Timing of administration of beta-blockers: effects on early de novo postoperative atrial fibrillation

Background Early postoperative atrial fibrillation (ePoAF) is the most common complication after cardiac surgery. Guidelines recommend use of perioperative beta-blockers for ePoAF prevention. Despite these recommendations and improvements of surgical skills and postoperative care, the occurrence of ePoAF did not alter over time. Purpose We examined the effect of pre- and post-operative beta-blocker’ usage on development of early (≤ 5 days after surgery) de novo PoAF. Methods Adult patients (N=415) with coronary artery bypass grafting (N=225), valve surgery (N=135) or a combination of both (N=55) were retrieved from an existing database to investigate the relation between beta-blocker’ usage and de novo ePoAF. Early de novo PoAF was defined as AF occurring ≤ 5 days after surgery. Sub analysis was performed in a subpopulation on patients who received pre- and post-operative beta-blockers and had undergone epicardial mapping of the right atrium in sinus rhythm (SR) (N=125) (Table 1). Unipolar potentials were classified as single, short or long double and fractionated potentials. Unipolar potential voltage, conduction velocity and conduction block (CB) prevalence were measured. Results De novo ePoAF occurred in 135 (33%) patients and mainly (46%) on the second post-operative day. Post-operative beta-blockers were administered in 13 (4%) patients in day 0, 163 (46%) patients in day 1, 103 (29%) patients in day 2, 42 (12%) patients in day 3, 23 (7%) patients in day 4, and 8 (3%) patients in day 5. Preoperative beta-blockers’ usage was not significantly associated with lower prevalence of de novo ePoAF (usage versus no usage 35% versus 27%; odds ratio 1.45 [95% CI, 0.93 – 2.25]). However, postoperative beta-blockers’ usage was significantly associated with lower prevalence of de novo ePoAF (usage versus no usage 20% versus 61%; odds ratio 0.16 [95% CI, 0.10- 0.26]). De novo ePoAF occurred in 22% of the patients using both pre- and post-operative beta-blockers. In the sub analyses of the epicardial mapping data patients without (N=91) and with ePoAF (N=34) were compared. ePoAF was associated with more low voltage areas and CB. Furthermore, patients with ePoAF had more long doubles, and fractionated potentials and subsequently less single potentials. (Figure 1). Conclusion Our results suggest that administration of beta-blockers after, but not prior to cardiac surgery, could be effective in preventing de novo ePoAF. However, this therapy is only moderately effective in preventing ePoAF. Among patients using both pre- and postoperative beta-blockers, patients with ePoAF had more extensive arrhythmogenic substrates prior to cardiac surgery compared to patients who remained in SR.