Is There Any Difference in Risk Factors between Male and Female Patients in New-Onset Atrial Fibrillation after Coronary Artery Bypass Grafting?

BACKGROUND: We aimed to investigate the risk factors of post-coronary artery bypass grafting (CABG) atrial fibrillation (AF) in male and female patients without any history of AF, to identify the sex-specific risk factors, and to examine the effect of sex-specific risk factors on the overall population. METHODS: This retrospective study was conducted using the hospital database with 4,758 patients who underwent CABG surgery. Among them, 2,836 patients with complete data participated in this study. The female patients were divided into two groups as patients who developed new-onset AF after CABG and those who did not. The relationship between the patients' variables and risk factors of post-CABG AF was examined. Multiple analysis revealed the independent risk factors for male and female patients, respectively: mean age (odds ratio [OR] = 1.057, OR = 1,076), age over 65 years (OR = 2.156, OR = 2.736), the European System For Cardiac Operative Risk Evaluation Scores (EuroSCORE) (OR = 1.13, OR = 1.218), COPD (OR = 1.589, OR = 1.789), BUN level (OR = 1.026, OR = 1.019), mean cardiopulmonary bypass (CPB) time (OR = 1.007, OR = 1.010), prolonged CPB time (OR = 1.604, OR = 2.309), mean cross-clamp time (OR = 1.009, OR = 1.017), need of inotropic support (OR = 2.249, OR = 2.731), and mean mechanical ventilation time (VT) (OR = 1.026, OR = 1.027).Low left ventricular ejection fraction (LVEF) (OR = 1.419), left ventricular (LV) aneurysm repair (OR = 1.533), carotid artery disease (OR = 1.750), prolonged VT (OR = 1.729), and use of intra-aortic balloon pump (IABP) (OR = 2.436) were found to be the risk factors only for male AF patients.Unstable angina (OR = 1.969), right coronary artery (RCA) disease (OR = 2), prolonged cross-clamp time (OR = 2.152), and the number of grafts per operation (OR = 1.298) were found to be the risk factors only for female AF patients. CONCLUSION: This study suggests that predictors of AF in the overall population may be due to isolated patient groups. Multiple regression analysis and artificial intelligence modelling should be performed on large-scale, isolated groups to make strong AF prediction.