• atrial fibrillation
• coronary artery disease surgery

Despite advances in surgical techniques, atrial fibrillation (AF) continues to be the most prevalent rhythm disorder after cardiac surgery, including coronary artery bypass grafting (CABG).1 Post-CABG AF typically presents between the second and third postoperative day. It is associated with adverse long-term outcomes including perioperative stroke, hospital length of stay and overall mortality.2 The prevalence of post-CABG AF has not significantly changed over the past two decades, despite advances in surgical techniques, and it continues to be the subject of ongoing research.

In their Heart manuscript, Filardo and colleagues report on the incidence of post-CABG AF using data collected from five high-volume surgical centres in the USA. The authors found that the unadjusted incidence of post-CABG AF was around 29.5% and 33.1% after adjustment for comorbidities captured using the Society for Thoracic Surgery score. The incidence of post-CABG AF did not significantly change over the study period which spanned from 2002 to 2010. The median time of onset of AF after CABG was 52 hours and the first episode lasted for an average of 7.2 hours.3

The study found some expected differences in the prevalence of cardiovascular comorbidities in patients who did and those who did not develop post-CABG AF. These include age, male sex, white race, hypertension, vascular disease, smoking, cerebrovascular disease and others. These same risk factors are associated with AF and stroke in the general population. Obesity and sleep apnoea have also been linked to AF in the general population. Weight loss has been shown to reduce the burden of AF in obese patients. It is notable that no difference in body mass index was seen in this cohort between patients who developed post-CABG AF and those who did not.

The role of drug prophylaxis using beta blockers and antiarrhythmic drugs including amiodarone has been evaluated in patients undergoing CABG deemed to be at an increased risk of postoperative AF. Both were shown to decrease incident postoperative AF with associated decrease in hospital stay and postoperative complications.4 Patients with ischaemic heart disease are commonly treated with beta blockers. Amiodarone use has not gained sufficient traction, likely due to concern over side effects associated with long-term use. Data on whether patients in this cohort study were prophylactically treated with amiodarone and how many were established on beta blocker prior to CABG are not available.

The left atrium has been the focus of attention in identifying triggers and a substrate for AF. On-pump surgery requires cannulation for cardiopulmonary bypass and this can be done through the right atrium or the vena cava. CABG does not typically require significant manipulation of the left atrial substrate. Therefore, the pathophysiology behind the high incidence of new-onset post-CABG AF is not completely understood. It is likely, however, that post-CABG AF shares common mechanisms with AF that is not associated with surgery. These include the firing of electrical triggers, a vulnerable atrial substrate composed of both electrical and structural tissue remodelling, all in the presence of a proarrhythmic autonomic milieu.

In a study by Yaksh et al,5 episodes of AF in a post-CABG population were noted to be preceded by an increase in the heart rate and the frequency of premature atrial depolarisations. The increase in heart rate is typically due to a combination of vagal withdrawal and sympathetic stimulation. This shift in autonomic balance is also accompanied by an increase in heart rate variability.6 Tightly coupled premature atrial beats in this autonomic milieu are more likely to induce AF.

Acute electrical repolarisation changes have been demonstrated to contribute in creating an electrical substrate for AF and have been observed after CABG. A wider dispersion of the atrial effective refractory period has been demonstrated by Soylu and colleagues following CABG.7 This change in the electrical properties of the atrial tissue is an indicator of heterogeneity in atrial electrical excitability, which increases the propensity for fibrillation.

Ischaemic heart disease can contribute to the creation of an atrial fibrotic substrate for AF through atrial ischaemia or infarction. Indeed, human and animal studies show that infarctions involving atrial branches of the right coronary artery and left circumflex branches of the left coronary artery are associated with later onset of AF.8 Elevated inflammatory markers, such as soluble vascular cell adhesion molecule 1, have also been associated with new-onset post-CABG AF indicating a contribution from a systemic inflammatory process.9

Postoperative AF, following either cardiac or non-cardiac surgery, is strongly associated with later recurrences of the arrhythmia, temporally removed from the acute surgical changes discussed above.10 This suggests that autonomic changes, electrical dispersion and a tissue substrate for AF may already exist preoperatively and likely continue to develop postoperatively. Morphological changes including increase in atrial size and P wave prolongation on 12-lead ECGs have been shown to be associated with new-onset AF following CABG.11 Similar associations of P wave duration and atrial size have been made with new-onset, non-postoperative AF.12 These morphological changes reflect an underlying tissue change of which atrial fibrosis is a major component. Clinical risk factors, including advanced age, systemic hypertension, diabetes, ischaemic heart disease, heart failure and sleep apnoea, are also associated with atrial fibrosis in patients with AF.13 Interventions aimed at risk factor control including hypertension, obesity and sleep apnoea have been shown to reduce the burden of AF.14 Whether these interventions, including treating ischaemic heart disease through revascularisation after CABG, reverse or slow the progression of fibrosis in human atria needs further investigation.

In summary, the study by Filardo and colleagues shows a relatively stable rate of post-CABG AF, with the expected associations with clinical risk factors. The likely cause of this stable incidence is that these risk factors are associated with the development of atrial fibrosis which creates a substrate for AF. The role of revascularisation and other treatments aimed at controlling these risk factors in reducing or halting the progression of fibrosis needs further investigation.