Previous studies reported that the use of natriuretic peptides (NPs) can effectively decrease arrhythmias. However, there is a lack of clinical evidence that recombinant human brain natriuretic peptide (rh-BNP) inhibits postoperative atrial fibrillation (POAF). This cohort aims to assess the effect of rh-BNP on POAF. This study retrospectively reviewed patients who underwent isolated coronary artery bypass grafting from January 2018 to January 2021. Patients were divided into 2 groups according to whether they received rh-BNP therapy within 5 days after surgery. A total of 1153 patients met the inclusion and exclusion criteria, of which 54 received rh-BNP therapy within 5 days. After propensity score matching, 53 patients were treated with rh-BNP, and 148 patients were not treated with rh-BNP. The incidence of POAF was lower in rh-BNP group than non-rh-BNP group (18.9% vs. 37.2%, odds ratio = 0.393, 95% confidence interval, 0.183–0.845, P = 0.017). There was no significant difference in the occurrence of ventricular arrhythmia (P = 0.4), hypotension (P = 0.763), and the risk of death (P = 0.14). rh-BNP could significantly reduce the occurrence of POAF after coronary artery bypass grafting, and rh-BNP did not increase the risk of ventricular arrhythmia, hypotension, and death. Accordingly, rh-BNP could be a potential safe medicine for preventing POAF.