Outcomes and Management of Stressor-Associated Atrial Fibrillation

Abstract

Stressor-associated atrial fibrillation (AF) is defined as new onset AF occurring in the setting of an acute and reversible physiologic stressor (eg, cardiac surgery, acute infection). Although historically viewed as transient and potentially benign, a growing body of evidence demonstrates that stressor-associated AF is prognostically important, with longterm rates of AF recurrence and AF-related adverse events (eg, stroke, heart failure) approaching those of AF occurring without a stressor. Nevertheless, key aspects of stressor-associated AF management (eg, prevention, surveillance, treatment) remain highly variable. In this state-of-the-art review, we provide a clinically oriented synthesis of the epidemiology, mechanisms, outcomes, and management of stressor-associated AF across the breadth of characterized precipitants. Within this context, we outline a pathophysiologic framework in which stressor-associated AF represents a spectrum spanning variable relative contributions of direct stressor effects, stressor-mediated substrate, and underlying predisposition (ie, modifiable and nonmodifiable AF risk factors). Stressor-associated AF episodes occurring in the setting of a strong underlying predisposition to AF (sometimes termed “provoked AF”) have a higher risk of recurrence, while episodes occurring in response to a profound physiologic stressor with minimal or no underlying predisposition (sometimes termed “reversible AF”) have a lower risk of recurrence. Using postcardiac surgery stressor–associated AF as an exemplar, we summarize recent advances translating greater pathophysiologic understanding (eg, role of local inflammation) into novel clinical interventions capable of reducing stressor-associated AF incidence (eg, posterior pericardiotomy). We summarize available evidence for clinical management of stressor-associated AF and offer a conceptual framework to inform care. Finally, we conclude by outlining key priorities for future research designed to facilitate precision management of stressor-associated AF, including identification of more effective preventive measures, development of models to stratify risk of recurrent AF and AF-related adverse events following a stressor- associated AF episode, and integration of patient-centered outcomes.

CENTRAL ILLUSTRATION Pathophysiology, Clinical Outcomes, and Future Directions in Stressor-Associated AF

This schematic summarizes the mechanisms and consequences of stressor-associated atrial fibrillation (AF). Stressor-associated AF can be conceptualized as a spectrum spanning “reversible AF,” in which stressor effects predominate, to “provoked AF,” in which underlying predisposition plays a larger role. Acute physiological stressors such as surgery, systemic infections, or ischemia act via systemic inflammation, autonomic dysregulation, and myocardial stress to initiate AF through ectopic firing through mechanisms that include: 1) decreasing the effective refractory period; and 2) stimulating delayed afterdepolarizations. Maintenance of AF depends on the presence of a vulnerable atrial substrate and is additionally influenced by stressor-mediated substrate. Recurrence rates of stressor-associated AF approach those of primary (non-stressor-associated) AF. Compared with individuals without stressor-associated AF exposed to the same stressor, patients with stressor-associated AF have increased risk of adverse outcomes including stroke, heart failure, and mortality. Future directions emphasize developing more effective prevention, personalizing management, and integrating patient-centered outcomes. The graphic was created using BioRender.

WHAT IS STRESSOR-ASSOCIATED ATRIAL FIBRILLATION?

The global burden of atrial fibrillation (AF) is increasing, now affecting nearly 60 million people worldwide.¹ In nearly one-third of cases, AF is first diagnosed in the context of an acute physiological stressor such as surgery or infection, and has been variably referred to as “triggered AF,” “precipitated AF,” “secondary AF,” “acute AF,” “provoked AF,” “peri- or postoperative AF,” or “AF occurring transiently with stress” (Table 1).^2–6 Here, we adopt the term stressor-associated AF to describe a new AF episode occurring in close proximity to a reversible and potentially precipitating factor.^2,3 We selected this term over alternatives to acknowledge inherent uncertainty in whether AF detected in the setting of a stressor is truly secondary to the stressor vs preexisting and detected incidentally on account of hospitalization.⁷ Stressor-associated AF contrasts with non-stressor-associated AF or “primary AF” which refers to new onset AF occurring without proximity to an acute precipitating factor. As we will discuss in detail and regardless of the term used, it is important to acknowledge that AF occurring in the context of a stressor is itself a heterogeneous entity with commonly unclear and potentially variable causal contributions of the underlying stressor to the occurrence of AF (Central Illustration). Contemporary guidelines now recognize a broad spectrum of potential stressors, including surgical stressors such as cardiac and noncardiac surgery, as well as nonsurgical (medical) stressors encompassing cardiac (eg, myocardial infarction [MI], myocarditis) and noncardiac (eg, acute infection) conditions, “emergency conditions” (eg, trauma), binge alcohol consumption, drug use (eg, methamphetamine, cocaine), and endocrine disorders (eg, thyroid, adrenal).^2,3

TABLE 1.

Glossary of Terms

Term	Definition
Stressor-associated AFa	New AF episode in close proximity to a reversible and potentially precipitating acute stressor.
Primary/non-stressor-associated AF	New AF occurring without proximity to an acute stressor.
Provoked AF	New AF unmasked in the setting of an acute stressor in the context of a strong predisposition to AF.
Reversible AF	New AF occurring in the setting of a profound physiological stressor in the context of little or no predisposition to AF.

^aOther terms that have been used in the literature to refer to this entity include “trigger-induced AF,” “secondary AF,” “provoked AF,” “precipitated AF,” “acute AF,” “peri- or postoperative AF AF,” or “AF occurring transiently with stress.”

AF = atrial fibrillation.

Although once considered benign and self-limited, a seminal analysis of the Framingham Heart Study showed that rates of recurrent AF and AF-related adverse events including stroke, heart failure, and death following stressor-associated AF approach those of primary AF.^4,5 Subsequent studies have reinforced the significant prognostic implications of stressor-associated AF, showing that individuals who develop stressor- associated AF face a higher risk of adverse outcomes compared with individuals exposed to the same precipitating event but who do not develop AF.^4,8,9 This review will provide an overview of the epidemiology, risk factors, and contemporary preventive and therapeutic management principles of stressor-associated AF, focusing on common surgical and medical stressors including acute infection, sepsis, MI, and toxic conditions like hyperthyroidism and acute alcohol consumption. While the review emphasizes these better characterized stressors, we acknowledge the clinical relevance of additional, lesser studied precipitants such as stimulant use, trauma, pulmonary embolism, and other endocrine disorders, which are not discussed in detail here on account of more limited data yet merit further investigation. Throughout, we will discuss existing knowledge gaps and highlight future avenues toward a precision approach to the management of stressor-associated AF

NARRATIVE REVIEW METHODS

We conducted a narrative review to synthesize current knowledge on stressor-associated AF across diverse clinical contexts (Supplemental Figure 1). Initial sources were identified from contemporary AF management guidelines and consensus statements issued by major cardiovascular societies, including the American Heart Association, European Society of Cardiology, and Canadian Cardiovascular Society. Additional literature was identified through: 1) manual review of reference lists; 2) targeted PubMed and Google Scholar searches using combinations of keywords and MeSH (Medical Subject Headings) terms (“atrial fibrillation,” “new onset,” “precipitated,” “postoperative,” “cardiac surgery,” “noncardiac surgery,” “sepsis,” and “myocardial infarction,” “anticoagulation”); and 3) expert input from study authors. We prioritized prospective studies, large cohort analyses, and systematic reviews or meta-analyses providing important clinical or mechanistic insights. Given heterogeneous terminology used across the existing literature, particular attention was paid to how AF was temporally and etiologically defined. The goal of this review is to distill practice-relevant themes, highlight remaining areas of uncertainty, and contextualize surveillance and management considerations for clinicians encountering stressor-associated AF. This review did not involve original human subjects research, and therefore ethical approval and Institutional Review Board oversight were not obtained.

INCIDENCE OF STRESSOR-ASSOCIATED AF

Analogous to the poststroke population, stressor-associated AF incidence is a function of monitoring frequency, duration, and intensity.^10–13 Consistent with expectations, stressor-associated AF incidence is substantially higher with routine continuous monitoring vs intermittent clinically driven monitoring in surgical and nonsurgical populations. For example, in a meta-analysis of prospective studies assessing stressor-associated AF after noncardiac surgery (27 studies; n = 5,829,758), studies that utilized continuous monitoring reported 7 times the rate of stressor-associated AF compared with those that did not (13.9% vs 1.9%).^14,15 Similarly, studies utilizing extended continuous monitors (eg, patch monitors, implantable loop recorders) have reported markedly higher incidences of stressor-associated AF (Table 2).^13,16,17 Taken together, these findings suggest that AF ascertainment on the basis of intermittent, short-term, or clinically driven monitoring (eg, symptom triggered, 12-lead electrocardiogram [ECG], Holter monitors) may substantially underestimate the true incidence of stressor-associated AF, and reported incidence rates must be interpreted cautiously in light of this limitation. It is plausible that in some cases, stressor-associated AF may actually represent previously undiagnosed paroxysmal AF that is incidentally detected through the heightened clinical monitoring typical of the hospital setting. Such misclassification may inflate the observed incidence of stressor-associated AF and bias event rates toward those typically reported for primary AF. Postoperative AF is a common complication of cardiac surgery, with an overall incidence of approximately 30% (Table 2). The incidence varies according to the type of surgical intervention, occurring in roughly 20% of patients undergoing isolated coronary artery bypass grafting (CABG), increasing to 40% following valve replacement or repair, and reaching up to 50% in cases involving combined CABG and valve procedures.^2,18–23 In contrast, the incidence of stressor-associated AF is lower following noncardiac thoracic surgery (5%-30%), and lowest after noncardiac, nonthoracic surgery (0.4%-15%).^{14,18,22,24–27}

TABLE 2.

