**Strategies for the Prevention of Postoperative Atrial Fibrillation in Cardiac Surgery**

Estella M. Davis1, Kathleen A. Packard1, Jon T. Knezevich1, Thomas M. Baker2 and Thomas J. Langdon2 *1Creighton University School of Pharmacy and Health Professions 2Alegent Health, Cardiovascular and Thoracic Surgery USA* 

### **1. Introduction**

204 Special Topics in Cardiac Surgery

Yeatman M., Caputo M., Narayan P., Ghosh A.K., Ascione R., Ryder I., Angelini G.D.

during off-pump coronary operations. Ann Thorac Surg 2002;73:1411-1417. Yokoyama T., Baumgartner F.J., Gheissari A., Capouya E.R., Panagiotides G.P., Declusin R.J.

Surg 2000;70:1546-1550.

Intracoronary shunts reduce transient intraoperative myocardial dysfunction

Off-pump versus on-pump coronary bypass in high-risk subgroups. Ann Thorac

Atrial fibrillation (AF) occurs in 15% to 50% of patients after cardiac surgery (Bradley et al., 2005; Dunning et al., 2006). Postoperative atrial fibrillation (POAF) most often develops between the second and fifth postoperative day, with a peak incidence in the first two to three days. While POAF can be self-limiting, it may also be associated with hemodynamic compromise, postoperative stroke, perioperative myocardial infarction (MI), ventricular arrhythmias, and heart failure (Echahidi et al., 2008; Kaireviciute et al., 2009). The development of POAF is associated with, on average, an additional hospital length of stay (LOS) of 1 to 1.5 days (Kim et al., 2001; Zimmer et al., 2003). Some studies, however, report that POAF increases hospital LOS by almost 5 days (Aranski et al., 1996; Gillespie et al., 2006). POAF is also associated with higher hospital costs with an average increase of \$10,000-\$12,600 per hospitalization (Gillespie et al., 2006; Aranski et al., 1996).

Practice guidelines for the prevention of POAF in patients undergoing cardiac surgery exist which include the American College of Chest Physicians (ACCP) 2005 POAF Guidelines, the ACCP 2005 Recommendations for the Role of Cardiac Pacing for POAF, the American College of Cardiology (ACC)/American Heart Association (AHA)/European Society of Cardiology (ESC) 2006 Atrial Fibrillation Guidelines, the ACC/AHA 2004 Coronary Artery Bypass Graft Surgery (CABG) Guidelines, the Canadian Cardiovascular Society (CCS) Consensus Conference Statements on AF, and the European Association for Cardio-Thoracic Surgery (EACTS) 2006 POAF Guidelines and updated ESC/EACTS 2010 AF Guidelines (Bradley et al., 2005; Maisel & Epstein 2005; Dunning et al., 2006; Fuster et al., 2006; Eagle et al., 2004; Mitchell et al., 2005; Kerr & Roy, 2004; European Society of Cardiology ([ESC], 2010) (Table 1).

The guidelines are consistent in that they all strongly recommend using beta-blockers to reduce POAF incidence (ACCP 2005 POAF Guidelines Strength A, ACC/AHA/ESC 2006 AF Guidelines and ACC/AHA 2004 CABG Guidelines Class I, Canadian Cardiovascular Society AF/POAF Consensus Class I, and ESC 2010 AF Guidelines Class I). The Surgical Care Improvement Project (SCIP) National Quality Measures also state that all patients undergoing cardiac surgery should receive a beta-blocker during the perioperative period if they were on a beta-blocker prior to arrival (Surgical Care Improvement Project [SCIP] Version 3.0a, 2009). Most institutions have incorporated this requirement into their prospective preoperative order sets for all patients without contraindications to beta-blockers.


ACCP= American College of Chest Physicians, ACC= American College of Cardiology, AHA= American Heart Association, ESC= European Society of Cardiology AF = atrial fibrillation, BB = beta-blocker, CABG = coronary artery bypass graft, CV = cardiovascular, EACTS = European Association for Cardiothoracic

Table 1. International Guideline Recommendations for Therapies for the Prevention of

Surgery, POAF= postoperative atrial fibrillation, pts = patients

POAF in Patients Undergoing Cardiac Surgery


ACCP= American College of Chest Physicians, ACC= American College of Cardiology, AHA= American Heart Association, ESC= European Society of Cardiology AF = atrial fibrillation, BB = beta-blocker, CABG = coronary artery bypass graft, CV = cardiovascular, EACTS = European Association for Cardiothoracic Surgery, POAF= postoperative atrial fibrillation, pts = patients

