Background: Patent foramen ovale (PFO) can unload left atrial pressure via interatrial shunt. We evaluated whether device occlusion of PFO is associated with subsequent risk of heart failure (HF). Methods: We enrolled 4,804 consecutive patients (median age: 68 years, 62.4% male) who underwent transesophageal echocardiography at tertiary medical centers in Korea between 2007 and 2019. The primary outcome was the 4-year risk of HF. The independent association of PFO closure with HF was evaluated using the Cox regression hazard model after multivariate adjustment for demographics, clinical risk factors, and echocardiographic parameters. Additional subgroup analysis was performed according to the presence structural heart disease defined as left ventricular (LV) ejection fraction < 55%, left atrial (LA) size > 35ml or LA volume ≥ 40ml/m 2 , LV mass index ≥ 115g/m 2 , E/e' ≥ 15, or presence of atrial fibrillation. Results: In total, PFO was observed in 981 (20.4%) patients, where 161 (16.4%) of them underwent device closure. Patients with PFO showed a lower risk of HF than those without (2.6 vs. 4.0%, adjusted hazard ratio [HR]: 0.65, 95% confidence interval [CI] 0.45-0.94, p=0.021). PFO closure group showed a significantly higher risk of HF than no closure group (5.5 vs. 1.2%, adjusted HR: 5.59, 95% CI 4.26-7.34, p<0.001) (Figure). In subgroup analysis, PFO closure was significantly associated with HF risk in patients with underlying structural heart disease (9.6 vs. 3.9%, adjusted HR: 2.55, 95% CI 1.95-3.33, p<0.001), demonstrating an increasing trend in proportion to the number of structural abnormalities. However, PFO closure showed no significant association with HF risk in patients without structural heart disease (1.7 vs. 1.5%, adjusted HR: 1.22, 95% CI 0.99-1.50, p=0.054). Conclusion: PFO closure may increase the risk of HF. Therefore, careful patient selection is needed, especially among those with underlying structural heart disease.