Wishing you good health and lots of NSR,
Hans

ABSTRACTS

BERGEN, NORWAY. Researchers at the Haukeland University Hospital have carried out a study to determine the afib recurrence rate after a PVI and the change in quality of life (QOL) associated with the procedure. Their study involved 59 men and 13 women with a mean age of 52 years who had been unsuccessfully treated with two or more antiarrhythmic drugs. Most (89%) had lone atrial fibrillation and 49% of all participants had an enlarged left atrium.

The patients underwent a PVI using a 10-polar Lasso catheter for mapping and a Celsius deflectable 7F quadripolar catheter for recording and ablating. An average of 3.1 veins were isolated per patient. A single procedure was performed on 60 of the patients, two procedures on 10 patients, and three procedures on 2 patients. The second procedure (touch-up) was done within 48 hours of the first procedure in 5 patients. One patient experienced cardiac tamponade, one had a stroke, and one later developed stenosis.

During an average follow-up of 10 months 27 patients experienced more than one afib episode. Quality of life was judged very good in 26 patients who had not experienced a recurrence; it was judged to be better in 30 patients (15 with recurrences), unchanged in 11 patients of which 10 had recurrences, and worse in 2 patients who both had recurrences.

The researchers noted that the risk of recurrence increased sharply with age and the number of years the afib had been present. As a matter of fact, they suggest that patients who have suffered from AF for more than 10 years may need a more extensive procedure than the standard PVI to resolve their problem. Overall, 60% of all patients involved in the study were free of afib during the 10 months follow-up and 72% reported that they were satisfied with the procedure.
Chen, Jian, et al. A clinical study of patients with and without recurrence of paroxysmal atrial fibrillation after pulmonary vein isolation. PACE, Vol. 28, suppl 1, January 2005, pp. S86-S89

GENEVA, SWITZERLAND. There is increasing evidence that essential hypertension (high blood pressure) is an important risk factor in the development of atrial fibrillation. What is less clear is whether treating the hypertension lessens the risk of afib. Swiss researchers now report that appropriate treatment does indeed reduce the risk and that the risk reduction is independent of the type of blood pressure reducing agent used. The study involved a group of 597 patients who, after having been diagnosed with hypertension (systolic blood pressure equal to or greater than 140 mm Hg and/or diastolic pressure equal to or greater than 90 mm Hg), were placed on antihypertensive therapy using ACE inhibitors (46%), angiotensin II receptor blockers (23%), calcium channel blockers (52%), and beta-blockers (21%) either alone or in combination.

After a 7-year follow-up the researchers found that the risk of developing atrial fibrillation decreased by 24% with a 12 mm Hg drop in systolic pressure after adjusting for age, gender, body mass, and pulse pressure (systolic blood pressure minus diastolic pressure). All blood pressure measurements were averages of 24-hour ambulatory measurements. Those of the patients who did develop atrial fibrillation were slightly older than those who did not, were more likely to be men, and to be overweight or obese. A decrease in pulse pressure also correlated with a decrease in AF risk, but this trend was not statistically significant. There was no indication that one class of blood pressure medications was superior in preventing the development of AF. The researchers conclude that an increased systolic pressure and pulse pressure may promote the onset of atrial fibrillation by modification of left ventricular diastolic function.
Ciaroni, S, et al. Prognostic value of 24-hour ambulatory blood pressure measurement for the onset of atrial fibrillation in treated patients with essential hypertension. American Journal of Cardiology, Vol. 94, December 15, 2004, pp. 1566- 69

NEW YORK, NY. Researchers at the Columbia University College of Physicians and Surgeons report that an enlarged pulmonary vein diameter is associated with an increased risk of atrial fibrillation. Their study involved 100 AF patients (42% with lone AF) who were scheduled to undergo a PVI and 24 age- and sex-matched controls with no AF. All participants underwent spiral-computed tomography prior to the PVI procedure. The researchers drew the following conclusions from their data:

• The average ostial pulmonary vein (PV) diameter was significantly larger in AF patients than in controls (150 mm versus 120 mm) and all four veins were similarly affected.
• An enlarged left atrium was associated with increased PV diameters.
• PV diameters in patients and controls with hypertension were larger than in patients and controls without hypertension.
• Patients with persistent AF had larger PV diameters than did patients with paroxysmal (intermittent) AF.

The researchers speculate that impaired left ventricular diastolic function is associated with a stretch-induced PV arrhythmia mechanism. They also suggest that the finding that persistent afibbers have larger vein diameters than paroxysmal afibbers would tend to support a progressive nature of PV enlargement and associated afib severity. They do acknowledge though that AF is a multifactorial disease and not solely associated with enlarged veins.

Dr. John McAnulty of the Legacy Heart Institute in Portland, Oregon speculates that the measurement of PV diameters may ultimately prove useful in deciding whether a specific AF patient is more likely to benefit from medication, an ablation or implantation of an ICD.
Herweg, B, et al. Hypertension and hypertensive heart disease are associated with increased ostial pulmonary vein diameter. Journal of Cardiovascular Electrophysiology, Vol. 16, January 2005, pp. 2-5
McAnulty, J. Probing for mechanism of atrial fibrillation: pulmonary vein ostia. Journal of Cardiovascular Electrophysiology, Vol. 16, January 2005, p. 6

TAIWAN, REPUBLIC OF CHINA. Researchers at the Columbia University Medical College of Physicians and Surgeons recently concluded that patients with atrial fibrillation tend to have significantly larger ostial pulmonary vein (PV) diameters than do people without AF. They also noted that larger PV diameters were associated with enlarged left atria.

Taiwanese researchers now report that successful pulmonary vein isolation (PVI) gradually remodels the heart toward normal by decreasing the diameter and degree of eccentricity of ablated veins while, at the same time, reducing the size of the left atrium.