Table 1. International Guideline Recommendations for Therapies for the Prevention of POAF in Patients Undergoing Cardiac Surgery

Strategies for the Prevention of Postoperative Atrial Fibrillation in Cardiac Surgery 209

HR= heart rate, SD= standard deviation, NADPH= nicotinomide adenine dinucleotide phosphate, CABG= coronary artery bypass grafting, ECM= extracellular matrix, CRP= C-reactive protein, IL-6=

interleukin-6, HSP= heat shock protein, PAI= plasminogen activator inhibitor Fig. 1. Pathogenesis of postoperative atrial fibrillation (Banach et al., 2010)

Though there are no studies examining POAF prophylaxis for patients intolerant of betablockers, effective alternatives include sotalol and amiodarone, depending upon the contraindication. The guidelines further specify that amiodarone may be given as an alternative or considered in patients at high risk for POAF (Fuster et al., 2006; Eagle et al., 2004; ESC, 2010; Mitchell et al., 2005a; Kerr & Roy, 2004). Only the previous 2006 EACTS and Canadian guidelines support the use of magnesium and state that it may be given in addition to other strategies to reduce POAF(Dunning et al., 2006; Mitchell et al., 2005a; Kerr & Roy, 2004). Additionally, the most recent ESC guidelines include consideration of corticosteroids for the prevention of POAF (ESC, 2010).

The practice guidelines also recommend utilization of non-pharmacologic strategies for the prevention of POAF in cardiac surgery patients (Table 1). The most common strategy referred to in the guidelines is cardiac pacing. The most recent 2010 ESC AF guidelines and ACCP statement from 2005 recommend that biatrial pacing should be considered for prophylaxis (ESC, 2010, Maisel & Epstein, 2005). The CCS statement also recommends that atrial pacing with or without a ventricular lead should be considered in patients with symptomatic bradycardia (Class 2A recommendation based on Level A evidence) and that atrial pacing should be considered if a patient is not on a beta-blocker before surgery (Class 2A recommendation based on Level B evidence) (Mitchell et al., 2005a; Kerr & Roy, 2004). Lastly, the CVS guidelines strongly recommend placing temporary ventricular epicardial pacing electrode wires at the time of surgery to allow for backup pacing as necessary (Class 1 recommendation based on Level C evidence) (Mitchell et al., 2005a; Kerr & Roy, 2004). Other non-pharmacologic strategies mentioned in the guidelines include the use of offpump CABG, posterior pericardiotomy, and introperative maze ablation (Mitchell et al., 2005a; Kerr & Roy, 2004; ESC, 2010).

### **2. Pathogenesis of POAF**

The underlying mechanisms for the development of POAF after cardiac surgery are not precisely known, but are thought to be multifactorial (Figure 1) (Banach et al., 2010). It has been proposed that certain causative mechanisms alter atrial refractoriness and slow atrial conduction which results in multiple reentry wavelets circulating within the atria (Baker & White, 2007a). Some of these mechanisms include pericardial inflammation, excessive production of catecholamines, and volume and pressure changes. Numerous predisposing factors such as advanced age, hypertension, diabetes, left atrial enlargement, left ventricular hypertrophy, intraoperative and postoperative factors such as atrial injury or ischemia, are all thought to impact the development of POAF. Once these conditions exist, a triggering event such as premature atrial contraction, electrolyte imbalance, and/or enhanced adrenergic or vagal stimulation initiates POAF. Neurohormonal activation is more widely recognized as a cause of POAF based on studies linking elevated norepinephrine and epinephrine concentrations to the development of POAF (Baker & White, 2007a; Kalman et al., 1995). Hence, the majority of interventions that reduce the incidence of POAF modulate sympathetic and parasympathetic systems or alter cardiac conduction (Table 1).

While the mechanisms involved in the development of POAF are multifactorial, there is increasing evidence that inflammation also plays a role. Such inflammation may be induced by extracorporeal circulation or cardiopulmonary bypass (CPB) with subsequent elevations of C-reactive protein (CRP), interleukin-6 (IL-6), and the complement system (Echahidi et al., 2008; Gaudino et al., 2003; Bruins et al., 1997; Canbaz et al., 2008). Angiotensin II has been

