**4.1 Study limitations**

364 Perioperative Considerations in Cardiac Surgery

Another antiarrhythmic agent used for the prevention of atrial fibrillation in cardiac surgery patients is sotalol which was shown to reduce the incidence of postoperative atrial fibrillation (Burgess et al., 2006; Crystal et al., 2004) compared to placebo or to other betablocker such as atenolol (Sanjuan et al., 2004), metoprolol (Parikka et al., 1998) or propranolol (Suttorp et al., 1990) but it had no impact on length of hospital stay, risk of strokes, or mortality (Crystal et al., 2004). However, the use of sotalol in postoperative atrial fibrillation is limited because of its significant side effects such as bradycardia and torsade de pointes, especially in patients with electrolyte disturbances. For these reasons, sotalol therapy for atrial fibrillation prevention in cardiac surgery patients is a class IIb indication in

Amiodarone and its beneficial effect in postoperative atrial fibrillation prevention was the subject of various studies and meta-analyses. Amiodarone decreased the incidence of postoperative atrial fibrillation (Burgess et al., 2006; Bagshaw et al., 2006) and significantly shortened the duration of hospital stay, and reduced the incidence of stroke and postoperative ventricular tachyarrhythmia (Burgess et al., 2006; Bagshaw et al., 2006), but not postoperative mortality (Bagshaw et al., 2006). The beneficial effects of amiodarone were observed irrespective of patients age, type of cardiac surgery (coronary artery bypass grafting only or valve surgery with or without coronary artery bypass grafting), and preoperative beta-blocker therapy. At present, amiodarone has a class IIa indication for atrial fibrillation prevention in patients undergoing cardiac surgery as recommended in the

the ESC Guidelines for the management of atrial fibrillation (Camm et al., 2010).

in the ESC Guidelines for the management of atrial fibrillation (Camm et al., 2010).

(Miller et al., 2005).

2006; Crystal et al., 2004) but results are controversial.

contractility, intracardiac conduction, or ventricular repolarisation.

Other pharmacologic agents used in clinical study for the prevention of postoperative atrial fibrillation were digoxin, which was not found to be effective for atrial fibrillation prevention (Kowey et al., 1992) or calcium channel blockers, of which non-dihydropyridines significantly reduced supraventricular tachyarrhythmias in a subgroup analysis of a metaanalysis (Wijeysundera et al., 2003). Hypomagnesaemia is an independent risk factor for postoperative atrial fibrillation. A meta-analysis of randomized trials showed that prophylactic i.v. magnesium reduced the probability of postoperative atrial fibrillation

From the non-pharmacologic interventions investigated for atrial fibrillation prevention in the postoperative setting, prophylactic atrial pacing reduced the incidence of post-operative atrial fibrillation regardless of the atrial pacing site or pacing algorithm used, (Burgess et al.,

Despite this relative large range of prophylactic interventions for postoperative atrial fibrillation, there are subgroups of patients with conditions that limit the use of betablockers or other antiarrhythmic drugs. Among such conditions are cardiac conduction abnormalities or severe left ventricular dysfunction, active bronchospasm. In these patients ivabradine, a selective sinus node inhibitor, could be a viable alternative. Ivabradine is a specic inhibitor of the If current in the sinoatrial node. Consequently, it is a pure heart-ratelowering agent in patients with sinus rhythm, without affecting blood pressure, myocardial

In BEAUTIFUL study, performed in patients with coronary artery disease and left ventricular systolic dysfunction (left ventricular ejection fraction of less than 40%), even if ivabradine failed to change the primary composite endpoint of cardiovascular death, admission to hospital for acute myocardial infarction, or admission to hospital for newonset or worsening heart failure in any of the subgroups analysed, in a subgroup of patients with baseline heart rate of 70 bpm or higher it reduced the incidence of endpoints related to One limitation of our study is the absence of an washout period. About 85% of patients had preoperative beta-blocker therapy and it was not stopped before the randomization. The practice in our department was to routinely continue preoperative beta-blocker therapy without any pause and changing the active principle according to the study group. Another limitation is the fact that about 30% of the patients with previous episodes of atrial fibrillation received prior to the inclusion in the study an antiarrhythmic agent such as amiodarone or sotalol. These limitations would induce a possible underestimation of some results.
