Postoperative atrial fibrillation (POAF) has traditionally been considered a minor complication with limited impact on patients undergoing cardiac surgery. Its incidence varies depending on the series and the underlying heart disease treated, ranging between 40-60% for valve surgery patients and 10-30% for patients undergoing myocardial revascularization. Given its frequent reversibility—either through therapeutic measures or spontaneously within two months post-surgery—a definitive causal link with other postoperative complications has yet to be established. However, numerous studies have explored this association beyond the logical connection with increased mortality and a higher incidence of cerebrovascular accidents in the short and medium term.

This meta-analysis represents the most extensive review to date on this topic, encompassing 57 studies over a broad time frame from 1997 to the present (analyzed to determine temporal trends) and including multiple nationalities (although 13 potentially eligible studies were excluded solely for not being in English). Altogether, this study compiles data on 246,340 patients in an analysis with ample statistical power, considering, for the first time, mortality and five additional adverse events in the early postoperative phase of cardiac surgery, as well as mortality and two more adverse events during follow-up. The meta-analysis design is sound from all formal aspects of bias control, though an excessive permissiveness in model heterogeneity stands out—a limitation that authors and reviewers should have acknowledged.

The main findings reveal that POAF is significantly associated with increased postoperative mortality (OR = 1.92), incidence of cerebrovascular events (OR = 2.17), perioperative myocardial infarction (OR = 1.28), postoperative renal failure (OR = 2.74), and a prolonged hospital stay, both in general and in intensive/critical care. During follow-up, POAF was significantly associated with higher mortality rates (RR = 1.54), incidence of cerebrovascular events (RR = 1.21), and persistent/permanent AF (RR = 4.73). In the analysis of temporal trends, the association between POAF and increased postoperative mortality remained stable regardless of the period from which the studies originated. The method of diagnosing POAF, whether through a single ECG or continuous telemetry monitoring, also did not influence its impact on early postoperative mortality. However, POAF was associated with higher early postoperative mortality in patients undergoing myocardial revascularization surgery (OR = 2.4), without reaching statistical significance in combined valve and coronary surgery or isolated valve surgery, although statistical power was more limited in the third subanalysis.

The authors conclude that POAF has a significant association with mortality and the development of complications in the short and long term following cardiac surgery. However, causality for some of these associations remains to be determined in future studies.

COMMENTARY:

This is the most comprehensive review to date on POAF, both in terms of the number of included studies and its statistical power, as well as in its novel investigation of POAF’s association with other complications previously suggested in individual studies.

POAF represents the most common cause of secondary AF. Its development involves various factors: preoperative (e.g., hypertension and, particularly, underlying ischemic or structural/valvular heart disease), perioperative (e.g., surgical trauma, local inflammation, fluid balance alterations, ionic disturbances), and postoperative (e.g., use of inotropic and vasoactive agents, lack of atrial pacing, respiratory complications). Given its high incidence and its demonstrated association with increased perioperative morbidity and mortality and extended hospital stays, the implementation of preventive measures could become one of the most efficient strategies in enhanced recovery recommendations for cardiac surgery patients.

The association between AF and increased perioperative morbidity and mortality, as found in this study, is also a matter of debate. The authors highlight the need to establish causal relationships that could substantiate the observed statistical significance. The pathophysiological link between AF and the increased incidence of cerebrovascular events is undeniable, which, combined with its association with other adverse events and postoperative morbidity, would justify its connection with extended postoperative stays and mortality. Moreover, the presence of a perioperative arrhythmogenic substrate could easily explain why these patients develop higher rates of persistent/permanent AF during follow-up, leading to further cerebrovascular events and mortality. Mortality during follow-up may also be justified by the association between AF and progressively worsening heart failure. However, the associations with renal failure or perioperative myocardial infarction are less clear, suggesting that AF might be a concomitant, non-causal factor, serving instead as a marker rather than a risk factor.

It is also noteworthy to distinguish de novo POAF, which occurs postoperatively, from undiagnosed paroxysmal AF at the time of surgery. This distinction could explain the survival impact in the group of patients undergoing myocardial revascularization surgery compared to those undergoing valve surgery, where silent paroxysmal AF might not have been considered and could act as a confounding factor. Lastly, the lack of superiority of continuous monitoring systems for POAF over daily ECGs is also notable. If continuous monitoring had shown superiority, it alone would have identified the statistical association with the adverse event, in this case, perioperative mortality.

The study by Caldonazo et al. will serve as a reference for future research, particularly to answer some of the questions raised by their analysis: What anticoagulation policy should be followed for patients developing POAF, given its association with future persistent AF, increased stroke risk, and follow-up mortality, as well as the better safety profile of new direct-acting oral anticoagulants? Could a subanalysis of the included studies help develop scoring models to identify patients at higher risk of developing POAF? Should protocols for the prevention of POAF be reviewed and applied systematically or selectively in high-risk patients? Finally, would reducing postoperative AF ultimately have a significant impact on short- and long-term mortality? This would ultimately confirm whether POAF acts as a factor or simply a marker of risk.

REFERENCE:

Caldonazo T, Kirov H, Rahouma M, Robinson NB, Demetres M, Gaudino M, et al; POAF-MA Group. Atrial fibrillation after cardiac surgery: A systematic review and meta-analysis. J Thorac Cardiovasc Surg. 2023 Jan;165(1):94-103.e24. doi: 10.1016/j.jtcvs.2021.03.077.