Abstract Background Atrial fibrillation (AF) is the more frequent arrhythmia in clinical practice and probably one of the more analyzed. Despite this, the clinical pattern classification not always correlates with the clinical course of AF. There are still a lack of biomarkers to predict recurrence and the clinical course of this disease. Purpose ST2 soluble (ST2S) biomarker has being previously analyzed by our group in patients with AF and electrical cardioversion (ECV) demonstrating utility to predict recurrence. We sought to analyze the clinical outcome of patients with AF who underwent an ECV or pulmonary vein isolation (PVI) and its correlation with the ST2S biomarker at mid-term follow-up. Methods We performed a prospective, observational clinical trial that included 250 patients with AF who were referred to our hospital for a clinical procedure including ECV or PVI from September 2016 to 2019. A total of 40 matched controlled patients were also included for the initial analysis. ST2S was analyzed from blood samples at baseline, 3- and 6-months follow-up. Results From 250 patients with AF: 94 underwent an ECV and 156 patients an PVI. Mean age was of 58.5 ±10.4 years. From the 156 PVI: 68 had paroxysmal and 88 persistent AF. Clinical follow-up was continued for 1 year and recurrence was of 65.9% in ECV and 22.4% in PVI patients. Clinical and echocardiographic characteristic are described in table1. The initial value of the ST2S biomarker was higher in AF cases with respect to controls (p<0.001). The patients with paroxysmal AF had higher levels of ST2S.0 (12113.4 pg/mL) respect to control cases but with no significant difference (p=0.234). Comparing the values of ST2S in persistent AF patients (15634.8 pg/mL) with controls the value was higher with a significant difference. ST2S baseline values were compared according to the rhythm and patients in AF had significantly higher levels with respect to those in sinus rhythm (SR). ST2S baseline values were also compared according to heart rate (HR) and those in tachycardia had higher levels respect to those who didn't(Table2). When analyzing ST2S levels in the complete cohort of patients with AF according to recurrence, there were no significant differences (Figure1). In multivariant analysis only AF rhythm and LA index volume were significant predictors of recurrence at 3 months follow-up and AF rhythm and age were predictors at 6 months follow-up. Conclusions ST2S baseline values were higher in persistent AF with respect to paroxysmal AF and only persistent AF patients had higher levels compared with controls. Initial ST2S biomarker levels did not have a predictive value in the global cohort of AF but do correlate with the type of AF and AF rhythm. With these results we consider that even though there is an association with AF and ST2S biomarker levels, when analyzing independently to the type of procedure ST2S values were not predictive of recurrence at mid-term follow-up.