AF ablation Atrial fibrillation

What’s the best medicine for AF? Part 1

What’s the best medicine for treating AF?

Now that is a frequently asked question.

You already know the answer. But perhaps the obvious deserves a few words?

First, let’s clarify what we are talking about when we use the term AF medicine. I’m speaking here about the so-called anti-arrhythmic drugs. Other names include “membrane-active” or “rhythm-control” medicines. These are the drugs that act directly on the electrical properties of atrial muscle cells. The most common ones are listed here:

  • Type IC-drugs: propafenone and flecainide
  • Type III drugs: sotalol and dofetilide.
  • Multi-channel blocking drugs: dronedarone and amiodarone

For the purposes of this post, let’s exclude anticoagulants (blood thinners) and the drugs that slow the ventricular rate (pulse) of AF. The so-called “rate-control” medicines act almost exclusively on the AV node—the connection between the atria and ventricle. Beta-blockers (metoprolol, carvedilol), calcium channel blockers (diltiazem, verapamil) and digoxin slow the rate of AF by slowing conduction of AF impulses through the AV node. They rarely suppress AF.

So let’s talk about AF suppressing drugs. First the disease AF:

One major rule of thumb for AF is that there are no rules. The disease is crazy complicated and diverse. It’s more common in older people, but young people get it too. It begets itself, but can also go just go away. AF is more common in patients with obesity and high blood pressure, but thin patients with low blood pressure also get it. Some patients fibrillate for years and never know it, while others are incapacitated by only brief episodes. Older patients with advanced heart disease can deteriorate rapidly with AF, and they benefit from emergency treatment. But for the majority of AF patients, such emergency treatment increases the risk of over-treatment.

At the start of my career, AF was easy to treat. There were hardly any medicines and ablation therapies had not yet been thought of as possible. AF patients mostly had to live with the disease. Our lack of tools meant there wasn’t a need for AF specialists and electrophysiologists. Fortunately that’s all changed now.

If you think AF is tough to figure out, let me tell you about anti-arrhythmic medicines. What a paradox.

There are three major problems with AF medicines:

Lack of Efficacy: The most recent trial comparing AF medicines to ablation revealed that 72% of patients in the medical arm had AF recurrences. Admittedly, using just 30 seconds of AF as the threshold for treatment failure is overly rigorous, but the fact remains that rhythm-control medicines fail more times than they work. They rarely eliminate all episodes of AF and rarely do they work over the long-term. In fact, when a patient who is taking an AF-drug reports being free of AF for years, I question whether it’s the drug or perhaps the AF is gone.

Pro-arrhythmia: All of the rhythm-control medicines can actually cause arrhythmia. We call this pro-arrhythmia. And no, that’s not a typo. Rhythm-control AF medicines harbor the potential to make things worse. Much worse. Take these vignettes:

Examples of pro-arrhythmia range from mild to fatal. In mild cases, many of the rhythm-control medicines predispose to, unmask or exacerbate low heart rates (bradycardia). This gets bumped to moderate when pacemakers are implanted to treat drug-related slowing. Other moderate examples of pro-arrhythmia include the properties of conduction slowing drugs like propafenone and flecainide. In some cases the drug suppresses AF; but in other cases the slowing effect converts AF to Aflutter—a slower more organized, but equally troublesome arrhythmia. And no, you cannot tell when this will happen or in whom.

The most severe pro-arrhythmia from AF drugs comes when drugs that prolong duration of activation (action potential), like sotalol and dofetilide cause QT interval prolongation. (The QT interval can be measured on an ECG.)  If QT prolongation goes unchecked, ventricular tachycardia (torsades de pointes) can occur. This predisposes the AF patient to sudden death. One of my first lectures in training reviewed a famous study of the old drug quinidine. Researchers found that quinidine reduced the burden of AF, but at the cost of an increased risk of death.

That’s the paradox: the same drug that can help can also cause harm. It’s always been this way for rhythm drugs.

Side Effects: In their tendency to cause side effects, rhythm-control medicines are no different from all other drugs. Fatigue, GI intolerance, blurred vision, funny taste in the mouth, and balance issues are all side effects that can occur with these drugs. Again the tradeoff: a patient that feels their AF is suppressed with, say propafenone, has to decide if the metallic taste is better than having episodes of AF. Tradeoffs. Treating disease is always about tradeoffs.

That’s why it is better to prevent preventable disease.

In the next post, I’ll explore the factors that go into choosing AF drugs.


11 replies on “What’s the best medicine for AF? Part 1”

Great start to a complex topic.

I hope you’ll add in part II some words on how device therapy may facilitate AF medical therapy. When appropriately indicated for CHF or bradycardia treatment, pacing can help mitigate all of medication downsides you’ve discussed above (lack of efficacy, pro-arrhythmia and side effects). I could host a huge picnic of all of my highly symptomatic AF patients now satisfied with pacing/sotalol combination therapy.

Modern pacemakers and ICDs give a wealth of information about AF control. Over the years, pacers have gotten cheaper and ablation has gotten more expensive. In a patient indicated for either pacer/meds vs. ablation, I bet the device option is quicker and cheaper.

Not to say, of course, that devices are indicated or appropriate in all AF patients — far from it. I just know that when we do get a device in a symptomatic AF patient,
life typically gets easier for patient and caregivers alike.