Incidence and Longitudinal Recurrence Rates of Stressor-Associated AF

Stressor	Stressor-Associated AF Incidence	1-year AF recurrence rate	1-y to 5-y AF Recurrence Rate	≥5-y AF Recurrence Rate	Additional Notes
All stressors
	15%-40%4,5	20%-32%4,5,47	41%-52%4,5	10 y: 55%-56%4,5	Stressor-associated AF incidence is highest after cardiac surgery.4,5 Cardiac surgery has the lowest recurrence rate relative to other stressors.4,5
Surgical stressors
Cardiac surgery	20%-50%18–22,141,173, 174,184,295,296	• 18%-34%205,245,297 • ILR: 30%-76%21,189,191	• 30%-40%4,20 • ILR: 35%191	45%-60%4,5,20	Stressor-associated AF incidence increases with surgical complexity and is highest for combined valve/CABG procedures (~50%).18,22,141 Recurrence rates highest in 0-3 mo after stressor-associated AF onset.21,189,297
Noncardiac surgery	• Noncardiac thoracic surgery: 10%-50%14,18,22,24–26,176,187,217 • Noncardiac, nonthoracic surgery: 1%-15%14,18,22,27	0%-37%14,26,204	8%-40%4,26	60%-64%4,5	Similar to cardiac surgery, incidence and recurrence rates increase with surgical complexity.14,24,26,298
Medical stressors
Myocardial infarction	2%-21%178,188,192,193,299,300	~20%192	15%-22%192,193	64%5	In selected heart failure populations receiving ILR, long-term incidence of AF after MI was 29%-58%.16,17
Sepsis	6%-46%8,28–30,301	36%-44%9,166	55%9	61%-65%4,5	Stressor-associated AF incidence increases with severity of disease (sepsis: 6%-10%, severe sepsis 10%-22%, septic shock 23%-46%).28–30
Pneumonia (hospitalized)	• Bacterial: 2%-10%32–35,169,302,303 • COVID-19: 3%-12%36–38,304,305 • Influenza: 4%38	Bacterial: 21%169	Bacterial: 33%169	–
Pulmonary embolism	1%-13%307–310	–	–	–
Acute respiratory distress syndrome	10%313	–	–	–
Myo/pericarditis	Myocarditis: 3%-13%306,312 Pericarditis: 4%313	–	Pericarditis: 34%313	–
Thyrotoxicosis	1%-15%114,306,314	23%-51%194,314	44%-60%195,196,199	–	Stressor-associated AF incidence increases with decreasing thyrotropin levels, including those in the normal range.111–113,315,316 Increased duration of both hyperthyroidism and associated AF significantly increases recurrence risk.195,196

AF = atrial fibrillation; CABG = coronary artery bypass grafting; ILR = implantable loop recorder; MI = myocardial infarction.

Sepsis is associated with stressor-associated AF incidence rates between 6% and 40% (Table 2).^{2,3,8,28–30} In addition to the impact of varying rhythm monitoring practices, differences in reported AF incidence in the sepsis population likely also reflect variation in sepsis ascertainment methods (eg, billing codes vs clinical definitions) and underlying illness severity.^8,28–30 In one of the largest prospective studies of patients with sepsis admitted to an intensive care unit (n = 1,782), the incidence of stressor-associated AF rose steeply with worsening hemodynamic parameters and organ function, from 10% to 40% in sepsis vs septic shock, respectively.²⁸ Relative to sepsis, rates of stressor-associated AF following bacterial and viral pneumonia are lower, between 5% and 10%, as demonstrated in studies using radiographic, culture data, and serum and urine antigen testing.^31–38 Notably, one study directly comparing COVID-19 with influenza observed nearly identical rates between the two groups.³⁸

Incidence of stressor-associated AF after MI is estimated around 6% to 11% (Table 2).^13,39–41 However, findings from the ARREST (Continuous Rhythm Monitoring in Patients After Acute Myocardial infaRction andpREServed Lef venTricle Ejection Fraction), CARISMA (Cardiac Arrhythmias and Risk Stratification After Acute Myocardial Infarction), and SMART-MI (Implantable Cardiac Monitors in High- Risk Post-Infarction Patients With Cardiac Autonomic Dysfunction) trials, which used implantable loop recorders for long-term AF detection in highly selected post-MI populations, reported substantially higher rates (28%-58%).^16,17,42

PATHOPHYSIOLOGY TO PROGNOSIS

DYNAMIC RELATIONSHIP OF STRESSOR AND SUBSTRATE.

The pathophysiology of stressor-associated AF is incompletely understood and likely varies by stressor. Nevertheless, evidence suggests that a unifying theme is likely to include acute physiologic changes that induce ectopic activity (initiation) in the context of an atrial substrate sufficiently vulnerable so as to sustain the re-entrant circuits characteristic of AF (maintenance) (Central Illustration).²² While nonmodifiable risk factors (eg, genetic predisposition, age) form the basis of AF risk, other modifiable risk factors (eg, hypertension, obesity) accelerate structural and functional abnormalities that predispose to AF such as atrial dilation and fibrosis (Figure 1).^2,3,43,44 When acute stressors occur, they not only act to induce ectopic activity, but may themselves contribute to enduring proarrhythmic changes during the period of exposure (ie, act as an AF “perpetuator”).^43,44 For example, as discussed in greater detail subsequently, MI can induce both direct stressor effects like ectopic activity and contribute to stressor-induced substrate such as atrial dilation secondary to increased pressure from left ventricular dysfunction.^45,46 Thus, in a patient with both hypertension and MI, modifiable risk factors and stressor-mediated substrate contribute synergistically to pathological remodeling of the atria, which in turn increase future risk of AF.⁴⁷

FIGURE 1. Spectrum of Stressor-Associated AF and the Role of Risk Factors.

This conceptual model illustrates the relative contributions of direct stressor effects, stressor-mediated substrate, modifiable risk factors, and nonmodifiable risk factors across different atrial fibrillation (AF) phenotypes. In addition to direct stressor effects (eg, triggered activity), stressors can result in changes to atrial substrate (stressor-mediated substrate), which may contribute to longer-term AF recurrence risk. In reversible AF, stressor effects predominate with relatively modest contributions of underlying predisposition (ie, modifiable and nonmodifiable AF risk factors), resulting in lower long-term recurrence risk. In provoked AF, the contribution of underlying predisposition is greater, resulting in higher long-term recurrence risk. Optimization of modifiable risk factors can reduce overall predisposition to AF, potentially preventing onset of stressor-associated AF. The bottom panel categorizes contributing factors by mechanism and modifiability. The graphic was created using BioRender.

Within this framework, it becomes apparent that stressor-associated AF, even within the confines of the same stressor, likely represents a spectrum spanning variable relative contributions of direct stressor effects, stressor-mediated substrate, and underlying predisposition. Two ends of the stressor- associated AF spectrum have been previously dichotomized into “provoked” and “reversible” AF (Table 1).^48–50 “Provoked” AF refers to AF unmasked in the setting of an acute stressor in the context of a chronically susceptible atrial substrate with a strong predisposition to AF. Further, in some cases “provoked” AF likely represents primary AF detected fortuitously by virtue of rhythm monitoring inherent to hospitalization. In contrast, “reversible” AF refers to AF occurring in response to a profound physiologic stressor occurring in the context of minimal or no underlying predisposition to AF. Intuitively, following resolution of the stressor, individuals with “provoked” AF are at substantially higher risk of AF recurrence compared with individuals with “reversible” AF.

LOCAL AND SYSTEMIC INFLAMMATION.

Systemic activation of the innate immune system by pathogen-associated or sterile inflammatory signals is an established precursor to non–stressor-associated AF.^51–53 NLRP3 inflammasome activation leads to the upregulation of proinflammatory cytokines, including C-reactive protein (CRP) and interleukin (IL)-6. These cytokines, in turn, alter the expression of membrane ion channels as well as intracellular ion channels known as connexins, which result in promotion of atrial triggered activity, as evident in animal models of sepsis and MI.^51–56 The clinical association of inflammatory markers with stressor- associated AF has been described in acute infection, sepsis, and MI.^28,36,57,58

In contrast to AF associated with medical stressors, studies investigating associations between systemic inflammatory markers (eg, white blood cell count, CRP, and IL-6) with postoperative AF have produced conflicting results.^22,59–65 While white blood cell concentration and peak levels have been correlated with timing of AF onset, more specific markers of inflammation including pre- and postoperative CRP and IL-6 levels have been inconsistently associated with postoperative AF.^{22,59,60,62–65} Notably, a meta-analysis indicated that reducing inflammation with colchicine is associated with reduced postcardiac surgery AF. However, current guidelines do not recommend colchicine prophylaxis due to adverse drug reactions and frequent drug discontinuation.^66–69

Although the contribution of systemic inflammation to stressor-associated AF following cardiac surgery remains incompletely defined, local atrial and pericardial inflammation secondary to accumulation of mediastinal blood in the pericardial space has been consistently observed to produce conduction abnormalities comparable to those observed in systemic inflammatory states.^70–73 These observations provided the theoretical basis for a randomized trial of posterior pericardiotomy for postoperative AF prevention, in which 420 patients undergoing cardiac surgery were assigned to either posterior pericardiotomy or usual care.⁷⁴ The trial found that pericardiotomy halved the risk of both pericardial effusion and stressor-associated AF, leading to the inclusion of posterior pericardiotomy prophylaxis in international guidelines (Table 3, Figures 2 and 3).^2,3,74,75

TABLE 3.