Forty-five afibbers participated in the study which involved magnetic resonance angiographic (MRA) imaging one month before and 12 months after the PVI. Thirty-five (78%) of the patients experienced no AF recurrence, while 10 experienced afib episodes more than one month after ablation. The researchers found that the diameters of the superior PVs in successfully ablated patients had decreased significantly one year after the PVI and also showed less eccentricity. They also noted a slight reduction (8%) in left atrium volume.

Among patients whose PVI had been unsuccessful the story was quite different. Twelve months after the PVI the right superior PV had increased in diameter by about 21% and the volume of the left atrium had increased by about 29%. The researchers conclude that a successful PVI is associated with desirable structural remodeling of the superior PVs and the left atrium. An unsuccessful PVI resulting in late recurrence of afib is, however, associated with continued left atrium enlargement.
Tsao, HM, et al. Morphologic remodeling of pulmonary veins and left atrium after catheter ablation of atrial fibrillation. Journal of Cardiovascular Electrophysiology, Vol. 16, January 2005, pp. 7-12

HERAKLION, GREECE. Greek researchers have carried out a long-term study to determine the relative effectiveness of sotalol (Betapace) and propafenone (Rythmol) in maintaining normal sinus rhythm in patients with recurrent atrial fibrillation. Their study included 245 afibbers who were randomly assigned to receive 300 mg/day of sotalol (in two divided doses), 450 mg/day of propafenone (in 3 divided doses), or placebo. After 18 months 81% of the patients on sotalol had relapsed into AF or experienced severe side effects as compared to 52% in the propafenone group after 26 months. In the placebo group 88% had relapsed into afib after 11 months. After a follow-up of 30 months the number of study participants in sinus rhythm was 47% in the propafenone group, 25% in the sotalol group, and 17% in the placebo group. The researchers conclude that propafenone is superior to sotalol for maintenance of sinus rhythm over the long term. Sotalol, on the other hand, is only slightly more effective than placebo.
Kochiadakis, GE, et al. Sotalol versus propafenone for long-term maintenance of normal sinus rhythm in patients with recurrent symptomatic atrial fibrillation. American Journal of Cardiology, Vol. 94, December 15, 2004, pp. 1563-66

BOSTON, MASSACHUSETTS. Brain natriuretic peptide (BNP), a cousin to atrial natriuretic peptide (ANP), is a hormone released from the walls of the ventricles during unusually strenuous activity. It is stored as a prohormone within secretory granules in the ventricles and is secreted as a N-terminal fragment, N-terminal pro- brain natriuretic peptide (nt-pro-BNP), and the smaller active hormone BNP. BNP has effects similar to those of ANP, that is, it decreases sodium reabsorption rate, renin release, and aldosterone release; it also increases vagal (parasympathetic) tone and decreases adrenergic (sympathetic) tone. Because nt-pro-BNP is easier to measure than BNP it is often used as a marker for BNP.

It is well established that BNP and nt-pro-BNP levels are elevated in heart failure and that the degree of elevation is directly proportional to the seriousness of the failure. Researchers at the Massachusetts General Hospital now report that lone afibbers also have elevated nt-pro-BNP values even when in sinus rhythm. Their study involved 150 participants with lone atrial fibrillation (LAF) and 75 afib-free controls matched according to age, gender, race, and ethnicity. The majority of participants (81%) were men, the average age at enrolment was 54 years, and the average age at first diagnosis was 45 years. The demographics of the study group thus closely mirrors that of the much larger groups involved in our own LAF surveys and, once again, puts "paid" to the still widely held notion that afib is a disease of old age, which it clearly is not. At the time of enrolment 130 afibbers had the paroxysmal variety, while 20 were in permanent AF.

Blood samples were obtained from all participants at enrolment. The researchers found that the median level of nt-pro-BNP was significantly higher among lone afibbers (even when in sinus rhythm) than among controls (166 versus 133 fmol/mL); they also observed that nt-pro-BNP levels were higher in afibbers with permanent LAF than in those with paroxysmal LAF (189 versus 157 fmol/mL), and that afibbers with high nt-pro-BNP levels at study entry were more likely to progress to the permanent version than were those with lower levels (197 versus 163 fmol/mL). There were no significant differences in ANP levels between afibbers and healthy controls, but levels in afibbers who later developed hypertension were significantly higher than in those who did not (3764 versus 1622 fmol/mL). The researchers speculate that BNP may be involved in sustaining fibrillatory rotors through its potentiating effect on vagal nerve impulses transmitted from the brain.
Ellinor, PT, et al. Discordant atrial natriuretic peptide and brain natriuretic peptide levels in lone atrial fibrillation. Journal of the American College of Cardiology, Vol. 45, January 4, 2005, pp. 82-86

Editor's comment: It is known that medication with ACE inhibitors produces an almost immediate, dose-dependent reduction in natriuretic peptide levels. Could this be another possible pathway by which ACE inhibitors may help prevent afib and, in particular, prevent it from becoming permanent?

DES MOINES, IOWA. Cryoablation, that is, ablation using a catheter cooled with liquid nitrogen rather than one heated with radiofrequency energy, is fairly well established in certain centers in Europe, but has had a slow start in the US. Despite the fact that, CryoCor Inc., the developer of the nitrogen-cooled catheter, is located in San Diego. Seven of the initially successfully ablated patients experienced serious afib breakthroughs within 3 months of the procedure and underwent a touch-up using regular RF ablation. The average incidence of afib episodes over the first 3 months for the remaining 22 patients who only had the cryoablation was 4.9/month, but this declined to 2.8/month in the 3- to 6-month period after ablation. At the end of 6 months 18 patients experienced no afib episodes at all. This corresponds to an 82% success rate among those afibbers who did not need a RF touch- up and a 62% success rate when considering the whole group. There were no indications of stenosis in CT scans performed 3 and 6 months after the initial cryoablation. However, there were a rather astonishing number of side effects reported. Among them – respiratory depression during conscious sedation, transient leg weakness, transient asymptomatic sinus pauses, femoral pseudoaneurysm, atrial defibrillator implantation, and bacterial endocarditis.

