ABSTRACTS

WASHINGTON, DC. The maximum heart rate (resting ventricular rate) experienced during an atrial fibrillation episode varies from person to person. Some afibbers barrel along at 200 bpm or more, while others experience afib at rates as low as 40 bpm. The LAF Survey V found that maximum rates varied between 55 and 280 bpm among 132 respondents.

A team of American researchers has just completed a study in which they investigated the relationship between the maximum heart rate experienced during afib (at rest) and other afib variables. The study included 4059 participants in the AFFIRM trial. This trial involved persistent afibbers with one or more risk factors for stroke who were randomized to a rate control group (no attempt made to convert to sinus rhythm) or a rhythm control group (attempts made to regain and maintain sinus rhythm through the use of electrical cardioversion and antiarrhythmic drugs).

The researchers found that the maximum rate during afib varied between 37 and 300 bpm. The rate tended to be higher among women and smokers and was also higher during the first afib episode than during subsequent ones. Afibbers with a normal left ventricular ejection fraction also tended to have higher heart rates during afib. Afibbers with a history of coronary artery disease, hypertension or myocardial infarction (heart attack), and patients with an enlarged left atrium and long episodes (greater than 48 hours), on the other hand, tended to have a lower maximum heart rate during episodes. This would tend to indicate that afibbers with healthier hearts have higher maximum heart rates during afib. The results of the study also showed that afibbers with a high maximum rate had a higher probability of achieving and maintaining normal sinus rhythm than did those with lower maximum rates.

The researchers conclude that the maximum heart rate during AF may be a useful indicator in deciding whether a persistent afibber should be steered toward rhythm control (those with high maximum heart rates) or rate control (those with low maximum heart rates).
Cooper, HA, et al. Relation of initial resting ventricular rate to the ability to achieve and maintain normal sinus rhythm in patients with atrial fibrillation. American Journal of Cardiology, Vol. 95, March 1, 2005, pp. 597-602

BAD KROZINGEN, GERMANY. The pulmonary veins are the main sources of the ectopic beats and tachycardias responsible for the initiation of AF. The rogue cells creating the ectopics are found in the myocardial sleeves extending from the left atrium into the veins. Current ablation practices therefore focus on electrically isolating the pulmonary veins from the atrium through the creation of lesions (with radiofrequency energy or freezing [cryoablation]), which interrupt the electrical pathways between the veins and the left atrium. There are two main approaches to achieving the isolation:

• The electrophysiological approach in which the electrical pathways are located using a mapping catheter (Lasso or similar) and then isolating by ablation.

• The anatomical approach in which the veins are isolated guided by an electroanatomical (non-fluoroscopic) mapping system (CARTO or Astronomer). In this method, the offending pathways are not specifically identified, but rather a contiguous, circumferential lesion is created which presumably stops all conduction between the pulmonary veins and the atrium.

German researchers now report that the anatomical approach, on its own, is not effective in preventing recurrence of AF. Their study involved 34 afib patients (average age of 52 years, 76% male) who had all proven resistant to at least 3 antiarrhythmic drugs. The majority (76%) had lone AF – 22 patients the paroxysmal variety (2 or more episodes per week) and the remaining 12 the persistent form. All patients underwent anatomically guided pulmonary vein ablation (PVA) after which the extent of remaining conduction between the veins and the atrium was measured with a 64-pole basket catheter. Only 46% of the isolated veins showed complete conduction block indicating that sole dependency on electroanatomical mapping during PVA is not sufficient to ensure a cure.

The remaining pathways were then isolated using electrophysiological guidance (basket catheter) and at the 12- month follow-up, 21 of the 34 patients (62%) were in sinus rhythm without the use of antiarrhythmics. Another 8 patients (24%) were in sinus rhythm with the aid of antiarrhythmics, which had been ineffective prior to the ablation. Nine patients (26%) needed a touch-up and 6 patients (18%) developed atrial flutter after the procedure.

The researchers conclude that the use of a combined anatomical/electrophysiological approach does not result in better clinical success than the use of the electrophysiological approach on its own. They also point out that the occurrence of left atrial flutter after ablation is highly unusual when the electrophysiological approach is used on its own, so the 18% incidence during their study was probably associated with the initial creation of the contiguous, circumferential lesion. NOTE: Electrophysiologists at the Hopital Cardiologique du Haut-Leveque in Bordeaux have also found that purely anatomically guided ablation leaves residual conduction in 45% of the supposedly isolated pulmonary veins[1].
[1] Hocini, M, et al. Prevalence of pulmonary vein disconnection after anatomical ablation for atrial fibrillation: consequences of wide atrial encircling of the pulmonary veins. European Heart Journal, Vol. 26, April 2005, pp. 696-704www.eurheartj.org/cgi/content/full/26/7/696
Arentz, T, et al. Effects of circumferential ostial radiofrequency lesions on pulmonary vein activation recorded with a multipolar basket catheter. Journal of Cardiovascular Electrophysiology, Vol. 16, March 2005, pp. 302-08

FORT COLLINS, COLORADO. Ten thousand years ago the agricultural revolution began a dramatic change in the human diet that continues to the present day. Estimates suggest that over 70% of dietary calorie intake in Western populations comes from foods which our Paleolithic ancestors rarely or never ate.

A group of experts from Colorado State University have assembled evidence to support their theory that this dietary change has produced many of the so-called diseases of civilization, such as cardiovascular disease, cancer, diabetes, osteoporosis, arthritis, and gastrointestinal disease. In a recent commentary they propose that changes in food production 10,000 years ago occurred too recently for the human genome (genetic code) to adjust. Instead, it is still adapted to the environment of our ancestors, as 10,000 years is short on an evolutionary time scale. The authors explain that intakes of dairy products, cereals, refined sugars, vegetable oils, alcohol, fatty meat and salt have risen significantly, and highlight seven main nutritional characteristics of our diet which have altered drastically. Intake of foods with a high glycemic load (blood glucose raising potential) has increased, raising insulin levels and increasing rates of the metabolic syndrome. Fatty acid intake has altered, with an excess of saturated and trans fats, raising cholesterol and contributing to cardiovascular disease.

Macronutrient composition has also shifted, as evidence suggests we consume less protein and more carbohydrate than our ancestors. This could explain some of the increase in cardiovascular disease, say the authors. Micronutrient density has also changed, with a greater consumption of refined grains containing fewer vitamins and minerals, producing deficiencies and subsequent illness. The acid-base balance of the 'Western' diet has altered in favor of acid, which may have a damaging effect on the kidneys and other organs. A huge rise in sodium through salt intake has modified the sodium-potassium ratio, which is linked to a wide range of chronic illnesses, the authors believe. Finally, fiber content has dropped to the extent that it could play a role in many gastrointestinal and circulatory problems.

The authors conclude that our genome clashes with our modern lifestyle, and together with genetic and environmental elements, virtually all of the 'diseases of civilization' have multifactorial dietary causes.
Cordain, L et al. Origins and evolution of the Western diet: health implications for the 21st century. American Journal of Clinical Nutrition, Vol. 81, February 2005, pp. 341-354

Editor's comment: Several afibbers have found that adhering to a diet similar to that of our ancestors (the Paleo diet) is effective in eliminating or reducing the incidence of afib episodes.

MIRANO, ITALY. The outcome of a pulmonary vein ablation (PVA) is usually determined by the results of a Holter recording and standard electrocardiogram (ECG) taken 1 to 4 months after the ablation. If no AF is observed during these tests the procedure is deemed to have been a success.

Italian researchers now question the validity of this approach. In a study of 72 patients who had undergone a seemingly successful PVA the researchers found that 13.9% of them showed evidence of AF episodes on Holter recordings or ECGs taken 30 or 120 days post ablation. However, when patients used a trans-telephonic electrocardiographic monitor (TT monitor) for a 90-day period (one 30-second recording per day plus recordings when palpitations were felt) it was clear that 27.8% of participants had evidence of AF episodes. Of particular interest is the finding that 8 of the 20 patients showing AF episodes were completely asymptomatic, while 10 patients had at least one asymptomatic episode.