Translation of Stressor-Specific Mechanisms to Directed Preventive Measures

Stressor(s)	Candidate Mechanism	Directed Preventive Intervention
Cardiac surgery	Pericardial inflammation	Posterior pericardiotomy
Cardiac surgery Noncardiac surgery	Atrial distension and ischemia	Optimization of perioperative fluid and blood product management
Cardiac surgery Noncardiac surgery Sepsis	Autonomic dysfunction	Judicious exposure to catecholaminergic vasopressors
Thyrotoxicosis	Direct effects of thyroid hormone on inotropy, chronotropy, and autonomic function	Selective beta-blockers, optimization of thyroid state
Acute alcohol intoxication	Autonomic dysregulation, atrial electromechanical dysfunction	Minimize alcohol intake

FIGURE 2. Prevention, Acute Management, and Longitudinal Considerations for Management of Stressor-Associated AF.

The figure outlines a comprehensive approach to stressor-associated atrial fibrillation (AF), beginning with exposure to acute stressors such as surgery, sepsis, or ischemia. Acute management includes prophylactic strategies (eg, perioperative beta-blockers or amiodarone), and rhythm control tailored to hemodynamic status. Longitudinal care emphasizes individualized anticoagulation decisions based on bleeding and thromboembolic risk, rhythm surveillance strategies according to recurrence risk, and broader management strategies including monitoring for new heart failure, and control of modifiable risk factors such as hypertension, diabetes, obesity, and alcohol. The graphic was created using BioRender. AF = atrial fibrillation; ECG = electrocardiogram.

FIGURE 3. Summary of International Guideline-Based Recommendations for Prevention and Management of Stressor-Associated AF.

Summarization of key recommendations from the American (American Heart Association [AHA]/American College of Cardiology [ACC]/American College of Clinical Pharmacy [ACCP]/Heart Rhythm Society [HRS]), Canadian (Canadian Cardiovascular Society [CCS]/CCHRS), and European (European Society of Cardiology [ESC]) guideline committees regarding prophylaxis, arrhythmia management, anticoagulation, and rhythm surveillance in the context of stressor-associated atrial fibrillation (AF). The strength of each recommendation is denoted according to the issuing body’s classification system: AHA/ESC Class I (strong), IIa (moderate), III (harm); CCS strong (S), weak (W). The graphic was created using BioRender. *If CI to both BB and amiodarone, consider posterior pericardiotomy, i.v. magnesium, colchicine, biatrial pacing (W). †From 2024 AHA/ACC/ACS/ASNC/HRS/SCA/SCCT/SCMR/SVM Perioperative Guidelines. AAD = antiarrhythmic drug; ACS = American College of Surgeons; ASNC = American Society of Nuclear Cardiology; BB = beta-blocker; CCB = calcium-channel blocker; CHRS = Canadian Heart Rhythm Society; CI = contraindication; DCCV = direct current cardioversion; i.v. = intravenous; LAA = left atrial appendage; OAC = oral anticoagulation; SCA = Society for Cardiovascular Angiography and Interventions; SCCT = Society of Cardiovascular Computed Tomography; SCMR = Society for Cardiovascular Magnetic Resonance; SVM = Society for Vascular Medicine.

Epicardial fat, increasingly recognized as a potential contributor to both primary AF and postoperative AF, is emerging as a therapeutic target for stressor- associated AF prevention.^76–81 Epicardial fat is hypothesized to contribute to atrial arrhythmias through 2 primary mechanisms: 1) fibrotic changes caused by local infiltration of adipose tissue into the atrial myocardium; and 2) upregulation of inflammatory cascades driven by paracrine signaling from adipocytes.^82,83 Clinical evidence demonstrating that suppression of epicardial fat activity with botulinum toxin reduces AF recurrence in patients with established AF has motivated study of its potential role in preventing stressor-associated AF.^78,80 The NOVA (Botulinum Toxin Type A [AGN-151607] for the Prevention of Post-operative Atrial Fibrillation in Adult Participants Undergoing Open-chest Cardiac Surgery) study, a randomized trial of 330 patients undergoing cardiac surgery randomized 1:1:1 to intraoperative epicardial fat pad injection with placebo, 125 U, and 250 U of botulinum toxin A (AGN-151607), did not show significant reduction of postoperative AF or rehospitalizations in the overall study cohort, although there was a signal toward potential benefit in older patients undergoing isolated CABG, which merits further investigation.⁷⁹

MYOCARDIAL ISCHEMIA.

Myocardial ischemic syndromes are thought to contribute to stressor- associated AF via ischemia of the atrial myocardium, which disrupts electrical conduction, and left ventricular dysfunction, which increases left atrial pressure and wall tension.^39,84–89 In a prospective study of 109 patients undergoing angioplasty of the right or circumflex coronary arteries (which typically supply atrial myocardium), those with inadvertent atrial branch occlusion following percutaneous intervention were found to have higher rates of periprocedural AF compared with those without atrial branch occlusion (12% vs 0%).⁸⁵ Concordantly, a retrospective study of 42 patients reported higher rates of inferior/right coronary artery MI in those with stressor-associated AF compared with matched referents without AF (86% vs 33%).⁸⁹ This study also demonstrated that elevated left ventricular diastolic pressure was associated with stressor-associated AF independently of systolic function—an observation supported by animal models in which acute increases in atrial pressure induced a reversible vulnerability to AF.^46,89,90 Perioperative factors associated with increases in left atrial pressure including positive fluid balance (and corresponding elevation in N-terminal pro-B-type natriuretic peptide), intraoperative transfusion, and prolonged ventilation have similarly been associated with stressor-associated AF risk.^24,91–94 Accordingly, optimizing fluid and blood product management may represent a modifiable strategy to reduce AF risk in this setting (Table 3, Figure 2).

Unlike the acute ischemic insult caused by coronary obstruction, cardiac surgery is primarily associated with ischemia-reperfusion injury, which promotes proarrhythmic remodeling through mechanisms such as tissue inflammation, mitochondrial dysfunction, and apoptosis.^22,95–97 Histopathological evidence of atrial tissue injury in individuals with postoperative AF following cardiac bypass provided the rationale for a trial of 2,103 patients randomized to on- vs off-pump coronary artery bypass surgery, which found equivalent rates of postoperative AF (25% vs 27%).^95,97–99 Therefore, further study is necessary to better understand putative mechanisms of ischemia-reperfusion injury in cardiac surgery and identify potential targets to mitigate injury (and perhaps reduce risk of stressor-associated AF) during cardiopulmonary bypass.

AUTONOMIC DYSFUNCTION.

Autonomic dysregulation is a well-recognized factor commonly present in multiple AF stressors including sepsis, MI, and postcardiac and noncardiac surgery, among others.^45,100–102 Activation of both the sympathetic and parasympathetic divisions of the autonomic nervous system can induce proarrhythmic changes that induce ectopic activity and alter atrial substrate. 1Sympathetic stimulation increases intracellular calcium release, promoting delayed afterdepolarizations and prolonging action potential duration. Clinically, this is evident in the association of catecholaminergic vasopressors, which elevate sympathetic tone, and their association with stressor-associated AF in sepsis and surgical populations (Table 3).^92,103 Vagal stimulation, meanwhile, shortens atrial refractoriness by enhancing IKACh activity, resulting in heterogeneity in atrial conduction.¹⁰⁴ Studies in patients with AF following MI and cardiac surgery have demonstrated a close temporal correlation between the onset of AF and markers of autonomic dysregulation, providing foundational evidence of their relationship.^45,105,106

TOXICOSES (THYROID HORMONE AND ALCOHOL).

Hyperthyroidism, even at subclinical levels, results in elevation of basal heart rate and increased supraventricular ectopic activity in pulmonary vein myocytes, which together predispose to AF.^107–114 While there is strong evidence of a causal mechanism linking acute alcohol exposure to AF in controlled settings, it has yet to be supported conclusively by large-scale epidemiological research.^115–119 Alcohol intoxication has been shown to acutely induce autonomic dysregulation and atrial electromechanical dysfunction, both of which are associated with AF onset.^120–122

GENETICS OF STRESSOR-ASSOCIATED AF.