Though there are no studies examining POAF prophylaxis for patients intolerant of betablockers, effective alternatives include sotalol and amiodarone, depending upon the contraindication. The guidelines further specify that amiodarone may be given as an alternative or considered in patients at high risk for POAF (Fuster et al., 2006; Eagle et al., 2004; ESC, 2010; Mitchell et al., 2005a; Kerr & Roy, 2004). Only the previous 2006 EACTS and Canadian guidelines support the use of magnesium and state that it may be given in addition to other strategies to reduce POAF(Dunning et al., 2006; Mitchell et al., 2005a; Kerr & Roy, 2004). Additionally, the most recent ESC guidelines include consideration of

The practice guidelines also recommend utilization of non-pharmacologic strategies for the prevention of POAF in cardiac surgery patients (Table 1). The most common strategy referred to in the guidelines is cardiac pacing. The most recent 2010 ESC AF guidelines and ACCP statement from 2005 recommend that biatrial pacing should be considered for prophylaxis (ESC, 2010, Maisel & Epstein, 2005). The CCS statement also recommends that atrial pacing with or without a ventricular lead should be considered in patients with symptomatic bradycardia (Class 2A recommendation based on Level A evidence) and that atrial pacing should be considered if a patient is not on a beta-blocker before surgery (Class 2A recommendation based on Level B evidence) (Mitchell et al., 2005a; Kerr & Roy, 2004). Lastly, the CVS guidelines strongly recommend placing temporary ventricular epicardial pacing electrode wires at the time of surgery to allow for backup pacing as necessary (Class 1 recommendation based on Level C evidence) (Mitchell et al., 2005a; Kerr & Roy, 2004). Other non-pharmacologic strategies mentioned in the guidelines include the use of offpump CABG, posterior pericardiotomy, and introperative maze ablation (Mitchell et al.,

The underlying mechanisms for the development of POAF after cardiac surgery are not precisely known, but are thought to be multifactorial (Figure 1) (Banach et al., 2010). It has been proposed that certain causative mechanisms alter atrial refractoriness and slow atrial conduction which results in multiple reentry wavelets circulating within the atria (Baker & White, 2007a). Some of these mechanisms include pericardial inflammation, excessive production of catecholamines, and volume and pressure changes. Numerous predisposing factors such as advanced age, hypertension, diabetes, left atrial enlargement, left ventricular hypertrophy, intraoperative and postoperative factors such as atrial injury or ischemia, are all thought to impact the development of POAF. Once these conditions exist, a triggering event such as premature atrial contraction, electrolyte imbalance, and/or enhanced adrenergic or vagal stimulation initiates POAF. Neurohormonal activation is more widely recognized as a cause of POAF based on studies linking elevated norepinephrine and epinephrine concentrations to the development of POAF (Baker & White, 2007a; Kalman et al., 1995). Hence, the majority of interventions that reduce the incidence of POAF modulate

sympathetic and parasympathetic systems or alter cardiac conduction (Table 1).

While the mechanisms involved in the development of POAF are multifactorial, there is increasing evidence that inflammation also plays a role. Such inflammation may be induced by extracorporeal circulation or cardiopulmonary bypass (CPB) with subsequent elevations of C-reactive protein (CRP), interleukin-6 (IL-6), and the complement system (Echahidi et al., 2008; Gaudino et al., 2003; Bruins et al., 1997; Canbaz et al., 2008). Angiotensin II has been

corticosteroids for the prevention of POAF (ESC, 2010).

2005a; Kerr & Roy, 2004; ESC, 2010).

**2. Pathogenesis of POAF** 

HR= heart rate, SD= standard deviation, NADPH= nicotinomide adenine dinucleotide phosphate, CABG= coronary artery bypass grafting, ECM= extracellular matrix, CRP= C-reactive protein, IL-6= interleukin-6, HSP= heat shock protein, PAI= plasminogen activator inhibitor

Fig. 1. Pathogenesis of postoperative atrial fibrillation (Banach et al., 2010)

Strategies for the Prevention of Postoperative Atrial Fibrillation in Cardiac Surgery 211

 <30 6 30-39 12 40-49 18 50-59 24 60-69 30 70-79 36 ≥80 42 History of AF 7 History of COPD 4 Concurrent valve surgery 6

 ΒB 6 ACEI 5

 Preoperative and postoperative -7 Postoperative -11 Preoperative and postoperative ACEI treatment -5

 Potassium supplementation -5 NSAIDs -7

aRisk Groups based on summative total point assignment using predictors from table:

ACEI = angiotensin converting enzyme inhibitor, AF = atrial fibrillation, BB = beta-blocker, CABG = coronary artery bypass graft, COPD = chronic obstructive pulmonary disease, NSAIDs = non-steroidal