Among the 7 patients who underwent a touch-up ablation with RF, one TIA, one femoral pseudoaneurysm, and one femoral hemorrhage was observed.

The participating electrophysiologists conclude with this somewhat surprising statement, "For paroxysmal AF, the clinical efficacy of cryoablation was comparable to RF segmental PV isolation, with repeat procedures required for AF recurrence in up to nearly 50% of patients. These results underscore the limitations of the segmental PV isolation approach, regardless of the energy source." NOTE: This study was funded by CryoCor Inc.
Hoyt, RH, et al. Transvenous catheter cryoablation for treatment of atrial fibrillation. PACE, Vol. 28, suppl 1, January 2005, pp. S78-S82

Editor's comment: Only one of the participating institutions made it to the "A List" in my recent tabulation of top-notch ablation centers (see Lone Atrial Fibrillation: Towards A Cure – Vol. II). This certainly shows in the results. A 62% initial success rate is well below expectations and the radiation exposure from fluoroscopy would also seem to be unacceptably high at an average of 131 minutes. The average fluoroscopy time at the Bordeaux Clinic in France is only 57 minutes and that at the Cleveland Clinic somewhere between 51 and 110 minutes. I would conclude that cryoablation, in the US at least, is still in its infancy and unless you fancy being a guinea pig you would be better off waiting or going for a RF ablation at one of the institutions on my "A List".

My first brush with afib was in late '99 while working a grim, low-wage cashier job involving a lot of overtime and too few days off to recuperate. I was 57 years old, seriously overweight, had that year gone through a lot of stressful life changes, was eating poorly [whatever I could pick up in the convenience store I worked in], it was hot weather and I had no air conditioning, and I was surviving on coffee. I had a couple of short episodes that went away before I could get to a doctor, and of course when I did get to the doctor he found nothing wrong. Then I had one that did not go away, and ended up in the hospital for 3 days. I changed jobs after that, and worked more normal hours, dropped the coffee and ate better [more vegetables, less junk], and had no more afib until August 2000, when I was hospitalized again with another "just-won't-go-away" episode. This again was associated with caffeine [green tea this time, dozens of cups of it, trying to stay awake at work] and hot weather, compounded by lack of sleep. After that I dropped caffeine altogether, and got an air conditioner.

For the next several years I had short episodes occasionally, but they always went away by themselves, and in any case, I was getting turned off by hospital emergency rooms. I had learned a little about using computers by that time, and was researching better nutrition. I retired and moved back to Maine, and eventually got my own computer, and found Hans' site. Here I found there were a lot of people taking various drugs, and none of these drugs seemed to be curing their afib. They were still getting afib attacks, trading drug advice, going on different drugs, and still getting afib. Some of them were talking about, and some even resorting to, heart surgery. I couldn't blame them for doing this, because their afib had started small and gradually increased until it ruled their lives. I was afraid mine would do that too.

Worse yet, by no means all of those ablation patients had gotten rid of their afib either. Two of them had had near-death experiences, and I was pretty sure that the reason there were not more stories like that was because most ablations that went bad had resulted in death, and of course, we are not very likely to hear from those people. And then there were 2 people posting who claimed to have gotten rid of their afib by diet and supplements. These were Fran and Erling.

Well, I thought, if these 2 people so different from one another can do that, maybe I can too. Food choices are something I can control. So I changed to a mostly paleo diet, and sent away for some Carlson's magnesium glycinate. At first I still did get some short, mild afib episodes, but then I began seeing posts about low sodium V8 juice, 850 mg potassium per 8 oz. glass. I was having trouble consuming enough vegetables and fruits to get in 3-5 grams K a day, and this seemed like just what I needed, and sure enough it was. I haven't had any more afib from that day forth, and that was December 2003.

Also, I need to mention that those last episodes, mild though they were, appeared right after use of a seasoning containing MSG. I had never had an afib episode that I could tie to MSG before, but then I had never been without it for any period of time before either. For all I really know, they could have all had to do with MSG, in combination with stress, hypoglycemia, dehydration, electrolyte deficiency, caffeine, and any of the other myriad stressors of modern life.

Any paleo diet purist will point out that I ingest a lot of stuff that isn't paleo. The V8 certainly isn't, and neither are the supplements I take. I do eat a little cheese, too, though not the plasticized processed cheese. I cannot afford organic food, so I make do with what I can find in the local supermarket, cheapest first. I go out to eat sometimes, and on those occasions I commit excesses like baked potato and gravy, or bread on sandwiches. I cheat outrageously sometimes, too, particularly with chocolate baked goods.

Speaking of bread, gravy, and bakery goodies, if I hadn't gone to paleo I would also never have realized that I have a bad reaction to wheat. Since taking up the paleo diet my antacid consumption has gone way down, except when I eat anything with wheat in it. That will have me eating antacids for a good 12 hours and sometimes more.

Another good thing about the paleo diet is that I fit the classic profile for insulin resistance - fat, high blood pressure, relatively inactive, cholesterol a bit on the high side - and the paleo diet is good for insulin resistance. I hope to avoid type 2 diabetes this way, or at least to slow it down.

For those concerned about whether my afib is "really cured", I do not think I can expect to be cured of needing proper nutrition, any more than cars are cured of needing gasoline. I don't think I am going to ever again be just like I was in my 20's either. To use the same metaphor, old cars are never again just like they were when new.

I think afib is one of the long latency deficiency diseases, and that is why, in most people, it does not appear until a relatively 'older' age, and why it appears in the context of stress so often. I am still old, fat, and lame in the knees, but I don't have afib any more. If I can do this, you can too.