Genetic studies have identified common variants associated with both stressor-associated AF and primary AF including PITX2 and NEURL, providing evidence of a partially shared genetic predisposition.^123–128 Furthermore, Kertai et al¹²⁹ demonstrated that applying a polygenic risk score (PRS) developed in a primary AF population improved prediction of postoperative AF in a retrospective cohort of 1,047 cardiac surgery patients.¹³⁰ When added to clinical factors, the PRS increased the model’s discrimination (change in C-statistic 0.04; 95% CI: 0.02–0.06) and calibration, suggesting that genetic predisposition to primary AF also influences risk of stressor-associated AF.¹²⁹ In related work, the inclusion of a primary AF PRS also improved prediction of stressor-associated AF after coronary revascularization (via percutaneous coronary intervention or CABG) compared with a model utilizing clinical factors alone, though the effect was more modest (change in C-statistic 0.007; 95% CI: 0.005-0.01).^131,132 Stressor-associated AF may also be associated with genetic variants distinct from primary AF, such as those identified near LY96, which plays a role in innate immune activation and is thought to mediate the inflammatory response to ischemia-reperfusion injury.^133,134 Further work leveraging larger cohorts of more diverse surgical and medical stressors is necessary to better understand the genetic basis of stressor- associated AF, which may in turn uncover new treatment paradigms (Figure 1).

PREDICTION OF STRESSOR-ASSOCIATED AF AND RELATED OUTCOMES

PREDICTION OF STRESSOR-ASSOCIATED AF.

A robust body of evidence, largely derived from retrospective stressor-specific populations such as those undergoing cardiac surgery or post-MI, demonstrates that established AF risk factors, including advancing age, male sex, hypertension, and heart failure, are associated with the development of stressor-associated AF among exposed individuals.^{22,24,29,40,135} These observations have in turn spurred significant efforts to refine methods for identifying individuals at risk of stressor-associated AF.^{25,28,35,93,136–143} Assessments of validated risk scores used in primary AF including CHA₂DS₂-VASc, APPLE, and HATCH have shown limited performance in cardiac surgery, MI, and pneumonia populations.^{136–139,142,144} In the postoperative population, more complex scores incorporating additional clinical, echocardiographic, or procedural parameters have demonstrated small improvements.^{93,141,145–148} Overall, modest discriminative utility, need for cumbersome calculation, and insufficient evidence for generalizability has limited widespread use of any existing method of stratifying risk for a stressor-associated AF episode.

The use of artificial intelligence (AI)-based approaches, which enable the integration of clinical information with novel data sources like telemetry and ECGs, have shown remarkable performance in the prediction of cardiovascular diseases like structural heart abnormalities and even AF.^149–152 Early work exploring the application of AI approaches to the prediction of stressor-associated AF and AF- related outcomes have established a promising foundation for future investigation.^153–155 Zhang et al¹⁵⁶ demonstrated that an AI model trained on continuous rhythm monitoring from 505 patients with AF and 885 patients without AF could predict the onset of stressor-associated AF within the subsequent 10-minute interval, achieving 93% sensitivity and 97% specificity. In a study of 27,546 patients undergoing cardiac and noncardiac surgery with no history of AF, an AI model incorporating preoperative 12-lead ECG waveforms and patient demographics predicted 7-day stressor-associated AF risk with a C-statistic of 0.83 and a negative predictive value of 99%, outperforming other contemporary models requiring detailed clinical data.^146,154 Future research investigating modeling approaches integrating additional high-dimensional data (eg, imaging, novel biomarkers, and polygenic risk) with clinical variables has potential to enhance stressor-associated AF risk prediction further.^{129,131,157–159} To this end, application of rigorous standards to AI models, built on core principles of generalizability, feasibility, interoperability, bias mitigation, and security, is essential for their successful integration into clinical practice.¹⁶⁰ As with all predictive models, the clinical relevance of stratifying risk of stressor-associated AF will be maximized by coupling predictions with clear downstream actions.

In summary, current strategies for predicting stressor-associated AF are limited by modest performance and require refinement and validation before widespread use in clinical practice.

THROMBOEMBOLIC RISK STRATIFICATION IN STRESSOR-ASSOCIATED AF.

Clinical risk scores such as CHA2DS2-VASc, CHA2DS2-VA, and CHADS-65 widely used in primary AF prognostication are similarly guideline-recommended for thromboembolic risk estimation after stressor-associated AF (Figure 3).^2,3,48 Retrospective data from the postoperative population suggests that greater CHA₂DS₂-VASc scores may be associated with increased risk of ischemic stroke following cardiac surgery, but evidence in noncardiac surgery is less robust.^161–164 Similarly, in medically stressor-associated AF, CHA ₂DS₂-VASc discriminates longitudinal risk of stroke following MI, thyrotoxicosis, and pneumonia, but has not yet shown utility in predicting stroke risk after sepsis.^165–170

CONSEQUENCES OF STRESSOR-ASSOCIATED AF

AF-RELATED ADVERSE CARDIOVASCULAR EVENTS.

Stressor-associated AF carries important prognostic implications, with cardiovascular adverse event rates approaching those observed with primary AF (Table 4).^5,171 Patients with stressor-associated AF, compared with those exposed to the same stressor but without AF, have higher resource utilization (eg, longer hospital stays and higher hospitalization costs) and face greater short- and long-term risk of AF-related major adverse cardiovascular outcomes, including stroke, systemic thromboembolism, heart failure, and mortality (Central Illustration).^{8,9,28,29,40,41,172–181} A post hoc analysis of 3,023 cardiac surgery patients enrolled in the Arterial Revascularization Trial reported cumulative incidence rates of 6.3% vs 3.7% for those with and without postoperative AF, respectively (HR: 1.53; 95% CI: 1.06-2.23), particularly for those with elevated CHA₂DS₂-VASc scores (HR: 4.05; 95% CI: 2.54-6.46).¹⁸² Similarly, analysis of 2,103 cardiac surgery patients in the On Versus Off Bypass trial found a significantly higher rate of 1-year all-cause mortality (5.2% in the stressor-associated AF group vs 2.3% in the no AF group; P < 0.05).¹⁸³ Findings from prospective studies have been confirmed in large retrospective studies including a registry study of 49,264 patients in The Society of Thoracic Surgeons database, which reported that stressor- associated AF after cardiac surgery was associated with a 2-fold odds of mortality (4.0% vs 1.7% for stressor-associated AF and no AF, respectively) with corresponding increases in intensive care unit length of stay (48 hours), hospital length of stay (3 days), and overall hospitalization cost ($9,000).¹⁷² Large meta-analyses of patients with cardiac surgery-associated AF have shown a 2-fold increase in both perioperative and long-term risks of stroke and mortality compared with patients without AF.^173–175 Overall, reported perioperative cumulative incidence rates range from approximately 1.9% to 5.3% for stroke and 1.8% to 6% for mortality, while longterm cumulative incidence rates range from 1.5% to 6.3% for stroke and 6% to 30.2% for mortality.^{100,172–176,182–186}

TABLE 4.

Event Rates and Associations of Stressor-Associated AF With Adverse Outcomes

	Event Rate With Stressor-Associated AFa		Event Rate With No Stressor-Associated AF		Effect of Stressor-Associated AF vs No Stressor-Associated AF
	0-90 d	Long Termb	0-90 d	Long Term	0-90 d	Long Term
Cardiac surgery
Stroke	1.9%-5.3%172,174,182,183,185	1.5%-6.3% 0.75-1.8 per 100 PY173,175,176,182,184,186	1.0%-2.4%172,174,182,183,185	0.4%-4.7% 0.5-1.3 per 100 PY173,175,176,182,184,186	OR 1.53-2.17172,174,182,183,185	HR 1.53-2.0173,175,176,182,184,186
Heart failure	–	2.1-6.6 per 100 PY184,185,281	–	1.1-4.4 per 100 PY184,185,281	–	HR 1.14-1.35184,185,281
Mortality	1.8%-6%172,174,182,183,185	6%-30.2% 4.52 per 100 PY172,174,182,183,185,186	0.7%-3%172,174,182,183,185	3.7%-18% 2.59 per 100 PY20,173,175,182–184,186	OR 1.44-2.04172,174,182,183,185	HR 1.08-1.3620,173,175,182–184,186
Noncardiac surgery
Stroke	2.1%317	1%-2% 1.9-5.6 per 100 PY164,175,176,187,317	0.7%317	0.6%-0.9% 1-1.8 per 100 PY164,175,176,187,317	OR 2.82317	HR 1.3-3.43164,175,176,187,317
Heart failure	–	6.6 per 100 PY281	–	1.7 per 100 PY281	–	HR 2.02281
Mortality	14.1%-15%217,317	4.3-31.4 per 100 PY164,175,187,318	2.1%-5.4%218,317	2.5-9.3 per 100 PY164,175,187,318	OR 1.72-3.36217,317	HR 1.51-2.51164,175,187,318
Myocardial infarction
Stroke	4.5%41	1.9%-15.9%41,319	1.5%41	0.9%-6.3%41,319	HR 3.0141	HR 1.67-2.0541,319
Heart failure	1.1%41	19.9%41	0.4%41	8%41	HR 2.7241	HR 1.7341
Mortality	10.9%- 14.3%41,188,299	18%-21.5% 9-17 per 100 PY40,41,177,299	0.5%-4.7%41,188,299	7.9%-18.7% 3.2 per 100 PY40,41,177,299	HR 1.81-2.3641,188,299	HR 1.31-1.9440,41,177,299
Sepsis
Stroke	2.6%8	2.1%9	0.5%8	2.0%9	OR 2.78	HR 1.229
Heart failure	–	5.9%9	–	3.9%9	–	HR 1.259
Mortality	29%-72%8,28,30,320	46.2%-61%9,28,29	14%-57%8,28,30,320	40%-41.5%9,28,29	OR 1.96-2.258,28,30,320	HR 1.07-3.289,28,29
Bacterial pneumonia (hospitalized)
Stroke	–	2.1% 2.2 per 100 PY169	–	0.8% 0.8 per 100 PY169	–	HR 1.45169
Heart failure
Mortality	17.9%-34.8%32,35	26.3%-49.8%169	2.9%-6.3%32,35	12.1%-28.7%169	HR 1.0432,35	–
Viral pneumonia (hospitalized for COVID-19 or influenza)
Stroke	1.6%-4.0%36,38	–	0.6%-1.3%36,38	–	RR 2.736,38	–
Heart failure	–	–	–	–	–	–
Mortality	COVID: 45.2%-54.3%36–38 Influenza: 16%38	–	COVID: 11.9%-37.2%36–38 Influenza: 10%38	–	COVID: RR 1.56-1.7736–38 Influenza: RR 1.7838	–
Pulmonary embolism
Stroke	–	–	–	–	–	–
Heart failure	–	–	–	–	–	–
Mortality	0%-13%308,310,311,321,323	11.8%-48.4%307,309,310,321	3.5%- 6.1%307,309,310,321,322	11.2%-14.2%307,309,310,321	OR 1.51-1.61307,309,310,321,322	HR 1.34-1.51307,309,310,321
Acute respiratory distress syndrome (mortality)
Mortality	43%311	–	19%312	–	OR 3.08311	–

