Valvular atrial fibrillation: rheumatic mitral valve disease

I have read with great interest the article by Molteni et al.1 It shows that there is a lack of clearness of concepts about valvular/non-valvular atrial fibrillation (AF) in the real world. From my position as cardiac surgeon in Latin America, I would like to express a few thoughts regarding this issue. Most often, valvular AF seen in the operating room is associated with mitral valve (MV) disease. In my practice of cardiac surgery, 75% of the MV operations have rheumatic aetiology. Particular attention should be paid to the fact that nearly three-quarters of them exhibit long-standing persistent AF. Furthermore, most often, the echocardiographic study of the MV shows typically a combined lesion type, stenosis and regurgitation. Scaring leaflet tissue, thickening, shortening, commissural, and chordae fusion are pathognomonic findings for non-active rheumatic MV disease. So, the mere fact of these echocardiographic findings on the MV with concomitant AF may be a strong indicator for valvular AF. On the other hand, when a standard full biatrial cut-and-sew Cox-maze procedure is performed in these cases, the surgeon immediately realises that the atrial myocardium is very different when compared with other MV cases operated on for another MV aetiology (e.g. Barlow disease). Atrial tissue in rheumatic MV disease is thicker and harder. This may be due to atrial fibrosis,2 especially in left atrium.3,4 In fact, there are some evidence that the atrial fibrosis is responsible of AF in these rheumatic cases.5 In addition, the classical concepts by Cox (large macro-re-entrant circuits) and Haissaguerre (pulmonary vein foci) as responsible of maintenance and origin, respectively, of AF become virtually unusable.5 Instead, any atrial point with fibrosis can generate AF. In conclusion, we can highlight several points. First, the aetiology of valvular AF must be understood in the context of rheumatic MV disease. Secondly, success rate for maze and catheter ablation procedures can be lower in these cases because of atrial fibrosis. Thirdly, the echocardiographic study on the MV might be suggestive to define the type of AF regarding valvular/non-valvular aetiology. Conflict of interest: none declared.