Incidence and predictors of de novo late-onset persistent atrial fibrillation in postoperative rheumatic heart disease patients (LOAF-RHD study)

Purpose Late-onset atrial fibrillation (LOAF) after valve surgery for degenerative mitral valve disease often with underlying mitral valve prolapse is a known phenomenon. However, there is no similar data for postoperative rheumatic heart disease (RHD) patients. We sought to assess the incidence and predictors of LOAF during postoperative follow-up in RHD patients. Methods This single-center retrospective case-control study included a total of 384 RHD patients with normal sinus rhythm (NSR) who underwent rheumatic valve surgery between 1st July 2008 and 30th June 2013. Patients detected with de novo persistent atrial fibrillation (AF) after 2 months of valve surgery were diagnosed as having LOAF. Presurgical demographic and echocardiographic parameters were compared between the LOAF and NSR groups to identify risk factors for LOAF. Results The incidence of de novo LOAF after rheumatic valve surgery was 9.63% at an average of 2.67 +/- 1.32 years follow-up. Age >= 32 years [OR 2.4 (95% CI 1.2-5.1); P = 0.01] and left atrial (LA) size >= 51 mm [OR 5.9 (95% CI 2.8-12.4); P < 0.0001] were the most significant and independent predictors of LOAF. Moreover, significant mitral valve disease was associated with a higher risk of LOAF than significant aortic valve disease (P = 0.037). LA size >= 51 mm at surgery showed a fair discriminative power [AUC = 0.75; sensitivity = 68%, specificity = 70%] to identify patients at high risk for LOAF. Conclusions Late-onset AF develops in almost a tenth of the RHD patients postoperatively following corrective valve surgery. Preoperative LA size can be used to identify patients at high risk for LOAF.