Compared with individuals exposed to the same stressor without stressor-associated AF. Additional key findings including observations in myocarditis (no known association between stressor-associated AF and transplant-free survival or death),³¹³ pericarditis (no known differences in long-term mortality or stroke in those with stressor-associated AF vs no AF),³¹⁴ and thyrotoxicosis (similar to possibly increased risk of stroke and systemic thromboembolism in AF with untreated hyperthyroidism compared with primary AF).^279,323,324

^aEvent rate refers to incidence rate or cumulative incidence.

^bLong term was defined as ≥1 year following the initial diagnosis of stressor-associated AF.

PY = person-years; other abbreviations as in Table 2.

Outcomes following stressor-associated AF after noncardiac surgery are similarly worse compared with individuals without AF. In a post hoc analysis of 18,117 patients enrolled in the POISE (PeriOperative ISchemic Evaluation) I and II trials, stressor-associated AF was significantly associated with adverse 1-year outcomes, including stroke (5.58 vs 1.54 events per 100 person-years [PY]; HR: 3.43; 95% CI: 2.00-5.90), all-cause mortality (31.37 vs 9.34 events per 100 PY; HR: 2.51; 95% CI: 2.01-3.14), and MI (2.62 vs 0.82 events per 100 PY; HR: 5.10; 95% CI: 3.91-6.64), compared with individuals without AF.¹⁸⁷ Of note, in a meta-analysis of 35 studies of postoperative AF (n = 2,458,010), patients with noncardiac surgery stressor-associated AF may have higher rates of stroke compared with patients with cardiac surgery stressor-associated AF, although the underlying driver of this observed difference is unknown.¹⁷⁵

In a follow-up study of the APEX-AMI (Pex- elizumab in Conjunction With Angioplasty in Acute Myocardial Infarction) trial, which enrolled 5,745 patients with ST-segment elevation MI, stressor-associated AF was strongly associated with 90-day outcomes including mortality (HR: 1.81; 95% CI: 1.06-3.09), heart failure (HR: 2.66; 95% CI: 1.74-4.06), and stroke (HR: 2.98; 95% CI: 1.47-6.04), with similar results observed in population-based studies (Table 4).^41,177,188 Earlier timing of AF onset relative to MI, and higher AF burden, may be associated with higher risk of adverse events.^40,178,179 In the prospective study of 1,782 patients with sepsis discussed previously, stressor-associated AF was associated with higher risk of death (29% vs 14%; HR: 2.10; 95% CI: 1.61-2.73), and likewise has been found to associate with higher risks of heart failure and stroke in registry-based studies and meta analyses.^{8,9,28,29,180,181}

AF RECURRENCE.

Population-based studies, including the Framingham Heart Study, relying primarily on clinically triggered ascertainment (eg, clinically overt AF), have reported AF recurrence rates comparable to those of primary AF, with cumulative incidence estimates of 20% at 1 year, ∼40% to 50% at 5 years, and ∼55% at 10 years (Table 2).^4,5 However, as seen with stressor-associated AF incidence, reports utilizing systematic monitoring for surveillance have shown even higher recurrence rates.^21,189,190 For example, a prospective study of 139 participants with medical and postoperative stressor-associated AF identified a 1-year recurrence rate of 32.3% using periodic 14-day rhythm monitoring compared with 3.0% in the usual care group.⁴⁷ As expected, invasive rhythm monitors yield the highest rates of AF recurrence. In a metaanalysis (8 studies; n = 185) of patients with cardiac surgery stressor-associated AF who received implantable loop recorders, the 1-year recurrence rate was 30.1% at 12 months, though recurrence rates have been as high as 76% in smaller prospective studies.^21,189,191 Data from intensive monitoring studies in the cardiac surgery population have further demonstrated that recurrence tends to occur in closer proximity to stressor-associated AF onset, but recurrence risk persists over subsequent years.^21,189

Even within the limitations of the surveillance methods employed by individual studies, recurrence rates of noncardiac surgery and nonsurgical stressor-associated AF are substantial. A meta-analysis of 13 studies (n = 25,726) evaluating postoperative AF after noncardiac surgery reported recurrence rates ranging from 0% to 37.3%. However, only one small study employed a systematic monitoring protocol, and few included routine outpatient ECG follow-up across the entire study cohort.¹⁴ In a large national registry study of nearly 50,000 Danish patients with infection-associated AF, the 1-year recurrence rate ascertained using high-specificity billing codes was approximately 36%, with variability by infectious etiology, likely reflecting underlying differences in disease severity.¹⁶⁶ Similarly, in a study of the same Danish registry, the 1-year recurrence rate after bacterial pneumonia was 20.5% (95% CI: 19.6%- 21.5%).¹⁶⁹ In studies of stressor-associated AF following MI utilizing comprehensive protocolized follow-up with providers and ECGs, observed rates ranged between 15% and 22%.^192,193 In the context of thyrotoxic AF, recurrence rates have been reported between 25% and 60%, with higher rates associated with longer AF duration prior to achieving a euthyroid state.^194–196

Recurrence rates likely reflect both the varying influence of stressor-induced substrate and the typical risk profiles of individuals exposed to specific stressors (Figure 1). Generally lower AF recurrence rates after cardiac surgery may be due to the localized nature of pericardial and atrial inflammation.^4,5,22,197 Similarly, prompt treatment of hyperthyroidism and maintenance of a euthyroid state is associated with a rapid conversion back to sinus rhythm in most patients.^198,199 In contrast, stressors with higher long-term recurrence rates such as sepsis^9,166 may result in more pronounced and durable stressor effects on atrial substrate, including electrical and structural remodeling from prolonged systemic inflammation.¹⁹⁷ Additionally, the characteristics typical of patients presenting with sepsis may inherently reflect a population with greater baseline susceptibility to AF. In other words, within our outlined pathophysiologic framework of stressor-associated AF, certain stressors (eg, cardiac surgery, thyrotoxicosis) are more likely to result in stressor-associated AF episodes closer to the “reversible AF” end of spectrum, while others (eg, sepsis, MI) are relatively enriched for AF episodes closer to the “provoked AF” end. One clinically recognized variant of stressor-associated “reversible AF” include early AF episodes occurring during the blanking period after ablation. Early AF recurrences within the first 4 to 8 weeks following catheter ablation are hypothesized to involve ablation-induced atrial inflammation and edema,^200,201 and as expected such episodes have not been reliably associated with subsequent AF recurrence.^202,203

Predicting risk of recurrent AF following a stressor-associated AF episode is essential for quantifying stroke risk and guiding individualized decisions regarding rhythm monitoring and long-term oral anticoagulation (OAC).⁷ However, clinical factors traditionally associated with primary AF risk including age and hypertension have demonstrated inconsistent associations with recurrence risk in the context of stressor-associated AF.^{47,193,204,205} For example, while some studies have suggested that older age is associated with recurrence after cardiac surgery (HR per 1-year increase: 1.05; 95% CI: 1.031.06) and MI (HR per 1-year increase: 1.05; 95% CI: 1.03-1.07), other prospective analyses have not confirmed these findings.^{47,193,204,205} The limited predictive utility of clinical risk factors alone was demonstrated in a study of 314 patients with stressor-associated AF after noncardiac surgery, where the authors found that clinical covariates included in the CHARGE-AF score poorly predicted future recurrence risk (C-statistic 0.59).^206,207 Although individuals with stressor-associated AF often exhibit a high burden of cardiac comorbidities, which may diminish the utility of traditional risk models, structural markers of atrial remodeling, particularly left atrial dilation, have shown some association with risk of recurrence.^{47,193,204,205} It is possible that models integrating echocardiographic parameters, additional biomarkers of left atrial distension like N-terminal pro-B-type natriuretic peptide, or other novel data types may improve predictive performance.

