Increased risk of atrial fibrillation was reported in patients with primary aldosteronism. However, data are limited regarding the time-dependent risk of atrial fibrillation in surgically or medically treated primary aldosteronism. From the National Health Insurance Claim database in Korea (2003–2017), a total of 1418 patients with primary aldosteronism (adrenalectomy [ADX], n=755, mineralocorticoid receptor antagonist n=663) were age- and sex-matched at a 1:5 ratios to patients with essential hypertension (n=7090). Crude incidence of new onset atrial fibrillation was 2.96% in primary aldosteronism and 1.97% in essential hypertension. Because of nonproportional hazard observed in new onset atrial fibrillation, analysis time was split at 3 years. Compared with essential hypertension, risk of new onset atrial fibrillation peaked at 1 year gradually declined but remained elevated up to 3 years in overall treated primary aldosteronism (adjusted hazard ratio [aHR] 3.02; P <0.001) as well as in both ADX (aHR, 3.54; P <0.001) and mineralocorticoid receptor antagonist groups (aHR 2.27; P =0.031), which became comparable to essential hypertension afterward in both groups (ADX aHR, 0.38; P =0.102; mineralocorticoid receptor antagonist aHR, 0.60; P =0.214). Nonetheless, mineralocorticoid receptor antagonist group was associated with increased risk of nonfatal stroke (aHR, 1.21; P =0.031) compared with essential hypertension, whereas ADX was not (aHR, 1.26; P =0.288). Our results suggest the risk of new-onset atrial fibrillation remained elevated up to 3 years in treated primary aldosteronism compared with essential hypertension, which declined to comparable risk in essential hypertension thereafter. Monitoring for atrial fibrillation up to 3 years after treatment, particularly ADX, might be warranted.