The History of Atrial Fibrillation: The Last 100 Years

Eric N. Prystowsky, M.D.


J Cardiovasc Electrophysiol. 2008;19(6):575-582. 

In This Article


Decades ago the treatment for AF was rather limited as detailed in two of the major textbooks on cardiology.[27,28] Cardioversion was suggested using either quinidine or a DC shock,[27,28] and digitalis was recommended for rate control, with adrenergic receptor blockade potentially useful in patients with resistance to digitalis.[28] Quinidine was recommended to maintain sinus rhythm if needed, and procainamide was just being introduced as an alternative to quinidine.[27] The anticoagulation studies on stroke prevention were decades away from being done; and in whom, when, and how to anticoagulate was unknown. Warfarin was suggested before cardioversion in some patients,[27,28] but long-term use was not advocated unless the patient had persistent AF and rheumatic heart disease.[27] How fortunate the modern-day clinician and patient are to have a cornucopia of treatment options and the knowledge derived from decades of research to be able to apply the best therapies to each patient, as detailed below.

Stroke Prevention

The results from a series of prospective randomized clinical trials were published in the 1990s that advanced our concepts on therapy to prevent stroke.[25] In aggregate, these studies made several key observations. They demonstrated: (1) there was no significant difference in stroke risk between paroxysmal and persistent AF; (2) the use of warfarin was needed for subgroups of high-risk patients; and (3) that aspirin in select patients was superior to placebo in reducing stroke. The recent ACC/AHA/ESC Management guidelines on AF have refined this further by using the CHADS2 scoring system for AF stroke risk.[25] The acronym CHADS stands for congestive heart failure (C), hypertension (H), age 75 or greater (A), diabetes mellitus (D), and stroke/TIA (S). Each risk factor gets one point except stroke/TIA, which receives two points. A score of two or more equates to high risk and requires warfarin use; one point is moderate risk and one can use aspirin or warfarin; and a score of zero is low risk, with aspirin the only suggested agent.

The 1980s ushered in the era of the application of transesophageal echocardiograpy (TEE) in the cardioversion of AF.[25] Initial observations showed that a thrombus in the left atrial appendage could be detected by TEE even though not seen by transthoracic echocardiography.[29] TEE-guided cardioversion of AF was demonstrated to be safe and effective,[30] and later the ACUTE trial concluded that the TEE-guided and traditional anticoagulation approaches to cardioversion were equivalent.[31] Thus, clinicians were now armed with a full set of safe and effective anticoagulation strategies for their patients with AF. Future research will likely focus on surgical and transcatheter techniques to exclude the left atrial appendage from the circulation and on oral anticoagulant agents that inhibit thrombin or related parts of the coagulation cascade.

Antiarrhythmic Drugs

When I was a cardiology fellow we had quinidine and procainamide to use for maintenance of sinus rhythm. I was fortunate to study the promising new agent disopyramide in patients with the Wolff-Parkinson-White syndrome, although no indication for AF was sought.[32] In fact, starting in the 1980s, a plethora of new drugs underwent study and received approval, though none were at first developed primarily for use in AF. Agents such as amiodarone, encainide, flecainide, propafenone, and sotalol initially sought approval for ventricular arrhythmias, while flecainide, propafenone, and sotalol later did appropriate clinical trials to get approved use for AF. Dofetilide was the last oral agent approved for use in AF. Interestingly, amiodarone, likely the most frequently prescribed drug for AF, has never been given approval for this indication by the FDA. Multichannel blockers such as dronedarone[33] are in late clinical trials, and atrial-selective blockers are far along in development.[34,35] So-called "upstream" therapies (ACE inhibitors, ARBs, statins) require more study to define their place in the armamentarium for AF prevention.[34]

In an attempt to improve safety in drug use for patients with AF, an algorithm was suggested that focused more on safety than efficacy (Fig. 3).[36] Certain drugs were considered safer to use in specific clinical situations, for example, heart failure and hypertension. It was an era when class Ia drugs were in common use, and flecainide was under the dark cloud of CAST.[37] Compare this with the most recent algorithm in the ACC/AHA/ESC guidelines (Fig. 4). Note that class Ia drugs are not initial or secondary recommendations, more clinical categories have been added, and catheter ablation has become a mainstream choice.

Figure 3.

Algorithm for selecting antiarrhythmic drug therapy to prevent atrial fibrillation.[36]

Figure 4.

Algorithm for maintenance of sinus rhythm.