PROGNOSTIC IMPORTANCE OF AF RECURRENCE AND AF BURDEN.

Recurrence, when it occurs, is clearly prognostically important. In a retrospective study of 315 patients with noncardiac surgery stressor–associated AF, Hyun et al²⁰⁴ found significantly higher rates of major bleeding (26.9% vs 4.1%) and thromboembolism (11.2% vs 0.8%) among individuals with vs without recurrent AF. Similarly, in a cohort study including 2,089 (19%) patients with stressor-associated AF among 10,723 patients with a new AF diagnosis, Wang et al⁴ found that timevarying AF recurrence was associated with markedly higher risks of adverse outcomes, with adjusted HRs of 4.2 (95% CI: 3.0-5.9), 1.7 (95% CI: 1.1-2.7), and 3.8 (95% CI: 3.2-4.5) for heart failure, ischemic stroke, and all-cause mortality, respectively.

Increased AF burden within a stressor-associated episode, which likely indicates a stronger underlying predisposition to AF, has been associated with greater risk of AF-related adverse outcomes, including AF recurrence.^40,193,208 A study of 278 individuals with MI-associated AF quantified inhospital AF burden with continuous monitoring and measured associations with long-term clinical out-comes.⁴⁰ They found that compared with low burden, an in-hospital burden >10% was associated with a 2-fold increase in the rates of ischemic stroke (2.20 events per 100 PY; 95% CI: 1.05-4.61 vs 4.42 events per 100 PY; 95% CI: 2.21-8.85), heart failure (11.35 events per 100 PY; 95% CI: 8.02-16.04 vs 31.67 events per 100 PY; 95% CI: 23.49-43.70), and allcause mortality (9.00 events per 100 PY; 95% CI: 6.26-12.96 vs 17.41 events per 100 PY; 95% CI: 12.3824.49).⁴⁰ However, future work is needed to assess how AF pattern (eg, paroxysmal vs persistent) and burden may modulate observed effects of AF recurrence on outcomes following stressor-associated AF and affect treatment.^209–211 Taken together, these data suggest that AF recurrence is common (occurring in 20% to 50% of individuals within 1 year), strongly associated with adverse outcomes, and remains challenging to predict.

DEMOGRAPHIC DIFFERENCES IN STRESSOR-ASSOCIATED AF INCIDENCE AND OUTCOMES

In addition to advancing age and male sex, selfreported White race has also been associated with higher risk of AF after cardiac surgery.²⁰⁷ Although the relative contributions of genetic vs environmental factors underlying this association are complex and incompletely understood, findings associating the degree of European ancestry with AF risk among self-reported Black individuals support the presence of a genetic component to stressor-associated AF risk.²¹² Consistent with these observations, self-reported White patients have also been observed to have increased risk of stressor-associated AF in cardiac and noncardiac surgery populations.^212–217 A study of 21,458 patients undergoing cardiac surgery, of which 1,176 (5.5%) were Black, found that White patients were more likely to develop stressor-associated AF than Black patients after propensity score matching for comorbidities, echocardiographic parameters (including left atrial size), and social factors (OR: 1.74; 95% CI: 1.70-1.78).²¹³ Though less well characterized, Hispanic patients may also have a lower risk of postoperative AF after cardiac and noncardiac surgery.^218,219 Despite adjustment for known clinical and social factors, it remains important to note that varying incidence of stressor-associated AF on the basis of race may be partially driven by factors affecting the likelihood of AF diagnosis (eg, monitoring practices, differences in symptom profile, access to clinical care, etc).²²⁰

While the observed incidence of postoperative AF may differ on the basis of race, associations with outcomes are less clear.^213,218,221 In a study of 1,215 patients undergoing elective cardiac surgery (13% Black), operative mortality among Black patients was higher in those with stressor-associated AF (adjusted relative risk [RR]: 5.0; 95% CI: 1.3-13), and Black patients with stressor-associated AF had a 2-fold higher risk of operative death compared with White patients with stressor-associated AF.²²¹ However, Rader et al²¹³ found that although Black patients appeared to have a higher rate of venous thrombosis, there were no differences in other AF-related adverse outcomes such as stroke, heart failure, or in-hospital mortality. Conversely, an analysis of 294,112 patients in the National Inpatient Sample undergoing noncardiac surgery reported that compared with White patients, both Black (OR: 0.70; 95% CI: 0.55-0.89) and Hispanic (OR: 0.62; 95% CI: 0.44-0.86) patients had lower risk of mortality, while nonprivate insurance status and lower income were associated with higher mortality.²¹⁸

The influence of sex on prognosis also remains less defined.^218,222,223 In a retrospective analysis of 21,568 patients undergoing cardiac surgery, approximately 40% of whom developed new onset AF, women had a lower odds of postoperative AF (OR: 0.85; 95% CI: 0.79-0.91) compared with men, but compared with men with postoperative AF, women with post-operative AF demonstrated a 12% greater hazard (95% CI: 1.02-1.23) of mortality.²²² In contrast, a large claims-based study of 294,112 individuals with post-operative AF did not find a significant sex difference in in-hospital mortality.²¹⁸ Clearly, a better understanding of the potential impacts of sociodemographic and other social determinants on the occurrence and prognosis of stressor-associated AF is needed.³⁹

PREVENTION AND MANAGEMENT OF STRESSOR-ASSOCIATED AF

MODIFIABLE RISK FACTORS IN STRESSOR-ASSOCIATED AF PREVENTION AND AF SCREENING.

Optimization of modifiable risk factors that accelerate pathophysiologic structural and functional atrial remodeling is essential to prevent both primary and stressor-associated AF (Figure 1).^{2,3,140,145,224} Such risk factors include heart failure, physical inactivity, arterial hypertension, type 2 diabetes, tobacco use, obesity, alcohol intake, and sleep quality (Figure 2).^2,3 While the sporadic and unplanned nature of medical stressors is such that onset often cannot be anticipated, planned surgical procedures may present an opportunity to target preoperative lifestyle modification efforts (eg, weight loss, smoking cessation) toward individuals at increased risk of stressor-associated AF and AF-related adverse events.

Beyond risk factor modification, there is no clear role for additional ECG screening for subclinical AF prior to surgery outside of current guideline-based preoperative cardiac risk assessment.^225,226 Potential barriers to broad screening in this context include: 1) likely limited diagnostic yield in low-risk patients not meeting existing criteria for ECG screening; 2) limited evidence of clinical benefit of AF detection in such populations based on screening studies in primary AF; and 3) absence of evidence supporting cost-effectiveness.²²⁷

PHARMACOLOGIC PROPHYLAXIS.

Perioperative beta-blockade or amiodarone may be considered to reduce stressor-associated AF incidence in patients undergoing cardiac surgery, and may be of particular importance in individuals at elevated risk of post-operative AF (Figure 2, Table 3).^2,3 For patients undergoing cardiac surgery, beta-blockers significantly reduced the occurrence of AF (16.3 fewer events per 100 person-years; RR: 0.50; 95% CI: 0.42-0.59; 40 studies, 5,650 participants) without significant bradycardia, hypotension, or MI.²²⁸ Although betablockade is similarly efficacious in reducing AF incidence in patients undergoing noncardiac surgery (26 fewer events per 1,000 PY; RR: 0.41; 95% CI: 0.21-0.79; 9 studies, 9,080 participants), guidelines warn against perioperative introduction due to increased risk of adverse events, including mortality.^2,3,225,226 The Perioperative Ischemic Evaluation Study, a clinical trial of 8,351 patients undergoing intermediate-to high-risk noncardiac surgery in which patients were randomized to new initiation of perioperative metoprolol or placebo, reported that beta-blockade initiated in the perioperative period significantly increased the risk of death (3.1% vs 2.1% in the placebo group) and stroke (1.0% vs 0.5%).²²⁹ In a large meta-analysis, beta-blockers significantly reduced the occurrence of AF (RR: 0.41; 95% CI: 0.21-0.79; 9 studies, 9,080 participants) at the expense of bradycardia (RR: 2.49; 95% CI: 1.74-3.56; 49 studies, 12,239 participants) and hypotension (RR: 1.40; 95% CI: 1.29-1.51; 49 studies, 12,304 participants).²³⁰

Substantial evidence supports the use of prophylactic amiodarone in cardiac surgery (Figures 2 and 3). The PAPABEAR (Amiodarone to Prevent Post-Operative Arrhythmias) trial, which randomized 601 patients undergoing nonemergent cardiac surgery to perioperative amiodarone vs placebo, reported a 50% decrease in the rate of atrial arrhythmias (16.1% vs 29.5% in placebo; HR: 0.52; 95% CI: 0.34-0.69).²³¹ The benefit of amiodarone was maintained in a pre-specified subgroup of 344 patients who received preoperative beta-blockade (27 [15.3%] vs 42 [25.0%]; HR: 0.58; 95% CI: 0.34-0.99). The efficacy of amiodarone is further supported by a meta-analysis of 18 trials including 3,295 patients, wherein amiodarone reduced the incidence of AF to 3.2% vs 19.8% in the control group (OR: 0.48; 95% CI: 0.40-0.57).²³² Notably, a meta-analysis of head-to-head trials comparing beta-blockade with amiodarone prophylaxis in patients undergoing cardiac surgery did not show a difference between AF incidence or hospital length of stay between the groups, though mandated withdrawal of beta-blockade in the amiodarone group may have biased these findings.²³³ Increased efficacy of combined therapy with amiodarone and beta-blockers has been suggested by smaller trials.²³⁴ There is currently no established role of amiodarone to prevent stressor-associated AF in noncardiac surgical populations, although smaller studies, including a trial of 130 patients undergoing pulmonary resection in which amiodarone reduced AF incidence from 32% to 14% (relative risk reduction 57%), support its use.^235,236