Table 2. Multicenter Study of Perioperative Ischemia Atrial Fibrillation Risk Index (Mathew

analysis was published in 2002 by Crystal et al. that included 27 randomized controlled trials with 3,840 patients (Crystal et al., 2002). Use of beta-blocker therapy decreased the incidence of POAF from 33% in the control group compared to 19% in the group receiving beta-blockade. This corresponded to a number needed to treat (NNT) of seven patients. A large retrospective analysis of the Society of Thoracic Surgeons (STS) database containing 629,877 patients, demonstrated a reduction in mortality rate with use of peri-operative betablockers (Ferguson et al., 2002). It has been shown that patients receiving perioperative betablockers have reduced mortality compared to control (3.4% versus 2.8%, OR 0.8, 95% CI 0.78 – 0.82; p<0.001). Efficacy of beta-blockade in the prevention of POAF has been theorized to decrease hospital LOS. However, two beta-blocker trials reporting effect on LOS demonstrated a non-significant reduction in LOS (-0.66 days; 95% CI, -2.04-0.72) (Cybulsky

The importance of beta-blockers is also affirmed by the two to five-fold increase in the incidence of POAF when beta-blockers are discontinued postoperatively (Kalman et al., 1995; Jideus et al., 2000; Ali et al., 1997). The increase in POAF is thought to be caused by

Low risk = Score < 14, Medium risk = Score 14-31, High risk = Score >31

anti-inflammatory drugs, POAF = postoperative atrial fibrillation

Age (Y)

BB treatment

et al., 2004)

Postoperative treatment

et al., 2000; Wenke et al., 1999).

Withdrawal of postoperative treatment

Predictor of POAF after CABG Risk Score Point

Assignment

= aTotal Points

shown to increase the production of proinflammatory cytokines, adhesion molecules, and selectins (Erlich et al., 2006; Boos et al., 2006). White blood cell count may also be a predictor of POAF (Lamm et al., 2006). The degree of inflammation postoperatively can negatively affect atrial conduction and duration of atrial fibrillation (Ishii et al., 2005; Tselentakis et al., 2006).

Oxidative stress has also been implicated in the pathogenesis of atrial fibrillation as the atrial tissue undergoes oxidative challenge during CPB (Rodrigo et al., 2008). Patients with POAF have been shown to have increased acute myocardial oxidation when compared to patients that did not experience POAF (Ramlawi et al., 2007). Specifically, nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, an enzyme associated with the formation of the reactive oxygen species, superoxide, was found to be independently associated with increased risk of POAF (Kim et al., 2008). This may be due to damage of cardiac myocytes through lipid peroxidation, breakdown of cell membrane, decreased mitochondrial function, calcium overload, and apoptosis (Elahi et al., 2008). Because NADPH is activated by numerous mediators including tumor necrosis factor- α (TNF-α) (Griendling et al., 2000), it has been proposed as a link between inflammation and oxidative stress in POAF.

Based on these newly identified pathways, emerging pharmacologic therapies for the prevention of POAF have been under investigation including HMG Co-A reductase inhibitors (statins), renin-angiotensin-aldosterone-system modulators (including angiotensin converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs)), corticosteroids, omega-3 fatty acids, ascorbic acid, N-acetylcysteine, and sodium nitroprusside.

The guidelines suggest additive therapies can be considered for patients at high risk of developing POAF. Risk factors that have been identified to increase the risk of POAF include advanced age, history of atrial fibrillation, COPD, valvular surgery, hypertension, poor left ventricular function, chronic renal insufficiency, diabetes mellitus, rheumatic heart disease, withdrawal of preoperative beta-blockers or ACEIs, and increased aortic crossclamp and CPB time (Mathew et al., 2004; Baker et al., 2007b; Nisanoglu et al., 2007). No simple criteria exist that allow patients to be classified as high risk for the development of POAF. A risk index model (Multicenter Study of Perioperative Ischemia Atrial Fibrillation Risk Index) (Table 2) was developed to identify subjects at high risk for POAF (Mathew et al., 2004). Patients receiving a risk score less than 14 were considered low risk, 14-31 were considered medium risk, and greater than 31 were considered high risk for developing POAF. Comparison of the predictive ability of the model revealed that the incidence of atrial fibrillation was similar in the derivation and validation cohorts across the three risk groups, and the area under the receiver operating characteristic curve applied to the final model was 0.77 (where >0.75 represents a model with good discriminate power). This risk scoring tool has been used to stratify patients into risk groups that may benefit from add-on prophylactic therapy (Barnes et al., 2006).
