LAUSR

Background To study the impact of type of atrial fibrillation on outcomes following transcatheter mitral valve repair. The development of atrial fibrillation (AF) in degenerative mitral regurgitation (MR) can be a sign of progression of MR and associated with adverse outcomes. However, the impact of type of AF in patients undergoing transcatheter mitral valve (MV) repair remains uncertain. Methods Patients 18 years or older who underwent TMVR procedure in 2016 and had a concurrent ICD-10 diagnosis of either paroxysmal or non-paroxysmal AF were included from Nationwide Readmission Database (NRD). The association between type of AF and mortality, stroke, readmission (cardiovascular and non-cardiovascular readmissions) and composite outcome (mortality, inpatient stroke or 30-day readmissions) was analyzed using multivariable logistic regression. Statistical Analysis System (SAS) software 9.4 was used to conduct the analysis. Results A total of 913 (weighted N=1,750) TMVR hospitalizations from NRD for year 2016 were included. Of these, 510 (weighted N=995) patients had non-paroxysmal AF and 403 (weighted N=755) had paroxysmal AF. Patients with non-paroxysmal AF were older than paroxysmal AF (82.53 vs. 81.27; P=0.0004). As compared to paroxysmal AF, those with non-paroxysmal AF had comparable odds of composite outcome of stroke, readmission, or mortality (OR 1.31; 95% CI: 0.77-2.23), as well as stroke (OR 0.43; 95% CI: 0.10-1.78), or mortality (OR 0.54; 95% CI: 0.21-1.37), in patients undergoing TMVR. Similarly, no differences were noted in the odds of cardiac readmissions (OR 1.38; 95% CI: 0.83-2.28), non-cardiac readmissions (OR 0.80; 95% CI: 0.49-1.32) and discharge to skilled nursing/short term care (OR 1.24; 95% CI: 0.66-2.36) in those with non-paroxysmal vs. paroxysmal AF. Conclusions Inpatient outcomes and readmissions were similar in patient with paroxysmal and non-paroxysmal atrial fibrillation in this study. Future studies exploring the effect of type of atrial fibrillation on long term outcomes are needed.