The late great French cardiologist Philippe Coumel, a giant in the field of arrhythmias who also wrote the foreword for Lone Atrial Fibrillation: Towards A Cure by Hans Larsen, was the first to describe vagally mediated atrial fibrillation (VMAF) about 20 years ago. His seminal patient went into AF upon reclining but could terminate the episode upon standing. It is difficult to believe that such a dramatic disease could go unnoticed prior, unless, of course, it didn't exist prior. This suggests that environment plays a significant role in its genesis. To quote from Hans' new book Lone Atrial Fibrillation: Towards A Cure - Volume 2, Coumel "developed the concept of the triangle of arrhythmogenesis". "There are always three main ingredients required for the production of a clinical arrhythmia, the arrhythmogenic substrate, the trigger factor and the modulation factors of which the most common is the autonomic nervous system." The role of the ANS in the genesis of AF is obvious to all. Identification of the arrhythmogenic substrate (the tissue in which the arrhythmia starts and propagates) and the trigger factor have been more problematic.

Arrhythmogenic Substrate (P cells)

Many LAFers are consumed by the search for their AF trigger, some specific item present or absent from their diet. Although this search will undoubtedly improve their general health, for the vast majority frustration will be a constant companion. And this is because of the arrhythmogenic substrate that all LAFers share.

P is for pole cells and they are the pacemaker cells of the heart. These have traditionally been described only in nodal tissue (SA node and AV node). However, in August of 2003 the Cleveland Clinic group was the first to describe P cells in human pulmonary veins (PVs) near their entry into the left atrium. They were found at autopsy in 4/4 AF patients and in 0/6 controls (one without history of tachyarrhythmia and five heart transplant donors). To date they have not been described anywhere else in the heart outside of nodal tissue.

Pacemaker cells are unique in that they slowly depolarize by themselves, hence their greater inherent automaticity. This is due to their unique Na (sodium) and K (potassium) ion channels. They are regulated by the opposing influence of sympathetic (adrenergic) and parasympathetic (vagal) stimulation. Both of these, but especially the vagus, cause shortening of the effective refractory period (ERP). The refractory period is the rest period following a contraction of the heart muscle. Individual heart cells do not respond to stimulation during this period so ectopic beats or afib cannot be initiated during the ERP. A shortened ERP thus increases the risk of ectopics and afib.

In September of 1998 the Bordeaux group published an electrophysiologic (EP) study of 45 patients with frequent paroxysmal AF. They looked at the location of the triggering atrial ectopic beats and found that 94% were located in the PVs (2-4 cm inside the veins). I'm not sure what per cent of those with paroxysmal AF qualify as lone (LAF), but I would guess that if one accepts any degree of hypertension in the definition of LAF, it would be a clear majority.

In October of 2002 this same group published another EP study of 28 patients with paroxysmal AF (average age about 50) v. 20 age-matched controls. They evaluated the effective refractory period (ERP) of cells in the pulmonary veins and the atrial ERP (AERP) in both groups. First they found that AERP did not differ between the two groups. However, in the AF group the PV ERP was SHORTER than the AERP, whereas in the control group it was LONGER. Furthermore, they found that there was greater dispersion of PV conduction velocity in the AF patients, and that in these same patients AF was more easily induced (by fast pacing) in the PVs (22/90) than in the left atrium (1/81). Consequently, this PV refractoriness and dispersion of conduction properties create a substrate favorable for reentry. It would appear to me that these PVs represent the oft-quoted but elusive "defective substrate" or "arrhythmogenic substrate" of LAF. This is what is responsible for "the loss of physiologic rate adaptation" present in AF, e.g., conditions that should slow down the heart (slower conduction velocity, shortened refractory period) often result in a tachyarrhythmia instead.

Putting these studies together paroxysmal AFers appear to have a problem with their PVs. It would seem logical to implicate P cells in this process. The increasingly successful reports of catheter ablation via pulmonary vein isolation (PVI) for LAF are certainly consistent with this interpretation. Whether these P cells have been damaged by free radicals or not remains to be determined, although delayed onset of LAF is suggestive of this. Free radical damage to PVs is further supported by the frequent occurrence of AF after surgery. The mechanism involves lipid peroxidation (damage to cell walls caused by ROS or reactive oxygen species, especially peroxynitrite). It is also called ischemic reperfusion injury. Please see pp. 137-138 of Hans' book Lone Atrial Fibrillation: Towards A Cure for an eloquent description of how this relates to AF and the PVs. Furthermore, the demonstrated abnormalities in refractory period, conduction velocity, and dispersion in the PVs of LAFers involves many sodium, potassium and calcium channels. This finding is much more consistent with indiscriminate damage to cell membranes due to ROS than that rendered by some specific genetically determined channelopathy. The $64 question: We all know that LAF is more frequently encountered in endurance athletes (five times more in one study). Is their primary problem an arrhythmogenic substrate created by such action of ROS or is enhanced vagal tone more contributory to their VMAF (or both)? Vitamin C and magnesium have both been shown to be instrumental in preventing damage due to ROS.

Trigger Factor (low potassium)

Information on a diurnal rhythm for potassium, if any, is hard to find in the medical literature. I initially assumed that because blood potassium is so intimately associated with aldosterone, any possible diurnal rhythm would follow that of aldosterone (and ACTH/cortisol). But such is not the case.

The diurnal rhythm of aldosterone secretion in healthy individuals parallels that of cortisol and is ACTH dependent. The lowest values are observed from midnight to 4AM. Values peak in the morning around 8AM after which there is a gradual decline throughout the day (assuming a normal sleep-wake cycle). http://www.emedicine.com/med/topic3193.htm

Furthermore, cortisol and ACTH are not always secreted uniformly throughout the day. Episodic spikes can occur when the body is stressed, e.g., fasting, anxiety, etc. Otherwise, in humans, urinary potassium excretion peaks in the early morning between 0530 and 0730 with a minimum at night from 2100 to 0530. Therefore, the diurnal rhythm of urinary potassium excretion seems to be controlled by the diurnal rhythm of cortisol and/or aldosterone. Not so for blood potassium.