The researchers conclude that TT monitoring is superior to Holter recordings and periodic ECGs in evaluating AF relapses after ablation. In this particular case, the short-term success rate of the PVA was only 72% when using TT monitoring as compared to 86% when using ECGs and Holter recordings. They also point out that 50% of the patients with relapses where completely asymptomatic during at least one AF episode.
Senatore, G, et al. Role of transtelephonic electrocardiographic monitoring in detecting short-term arrhythmia recurrences after radiofrequency ablation in patients with atrial fibrillation. Journal of the American College of Cardiology, Vol. 45, March 15, 2005, pp. 873-76

PARIS, FRANCE. The presence of SEC (spontaneous echocardiographic contrast) in a transesophageal echocardiogram has been shown to predict the ischemic stroke risk in patients with AF. A team of French and Italian researchers has just completed a study aimed at determining the extent of SEC and other thromboembolic risk markers in patients with AF. Their study included 82 patients with lone atrial fibrillation (LAF) and 289 patients with AF and underlying heart disease. The definition of LAF used by the researchers differed somewhat from the conventional one in that it excluded AF with any of the following features – a history of stroke, coronary artery disease, congestive heart failure, valvular heart disease, cardiomyopathy, cardiomegaly, hypertension (controlled or uncontrolled), hyperthyroidism, diabetes, chronic obstructive pulmonary disease (emphysema and chronic bronchitis). The definition of LAF also excluded AF occurring only during trauma, surgery, or an acute medical illness. The definition used for persistent AF was AF that required electrical or pharmacologic cardioversion or lasted more than 7 days.

Paroxysmal AF was defined as episodes that self-terminated within 48 hours or lasted less than 7 days. Left atrial appendage (LAA) abnormalities were defined as a LAA area greater than 5 square centimeters, an emptying or filling velocity less than 25 cm/second, or the presence of a clot (thrombus) in the LAA or left atrium. All patients underwent conventional transthoracic echocardiography as well as transesophageal echocardiography (TEE) in which the ultrasound probe is placed in the esophagus rather than on the outside chest wall.

The researchers found that 29.3% of LAF patients showed signs of SEC as compared to 49.8% of non-LAF patients. They also observed that LAF patients over the age of 60 years were more likely to have SEC (39.5%) than were patients 60 years or younger (17.9%). Paroxysmal LAF patients were least likely to show SEC (5.9%), while 45.8% of persistent LAF patients had signs of SEC. It was also clear that paroxysmal LAF patients had a significantly smaller left atrial diameter (37.4 mm) than did persistent lone afibbers (40.9 mm). In general, the researchers found the lowest incidence of thromboembolic risk markers in paroxysmal lone afibbers below the age of 60 years. Older age, persistent afib, and the presence of cardiovascular disease or other risk factors markedly increased the incidence of SEC and LAA abnormalities.
Di Angelantonio, E, et al. Comparison of transesophageal echocardiographic identification of embolic risk markers in patients with lone versus non-lone atrial fibrillation. American Journal of Cardiology, Vol. 95, March 1, 2005, pp. 592- 96

MILAN, ITALY. Many patients with artificial mitral valves have or will eventually develop AF. Up until now pulmonary vein ablation (PVA) in these patients has been thought to be a somewhat dicey proposition due to the inherent danger of damaging the valve during the PVA and also because patients with artificial valves tend to have enlarged and scarred atria.

Dr. Carlo Pappone and his colleagues at San Raffaele Hospital now report that PVAs can be successfully performed in patients with mitral valve prostheses (MVP). Their study involved 26 MVP patients and 52 controls matched for age, gender, left atrial diameter, type of afib (paroxysmal or chronic), history of hypertension, and left ventricular ejection fraction. Thirteen of the MVP patients had paroxysmal AF and 13 had the permanent (chronic) variety. A transesophageal echocardiogram (TEE) was performed prior to the procedure to ensure the absence of thrombus in the left atrial appendage.

The patients and controls all underwent circumferential pulmonary vein ablation with 81% of patients in both groups receiving additional linear lesions between the lateral mitral valve annulus and the left inferior pulmonary vein as well as posterior lines joining the contralateral superior and inferior veins. The procedure was performed in whatever rhythm (sinus or fibrillation) the patient was in at the start of the procedure.

