Demographic, clinical, and echocardiographic factors predicting maintenance of sinus rhythm following first elective cardioversion

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background/Introduction Atrial fibrillation is common. Cardioversion to restore sinus rhythm is successful in over 90% of cases, but reversion to atrial fibrillation within a year occurs in the majority. Although predictors of maintenance of sinus rhythm following cardioversion have been described, to our knowledge no study has looked specifically at the population of patients undergoing first elective cardioversion for atrial fibrillation. Purpose We aimed to examine the demographic, clinical, and echocardiographic parameters that predict the maintenance of sinus rhythm to one year following first elective cardioversion, with the goal of generating a prediction tool to aid patient selection for the procedure. Methods Coding data was used to identify all patients who had undergone elective cardioversion at two centres in central New Zealand over a five-year period. Those with previous cardioversion were excluded. A range of demographic, clinical, and echocardiographic variables were collated, and univariate followed by multivariate analysis conducted to identify predictors of maintained sinus rhythm at one year, evidenced by documented recurrence of arrhythmia in the clinical record. Results 287 patients underwent first elective cardioversion across both sites, restoring sinus rhythm in 86% with 39% in sinus rhythm at one year. Atrial flutter (66 patients) had higher rates of successful cardioversion (96%), and maintained sinus rhythm (64%) than atrial fibrillation (83% and 32%), whilst cardioversion in one centre (52% vs. 31%) and the absence of paroxysmal symptoms (41% vs. 26%) were both associated with higher likelihood of maintained sinus rhythm at one year. Only the presence of atrial flutter (OR 1.34, P<0.001), and cardioversion in one of the centres (OR 1.19, P=0.002) increased the odds of maintained sinus rhythm at one year in multivariate analysis. A range of other demographic (age, gender, BMI), clinical (duration of symptoms, antiarrhythmics, renal function), and echocardiographic (left atrial volume, left ventricular ejection fraction, mitral valve dysfunction) variables failed to predict restoration and maintenance of sinus rhythm. Conclusion In patients undergoing first elective cardioversion, the presence of atrial flutter and site of cardioversion both predict maintained sinus rhythm at one year. A range of other demographic, clinical, and echocardiographic variables shown elsewhere to predict sinus rhythm after cardioversion fail to do so in this study. This negative outcome likely reflects low numbers of patients in sinus rhythm at one year following first cardioversion, significant population heterogeneity, and small effect size of contributor variables. Overcoming this would require a substantially larger dataset, something that could be overcome through the use of a national dataset for patients undergoing cardioversion.