Plasma potassium follows a diurnal rhythm with a peak at noon and a trough at midnight with an average peak- to-trough difference of 0.62 +/- 0.05 mmol/L http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&list_uids=1714003&dopt=Abstract

In other words blood potassium is lowest when aldosterone secretion is lowest and blood potassium climbs as blood aldosterone/cortisol are peaking. This seems somewhat contradictory. Is this excreted potassium coming from inside the cells? If aldosterone/cortisol are not driving this drop in blood potassium, what is? A partial answer may be insulin.

As an aside, blood magnesium peaks around 0330 and reaches its lowest point around 1530. Its diurnal variation is greater than that of blood potassium. We all know how critical magnesium is to the maintenance of intracellular potassium http://www.afibbers.org/conference/PCMagnesium. pdf

According to one study, the frequency of hypokalemia (potassium less than or equal to 3.5 mmol/L) is related to the time at which the blood potassium is measured. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=1714003&dopt=Abstract

Not many blood samples for evaluation of potassium are drawn in the evening, especially around midnight (its diurnal nadir). How many people with lower range blood potassium values on specimens drawn during a daytime visit to the doctor's office are frankly hypokalemic at midnight?

Blood potassium values decrease postprandially because of insulin released in response to an ingested carbohydrate load. Blood potassium after our largest meal of the day (dinner in America) slowly declines due to insulin.

Insulin induced cellular uptake of glucose affects blood potassium in the following manner: Similar to Na, entry of glucose into a cell brings water with it, thereby decreasing the concentration of intracellular potassium. The ATP (requires magnesium) sensitive Na+/K+ pump is then stimulated to increase intracellular potassium. Blood potassium consequently drops.

This insulin-induced uptake of glucose (and potassium) occurs primarily in fat, liver and muscle, including cardiac muscle. However, cardiac muscle (v. skeletal muscle) is relatively less dependent on glucose generated ATP and relatively more dependent on oxygen generated ATP. This latter process occurs in the mitochondria and is called cellular respiration. Heart muscle cells have the greatest concentration of mitochondria at about 5,000 per cell. By weight 40% of each heart cell is mitochondria. [This is one reason why CoenzymeQ10 (protects mitochondria from oxidative damage) deficiency associated with statin therapy is causing an epidemic of heart failure.] Also, the cell membrane of the heart is HIGHLY permeable to potassium ions (there are many passive potassium channels). Therefore, it would seem that beneficial glucose and potassium uptake by the heart is relatively less helpful and that the potassium concentration gradient for the heart becomes relatively more problematic (v. skeletal muscle, smooth muscle, liver and fat cells). For spontaneously depolarizing P cells (specialized nerve cells) gradient rules!

Furthermore, aerobic training is associated with enhanced insulin sensitivity in skeletal muscle but diminished insulin-stimulated glucose (and potassium) uptake in the heart. http://ajpendo.physiology.org/cgi/content/full/276/ 4/E706

Perhaps such conditioning enhances oxygen exchange so much so that glucose is relegated to an even less significant role as an energy substrate for the heart. Aging further diminishes this insulin sensitivity in the heart. So, especially in the physically fit and the elderly, low blood potassium is more likely to cause arrhythmia.

In one study comparing response to orthostatic challenge between those with high vagal tone and those with low vagal tone "a significant increase in plasma renin activity was found during LBNP (lower body negative pressure) in the HI responders only". http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=926361 9

In other words prolonged standing or the equivalent causes a greater release of aldosterone in those with high vagal tone. During this time, e.g., a round of golf, aldosterone secretion is increased. In VMAFers this is a mixed bag, aldosterone is vagolytic (lengthens ERP) but the potassium loss it induces shortens ERP. While relaxing afterward blood aldosterone drops and its protective effect is lost but not the damage it has done to the potassium gradient. Combine this with a concomitant drop in blood glucose and AF risk is maximized. Much of this was discussed in Session 33 of the Conference Room Proceedings.http://www.afibbers.org/conference/session33.pdf

Since a one mmole reduction in blood potassium generally translates to a deficit of about 300 mmoles of total body potassium, about 7 grams of potassium must move from the intracellular compartment to the extracellular compartment between its blood concentration trough and its peak.http://www.uhmc.sunysb.edu/inter nalmed/nephro/webpages/Part_D.htm

Part of the intracellular deficit so created is addressed by dietary potassium, but the RDA for potassium is only 3.5 gm in both America and Europe. And this dietary intake is offset by that lost in the urine.

This diurnal nadir of blood potassium certainly correlates well with the preponderance of nighttime episodes. Indeed the diurnal contribution of the PNS (vagal tone increases during the night, assuming a normal sleep wake cycle) seems to have camouflaged the diurnal contribution of low blood potassium to nighttime episode.

What to do about it?

Last year Hans brought to my attention an article by Dr Allan Struthers ("What Is the Optimum Serum Potassium Level in Cardiovascular Patients?" - 2004) in which he states that potassium supplementation is pretty useless at least in heart failure patients in the absence of an aldosterone antagonist, e.g., spironolactone or eplerenone. "A serum potassium increase of 0.25 mmol/L elevates serum aldosterone concentrations by 50% or 100%." Since most of supplemental potassium is absorbed and then distributed in the blood (total blood volume is about 5 liters), just over 1 mmole or about 50 mg of ingested potassium should result in a 50-100% increase in aldosterone.

According to Struthers et al., "Clinicians can be comforted by the fact that hyperkalemia does not typically occur in patients with normal renal status, because large potassium loads are efficiently and rapidly excreted." http://www.medscape.com/viewarticle/438088

Dr. Michael Lam (http://www.lammd.com) on p. 246 of his book How to Stay Young and Live Longer indicates 15 gm daily potassium as safe in a healthy adult. In a study of eight patients with long QT syndrome the equivalent of 250 mg of spironolactone and 9 gm of supplemental potassium daily (70 kg person) increased blood potassium from 4.0 to 5.2 mmoles/L. Four weeks of this therapy resulted in no serious complications.http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14642 687

So, the combination of an aldosterone antagonist with potassium is certainly effective and does not seem to pose excessive risk. However, although I'm a physician, I'm not your physician and this is not a blank endorsement of the above combination.

Furthermore, these aldosterone antagonists appear to be more effective in increasing blood potassium in the morning and ineffective in the evening. This latter finding is certainly consistent with the diurnal rhythm of aldosterone. It's hard to block aldosterone, if it's not being secreted (evening). Angiotensin converting enzyme inhibitors (ACEIs) decrease urinary potassium excretion and, according to most studies, increase blood potassium. http://home.caregroup.org/c linical/altmed/interactions/Nutrients/Potassium.htm

Although decreased urinary potassium excretion translates to increased blood potassium, changes in blood potassium directly stimulate aldosterone secretion without the renin angiotensin system (RAS). And ACTH, responsible for the diurnal rhythm, also works independently of the RAS.

As another aside, simultaneous supplementation of magnesium with potassium and an aldosterone antagonist increases cellular uptake of both potassium and magnesium.

It is important to remember that blood potassium levels may vary for reasons other than diurnal variation. Insulin and catecholamines cause transcellular shifts of potassium. The latter also causes urinary potassium wasting. Orthostatic challenge, e.g., prolonged standing or exercise, can also cause urinary potassium wasting via increased RAAS induced aldosterone. This would be more common amongst VMAFers, since they hyper respond with aldosterone during such activities (see above). And, of course, stress does the same for ALAFers (adrenergic lone atrial fibrillation).

On pp. 63-64 of Hans' book it is reported that excessive vagal tone is associated with a flat GTT (glucose tolerance test) curve. In other words VMAFers may have a more prolonged response to insulin. Perhaps the explanation for this lies in hepatic insulin sensitizing substance (HISS). This substance is secreted by the liver under the control of the vagus nerve. http://ajpgi.physiology.org/cgi/content/full/281/1/G29

Increased insulin sensitivity would certainly be useful in those whose caloric intake was occasionally insufficient for their daily energy expenditure (endurance athletes). Many LAFers have reported an increase in episodes during weight loss.

So, in VMAF insulin or hypoglycemia (transcellular shifts or urinary potassium wasting respectively) may be the predominant determinant for triggering a daytime episode. In ALAF it might be stress-induced cortisol/aldosterone. All appear to work by increasing the potassium gradient. Emphasis on steady potassium supplementation throughout the day from the moment we arise to bedtime would seem to be a good idea for all LAFers. Decreasing salt intake might also prove beneficial, since it causes urinary potassium wasting. Addition of a potassium sparing diuretic might improve not only this gradient but also magnesium balance as well. In ALAF the contribution of the potassium gradient may be relatively greater than in VMAF, where vagal tone is more critical (see below equation). Obviously under appropriate conditions (a round of golf – see above) aldosterone may aggravate VMAF and insulin can do the same for ALAF.

One brief word on "Waller water". This is an aqueous magnesium preparation divined by our own Erling Waller. He was one of the first to realize that many LAFers may owe their malady to magnesium deficiency, at least in part, since it is inextricably entwined with maintenance of intracellular potassium. He created the recipe (soda water and milk of magnesia), which can be found at http://www.afibbers.org/Wallerwater.pdf One word of caution concerning supplementing KCl or any powdered organic potassium preparation with magnesium. Potassium inhibits magnesium absorption and you may easily exceed your bowel tolerance for magnesium.

In my opinion the common denominator linking VMAF and ALAF appears to be low potassium, at least in part. P cells, as described above, would be the other link. As the Bordeaux group has shown, those with paroxysmal AF have inexplicably low PV ERP. Both arms of the ANS cause shortening of the ERP, as does low blood potassium. So either arm in combination with low potassium can trigger an episode in the arrhythmogenic heart. ERP x ( P cells + Potassium + ANS ) => AF Risk

My own personal experience with LAF episodes suggests that vagal tone and low potassium work in concert. My 9AM potassium values are usually 4.5 or less. That would put me well under 4 during the night. The simultaneous appearance of high vagal tone and low potassium would accentuate the shortening of ERP. This would conveniently explain my typical middle of the night episodes.

However, during the late morning or afternoon I've had occasional episodes (less than 10% of the total) that appear to be related to possible dehydration and/or hypoglycemia. Both of these stimulate not only catecholamine but also ACTH and aldosterone release (as does physical or emotional stress). Oftentimes I've had just a pastry for breakfast. Talk about an open invitation to an insulin surge. My HR is usually over 70 with low vagal tone at the time, but the episode is nonetheless triggered by a vagal maneuver. Presumably the insulin and fasting state induce the hypoglycemia. According to one study, ERP is shortest under hypoglycemia (v. hyperglycemia) in the left atrium (v. the right atrium). http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=8501411&dopt=Abstract

Might this be P cell related? A greater potassium gradient would presumably augment this shortening. Another study in rats has shown that insulin-induced hypoglycemia directly stimulates vagal neurons in the brainstem. http://jn.physiology.org/cgi/content/short/00791.200 3v1
Perhaps hypoglycemia can induce AF via both vagal stimulation and increased potassium gradient (catecholamine induced).

On very rare occasions I've triggered an episode by a short sprint. Here again, the timing of the episodes suggests low blood sugar and underscores the arrhythmogenic risk posed at the extremes of autonomic tone. During heart attacks the humoral catecholamine surge that can occur causes a rapid, transient transcellular shift of potassium, resulting in a short-lived but dramatic fall in blood potassium of approximately 0.5-0.6 mmol/L or more. http://www.medscape.com/viewarticle/438088_3
Although these shifts are evanescent and readily reversible, the transient drop in blood potassium can trigger PACs (and PVCs) and sometimes AF. This, of course, is in addition to the urinary potassium wasting that catecholamines cause.

I've also had episodes that are postprandial, but only in the evening (presumably because there is more reinforcing vagal tone at this time). Initially I thought this was due primarily to the alkaline tide associated with a meal (production of gastric acid by the stomach reflexively causes blood alkalosis) and subsequent urinary potassium loss (low blood potassium both causes and is caused by alkalosis). Then I thought that it was due to the effect of insulin and loss of cardiac intracellular potassium due to the increased concentration gradient. Then I thought I might have a mild problem with gastroesophageal reflux (GERD)/lower esophagal sphincter (LES). GERD is increased in athletes, especially in those that run, which I do (no more marathons, however). However, now I'm inclined to think that evening meals with poor K/glucose and K/Na ratios are the primary problem. I once thought that seafood (lots of salt) at dinner was a trigger. Looking back on such episodes, these meals were often low on the veggies and high on the simple carbs (love my desserts).

Hans went from typical stress related ALAF to typical vagally mediated AF, when he briefly took spironolactone, which has a vagotonic effect (in addition to being a potassium sparing diuretic). So, it seems that LAF can appear anywhere along the spectrum of autonomic tone. And remember, Hans always ran right at the lower limit of normal with his blood potassium. And his aldosterone, a vagolytic, and cortisol were always elevated (both aldosterone and cortisol bind to mineralocorticoid receptors, i.e., cause urinary loss of potassium). Furthermore, although there was never much change in his blood potassium (daytime values of 3.5 – 3.7 mmoles/L), his urinary potassium (and magnesium) excretion continued to escalate, as he approached the next episode. Obviously there was continual leakage of potassium from the intracellular compartment to maintain the constant blood potassium concentration in the face of escalating urinary loss.

Many LAFers have commented on what seems to be a repeating periodicity to their episodes. It seems that the length of my episodes was directly proportional to the time interval before the next episode. On the Bulletin Board (BB) in late 2002 before choosing the inaugural topic for the Conference Room (CR) (www.afibbers.org/conference/session1.pdf) Hans queried as to why there always seemed to be an abundance of PACs just prior to an episode and none just after termination. He surmised that ANP (atrial natriuretic peptide) might explain this. ANP is secreted during episodes via a mechanism that involves atrial cell stretch. ANP inhibits aldosterone synthesis and renin release thereby conserving blood potassium and helping replete intracellular levels. At some point ERP shortening due to the potassium gradient is sufficiently lengthened and the episode terminates. This point is usually in the AM when vagal tone is low and its associated ERP has also lengthened.

During this natriuresis (sodium in urine) the consequent drop in blood volume and increase in blood K/Na stimulates aldosterone secretion to oppose circulating ANP. Additionally during AF cardiac output drops by about 30% and with it the hydrostatic pressure sensed by the renal baroreceptors (the juxtaglomerular apparatus). This is another reason why aldosterone should be elevated at the end of an episode. I've often noticed on my Polar S810 HR monitor that immediately after termination of an episode my HR albeit NSR is always inappropriately high and my HRV is always inappropriately low for several hours. The vagolytic state caused by elevated aldosterone would easily explain this. It would also explain the complete absence of PACs. The half-life of aldosterone is about 15 minutes. The more aldosterone (vagolytic) present at the end of an episode, i.e., the longer the episode of AF, the longer this post episode PAC free period will last. But it comes with a cost and that is accelerated urinary potassium wasting after the protection of ANP has been removed. Life style and diet might then combine to slowly deplete these intracellular potassium stores until blood potassium level cannot be maintained and some threshold gradient value (for P cells) is breached and another episode is triggered. In this scenario the potassium gradient may be more integral in triggering ALAF episodes and vagal tone may be more integral in triggering VMAF episodes.

If one assumes the potassium gradient as vital to triggering an episode, then one can continue further along this line of thinking. Typical nighttime (vagally mediated) episodes rebalance potassium stores and protect against typical daytime (adrenergic) episodes and vice versa. However, if vagal tone in VMAF is lowered by medication, then daytime episodes should become relatively more frequent. And if potassium loss in ALAF is lowered by medication, then nighttime episodes should become relatively more frequent. I've experienced the former with disopyramide and Hans has experienced the latter with spironolactone.

LAFers exist all along the spectrum of both autonomic tone and the potassium gradient. Our arrhythmogenic substrate is a given and separates us from "normals". Some LAFers have minimal tone or gradient problems (fortunate) but enough of both to trigger LAF (unfortunate). They are fortunate because either by medication for autonomic tone or lifestyle and diet for potassium gradient they have been able to avoid the beast. Counter- regulating mechanisms (RAS, blood K/Na, ACTH) otherwise make the latter a very difficult proposition. The real question is could this "fortunate" category of LAFers be expanded by addition of medication for better potassium balance? Inbred resistance to combining increased potassium intake with a potassium sparing diuretic makes this a very difficult proposition. Triamterene or amiloride and NOT spironolactone or eplerenone would seem to be the best choices on this count (see below). Neseritide (synthetic BNP) or carperitide (synthetic ANP) might be even better, but are only available by the intravenous route. According to the medical literature ACEIs are vagotonic, but perhaps not all are. For example, enalapril but not captopril significantly inhibits plasma aldosterone concentration and urinary aldosterone excretion. Since aldosterone is vagolytic, perhaps enalapril is vagotonic and captopril is not. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=7881704&dopt=Abstract

Another interesting question is why proton pump inhibitors (PPI) not only relieve GERD but also appear to relieve AF. Is it only because there is less irritation of the lower esophagus (and less vagal stimulation)? Or is it also because of an improvement in the potassium gradient? Gastric acid production causes a simultaneous blood alkalosis (alkaline tide). Hypokalemia both causes and is caused by alkalosis. By inhibiting the proton pump (H+ is no more than a proton) less H+ is lost in the gastric juice. Less potassium is lost in the urine, because the blood is less alkaline (since the gastric fluid is less acidic).

