% of cases as compared to only 63% among cardiac surgery patients. It is also noteworthy that, while 18% of controls experienced atrial flutter within the first two months after the PVAI, only 4% continued to do so long term (1.2% experienced left atrial flutter). In the group of former heart surgery patients 60% experienced aflutter during the first two months and 33% needed an ablation to eliminate it.

The researchers conclude that patients who have undergone cardiac surgery previous to their PVAI should be warned of the possibility of developing atrial flutter after the procedure.
Kilicaslan, F, et al. The need for atrial flutter ablation following pulmonary vein antrum isolation in patients with and without previous cardiac surgery. Journal of the American College of Cardiology, Vol. 45, March 1, 2005, pp. 690-96

BORDEAUX, FRANCE. Conventional medical wisdom has it that left ventricular dysfunction can be a causative factor in atrial fibrillation (AF). Now researchers at the Hopital Cardiologique du Haut Leveque provide tantalizing evidence that it may actually be the other way round, ie. that AF can result in left ventricular dysfunction.

Their study included 38 men and 10 women with paroxysmal or chronic (persistent or permanent) symptomatic AF. Their type of AF was classified as isolated, that is AF in the absence of moderate or severe mitral regurgitation and/or mitral stenosis, mitral annular calcification, coronary artery disease, chronic pulmonary disease, systemic hypertension (greater than 140/90 mm Hg), thyroid disease, and diabetes.

This classification would seem to be quite similar to "lone" afib or perhaps more precisely "idiopathic" AF. The participants were studied in sinus rhythm (when possible) prior to undergoing a PVI procedure with additional lesion lines as required. The researchers observed a pronounced tendency to left ventricular dysfunction in the AF patients as compared to controls. They found a significant reduction in lateral early diastolic peak velocity in 37% of paroxysmal afibbers and in 48% of chronic afibbers when compared with healthy controls.

Twenty-nine (78%) of the 37 paroxysmal patients and 6 (54%) of the 11 chronic patients were successfully treated without AF recurrence. All patients with recurrence were classified as partial success and pooled with patients with complete success. None of them were considered to have had a failure of ablation (defined as less than 70% improvement) and none required a second procedure. The patients were followed for 12 months after their PVI. The following major changes were observed at the final (12-month) examination:

• Left atrium diameter had decreased by an average of 11% in both paroxysmal and chronic afibbers. Left atrium area had decreased by an average of 18% in paroxysmal and by 23% in chronic afibbers.

• Left ventricular diastolic function had improved significantly with an average increase in lateral early diastolic peak velocity of 29% in paroxysmal afibbers and a 46% increase in chronic AF patients.

• Left ventricular ejection fraction also increased significantly over the observation period (by an average of 7.7% in paroxysmal afibbers and 18.8% in chronic ones).

The researchers conclude that remodeling of the left atrium and ventricle does indeed take place after a successful PVI and that this may suggest that afib may be partly the cause rather than a consequence of diastolic dysfunction.
Reant, P, et al. Reverse remodeling of the left cardiac chambers after catheter ablation after 1 year in a series of patients with isolated atrial fibrillation. Circulation, Vol. 112, November 8, 2005, pp. 2896-2903

BORDEAUX, FRANCE. Pulmonary vein isolation (PVI) has been shown highly effective in eliminating paroxysmal (intermittent) atrial fibrillation with some specialized centers showing cure rates of close to 90%. The elimination of persistent and permanent afib has, however, proven much more difficult with success rates closer to 60%. Researchers at the Hopital Cardiologique du Haut Leveque in Bordeaux now report the development of a new ablation procedure resulting in a cure rate of 95% for patients with persistent afib.

Their study involved 53 patients with persistent afib (episodes longer than 7 days, but amenable to electrical cardioversion) and 7 patients in permanent afib. The patients were between the ages of 44 and 62 years (mean age of 53 years) and had endured afib for a median of 12 months. They also had failed an average of 3.3 antiarrhythmic drugs. The patients underwent a highly complicated procedure in which several distinct areas of the left and right atria were ablated in random order. The ablations were all carried out using the Thermacool catheter (an irrigated tip ablation catheter with a distal 3.5 mm tip and three 1-mm electrodes separated by 2-5-2 mm interelectrode spacings). Maximum power used inside venous structures was 20-30 watts, while 30-40 watts was the maximum when atrial structures were ablated.

