Background: Reverse remodeling of the left atrium (LA) following successful pulmonary vein isolation (PVI) in patients with atrial fibrillation (AF) has been well documented. However, mitral regurgitation (MR) recovery after successful PVI has never been demonstrated systematically. The objective of our study was to retrospectively analyze the effectiveness of PVI in patients with AF on recovery of MR using cardiac magnetic resonance (CMR) imaging.
Methods: Prior to PVI, patients underwent a clinically indicated CMR imaging. Post-PVI (6 ± 2 months), patients underwent a follow-up MRI and were classified into two groups-responders (R) and non-responders (NR) to PVI-as assessed by cessation of AF at the end of the prespecified 6-month (14-day "P" sensitive event monitor defined) follow-up period. Furthermore, CMR was used to evaluate the severity of MR (0 to 4+) and to relate changes in MR to LA volumes as well as mitral apparatus geometry. Patients who had mild and higher MR (2+) on baseline CMR and had a post-PVI CMR were selected for final analysis.
Results: Out of the consecutive 122 patients with AF who underwent PVI, 74 patients that had mitral regurgitation on initial CMR were included in the study. Of these74 patients with AF with MR, 52 (70 %) were classified as R and 22 (30 %) were classified as NR. Baseline demographics were similar between the groups. In the subgroup with mild to severe MR, pre vs. post in the R group MR severity significantly improved (mean = 2.3, median = 2.0 vs. mean = 1.0, median = 1.0, p < 0.0001) and was matched by favorable reverse remodeling of the mitral apparatus geometry (annulus = 35 ± 4 vs. 33 ± 3 mm, p < 0.002; tenting area = 175 ± 56 vs.137 ± 37 mm(2), p < 0.003; tenting height = 8 ± 2 vs.7 ± 2 mm, p < 0.02; and tenting angle = 129 ± 10° vs. 131 ± 11°, p = 0.1). However, in the NR subgroup, MR failed to improve (mean = 2.2, median = 2.0 vs. mean = 1.5, median = 1.0, p = NS) and paralleled general failure of mitral geometry reverse remodeling (annulus = 35 ± 4 vs. 35 ± 4 mm, p = 0.2; tenting area = 153 ± 39 vs. 152 ± 34 mm(2), p = NS; tenting height = 7 ± 1 vs. 7.0 ± 2, p = 0.1; and tenting angle = 131 ± 11° vs. 133 ± 10°, p = NS). In those with lesser degrees of MR, favorable remodeling was predicated on responder status to PVI. Similarly, other cardiac dimensions pre- to post-PVI favorably improved in the R group, but not in the NR group.
Conclusion: In those with durable maintenance of normal sinus rhythm (NSR), cardiac reverse remodeling demonstrated by 3D CMR occurs and is matched by marked improvements in MR and mitral apparatus, likely contributing to continued maintenance of NSR.