At the 12-month follow-up 73% of the MVP patients were in sinus rhythm as compared to 75% among controls. Fluoroscopy times were, however, longer for MVP patients than for controls – 35 minutes versus 21 minutes, and 3 complications (one TIA) were observed in the MVP patients versus none in the controls. MVP patients were also more likely to develop atrial tachycardia after the procedure (6 MVP patients versus 1 control) and 3 MVP patients needed a repeat procedure to correct this problem. The Italian EPs conclude that the presence of a MVP does not affect the success rate of a circumferential pulmonary vein ablation, but is associated with longer fluoroscopy time and a greater risk of complications and the subsequent development of atrial tachycardia.
Lang, CC, et al. Transcatheter radiofrequency ablation of atrial fibrillation in patients with mitral valve prostheses and enlarged atria. Journal of the American College of Cardiology, Vol. 45, March 15, 2005, pp. 868-72

COPENHAGEN, DENMARK. The risk of developing atrial fibrillation is increased in patients with uncontrolled high blood pressure (hypertension). A team of Scandinavian and US researchers has just released the results of a study aimed at determining whether blood pressure reduction treatment with angiotensin II receptor blockers (ARBs) is superior to treatment with beta-blockers in preventing AF.

Their study involved 8851 patients with hypertension but no AF at baseline. The patients were randomized to receive losartan (an angiotensin II receptor blocker) or atenolol (a beta-blocker) plus a diuretic, if necessary, as required to reach a target blood pressure below 140/90 mm Hg. During an average 4.8 years of follow-up, 150 patients (0.7%/year) in the losartan group developed AF as compared to 221 patients (1.1%/year) in the atenolol group; a relative risk reduction of 33% in the losartan group as compared to the atenolol group.

The incidence of stroke in the group that developed AF was 3 times higher than in the group that did not (3.2%/year versus 1.0%/year) and the incidence of heart failure was 5 times higher among AF patients than among patients who remained in sinus rhythm (2.7%/year versus 0.5%/year). The most important risk factor for the development of AF was increasing age, followed by male gender (56% increase in risk compared to women), and a high systolic blood pressure.

AF patients treated with losartan had half the incidence of stroke as compared to atenolol-treated patients, but there was no difference in overall cardiac mortality. However, atenolol-treated AF patients had fewer hospitalizations for heart failure and a lower incidence of sudden cardiac deaths. The researchers conclude that losartan (Cozaar) is superior to atenolol (Tenormin) in preventing the development of AF in patients with hypertension and that losartan is also superior when it comes to stroke prevention among hypertensive AF patients. NOTE: This study was funded by Merck and Co., the manufacturer of losartan.
Wachtell, K, et al. Angiotensin II receptor blockade reduces new-onset atrial fibrillation and subsequent stroke compared to atenolol. Journal of the American College of Cardiology, Vol. 45, March 1, 2005, pp. 712-19
Bourassa, MG. Angiotensin II inhibition and prevention of atrial fibrillation and stroke. Journal of the American College of Cardiology, Vol. 45, March 1, 2005, pp.720-21 (editorial)

BALTIMORE, MARYLAND. Cardiologists have long debated whether it is best for persistent afibbers to stay in fibrillation and just take beta or calcium channel blockers to keep their heart rate down (rate control) or whether it is best to attempt to convert them to sinus rhythm and maintain them in sinus rhythm through the use of antiarrhythmic drugs (rhythm control). The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) trial concluded that rate control is at least as effective as rhythm control. The trial involved 4060 AF patients with a mean age of 70 years. Seventy-one per cent had a history of hypertension, 38% had coronary artery disease, and 26% had impaired left ventricular function. Only 12% had lone AF. Half the patients were randomized to rate control plus anticoagulation while the other half were randomized to rhythm control plus anticoagulation. After 5 years of follow-up 21.3% of the patients in the rate control group had died as compared to 23.8% in the rhythm control group[26].

To quote the authors of the study, "the results probably cannot be generalized to younger patients without risk factors for stroke (i.e. patients with primary, or "lone" atrial fibrillation), particularly those with paroxysmal atrial fibrillation."

Quite apart from the fact that the study is not particularly applicable to lone afibbers especially paroxysmal ones, I believe it had several serious flaws:

• The most "popular" drug used in the trial was digoxin. Over 70% of the people in the rate control group had used this drug at one time or another. Digoxin had been used by 54% of the participants in the rhythm control group as well. So as far as digoxin use is concerned, there was little difference between the two groups.
• Beta-blockers were used liberally in both groups as well – 68% in the rate control group and 50% in the rhythm control group.
• The main antiarrhythmic used was amiodarone (Cordarone). This drug was used by 63% of the patients in the rhythm control group and by 10% in the rate control group.
• The second most popular "antiarrhythmic" used in the rhythm control group was sotalol (Betapace) – this drug was used by 41% of patients despite the fact that it is well known that it does little, if anything, to maintain sinus rhythm, although it may help control the heart rate during an afib episode.
• Propafenone, flecainide and disopyramide had been used by only 4-15% of patients in the rhythm control group. It is impossible to say whether any of these drugs were beneficial or detrimental because of the way the data is reported.

The AFFIRM trial investigators have now compiled and reported the results of a sub-study aimed at determining whether quality of life (QoL) is significantly different between rate control and rhythm control patients. After a 4- year follow-up they conclude that QoL scores were similar in the rate- and rhythm-control groups at baseline as well as after 2 months, 12 months, and annually. They also conclude that QoL scores were similar irrespective of whether the participants were in sinus rhythm or in afib at the time the QoL scores were determined. They also suggest that attempts to improve QoL by restoring sinus rhythm will usually be unsuccessful.

Dr. David Newman from the University of Toronto, in an accompanying editorial, points out that participants in the AFFIRM trial were old and already suffering from other heart-related conditions; they were also significantly less symptomatic than the average afib population. Both these factors could well have affected their QoL scores. He also points out that a similar, but somewhat smaller trial (RACE) concluded that sinus rhythm at follow-up was strongly correlated with an improved QoL score.
Quality of life in atrial fibrillation: the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study. American Heart Journal, Vol. 149, January 2005, pp. 112-20
Newman, D. Atrial fibrillation and quality of life: clarity or evidence-based confusion? American Heart Journal, Vol. 149, January 2005, pp. 4-6 (editorial)

Editor's comment: It would be unfortunate indeed if these latest conclusions by the AFFIRM investigators concerning quality of life for afibbers were to be accepted unquestionably by EPs and cardiologists. For most of us being in sinus rhythm is associated with an immeasurably better quality of life than being in afib and again most of us will go to considerable lengths (like having an ablation or maze procedure) to achieve NSR. It is probably safe to assume that none of the AFFIRM investigators were suffering from symptomatic afib themselves or were living with someone who did.

American College of Cardiology Conference

Pretreatment improves conversion rates
Afibbers with the persistent variety require electrical cardioversion in order to regain normal sinus rhythm (NSR). Greek researchers now report that pretreatment (for 4 weeks) with amiodarone (Cordarone) or the beta-blocker carvedilol (Coreg) prior to cardioversion improves the acute success rate and lengthens the time in NSR after the conversion. Their randomized study of 145 patients with persistent AF found that the conversion rate among those pretreated with amiodarone was 93.3% as compared to 91.5% for carvedilol-treated patients and 73% for those given a placebo. The percentage of patients whose afib recurred within the first month after cardioversion was 16.6% for amiodarone, 27.9% for carvedilol, and 39.4% for placebo. The researchers noted that patients on amiodarone experienced significantly fewer ectopic beats after conversion than did those in the other two groups.
Abstract 1021-253, p. 92A

Men more susceptible to AF
Researchers at the Mayo Clinic report that men tend to develop lone AF at an earlier age (average age of 45.2 years) than women (average age of 52.4 years). They also confirm at least 15% of lone afibbers may have a genetic (familial) disposition toward the disorder.
Abstract 1050-268, p. 97A

Left atrium changes after ablation
While radiofrequency (RF) ablation can be highly effective in eliminating AF, it is not clear what effect the creation of extensive lesions on the atrium wall has on the function and performance of the left atrium. American researchers have used contrast-enhanced magnetic resonance angiography and cine magnetic resonance of the left atrium to measure changes occurring in 11 patients who underwent RF ablation. They found that the left atrium volume had decreased by an average 8% 3 months after the ablation. This positive remodeling change, however, was offset by a 17% reduction in left atrial ejection fraction (from an average 52% to 38%). The researchers conclude that left atrial function may worsen after an extensive ablation.
Abstract 1164-248, p. 123A

Antiarrhythmics improve ablation success
While success rates for pulmonary vein isolation (PVI) now approach 80% or better for paroxysmal afibbers, the success among persistent afibbers is still relatively poor. Researchers at St. Luke's-Roosevelt Hospital Center in New York now report that electrical cardioversion and pre- and post-treatment with antiarrhythmics (dofetilide, amiodarone or sotalol) for one month before and after the PVI can markedly improve long-term success rate of PVI among persistent afibbers. They found that 81% of 43 persistent afibbers were still in sinus rhythm almost 2 years after their ablation compared to 88% among 142 paroxysmal afibbers.
Abstract 1164-250, p. 124A

Cost effectiveness of ablation
There are basically three management approaches to atrial fibrillation:

• Left atrial catheter ablation (LACA)
• Rhythm control
• Rate control with thrombotic therapy.

The following is an account of my ablation for atrial fibrillation carried out at the Hospital of Haut-Leveque, Bordeaux, France, in the week of March 14th-18th 2005.

I was admitted to the hospital at 8:30 am on March 14th. The cardiology building is a rather ugly concrete edifice of about 7 or 8 floors situated on a sort of campus of other unprepossessing hospital buildings in Pessac, a suburb of Bordeaux. I was shown to a very clean but spartan room with two beds and a bathroom and TV where my wife and I were to spend the next four nights.

After changing into hospital garb, usual affair, I was wheeled to a lower floor to have an echocardiogram and a TEE (trans oesophagal examination). The result was good in that I didn't have a clot. The TEE, which others have described as 'uncomfortable', was for me the worst experience of the week.

At about 12:30 I was again taken downstairs to the operating theatre and placed on a table surrounded by a mass of monitors, etc. I was prepped for surgery (having already been shaved); getting me ready took about half an hour. I was to be sedated but to remain conscious during some of the operation. The drugs were commenced and the sedation was such that I did not really notice the catheters being introduced into the main vein in my right groin. They start by inducing fibrillation. The procedure after that took about four hours. I must have dozed quite a bit, occasionally felt a little pain (which was quite bearable) at which point they would ask if they could continue a bit more. After a while I became quite interested in the screens, one of which showed a scene rather like a fairground where you look in a box and work levers trying to lift something out. There were three little loops being dangled about. Another screen showed what I assumed was my pulse beating at around 90-130. I then heard an exclamation, 'finally!' and saw my pulse rate drop to 65-66 and hold and realised that fibrillation had stopped.

After that I became quite confused, as my pulse rate alternated quickly between 50, 100, 130 and so on. I was then told that the procedure was over and my main doctor who carried out the procedure, Dr Pierre Jais, told me that it had gone well. Later Dr Jais who possibly does the procedure at least twice daily or maybe 400 to 500 times a year, told me my case was 'very difficult'.

A nurse put a bag of sand on my groin wound and held it in place by wrapping a cloth around my leg several times. I was then wheeled upstairs at about 6:30 pm. I was still on a drip. I had had nothing to eat since 8:00 pm the night before and with the relief that the procedure was over felt quite hungry. However I was not allowed anything until 11:30 pm and then was only allowed a yogurt and a fruit compote. I felt quite beaten up, but it was tempered with pleasure that the worst was over.

At about 8:30 pm Dr Jais came to see me in my room and explained how the procedure had been much more complex than he had anticipated. Having initiated fibrillation, he isolated the pulmonary veins. In many cases at this point fibrillation ends spontaneously but not in my case. He then made two linear lesions in the left atrium. Fibrillation still continued and he then had to tackle other areas of activity. 'Finally' was when the fibrillation stopped. At this point he attempted to induce fibrillation – that was the explanation for me seeing my pulse jump about – but could not do so. He found this inability to induce fibrillation to be very positive. At this stage his optimism about the outcome for me seemed quite guarded. [He did say that the most likely time for fibrillation to restart was in the first 24 hours after the operation].

