Discussion
Overall AF following cardiac surgery was detected in 24% of patients. POAF (AF following cardiac surgery without history of AF) was detected in 15.5% of patients without history of AF prior to surgery. Among patients with history of AF, AF following surgery was detected in 70%. History of AF and age were independent predictors of AF following cardiac surgery.
The true incidence of POAF following cardiac surgery remains unclear due to variability in methods of detection and definitions of POAF [12]. The majority of studies reported POAF rates between 20-40% following CABG, 40-50% after valve surgery, up to 60% following combined valve and CABG surgery, and as high as 80% following multiple valve surgery [11]. Mariscalco et al. reported that among 17,262 cardiac surgery patients, 4,561 (26.4%) developed POAF, primarily within 2 days of surgery [13] . Bessissow et al. reported that the incidence of POAF among adults aged 45+ ranges from 20% to 40% among those undergoing thoracic or cardiac surgery [14]. In our study, the overall AF following cardiac surgery was 24%, and POAF was 15.5%. This rate (15.5%) of POAF seems to be lower than the rate reported in the literature if we take into account only patients with no history of AF. The discrepancy could be attributed to differences in definition and detection of POAF. Part of patients undergoing cardiac surgeries who are considered to have no history of AF actually have undetected and or asymptomatic paroxysmal AF before surgery as many of them have risk factors for AF. Thus, to overcome this limitation, we decided to investigate the overall AF following cardiac surgery irrespective of history of AF before surgery.
Many risk factors for POAF were reported in different studies. In general, older patients and those with more preoperative comorbidities are more likely to develop POAF. These risk factors were used to create predictive scoring systems and risk models [2]. Risk models are based on known risk factors such as age, race, CHF, a high EuroSCORE rating, COPD, emergency operation, decreased preoperative left ventricular EF, and decreased estimated glomerular filtration rate (eGFR) [13, 15-17]. However, Current risk prediction models for POAF are derived from epidemiological studies and are not based on pathophysiologic mechanisms. Due to that they are moderately accurate at best, these models are infrequently used in clinical settings [9]. Currently, no validated, evidence-based threshold exists to stratify patients according to risk of developing POAF, and there are no published studies that delineate risk factors for POAF based on significantly differentiating factors.
In general, the major risk factors for POAF among both non-cardiac surgery and cardiac surgery patients are male sex, advancing age, CHF history, and hypertension [14,18]. Among cardiac surgery patients, specifically, history of arrhythmias and AF, history of vascular and coronary artery diseases, valvular heart disease, decreased preoperative kidney function, and type of surgery are also implicated [11,13,14,16,19-23]. Our results are consistent with the previous reports in some aspects. In univariate Cox regression analysis, age and history of AF were found to be predictors of AF post-surgery. Type of surgery and hypertension tended to predict AF following cardiac surgery. In multivariate Cox regression analysis, age and history of preoperative AF were independent predictors of AF following cardiac surgery. Indeed, history of AF was previously reported to be strongly associated with POAF complicating cardiac surgery [24]. However, this report emphasizes the importance of cautious interpretation of the data from different studies as some studies included patients with history of arrhythmias and AF. The risk of POAF increases non-linearly with age, and a large retrospective study of nearly 15000 patients who have undergone cardiac surgery showed that POAF incidence increases at higher rates over the age of 55 years [25]. Recently, Eikelboom et al. reported in systemic review and meta-analysis that POAF occurred in 36,988 (23.7%) out of 155,575 patients [9]. Patients with POAF were older than those without it (mean, 68.7 vs 63.7 years). There were no sex differences between the groups. Hypertension, dyslipidemia, diabetes, and smoking occurred equivalently in patients who did and did not develop POAF [9].With respect to intraoperative factors, there was no difference between patients with and without AF post cardiac surgery, except for aortic clamp time which tended to be longer among patients with AF post cardiac surgery (p=0.07). In some studies, procedural factors such as the duration of the aortic cross-clamp time, location of the venous cannulation, and duration of cardiopulmonary bypass have been shown to affect the rates of POAF [4].
Finally, POAF after cardiac surgery was reported to be associated with increased short-term and long-term morbidity and mortality, including an eightfold increase in the risk of subsequent AF, as well as a higher rate of stroke and cardiovascular death [3,7,9,26,27]. In our study, there was no difference in complications between the two groups, except for readmission within 30 days, which tended to be more frequent among patients with AF following cardiac surgery.