A pacemaker may be quicker and cheaper than an ablation initially, but it also carries a 20% to 25% risk of significant tricuspid valve damage, along with risks of endocarditis and lead problems. A patient who gets much ventricular pacing will be at greatly elevated risk for heart failure and more AF. A pacemaker also will not protect against any of the other side effects of antiarrhythmics (e.g., for amiodarone, catastrophic lung damage, eye, liver or thyroid damage) or negate the increased death rate associated with such drugs (tripled for sotalol, in one major study, IIRC). How much of the patient’s time, money and emotional well-being will be spent during a lifetime of follow-up and treatment, assuming he agrees to pursue that route, of any problems caused by the drugs and/or pacemaker? Your patients are initially satisfied (“life gets easier”), but over their lifetimes, what percentage of them will end up with major side effects?

Hi John. GREAT basic overview post on the arrhythmic drugs for AFib. I’d add a number of comments to what you’ve already wrote. KEEP UP THE GREAT WORK!

i) Most AFib episodes are “silent”. Holter studies have shown ~90% of all AFib is without symptoms! This is KEY – because as John says – the patient may NOT be aware of their AFib. Physicians see the patients who come in with symptoms! This also makes it difficult to track “if AFib is gone” – since unless you do long-term Holter studies – you’ll NEVER really know …

ii) Even in patients who say they are aware of their AFib episodes – one always has to wonder how many more episodes of AFib they are having that they are not aware of ….

iii) “Success” for treating AFib may be measured in some patients in “relatives”. By that I mean that IF antiarrhythmic drugs reduce both the number and duration of intermittent AFib episodes – they may significantly improve the situation for some patients in a manner that the patient will be satisfied. That said – as John emphasizes – you don’t “cure” AFib – as it virtually always recurs. The question is just when and how often ….

iv) Proarrhythmia is a “real” and problematic issue. My impression has been that Beta-blockers are perhaps the least likely drug class to cause proarrhythmia. While NOT a “rhythm converting” agent per se – there may be some patients whose AFib episodes are induced by excess sympathetic tone – for whom beta-blockers may be effective in minimizing episodes without significantly increasing risk ….

v) Re SIDE EFFECTS – Key goal of the clinician is to ensure “that the treatment is not worse than the disease”. At times – it may be reasonable to initiate antiararhythmic therapy (because the treatment initially is not worse than the disease) – but IF AT ANY TIME the treatment becomes worse than the disease – then it is time to stop the treatment (= stop the drug).

vi) The BEST chance to convert a patient and maintain sinus rhythm (at least for a certain amount of time) is following that 1st episode – especially if the patient has a normal (or relatively normal) heart – and especially if there is a potential predisposing cause of the AFib that can be correcrted (severe medical illness, lyte abnormality, hypoxemia, pulmonary embolism, sympathomimetic drug use). Thus there IS a rationale to consider a trial of medical therapy with antiarrhythmics in certain selected patients. At times – such treatment may be intermittent in selected patients (“antiarrhythmic cocktail”).

vii) Given that rhythm control doesn’t really work well longterm – it is also now reasonable NOT to even start drug treatment with certain patients (esp. older patients with longer term AFib when rate is controlled and they are asymptomatic … ). Such patients of course still need to be anticoagulated ….

viii) IF your patient is symptomatic and doesn’t respond to drugs – OR – in 2012 if your patient simply wants to be potentially “cured” of their AFib problem – REFER TO YOUR FRIENDLY EP SPECIALIST (who hopefully will be someone like Dr. John!).

THANKS for listening (from a longterm primary care educator with a passion for arrhythmias) – : ) Ken

Spent a lot of time working on ILR’s (protocols,guidelines,research,etc.) Even with the first crude discrimination algorithms and only patient sampling (random & symtomatic) ; patients and physicians greatly underestimated the total burden — symptomatic & asymptomatic . Further, without longer term monitoring , it’s difficult to document onset and/or termination. A plus was the tracking of pharma effects – titration changes, ‘cocktails’, burden, sustainability,post ablation,,,,,,,
Thanks for great site .

So why even consider drugs when ablation is clearly superior? I know I would want whatever is causing AF to be gone if in fact ablation can actually stop AF in its tracks.

Because all surgeries carry risk. Ablation carries the risk of damage to the phrenic nerve and esophagus, and pulmonary vein stenosis. Stats published in this blog suggest that frequently a second procedure is necessary. It’s a difficult decision.

@Phil – Many facets go into the answer to your question (which is why Dr. John will never lack for patients in need of his services). Much depends on specifics of the case – underlying disease, intermittent or persistent AFib, degree of symptoms (if any) – experience with other treatment (drug adverse effects or not – and on occasion onset of AFib due to some correctable cause. Then there is choice to be made by the informed patient about pros and cons of all approaches … (with some Cons noted by Allison).

Dear Dr John,

So pleased to find your posts. I have only scanned this one as I was originally interested in the af ablation topic but too late to reply.
My mother is eighty years old and an afibber for 6 years. The last year being persistent after the three months post her first catheter ablation. She had her second this May and still in rhythm but some annoying ventricular and super ectopic beats….real thuds sometimes on activity, and odd trills of tachy. We have just read that post ablation tachy is not a bad thing and can be remedied easily with short ablation. Her main reason for ablation was to get off the drugs, which make her life a misery more than the condition but GP’s are so reluctant to admit that they can cause other problems.

I know she will be thrilled to bits to read your articles.

Thanks again

Helen Nock

i have af permanently am on warfarin bisoporol and digoxin feel tired and ill most of the time

Comments are closed.