Predictors and outcomes of late atrial fibrillation following coronary artery bypass grafting

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background Recent studies have indicated that postoperative atrial fibrillation (POAF) is not a transient phenomenon; rather, it is associated with an increased risk of late AF following coronary artery bypass grafting (CABG). While predictors and outcomes of POAF are more well-studied, evidence is lacking concerning late AF after CABG. Purpose This study evaluated the predictors and five-year outcomes of late AF occurrence following CABG. Primary outcomes were all-cause mortality and major adverse cardio-cerebrovascular events (MACCEs, including death, acute coronary syndrome (ACS), cerebrovascular events (CVE), repeated revascularization, and heart failure (HF) admission). Secondary outcomes were all ACS, CVE, and HF admissions. Method From 2012 to 2016, 5175 patients without a previous history of AF were included. Multiple predictors of late AF occurrence were identified in a competing risk setting with death as a competing event. The independent association of late AF with the outcomes was assessed using Cox proportional hazard models. Results During a median follow-up of 60 [59.3-60.7] months, 85 (1.64%) patients developed late AF. The following best subset was identified as predictors of late AF occurrence: age (Subdistribution-HR (SHR):1.04, CI: 1.02-1.07), off-pump CABG (SHR:1.8, CI: 0.96-3.37), length of stay (SHR: 1.02, CI: 1.01-1.04), and POAF (SHR: 4.14, CI: 2.58-6.64). Late AF was not associated with all-cause mortality and MACCE at either the unadjusted or adjusted level (aHR: 0.90, 95%CI: 0.56 – 1.43, P: 0.652; aHR: 0.77, 95% CI: 0.50-1.18, P:0.227, respectively). However, it significantly increased the unadjusted risk of CVE (HR: 2.28; 95% CI: 1.07 – 4.87), which got weaker after adjustments with a trend towards statistical significance (aHR: 1.70; 95% CI: 0.78 – 3.68). Furthermore, late AF demonstrated some evidence of a trend towards higher HF admission at the univariable level (HR: 2.59; 95%CI: 0.81 – 8.32), which disappeared after adjustments (aHR: 2.05, 95%CI: 0.64-6.61). Conclusions POAF was identified as the most hazardous predictor of late AF occurrence, increasing its odds by more than four times. Therefore, POAF should not be considered a benign entity following CABG. Late AF was not associated with all-cause mortality and MACCE. Nevertheless, given the small number of events leading to possibly reduced power, we found weak evidence of its association with increased risk of CVE and HF admission. Future highly-powered studies can further elucidate these associations. Taken together, more intense monitoring for detecting late AF occurrences, particularly among patients with prior POAF, seems imperative. Surveillance for late AF occurrence can help identify patients who developed HF and those at higher risk of stroke sooner to consider the timely initiation of anti-coagulation and guideline-adherent HF therapy.