The initial ablation step in 22 patients was a PVI, isolation of the superior vena cava and the coronary sinus (thoracic veins) was the initial step for 19 patients, and the remaining 19 patients started out with ablation of areas in the left atrium showing unusual electrical activity. After this first "round", 3 patients (5%) converted to sinus rhythm. It is noteworthy that only 1 out of the 22 patients (5%) having the PVI achieved sinus rhythm as a result of just this procedure. Seventeen of the remaining 57 patients still in afib were then treated with a PVI (assuming they had not undergone one in the first step), 19 received thoracic vein ablation, and 21 received an atrial ablation. This brought another 12 patients into sinus rhythm for an overall success rate of 25%. The third step for the remaining 45 patients was a PVI in 15 patients, thoracic vein ablation in 17, and atrial ablation in 13. This resulted in termination of afib in 17 patients increasing the success rate to 53%.

Finally, linear ablation (involving the cavotricuspid isthmus and the left atrial roof) of the remaining 28 patients brought 20 into sinus rhythm resulting in a total success rate of 87%. In most cases the conversion to sinus rhythm went through a stage of atrial tachycardia, which had to be ablated during the procedure as well. Total average procedure time was 264 minutes (4.5 hours) with an average fluoroscopy time of 84 minutes.

The ablation site resulting in conversion to sinus rhythm or atrial tachycardia was the pulmonary veins in 18% of patients, the coronary sinus region in 17%, anterior left atrium in 15%, atrial roof in 10%, mitral isthmus in 8%, and the septum in 10%. Only 3% of patients converted after ablation around the superior vena cava.

Based upon their results the Bordeaux researchers propose the following sequence for ablation of persistent afibbers. PVI followed by linear ablation of the left atrial roof and then atrial ablation along the left atrial appendage and the coronary sinus. If afib persists then other atrial locations should be targeted with the final site being the mitral isthmus line.

Five of the 8 patients whose first ablation was unsuccessful underwent a successful second ablation bringing the total cure rate to 95%. All study participants were hospitalized for 1 day at 1, 3, 6 and 12 months after the last procedure for clinical review and ambulatory monitoring. Twenty-four patients developed atrial tachycardia during the first 3 months of follow-up and were successfully ablated for this after-effect. None of the patients developed pulmonary vein stenosis. The Bordeaux center has now performed ablations on over 4000 patients and has experienced no incidences of atrio-esophageal fistula.
Haissaguerre, M, et al. Catheter ablation of long-lasting persistent atrial fibrillation: Critical structures for termination. Journal of Cardiovascular Electrophysiology, Vol. 16, November 2005, pp. 1125-37
Haissaguerre, M, et al. Catheter ablation of long-lasting persistent atrial fibrillation: Clinical outcome and mechanisms of subsequent arrhythmias. Journal of Cardiovascular Electrophysiology, Vol. 16, November 2005, pp. 1138-47
Tse, Hung-Fat and Lau, Chu-Pak. Catheter ablation for persistent atrial fibrillation: Are we ready for "prime time"? Journal of Cardiovascular Electrophysiology, Vol. 16, November 2005, pp. 1148-49

Editor's comment: This study clearly shows that the elimination of persistent and permanent afib requires a far more comprehensive and complicated approach than does the elimination of paroxysmal (intermittent) afib where a simple PVI is usually enough to do the job. This may, at least partially, explain the wide variation in success rates reported by various centers. It would seem that the extensive destruction of atrial tissue resulting from this comprehensive procedure could affect the atria's efficiency in acting as booster pumps for the ventricles. The Bordeaux researchers are currently evaluating this aspect, but so far have found no indication that exercise capacity (duration and maximum workload) are decreased following the comprehensive procedure. My own conclusion would be that persistent and permanent afibbers need to seek out the very best EPs and centers if they are to have a half decent chance of being cured.

MORIOKA, JAPAN. Atrial fibrillation (AF) is a common arrhythmia in Japan although the prevalence is lower than in Western countries. Japanese scientists have just completed a study to determine the actual level of AF found in Japan and to estimate the future burden of the condition on the health care system. The study involved 10,042 men and 13,671 women who were examined (12-lead electrocardiograms) in 1980, 1990, or 2000.

