Aims This study sought to explore the relationship between plasma galectin-3 (Gal-3) and persistent atrial fibrillation (PsAF), and investigate whether Gal-3 predicts clinical outcomes in patients with PsAF undergoing catheter ablation. Methods Fifty consecutive PsAF patients without coexisting structural heart disease undergoing first-time catheter ablation and 46 healthy controls were included. Blood samples were collected on admission for analysis of plasma Gal-3. Pre-ablation clinical and laboratory data were also recorded. Persistent atrial fibrillations patients were followed after ablation and AF recurrence was defined as episodes of AF or atrial tachycardia lasting >30 s after the blanking period. Results Plasma Gal-3 concentrations were higher in PsAF patients than in healthy controls (P<0.001). In PsAF group, those with AF recurrence had higher plasma Gal-3 than did those without recurrence (P = 0.007). Both Gal-3 (hazard ratio 1.28, P = 0.006) and left atrial diameter (LAD) (hazard ratio 1.1, P = 0.025) were independent predictors of AF recurrence after ablation. Moreover, adding Gal-3 to LAD had an incremental predictive value for ablation outcomes (global chi(2) of LAD alone: 8.2; LAD and Gal-3 concentrations: 15.7; P = 0.006). Conclusion Plasma Gal-3 concentrations are elevated in PsAF patients without structural heart disease and independently predict AF recurrence after ablation. Plasma Gal-3 concentration may be helpful in identifying appropriate candidates for AF ablation.