Nonpharmacologic Approaches

The initial nonpharmacologic treatment for atrial fibrillation was directed at rate control and involved catheter ablation of the AV junction.[38,39] Direct current energy was the initial one chosen for this, but it has been supplanted by radio-frequency catheter ablation. This has been a very effective treatment for rate control but requires permanent pacemaker implantation, and more recent data have demonstrated that some of these patients may go on to congestive heart failure, presumably due to a pacing-induced cardiomyopathy.

Prior to catheter ablation attempts to cure atrial fibrillation, a surgical approach was developed by Dr. Cox.[40] The approach involved multiple incisions in the right and left atria that walled-off various sections of the atria, including the pulmonary veins, and produced a pathway for sinus impulses to reach the AV node. He appropriately termed this the MAZE procedure. This procedure has been quite effective and has undergone numerous modifications over the years, and some current approaches are directed predominantly to isolating epicardially the pulmonary veins. With our recent knowledge of the importance of the pulmonary veins in the initiation of atrial fibrillation, it appears in retrospect quite fortunate that part of the MAZE procedure also involved isolation of the pulmonary veins. Although this is an often successful procedure, its acceptance as primary therapy for atrial fibrillation has been sparse at best, and it is applied most frequently as an add-on procedure during other cardiovascular surgeries, for example, bypass graft surgery or valve replacement.

We owe much to Haissaguerre and his team in Bordeaux, France, for opening a new door into the endocardial catheter ablation approach to cure atrial fibrillation.[15] These investigators recognized the preeminence of pulmonary vein foci as initiators of atrial fibrillation, and through their work as well as many other laboratories, catheter ablation to cure atrial fibrillation has become a mainstream therapy for our patients.[15,41–53] Were there clues in the literature to the importance of pulmonary vein ectopy as a possible cause of atrial fibrillation in humans? Indeed, there were. In an excellent study from Cheung,[54] the presence of rapidly firing foci in pulmonary veins in guinea pig atria was demonstrated including block from the veins to the right atrium. Further, the musculature around the pulmonary veins was described very nicely by Nathan and Eliakim in 1966.[55] These seemingly isolated observations have new meaning now that we understand the importance of pulmonary veins in initiation and often maintenance of atrial fibrillation in humans. Initial ablation was targeted at paroxysmal atrial fibrillation and subsequently persistent atrial fibrillation, and most recently attempts to cure long-standing persistent atrial fibrillation have been reported.[33,56,57] Many investigators around the world are searching for newer and better nonpharmacologic approaches to cure patients with paroxysmal, persistent, and long-standing atrial fibrillation, including combined endocardial and epicardial catheter ablation approaches[58] as well as limited thoracotomy surgical approaches.[59]

Rate Versus Rhythm Strategies

For decades it was assumed that patients would have better outcomes in sinus rhythm compared with atrial fibrillation. This concept was studied in several randomized prospective clinical trials, the largest of which was AFFIRM.[60,61] In general, for most of these trials, entry criteria included patients who were elderly and typically at high risk for stroke. Thus, the data are directed to this patient group and not to all patients who have atrial fibrillation. These studies showed that if therapy is applied correctly, there appeared to be no difference in mortality or stroke risk regardless of whether one chose a rhythm or rate control strategy. Thus, in appropriate patients the clinician should choose sinus rhythm for symptom relief if it is not achieved with rate control only.

Endpoints for therapy have also been points of contention over the years. For a rhythm control strategy, some have advocated abolition of all episodes of atrial fibrillation, while others feel that reduction in the atrial fibrillation "burden" is enough, that is, a reduction in the frequency and duration of episodes of atrial fibrillation such that the patient has a good symptomatic outcome. In reality, even those most ardent seekers of total arrhythmia control must now yield to the fact that many patients who apparently have "complete" rhythm control actually have episodes of asymptomatic atrial fibrillation.[47,62] Therefore, the conventional thinking is to improve symptoms with the realization that patients will likely have asymptomatic episodes of atrial fibrillation and, importantly, the decision for maintenance of anticoagulation depends on the risk for stroke and not the rhythm versus rate strategy undertaken.[25] A particularly vexing question for clinicians is how to approach anticoagulation in patients who have never had symptomatic episodes of atrial fibrillation but their arrhythmia has been detected only during interrogation of an implantable pacemaker or defibrillator. To date, there are no prospective trials to guide therapy in such patients, but therapy can be individualized by risk stratification for stroke and duration of episodes.[63]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: