What Is "Non-Valvular" Atrial Fibrillation?

With the recent heavy marketing of the relatively new novel oral anticoagulants dabigatran, rivaroxaban, and apixaban, a new marketing phrase has been born: "non-valvular" atrial fibrillation.

What, exactly, is "non-valvular" atrial fibrillation?

Is it atrial fibrillation without any valvular heart disease like a teeny, tiny bit of functional mitral insufficiency? Or should doctors "ignore" the degree of mitral insufficiency when prescribing these medications? What about mitral valve prolapse patients with severe prolapse?

Is it atrial fibrillation without the presence of any prosthetic heart valve? What about a valve ring placed when a mitral valve is surgically "repaired?"

Is it atrial fibrillation without any rheumatic heart disease? What about mild mitral stenosis compared to moderate or severe mitral stenosis? If there's a difference, what valve area should we use to judge safety of prescribing the novel oral antiocoagulants?

Or is it some combination of one or more of these above patient groups?

For doctors who manage patients with atrial fibrillation and are considering if they should offer a novel oral anticoagulant to a patient in lieu of warfarin, this issue is not a trivial question.

To answer some of these questions, we should turn to the RE-LY, Rocket AF, and Atristotle Trials. But these trials offer only minimal guidance to today's practicing physicians.

For instance, in the RE-LY trial, only patients without "History of heart valve disorder (i.e., prosthetic valve or hemodynamically relevant valve disease)" were studied. What, exactly, do they mean by "hemodynamically relevant heart valve disease?" Does any valve qualify or just the mitral valve?

The Rocket AF trial describes their "non-valvular" heart disease patients a bit better as those with "(1) Hemodynamically significant mitral valve stenosis or a (2) a prosthetic heart valve (annuloplasty with or without prosthetic ring, commissurotomy and/or valvuloplasty WERE permitted)" but were mild mitral stenosis patient's included? What, exactly, defined "hemodynamically-significant" mitral stenosis patients?

The Aristotle trial defined their excluded valvular heart disease as patients as those with "moderate or severe mitral stenosis, or conditions other than atrial fibrillation that required anticoagulation (e.g., a prosthetic heart valve). (The reader must assume that surgically-repaired mitral valves were okay, but were they included or excluded from this trial - we're not sure.)

For doctors on the front line of medicine who might want to prescribe these new drugs to their patients, the term "non-valvular atrial fibrillation" seems to mean different things to different people.

Common sense would dictate that any patient with mitral stenosis (of any severity, in my opinion (be it rheumatic or post-surgical) or patients with prior placement of a prosthetic heart valve (either bioprosthetic or mechanical) should not be considered for these agents. But this is just my wild-ass guess. After all, there is no clear consensus on what really defines "non-valvular" atrial fibrillation, especially when we examine the evidence-based data available to doctors on this issue.

But beyond this, as researchers testing new therapies, we should be careful not to coin confusing new marketing terms to describe a complicated constellation of patients. Otherwise, we might risk injuring those we really are trying to help.

-Wes