The prevalence of AF among men and women over the age of 30 years increased from 0.7% in 1980 and 1990 to 0.9% in 2000; an increase of 29%. The prevalence among men (1.0%) was higher than that among women (0.6%). The prevalence of AF increased markedly with age; among men aged 70 years and older it was 3.5% in 2000 as compared to 2.1% for women. The researchers estimate that the number of Japanese citizens suffering from AF will exceed 1 million within another 10 years. NOTE: The survey did not distinguish between lone atrial fibrillation and AF associated with heart disease.
Ohsawa, M, et al. Rapid increase in estimated number of persons with atrial fibrillation in Japan. Journal of Epidemiology (Japan), Vol. 15, No. 5, September 2005, p. 194-96

Editor's comment: AF is clearly a significant health concern in Japan; however, it is likely that its prevalence was underestimated in this study. Whether or not a participant had afib was based on just one 12- lead electrocardiogram; it is thus likely that many cases of paroxysmal afib were missed.

MAASTRICHT, THE NETHERLANDS. A large team of European cardiologists has just released the results of a study aimed at determining just how well European cardiologists are following the official European/North American guidelines for the management of atrial fibrillation (AF). The guidelines can be found at www.circ.ahajournals.org/cgi/content/full/104/17/211 8

The study included 5333 AF patients enrolled in 182 centers in 35 countries. About 18% of the participants were enrolled after experiencing just one initial episode, 28% had paroxysmal afib when enrolled, 22% had the persistent variety, and 29% were in permanent afib. The status of the remaining 3% is unknown. About 10% of all participants had the lone (idiopathic) type of AF with the prevalence of lone afib being highest in the paroxysmal group (15%). About 65% of participants were hypertensive and 28% of paroxysmal and 52% of permanent afibbers had heart failure or a left ventricular ejection fraction below 35%. Thirteen per cent of patients with permanent AF had mitral valve stenosis as compared to only 3% among paroxysmal afibbers. The average left atrial diameter was 43 mm for paroxysmal, 48 mm for persistent, and 51 mm for permanent afibbers.

Highlights of the study are as follows:

• About 33% of patients who should have been on anticoagulation (warfarin) therapy according to the guidelines were not. Perhaps more disturbing is the finding that 49% of patients who did not need anticoagulation were receiving it. This despite the fact stated in the report, "In patients with low risk for stroke, the bleeding risk of anticoagulation therapy outweighs the benefit of stroke prevention, but aspirin is recommended in these patients." The report draws the following conclusion in regard to anticoagulation, "Altogether it seems that in daily practice selection of stroke prevention therapy is not strongly determined by the clinical indications as recommended by the guidelines."

• Only half of the patients had their thyroid hormone levels checked. This despite the fact that thyroid problems are a well established underlying cause of afib. Even more disturbing, 40% of all patients on amiodarone never had their thyroid function determined even though amiodarone is known to damage the thyroid gland.

• The majority (69%) of participants had symptomatic afib with the most common symptoms being palpitations and fainting.

• About 40% of all patients received rhythm control medication (primarily amiodarone and sotalol) even though almost half (44%) of them had no symptoms warranting the use of dangerous antiarrhythmics. This is in direct conflict with the guidelines, which state that rhythm control should only be prescribed for symptomatic patients. For most patients, taking antiarrhythmics did not eliminate their symptoms indicating that, "available rhythm control strategies are inadequate and that there is at present an unmet need for safe and efficacious antiarrhythmic drugs for control of AF." I think most of us can probably agree with this statement. The report, unfortunately, did not distinguish between medication schemes for lone afibbers and those for afibbers with underlying heart conditions.

The researchers point out that most of the centers involved in the study specialized in AF. Thus clinical practices in less specialized centers and private practices are likely to deviate even more from the guidelines.
Nieuwlaat, R, et al. Atrial fibrillation management: a prospective survey in ESC member countries – The Euro Heart Survey on Atrial Fibrillation. European Heart Journal, Vol. 26, November 2005, pp. 2422-34
Wyse, DG. The Euro Heart Survey on atrial fibrillation: a picture and a thousand words. European Heart Journal, Vol. 26, November 2005, pp. 2356-57 (editorial)

Editor's comment: The finding that inappropriate prescription of warfarin and antiarrhythmics is fairly common is fully in line with the results of our surveys, so there is no reason to believe that adherence to the guidelines is more pronounced in the US and Canada than it is in Europe.

It is clear that further studies are urgently required to clearly establish the benefits and disadvantages of LAA removal and equally clear that such studies, to be of value, must distinguish between afibbers with heart disease and those without.

References

LAF Survey 9 – Follow-up

Since the publication of the original LAF Survey-9 results in September I have received additional responses relating to procedures performed prior to August 2005. Data is now available covering the outcome of 215 RF (radiofrequency) ablation procedures performed on 165 individual afibbers. The success rates for the individual procedures are given below:

(1) No afib episodes, no antiarrhythmics or blockers
(2) No afib episodes, but still on antiarrhythmics or blockers
(3) Recurring afib episodes

It is clear that only about one third of all RF ablation procedures covered in the survey have a fully successful outcome, while just over 50% do not result in the elimination of afib episodes even with the continued use of medications. About 30% of survey respondents underwent a second or third procedure.

Of course, the most important aspect for an afibber undergoing RF ablation is whether or not they will ultimately be cured – no matter how many procedures it takes. The overall outcome for the 165 survey participants is given below:

The actual, objectively determined, complete success rate of 44% is disappointing and is lower than the percentage of afibbers who subjectively felt that their ablation has been a success (56%).

Whichever success rate one looks at, it is clearly much lower than the success rates claimed by top institutions such as the Cleveland Clinic and the Hopital Cardiologique du Haut Leveque in Bordeaux.

Thus, it would be of interest to analyze the data a bit closer to account for the influence on success of the electrophysiologist performing the ablation and the institution where it is done.

Procedural Performance Rating

In order to provide some guidance in regard to the chance of undergoing a successful and safe ablation at a particular institution, I have developed a Performance Rating scheme. This rating takes into account the success rates and adverse event rates reported by afibbers treated at specific institutions. The outcome factors entering into the Performance Rating are as follows:

The outcome factor adds the scores (for successes and adverse events) for each RF ablation and the performance rating then averages these outcome factors for each individual institution. For example, if a procedure is fully successful with no adverse events, than the outcome factor is +10. If a procedure is a failure and accompanied by the creation of left atrial flutter or tachycardia, then the outcome factor is –5. The primary performance rating does not take into account that a large number of adverse events are resolved within a few months after the procedure. To acknowledge this, an adjusted performance rating is also shown in which the effect of resolved adverse events has been taken into account.

Please note that in this particular evaluation a RF ablation procedure is not considered a failure unless followed by another RF ablation or continued afib episodes. The subsequent occurrence of left or right atrial flutter or tachycardia is treated here as an adverse event and not as an ablation failure.

It is clear that a performance rating is not very indicative in cases where just one or two procedures have been performed. Thus, performance ratings have only been established for institutions where results for 4 or more procedures were available. Based on the adjusted performance rating the various institutions stack up as follows:

The first 9 institutions (Group A) in the above table account for close to 50% of all ablation procedures performed; their performance is evaluated in detail below:

The electrophysiologists performing the procedures in the above 9 institutions are as follows:

The average procedural success and adverse event rates for the remaining centers are given in the following table.

Group B contains 6 institutions for which reports of 4 or more ablations are available
Group C contains 5 institutions for which reports of 3 ablations are available
Group D contains 48 institutions for which reports of 1 or 2 ablations are available

Combining the 9 top-ranked institutions with the 59 other ones yields the following results:

The above statistics are indeed sobering and do no support the popular "myth" that a RF ablation is "a piece of cake". Undergoing an ablation at an institution other than one of the 9 top-ranked ones is associated with an average complete success rate of 16%, a partial success rate of 13%, and a failure rate of 71%. This is accompanied by an average adverse event rate of 38%.

While it is fairly clear that an institution where 3 out of 4 procedures were failures belongs in the bottom-half, it is much more difficult to rate institutions where data for only one or two procedures are available. Thus, the following information may be of interest.

Institutions performing two procedures of which one was a complete success at first attempt are:

• Beaumont Hospital, Detroit, MI – Dr. David E. Haines
• Deaconess Medical Center, Spokane, WA – Dr. Gerhard Mulheims
• Mayo Clinic, Rochester, MN – Dr. Douglas L. Packer
• Riverside Methodist Hospital, Columbus, OH – Dr. John Hummel

Institutions performing one procedure with a 100% complete success rate at first attempt are:

• Lahey Clinic Medical Center, Burlington, MA – Dr. Roy M. John
• Prince of Wales Hospital, Sydney, Australia – Dr. R.M. Allen
• Utah Valley Regional Medical Center, Provo, UT – Dr. Chun Hwang
• St. Vincent's Hospital, Portland, OR – Dr. Blair Halperin
• Tulane University, New Orleans, LA – Dr. McKinnie
• Valley Hospital, Ridgewood, NJ – Dr. Jonathan Steinberg

The scarcity of data clearly means that nothing can be concluded about the general success rate of the above institutions or EPs. However, it may provide comfort to someone considering an ablation there, that they have had at least one success at first attempt.

Success Rate on First Attempt

The success rates of the initial procedure were as follows for institutions in Group A.

Initial success rates for institutions in Groups B, C, and D are presented below.

It is clear from the above that the best chance of being cured of afib with just one procedure can be found at Marin County Hospital (Dr. Natale), the Cleveland Clinic or Loyola University Medical Center (Dr. David Wilber)

Ultimate Success Rate

The ultimate measure of success for the individual afib patient is, of course, whether or not they are cured of afib irrespective of whether it takes 1 or 5 separate procedures. About one third of all initial procedures were followed by one or more additional RF ablations. Overall final results for the top 9 institutions (Group A) are presented in the table below.

(1) 4 of the 7 patients having repeat ablations did not have their initial procedure at Cleveland Clinic. NOTE: No significant differences in success rates were observed for the 4 EPS doing PVAIs at the CC (Drs. Natale, Saliba, Schweikert, and Tchou)
(2) 1 of the 5 patients having repeat ablations did not have their initial procedure at the Hopital Cardiologique du Haut Leveque.
(3) The repeat rate applied to the number of patients having their initial procedure at the institution.

The average repeat rate was 21%. However, if Marin County Hospital and the Cleveland Clinic, which both have very low repeat rates, are omitted then the repeat rates rise to 36% indicating that even EPs at the top institutions often need to perform two procedures to achieve a cure.

Comparative data for the institutions in Groups B, C, and D is given below.

Combining the results for all groups produces the following final outcome (after repeat ablations).

A recently published survey involving 8745 patients treated at 90 different institutions concluded that the overall complete success rate (no afib – no drugs) was 52% - including 27% who needed a touch-up procedure[1]. Institutions having performed more than 300 ablation procedures had an overall complete success rate of 63.8% - not substantially different from the 64% complete success rate observed for the 9 top-ranked institutions in this survey.

The overall repeat rate observed in our survey is identical to the one observed in the published survey (27%).

The success rate among institutions having performed less than 30 ablation procedures was 30% - somewhat higher than the 23% complete success rate observed for other than the 9 top-ranked institutions. I believe the reasons for this difference is the recent enormous increase in the number of ablations conducted and the fact that many of these ablations are performed by electrophysiologists operating during the very early stages of the learning curve.

Conclusion

I have made every effort to ensure that the calculations and conclusions made in this survey are correct. I have observed good internal consistency in the data and am comforted by the fact that the average rate of complete success for top-ranked institutions found in this survey (64%) is identical to the one reported in the Circulation study (63.8%). Similarly, the repeat (touch-up) rates found in the two surveys are identical at 27%.

The LAF Survey-9 results are based on a total of 215 procedures involving 185 individual patients, not an overly large number, but enough to draw reasonably valid conclusions in general terms. Where the survey results become somewhat less "solid" is in the evaluation of the success rates of individual institutions. A sample of just 4 or 5 procedures is clearly not very significant in a statistical sense and it is quite possible that a larger sample would produce different results. However, based on conversations with hundreds of afibbers, perusal of hundreds of articles relating to RF ablation, and my own instinctual feeling, I have no hesitation in recommending the 9 top-rated institutions presented in this survey. There may well be other institutions and individual EPs that deserve top ranking, but I have no compelling evidence that this is indeed the case.

To summarize, the inescapable conclusion of this survey is that RF ablation for atrial fibrillation is still an emerging technology and that a half decent chance of success can only be expected in top-rated institutions. To go anywhere else, at this point in time, is likely to lead to disappointment.

[1] Cappato, R, et al. Worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circulation, Vol. 111, March 8, 2005, pp. 1100-05