Although beta-blockers, amiodarone, and posterior pericardiotomy appear to reduce risk of stressor-associated AF after cardiac surgery, a Cochrane review found no association of these preventive interventions with lower risks of stroke, cardiovascular mortality, or all-cause mortality.²³⁷ Furthermore, the use of perioperative AF prophylaxis in the cardiac surgery population remains low, with less than one-third of providers utilizing perioperative beta-blockers or amiodarone in high-risk patients.²³⁸

Data supporting preventive measures for medically stressor-associated AF are comparably limited (Figure 2, Table 3). The CAPRICORN (CArvedilol Post-infaRct survIval COntRolled evaluatioN) trial randomized 1,959 patients with heart failure with reduced ejection fraction and acute MI in the preceding 3 to 21 days to carvedilol or placebo and found that carvedilol significantly reduced the incidence of AF (2.3% vs 5.4% in placebo group; HR: 0.41).²³⁹ While a substudy of the TRACE (Trandolapril Cardiac Evaluation) trial also suggested some benefit of ACE-inhibition in preventing AF after acute MI in patients with heart failure with reduced ejection fraction, larger studies using population-based data have not supported this finding in either heart failure with reduced ejection fraction or acute MI populations.^240–242

Choice of vasopressor in sepsis appears to impact AF risk: in a meta-analysis including 1,462 patients, the addition of vasopressin to catecholamine vasopressors was associated with a lower risk of stressor-associated AF (68 fewer incident episodes of stressor-associated AF per 1,000 PY in vasopressin group vs nonvasopressin group; RR: 0.77).²⁴³ In a trial of 1,679 patients with septic shock randomized to dopamine or norepinephrine, use of dopamine significantly increased the rate of arrhythmias, particularly AF (21% vs 11%), although patients were not excluded based on a prior history of AF. ¹⁰³ In general, however, our understanding of the potential role of approaches to prevent stressor-associated AF in the setting of nonsurgical stressors is limited.

LONGITUDINAL RHYTHM SURVEILLANCE.

Current recommendations for short-term (ie, 30-60 days) rhythm monitoring following an episode of stressor-associated AF after cardiac surgery are based largely on a trial of 523 patients with cardiac surgery stressor–associated AF who were randomized to a rhythm or rate control strategy (Table 3).^2,3,19 The study observed that among patients with a non-AF rhythm from discharge to 7 days, 84.2% in the rate control group and 86.9% in the rhythm control group remained free of AF through 60-day follow-up, indicating that many patients with non-AF rhythm following an episode of AF associated with cardiac surgery will revert to sinus, at least in the short term. However, a growing body of evidence indicates that more than 20% of recurrences following stressor-associated AF occur beyond the initial 30 to 60 days of monitoring, suggesting that this guideline-endorsed period is likely insufficient.^{4,5,20,21,191,205} Given the high long-term recurrence rates of stressor-associated AF, potential underestimation of recurrence rates in the absence of continuous monitoring, and consistent associations between AF recurrence and AF-related adverse events, it is reasonable to consider more intensive monitoring strategies (Figure 2). Such strategies may include periodic ECG, patch monitors, or implantable loop recorders, particularly for patients with stroke risk factors who are not receiving OAC. Given the substantial resource demands of implantable loop recorders and the growing availability of digital health technologies capable of detecting AF (eg, consumer-facing single-lead ECG wearables), evaluating the effectiveness of these tools for monitoring after stressor-associated AF represents an important research priority.^3,244–249 Importantly, although current evidence supports the ability of many consumer and wearable devices to detect AF, real-world positive predictive values remain only moderate, and confirmation with clinical-grade monitoring is generally advisable.^2,3 In the future, more accurate risk stratification tools may better prioritize intensive monitoring (eg, implantable loop recorders, ECG-capable wearable monitoring) toward individuals at highest risk of AF recurrence.

ARRHYTHMIA MANAGEMENT.

Current guidelines support the use of rate control (eg, beta-blockers), or rhythm control (eg, anti-arrhythmic drugs) in hemodynamically stable postoperative patients (Figures 2 and 3, Supplemental Table 1).^2,3 In a meta-analysis of randomized control trials evaluating rate vs rhythm control in cardiac surgery stressor–associated AF (8 studies; 990), there was no difference in AF recurrence at 3 months or perioperative mortality, although overall event rates were low with 37 recurrences and 25 deaths, respectively.²⁵⁰ While there are relatively less data comparing rate and rhythm control in the noncardiac surgery population, rhythm control appears more effective in conversion to sinus rhythm (RR: 1.93), as expected.²⁵¹ In patients with poor tolerance of AF, such as those experiencing hemodynamic instability or heart failure, synchronized cardioversion is recommended. This may be accompanied by consideration of antiarrhythmic drugs and imaging of the left atrial appendage to rule out thrombus before performing cardioversion.^2,3

Recommendations for the specific rate and rhythm control agents in sepsis are limited by a lack of conclusive randomized trial evidence and mixed results of retrospective cohort studies.^252–257 While hemodynamically unstable patients warrant synchronized cardioversion, intravenous medications for stable or semi-stable patients can also be considered.^2,3 Early randomized data suggested that amiodarone improved arrhythmia control with lower rates of drug discontinuation compared with calcium-channel blockers.²⁵⁷ However, contemporary evidence indicates that compared with intravenous amiodarone, intravenous beta-blockers may have equivalent or even superior efficacy in sepsis patients for both heart rate and rhythm control, as well as equivalent or lower rates of adverse events including hypotension.^254–256

In hyperthyroidism-associated AF, selective beta-blockers are recommended, which reduce systemic symptoms (eg, palpitations) and inhibit some degree of peripheral conversion of T4 to T3 (Supplemental Table 1).^2,3,258 In Graves’ disease in particular, thyroidectomy has also been shown to resolve arrhythmias, and may be more effective than radioiodine therapy in achieving arrhythmia resolution in this population.^259,260

Strategies to mitigate long-term risk of AF recurrence after stressor-associated AF are comparably lacking, and future research is needed to evaluate the comparative effectiveness of rate vs rhythm control strategies (medical or catheter based), for both postoperative and medical stressor-associated AF.²⁶¹

ANTICOAGULATION IN POSTOPERATIVE AF.

Although current guidelines state that OAC is reasonable in patients with postoperative AF who are at elevated thromboembolic risk, conflicting results from retrospective studies, and a lack of randomized trial evidence have led to ongoing uncertainty (Figures 2 and 3).^{2,3,20,27,146,161,163,262,263} In a retrospective study of 2,985 patients with postcardiac surgery AF, of whom 612 (20.5%) were discharged on warfarin, OAC was associated with lower adjusted mortality (HR: 0.78; 95% CI: 0.66-0.92) over a mean follow-up of 6 years.²⁰ However, in a populationbased study of 7,368 Swedish patients with postcardiac surgery AF followed for 5 years, OAC initiated within 30 days was not associated with lower risks of stroke (2.0% vs 1.7%), thromboembolism (2.6% vs 1.7%), or mortality (3.0% vs 2.8%) but did increase major bleeding (2.4% vs 1.9%) compared with no OAC. Notably, only one-half of patients remained on OAC after 1 year, and <40% remained on OAC at the completion of follow-up.¹⁸⁴ Furthermore, a propensity-score matched analysis of 38,936 patients with postoperative AF from the Society of Thoracic Surgeons Database found that OAC was associated with higher mortality at both 1 year (5.4% vs 4.2%) and 5 years (21.1% vs 19.5%) compared with no OAC (HR: 1.16; 95% CI: 1.06-1.26), a finding that remained consistent across CHA₂DS₂-VASc score strata.¹⁶³ There was also a trend toward higher risk of thromboembolism at 5 years in the OAC group, though this association was nonsignificant.¹⁶³ In keeping with inconsistent findings, various meta-analyses, including a recently published study including nearly 1.7 million patients, have been inconclusive.^262–264

Studies evaluating the efficacy of OAC in noncardiac surgery stressor–associated AF have been similarly discordant.^{27,146,204,263,265} In a nationwide registry study of 3,830 patients with AF following noncardiac surgery matched to 15,320 patients hospitalized with newly diagnosed nonvalvular primary AF, both groups exhibited similar long-term risks of thromboembolism (3.2 vs 3.0 events per 100 PY) and mortality (13.3 vs 10.9 events per 100 PY), with comparable benefits of OAC.²⁷ Similarly, in a study of 251,837 patients undergoing noncardiac surgery, stressor-associated AF was strongly associated with stroke compared with no AF (3.6% vs 2.3%; RR: 1.60; 95% CI: 1.37-1.87), particularly in patients not receiving OAC (RR: 1.81; 95% CI: 1.44-2.28), but this association was attenuated in the subset of stressor-associated AF patients on OAC (RR: 1.04; 95% CI: 0.71-1.51).¹⁴⁶ However, other studies and meta analyses have conversely suggested that OAC after noncardiac surgery stressor–associated AF is associated with higher bleeding risk and trends toward greater mortality.^204,263,265

