LAUSR

Functional mitral regurgitation (FMR) is a common problem, especially in patients with heart failure. Conventional FMR caused by left ventricular (LV) dilatation or LV dysfunction and mitral valve tethering (termed ventricular FMR (VFMR)) is widely recognised. However, with the ageing of the heart failure population, the concept of atrial FMR (AFMR) caused by left atrial (LA) enlargement but with normal LV size and function was recently proposed. Although AFMR generally occurs in patients with atrial fibrillation (AF) and/or heart failure with preserved ejection fraction (HFpEF), little is known about its pathophysiology, epidemiology, prognosis and treatment options. Sébastien Deferm and colleagues report the efficacy of mitral valve annuloplasty (MVA) for patients with AFMR (n=97) compared with those with VFMR (n=119). They defined AFMR as mitral regurgitation (MR) caused by isolated annular dilatation in the absence of intrinsic valvular and LV disease, with invariably normal LV volume and global/ regional systolic function (ie, LV ejection fraction (LVEF) ≥50%). VFMR also included patients with subvalvular leaflet tethering, with reduced LVEF (<50%) and/or global or regional alterations in LV geometry. They showed that patients with AFMR were typically women, with a history of AF and with a larger LA size compared with patients with VFMR. During a followup of 3.3 years, prognosis after MVA for treatment of AFMR was better than that after MVA for treatment of VFMR, as reflected by lower allcause mortality and recurrence of moderate or greater MR independent of baseline differences. This is the first study to show the longterm efficacy of MVA for AFMR. Since first described by Gertz et al, AFMR has been widely discussed as a new concept. However, even the definition of AFMR remains ambiguous. According to this issue of Heart and several studies, the current mainstream definition of AFMR is that it is caused by isolated dilatation of the mitral annulus, but with a normal LV volume and systolic function (LVEF >50%). 3 There are also various reports on the pathophysiology of AFMR (figure 1). Gertz et al and several other studies described isolated mitral annulus dilatation resulting from LA enlargement as the main cause of AFMR in patients with AF. HFpEF may also produce a similar situation of AFMR via LV diastolic dysfunction and mitral annular dilatation. Kim et al reported that insufficient leaflet growth in response to annular stretch (termed the annulus annular–leaflet imbalance) is the next main cause of AFMR. Patients with AFMR showed significantly smaller mitral leaflets compared with mitral annulus, and the degree of leaflet remodelling was associated with the severity of AFMR. Another possible mechanism of AFMR, termed atriogenic leaflet tethering, was also reported. The posterior portion of the mitral annulus is attached to the junction of the LA and free wall of the LV, which is stretched outward when the LA and mitral annulus enlarge. Expansion of the LA wall leads to deviation of the posterior annulus towards the outside of the myocardium, causing tethering of the posterior leaflet. However, there are many unanswered questions about the pathophysiology of AFMR, such as which patients with AF and/or HFpEF have AFMR and which patients with AFMR have concomitant tricuspid regurgitation (TR). Many of these questions may eventually be answered by genomic studies in the future. The prevalence and prognostic implications of AFMR are also poorly understood. Although the prevalence varies widely between studies, this may relate to differences in the methods used for MR grading and the study population. AFMR was present in 4%–8% of patients with AF and the prevalence increased with duration of AF—up to 28% in patients with AF for >10 years. Moreover, the prevalence of AFMR increased up to 53% in patients with HFpEF. With respect to prognostic implications, we previously reported the prognosis of hospitalised patients with heart failure with AFMR. Patients with AFMR had a significantly higher prevalence of adverse events during the followup period, especially rehospitalisation for heart failure, compared with patients with no/mild MR. Furthermore, there were no differences in the composite endpoints of cardiac death and