With breathtaking speed, atrial fibrillation has gone from “Huh?” to parlance. “A-fib”, a common cardiac cause of palpitations, is now in the front ranks of evils lurking to smite our well-being. There is no mystery to this transformation. In 2013, the Food and Drug Administration licensed three new drugs to prevent a stroke, the fearful complication of A-fib: apixaban (Eliquis), rivaroxaban (Xarelto) and dibatigatran (Pradaxa). This unleashed the full might of pharmaceutical marketing: the scientific data for efficacy that convinced the FDA is tortured till it convinced “thought leaders” whose opinions convinced influential journalists. Sales pitches populate print, broadcast and social media. A-fib is now more than a worrisome disease; it is a product line.

Nonetheless, A-fib can be scary for those afflicted. There are lots of choices to make and a lot of confusing, conflictual and counter-intuitive advice to deal with. Troubled by this situation, Mr. X recently sought me out to have a dialogue about his situation.

Mr. X is a 70-year-old business executive who has enjoyed robust good health and is on no prescription drugs. He exercises vigorously and is a consumer of various over the counter treatments purveyed as salutary. Like many “health-wise” Americans, he also takes 80mg of aspirin a day unaware that the benefit is minimal at best and is counterbalanced by a similar magnitude of risk. A month earlier he had the sudden onset of palpitations, a fluttering in his chest that made him exceedingly anxious and somewhat breathless. He waited an hour before asking his wife to drive him to the local emergency room. By the time he was first seen, another hour passed. The diagnosis of A-fib followed. Another hour passed to find the consulting cardiologists debating whether to convert the A-fib to a normal rhythm by using drugs or an electrical shock. Before they could decide, Mr X’s heart decided to behave again; he was back in a normal rhythm. It was a frightening experience for Mr. and Mrs. X. He left the ED shaken and shaky.

He also left with a follow-up appointment with a cardiologist who specialized in rhythm disorders and a prescription for a drug that slowed the conduction of electrical impulses initiating in one heart chamber, the right atrium, and traversing the heart. The normal “pacemaker” is a specialized cluster of muscle cells in the right atrium that discharges at regular intervals, initiating a current that causes the heart to contract in the synchronized fashion termed Normal Sinus Rhythm. In A-fib, for reasons that are poorly understood, multiple pacemakers form in the right atrium leading to chaotic discharging and circular currents in the right atrium. How many of these impulses manage to exit the atrium to traverse the heart depends on the capacity of the conducting tissues; most just stay confined to the atrium causing it to quiver ineffectively.

In retrospect, Mr. X recalls multiple episodes of palpitations in recent months, though few were as prolonged and disturbing as the episode that drove him to the ED. He noted several more before he arrived at the follow-up appointment. If he wasn’t anxious before, that interview did the trick. He was relieved that he basically had a normal heart with one serious problem; he had “PAF”, paroxysmal atrial fibrillation. This intermittent A-fib was probably more dangerous than persistent, “chronic” A-fib. The danger in PAF was not that he felt awful during attacks; those symptoms should respond to drugs that control the heart rate. The urgent issue is that the chaotically beating chamber is prone to blood clots, pieces of which can break off as “emboli” and travel to the brain causing an embolic stroke. The cardiologist wanted to start a drug designed to keep Mr. X in normal sinus rhythm and to start one of the new blood thinners to decrease the likelihood of clots forming and thereby decrease the likelihood of an embolic stroke. If that didn’t do the trick, the cardiologist mentioned the possibility of other drugs or even of “ablation”. This is where a catheter is introduced in the heart to identify bundles of muscle cells that were misfiring and burn them thereby creating tiny therapeutic heart attacks – a procedure that has risks, works most of the time, and is plagued with recurrences.

The plan that made sense but not enough sense for Mr. X to agree without first researching it’s the rationale, the benefit/risk ratios, and the alternatives. The cardiologist seemed to think the decisions were obvious. Mr. X knew they were not. The episodes of palpitations were still anxiety provoking but more in the sense of their portent than the associated discomfort. Mr. X needed to place the cardiologist’s advice into to a more personal context, one in which he could exercise his own values as to risk aversion.


