By Michael H. Crawford, Editor

SYNOPSIS: A large observational study of patients with severe mitral valve regurgitation due to flail leaflets revealed that atrial fibrillation at entry was associated with excess mortality. Surgery to correct regurgitation was associated with better survival vs. medical therapy; however, atrial fibrillation negatively influenced post-surgical outcomes.

SOURCE: Grigioni F, Benfari G, Vanoverschelde JL, et al. Long-term implications of atrial fibrillation in patients with degenerative mitral regurgitation. J Am Coll Cardiol 2019;73:264-274.

Atrial fibrillation (AF) complicating the course of degenerative mitral regurgitation (DMR) is unwelcome but not a clear indication for mitral valve (MV) repair or replacement due to inadequate knowledge. Investigators from the Mitral Regurgitation International Database (MIDA) sought to define the prevalence, clinical context, and prognostic implications of AF in patients with DMR and a documented flail leaflet(s). Patients with functional MR or significant comorbidities were excluded. The primary endpoint of this prospective, observational study was all-cause mortality. The secondary endpoint was death from cardiovascular (CV) causes. The database included 2,425 patients (mean age, 67 years), of whom 1,646 were in sinus rhythm at entry, 317 had paroxysmal AF (pAF), and 462 had persistent or chronic AF (cAF). Entry left ventricular ejection fraction (EF) was 64 ± 10%. Most patients had a normal EF, were asymptomatic initially, and had severe MR (94%).

Over a mean follow-up of nine years, there were 933 deaths, of which 64% were CV-related. New AF and pAF were 24% and 36% at 10 years and were predictive of 10-year survival: sinus rhythm, 74%; pAF, 59%; and cAF, 46% (P < 0.0001) and were independent of all baseline characteristics. Surgery (88% repair) was associated with improved survival independent of baseline characteristics and rhythm (adjusted hazard ratio [HR], 0.26; 95% confidence interval, 0.23-0.30; P < 0.001). However, post-surgical survival was related to AF: 10-year survival in sinus rhythm was 82%; pAF, 70%; and cAF, 57% (P < 0.0001). The authors concluded that detection of AF of any type should prompt consideration of surgery in patients with severe DMR.


Mitral valve surgery for severe DMR with new onset AF is a class IIa-B recommendation if repair at a low risk is highly feasible. The lack of data has kept AF from attaining a class I indication. The results of this study suggest that a class I indication may be warranted now in patients with severe DMR due to a flail leaflet who develop AF. The results of previous observational studies have suggested that flail leaflets alone with significant MR carry a high mortality that can be mitigated by surgery. Indeed, this MIDA registry study confirmed this. Those who underwent surgery (88% repair) exhibited improved survival independent of baseline characteristics and rhythm. However, even after surgery, AF before or after surgery increased 10-year mortality. Perioperative mortality was low at 2% but seemed to be related to AF, too: sinus rhythm HR, 1.67; pAF, 2.1; and cAF, 3.3% (P = 0.16). In aggregate, these data suggest that early surgery before cAF develops should be considered in patients with severe MR and a flail leaflet and that pAF should strengthen this decision. In the MIDA registry, no systematic hunt for AF was performed, but it raises the question of whether there should be one. Since pAF can be asymptomatic, perhaps ambulatory ECG monitoring should be performed in all patients with severe MR and a flail leaflet, especially if there is a reluctance to advise or accept surgery.

There were limitations to the MIDA database analysis beyond its observational nature, where residual confounding cannot be excluded completely. The enrollment period spanned from 1980 to 2005. Certainly, there were advances in surgical techniques during this period. There was a high incidence of systemic hypertension in these patients (40%), which raises the question of what role blood pressure plays in such patients. Also, there were very few ablations or Maze procedures performed (5%), which raises the issue of whether more invasive approaches to AF pre- and post-operation would have altered the results favorably. In addition, fewer patients with AF preoperatively underwent MV repair, which certainly decreased the operative and long-term survival of the ones who underwent valve replacement. Finally, it is unclear whether the presence of AF in patients without flail leaflet(s) and with more moderate MR would affect outcomes. At this time, I am going to add ambulatory ECG monitoring to my evaluation of patients with moderate to severe MR. I will use its presence to bolster my argument for earlier surgery in appropriate patients.