Jackie Burgess, well known and respected by all of us that frequent the BB and CR, has suggested a balanced snack of protein, complex carbohydrate and healthy fat two hours before bedtime. And, of course, magnesium and potassium are always welcome at anytime. This approach would impede any insulin induced hypokalemia/hypoglycemia and possibly AF as well. She also has reiterated for us an old adage "if you can feel your heart beating at night when lying in bed, you may be low in potassium". This undoubtedly is due to the mild BP elevating effects of low blood potassium. The heart has to work just a little harder, enough to make you aware of its beating, when all else is quiet.

Due to the insulin induced drop in blood potassium that occurs after a carbohydrate meal, it would seem prudent to always ingest potassium with your carbohydrates. Jackie has also pointed out that KCl can cause gastric irritation, at least if taken on an empty stomach. It can also elevate blood pressure in the salt sensitive. However, potassium in fruit without chloride is less absorbable. According to one source only 40 percent of the potassium in a banana is retained. This is one reason why MDs prescribe potassium as KCl (about 800mg K as KCl per tab of K-Dur). KCl also addresses hypochloremia, which often accompanies hypokalemia. If KCl irritates your stomach, then you should never let a meal go by without potassium supplementation, thereby exploiting its buffering effect in this regard. K-Dur is also a sustained release formulation, most convenient at bedtime to address the midnight diurnal nadir of blood potassium.

I heartily agree with the general recommendation to shift from simple to complex carbohydrates in our diets. I used to think this advice was better directed at those that struggle with their weight. However, given my problems with episodes being triggered when I skip or delay meals, I think thin people can also benefit from it. Eat properly and don't skip meals. Graze rather than gorge. Earlier is better than later.

In a past post on the BB I suggested that a portable potassium meter might be a very useful item for a LAFer. Horiba and Hoskin Scientific make good ones, but they are not quite ready for prime time, at least not in humans. I recently purchased an Omron BP monitor (less than $50 at Costco) and have found that relative evening BP (slightly higher systolic than normal) and/or the presence of PACs when lying on my right side both provide feedback on probable intracellular potassium. www.afibbers.org/conference/session36.pdf
Although this is only an indirect approach at best, it may be the difference between PM AF and normal sinus rhythm (NSR).

Being on top of my daily potassium supplementation definitely decreases my evening PACs and BP. This also requires awareness of activities that cause transcellular shifts of potassium and sometimes urinary potassium (and magnesium) wasting as well. Appropriate countermeasures are needed, i.e., more fluid intake, small potassium rich complex carb snacks, e.g., a banana, etc.

My Experience with Spironolactone

I have experimented extensively with potassium and spironolactone. A combination of 3 g/day of potassium and 100 mg/day of spironolactone increased my blood level of potassium to 4.7 mEq/L. Increasing my potassium intake to 6 g/day and my spironolactone to 200 mg/day (blood potassium over 5 mmoles/L) increased the frequency and duration of my afib episodes despite also taking 500 mg/day of disopyramide (Norpace). I have now concluded, and Hans' experience supports this, that spironolactone is unlikely to be beneficial for either adrenergic or vagal afibbers for the simple reason that its detrimental vagotonic effect clearly outweighs its positive effect in regard to potassium conservation. I have therefore terminated my experiment with spironolactone. My next experiment will involve aggressive potassium supplementation starting in the afternoon and escalating until bedtime as well as taking K-Dur (800 mg of elemental potassium in a sustained release formulation) at bedtime. If that produces insufficient improvement I plan on investigating triamterene (Dyrenium) and amiloride (Midamor). Most, if not all, ACE inhibitors and angiotensin receptor blockers (ARBs) are decidedly vagotonic so I intend to give them a miss.

Triamterene and amiloride, both potassium-sparing diuretics, work at the level of the distal convoluted tubule in the kidneys, but do not bind to mineralocorticoid receptors. They impair sodium reabsorption in exchange for potassium and hydrogen. Thus, unlike spironolactone and eplerenone, they should not directly increase vagal tone. In fact, they may modestly increase blood potassium and thereby elevate the K/Na ratio. This would increase aldosterone secretion and thereby increase vagolysis. This, in turn, would increase the component of total blood aldosterone due to K/Na compared to the other two sources of aldosterone (RAAS, major and ACTH, minor). This might prove beneficial because increased potassium and decreased sodium intake would have greater impact on aldosterone secretion, in effect providing greater control over aldosterone secretion via diet and/or supplements (especially useful for a VMAFer).

Conclusion

You can't directly control the arrhymogenic substrate (PVs) except through PVI. You can't effectively control vagal or sympathetic tone except through meds. That leaves low potassium. If you want to get serious about controlling your episodes, then you must get serious about your potassium intake. You must address not only how much you ingest but when you ingest it. Either avoid those situations that assault your blood potassium gradient (stress, hypoglycemia, dehydration, etc) or increase your daily potassium (and magnesium) intake with appropriately time-targeted supplementation. Although food sources are best for most of what we need, for potassium repletion IMHO supplements are superior (fish oils and omega 3s are another such exception). Intake of potassium through diet alone is less convenient, less quantifiable and less absorbable. And then there's the glycemic load problem posed. Presently there is great resistance within mainstream medicine to combining potassium supplementation with a potassium sparing diuretic. Hyperkalemia and life threatening ventricular arrhythmias are of great concern. However, with the pioneering work of Drs. Struthers, MacDonald and others this overemphasized concern may soon take a back seat to a rational combination regimen. In my view it is quite plausible that this LAF epidemic might be reversed, if such an aggressive regimen were pursued by those afflicted. The study that needs to be done is one similar to that referenced above on LQTS (long QT syndrome), but with amiloride or triamterene and on LAF patients instead. But until then please remember my above disclaimer and make sure you have good renal function. My BUN (blood urea nitrogen) and creatinine are well within normal limits, but I still routinely monitor my blood potassium.