How likely is an embolic stroke in the setting of A-fib?

There is no simple answer to this question. First of all, there has been a very impressive decrease in the incidence of fatal strokes in recent decades, and probably a decrease in the incidence of non-fatal strokes, too. That makes “old” data less and less relevant. Secondly, the epidemiology of A-fib is difficult to define because most A-fib, PAF or chronic A-fib, is asymptomatic. Furthermore, most palpitations occur while we are in Normal Sinus Rhythm. No doubt most estimates of the risk of stroke in the setting of A-fib are overestimates; most people with A-fib never become patients with A-fib.

So, the relevant question is what is the risk for stroke in patients newly diagnosed with A-fib? However, the incidence of stroke in the setting of A-fib increases with age and with complicating cardiac and other diseases. Hence, the more relevant question for Mr. X is what is the risk for a 70 year old man who is otherwise well? There is a study addressing this question in some 80,000 patients over age 65 who presented to hospital with new onset A-fib. If we follow 100 men with A-fib who are 70 years old and otherwise well for a year, one will suffer a stroke that is recorded in medical records. This is not trivial, nor is it as alarming as Mr. X was lead to believe.

How much does blood thinning reduce this risk?

There are three ways to interfere with blood clotting that are relevant. One is low dose aspirin, which Mr. X is already taking. The second is the time-honored standard blood thinner, warfarin (Coumadin), which requires laboratory monitoring and carries a risk of bleeding complications. The third are the new drugs which are more convenient since they do not require monitoring but have their own baggage. In terms of efficacy, there is no data that the new drugs have anything to offer over warfarin. So the first issue for Mr. X is whether taking low dose aspirin each morning is good enough?

Most of the recent drug trials do not address this question; they are comparing the new anticoagulants with warfarin. However, there was a considerable earlier literature comparing aspirin with warfarin for A-fib. Some of the individual trials are better designed and particularly compelling1. But the literature is sufficiently robust that there have been many systematic attempts to cull a consensus from the accumulated research2. There is no doubt that low dose aspirin therapy is effective but not as effective as warfarin. There is also no doubt that low dose aspirin therapy leads to fewer adverse effects. It follows that the higher the risk for an embolic stroke, the more the benefit is apparent and outweighs the harms. Mr. X has a low risk for embolic stroke; about 1% in a year without aspirin. Taking warfarin can reduce that risk by about 10%, to about 0.9% in a year (i.e. 1 in 110 might suffer an embolic stroke).

What are the hazards of reducing the risk of an embolic stroke with blood thinners?

Low dose aspirin causes easy bruising but major bleeding complications are rare. They are rare with warfarin if there is no overdosing, hence the monitoring, and with the newer oral anticoagulants – although bleeding is particularly difficult to treat as a complication of the newer agents.

But that’s the least of the risks to consider. The reason for aspirin or warfarin anticoagulation is to prevent clots and thereby embolic strokes. Both work, warfarin better than aspirin, but neither agent works perfectly. However, when aspirin fails the stroke is likely to be an uncomplicated embolic stroke; a blood clot blocks a vessel in the brain causing damage to the brain downstream by depriving that area of the brain of oxygen-carrying blood. Most recover partially or fully from embolic strokes. But when the blood is thinned by warfarin, there is a risk of bleeding into the damaged brain converting a simple embolic stroke into a catastrophe with little likelihood of recovery. That’s what happened to Ariel Sharon.


Now Mr. X is an informed patient. He is in a position to make his decision. What would you do?


  1. Singer DE, Chang Y, Fang MC, et al. The net clinical benefit of warfarin anticoagulation in atrial fibrillation. Annals of Internal Medicine 2009;151:297-3005
  2. Aguilar MI, Hart R. Oral anticoagulants for preventing stroke in patients with non-valvular atrial fibrillation and no previous history of stroke or transient ischemic attacks. Cochrane Database of Systematic Reviews 2005, Issue 3. Art. No.: CD001927.DOI:10.1002/14651858.CD001927.pub2.