That night there were many nocturnal visits by pleasant nurses taking blood and re-adjusting my drips. I was on two drips, one of heparin, an anti-coagulant, and the other the usual glucose drip. I was also rigged up with a monitor for my heart-beat which transmitted to the office at the other end of the corridor.

The nurses were most concerned about my lack of desire to pee, as I had absorbed four litres of fluid during the procedure. The following morning they gave me a diuretic and admonished me to pee as much as I could. The quantity of my urine was then monitored for a couple of days.

That same morning I had another echocardiogram and was also taken to have a chest x-ray. Thereafter it was a day in bed, still feeling beat up and having to keep my right leg straight to promote healing of the groin entry which was still under pressure from the bag of sand. I did not have time to be lonely as I had continual visits from well-dressed and cheerful nurses to take blood, etc. Moving around or even sitting in a chair at this stage was not easy with my drips and bag of sand. These were both removed at 8.00 pm that night. Being off the drips meant that I was mobile at last but the heparin drip was replaced by injections of heparin (Calciperin). These took place every eight hours, including one at 2:00 am, followed by extraction of a blood sample at 6:00 am. This routine was to continue until Friday.

On Day 3 I was feeling much better. The main activity of the day was to go on an exercise bike and have what seemed like a full physical workout, pulse up to 135. After 15 minutes on the bike I had worked up a big sweat and was quite tired afterwards, but was able to have a bath and a shave. In the early afternoon Dr Jais came to visit again. He appeared much more confident of my eventual outcome and told me that all the tests had been quite satisfactory. He explained that one risk was that during the healing process one of the scars might heal itself. If that happened, I would have to come back for a 'touch-up'. He was surprised that I was not in more discomfort, as my heart was quite inflamed due to the amount of burning. He prescribed some anti-inflammatory drugs and recommended that I stayed in the hospital until the Friday as scheduled. He was off to Vienna for a conference for a few days.

Day 4, and after a good sleep – one gets used to the nightly visits – I was feeling fully recovered. No activities were planned. I was going to learn how to inject myself with the anti-coagulant so that I could continue the injections on the journey home, [but decided not to. My wife was also not willing to learn how to do the injections].

In general, though the first sight of the hospital was a bit depressing, I was astonished at the level of care I received. The staff, doctors, nurses, and nurses' aides were all immaculately turned out and very nice to me. (Doctors in white coats, nurses in blue and white striped coats, and assistants and orderlies in green and white stripes). The nurses work in a team, and there was no feeling that they were understaffed. The wards have two beds, and one can have one's spouse occupy the second bed at a little extra charge. Everywhere was spotlessly clean; with the linoleum floors washed and polished each day until they glistened.

The food is plentiful except at breakfast. Both lunch and dinner are three or four courses, once you are well enough to eat them all.

The advantages of Bordeaux for AF ablation are:

1. It is certainly in the top one or two institutions in the world for treatment of afib.
2. The unit has the best of equipment; they perform as a large team; and do one to three ablations each weekday.
3. The institution provides wonderful aftercare. The ward has a large nursing team which specialises in ablation aftercare.
4. If necessary, they proceed beyond the initial stage of PV ablation and attempt to eliminate all tissue which is conductive of AF. This is certainly what happened in my case. Had I gone to a hospital in Canada or somewhere else where the doctors are less experienced, the procedure would likely have ended after PV isolation. I would almost certainly have continued with afib in this case. Now, according to the hospital's statistics, I have an 80 per cent chance of being afib free.

The disadvantages are:

1. The cost is not covered by private insurance. It costs 7.500 euros for those covered by EU medical cover or 10,600 if paying personally. There are no extra charges and all medications and tests are included in the hospital fee of about 650 euros a day.
2. We found ourselves speaking French to everyone except Dr Jais. Some of the nurses may speak some English, but they were not letting on to us. This was not a problem for us, but might be of others.

POSTSCRIPT

I am still afib-free after 6 weeks and am off all meds and hoping for the best. It took me a month to get my strength back, but am now back to all my usual golf and hiking, etc.