The variable effects of OAC seen across studies of patients with postoperative AF may relate to several factors including selection bias and heterogeneous patient populations, confounding effects (eg, OAC utilized in individuals perceived to be at higher risk of adverse outcomes), use of warfarin as the pre-dominant form of OAC (which has been shown to impart less stroke benefit with higher risks of bleeding compared with nonwarfarin anticoagulants in the postoperative setting), and limited data on postdischarge continuation and adherence.^266–268 Overall, the frequency of OAC in the postoperative setting appears low, and substantially lower than that observed among patients with nonsurgical AF with comparable CHA₂DS₂-VASc scores (OR: 0.31; 95% CI: 0.19-0.52).^162,269,270 However, within the limitations of the available literature, the use of clinical risk scores (eg, CHA₂DS₂-VASc, CHA₂DS₂-VA, CHADS-65) to guide anticoagulation in the post-operative population (as in other settings) is recommended, with an emphasis that decision making should be individualized based on the patient’s bleeding risks and preferences (Figures 2 and 4).^{2,3,7,48,161,271} Furthermore, in light of compounding evidence supporting AF recurrence as a strong risk factor for AF-related adverse events, particularly systemic thromboembolism, recurrence should prompt reconsideration of stroke prophylaxis for patients not receiving OAC similar to primary AF.^4,204

FIGURE 4. Conceptual Framework for Anticoagulation and Rhythm Surveillance After Stressor-Associated AF.

Depicted is a conceptual framework to guide surveillance and anticoagulation management following stressor-associated AF. Stroke and bleeding risk assessments should incorporate both patient- and stressor-specific factors. Surveillance intensity may vary depending on the patient’s risk profile and available monitoring strategies. We note that this conceptual framework does not constitute a formal evidence-based management algorithm, and relevant decisions should be made on a case-by-case basis. The graphic was created using BioRender. *See Figure 3 for summary of stressor-specific guideline recommendations for anticoagulation. †Surveillance intensity should also account for risk factors for AF (eg, CHARGE-AF, C2HEST, LA size) and heart failure (eg, obesity, diabetes, valvular disease). ††Management of recurrence in the setting of a second acute episode remains a significant area of uncertainty. Abbreviations as in Figure 3.

The demonstrated efficacy of left atrial appendage occlusion (LAAO) in reducing thromboembolic risk among patients with established AF has led to growing interest in its prophylactic use in individuals without known AF undergoing cardiac surgery.²⁷² In a retrospective propensity-matched analysis of 2,067 patients undergoing cardiac surgery with no prior history of AF, LAAO was associated with a lower incidence of postoperative stroke among those who developed acute AF. However, LAAO was also linked to a significantly higher rate of postoperative AF (0 of 145 events in the LAAO group vs 7 of 115 in the non-LAAO group; P < 0.05; OR for new onset AF: 1.36; 95% CI: 1.03-1.80).²⁷³ While multiple large retrospective studies have corroborated the association between LAAO and increased incidence of post-operative AF (which has been attributed to an acute decrease in left atrial compliance and resultant distension), they have not consistently demonstrated reductions in stroke or mortality.^274–276

Several randomized trials are underway to clarify the role of OAC and prophylactic left atrial appendage closure in cardiac and noncardiac surgery stressor–associated AF including: 1) PACES (Anticoagulation for New-Onset Post-Operative Atrial Fibrillation After CABG; NCT04045665), which is comparing the efficacy of adding OAC to background antiplatelet therapy in stressor-associated AF after cardiac surgery; 2) LEAAPS (Left Atrial Appendage Exclusion for Prophylactic Stroke Reduction Trial; NCT05478304) which is randomizing patients at increased risk for incident AF and stroke to receive prophylactic left atrial appendage ligation or no ligation during cardiac surgery; and 3) ASPIRE-AF (Anticoagulation for Stroke Prevention In Patients With Recent Episodes of Perioperative AF After Noncardiac Surgery; NCT03968393), which is randomizing patients undergoing noncardiac surgery with postoperative AF and high CHA₂DS₂-VASc scores to OAC vs no OAC.

ANTICOAGULATION IN NONSURGICAL STRESSOR-ASSOCIATED AF.

While there is strong evidence supporting the routine use of OAC in AF associated with MI or thyrotoxicosis, retrospective data regarding its use in sepsis and the broader population with infection remain mixed (Figure 3, Supplemental Table 1).^{2,3,7,277,278} In thyrotoxicosis, guidelines generally recommend initiation of OAC according to estimated stroke risk.^2,3 However, given evidence suggesting stroke risk associated with thyrotoxicosis-associated AF may be underestimated by traditional risk scores, consideration of OAC in selected patients with low CHA₂DS₂-VASc scores may be reasonable.^7,279 In a study of 3,992 patients with sepsis-associated AF, of whom 20% filled anticoagulation prescriptions within 30 days, OAC was associated with higher risk of ischemic stroke at 1 year (5.7% vs 2.3%; RD [risk difference]: 3.37%; 95% CI: 0.36-0.38).²⁸⁰ However, there are some data to support anticoagulation in less severe infections: in a nationwide registry study of 24,607 Danish patients with infection-related AF, anticoagulation was strongly associated with decreased risk of thromboembolism (HR: 0.75; 95% CI: 0.68-0.83) and death (HR: 0.53; 95% CI: 0.50-0.56), without substantially increasing bleeding (HR: 1.08; 95% CI: 0.98-1.18).¹⁸¹ The study further suggested that patients with infection-associated AF have higher rates of thromboembolism and death compared with non–infection-related AF irrespective of OAC treatment.¹⁸¹ In light of these findings, concordant with guidelines, anticoagulation should be guided by assessment of individualized stroke and bleeding risk as well as evidence for AF recurrence (Figures 2 to 4).^2,7,277

BEYOND STROKE: THE OVERLOOKED BURDEN OF HEART FAILURE AFTER STRESSOR-ASSOCIATED AF.

Although a strong association between stressor-associated AF and long-term risk of heart failure has been demonstrated in both retrospective and prospective studies, this relationship remains relatively understudied.^{9,41,184,185,281} Available evidence suggests that, similar to primary AF, the incidence of heart failure following stressor-associated AF may exceed that of stroke, with some studies reporting a twofold higher rate of heart failure events compared with stroke.^{9,41,184,282,283} While substantial efforts have focused on quantifying stroke risk and evaluating the role of OAC in stressor-associated AF, comparatively little attention has been directed toward understanding the incidence and short-term risk of heart failure across various stressors, or toward defining optimal strategies for heart failure risk stratification, prevention, and management in this population.

MANAGEMENT CONSIDERATIONS IN STRESSOR-ASSOCIATED AF.

Current evidence remains insufficient to provide specific recommendations for surveillance and treatment of stressor-associated AF. However, based on available evidence and general risk stratification principles, we provide a conceptual framework to inform clinical decision-making regarding longitudinal monitoring and management of stressor-associated AF (Figure 4).

In alignment with international guidelines, stroke risk should be estimated using a validated framework (eg, CHA₂DS₂-VASc, CHA₂DS₂-VA, or CHADS-65), and values equating to a ≥2% annual stroke risk should merit consideration for antithrombotic treatment.^2,3,284 Similarly, bleeding risk may be assessed on an individual basis, incorporating both: 1) patient-level factors such as age, prior bleeding history, concurrent use of antiplatelet therapy, liver and kidney function, and concomitant use of antiplatelet therapy; and 2) any stressor-specific considerations (eg, postoperative bleeding risk, sepsis).^285–287

Decisions regarding OAC initiation should be made within the context of shared decision making founded on: 1) balancing estimated stroke risk and minimizing bleeding risks (including potential transience of elevated bleeding risk); 2) pertinent clinical guidelines; and 3) acknowledgment of the uncertainty that surrounds both the use of standard clinical risk scores and the evidence base for OAC in stressor-associated AF (Figure 4). Given the transient nature of some episodes of stressor-associated AF, and the critical importance of recurrent AF as an indicator of subsequent adverse event risk, surveillance is paramount. Surveillance intensity should be tailored to an individual’s profile of AF, stroke, and heart failure risk factors.^2,3,48,288 Higher risk may warrant more intensive rhythm monitoring strategies ranging from periodic rhythm monitors to implantable loop recorders to facilitate early detection of recurrence. In cases of recurrence, OAC should be re-evaluated in keeping with established approaches to primary AF. Conversely, a prolonged period of freedom from AF recurrence may support a reassessment of the benefits of continued OAC use, particularly in patients with elevated bleeding risk. Although likely uncommon, we acknowledge that recurrent AF episodes that are again stressor-associated (either due to another episode of the same stressor, or a different stressor) represent an area of particular uncertainty.

Baseline risks of AF and heart failure may similarly inform longitudinal arrhythmia management. However, evidence to demonstrate benefit of prolonged rate or rhythm control in preventing AF recurrence or reducing risk of heart failure or other AF-related adverse outcomes following stressor-associated AF remains lacking.

PATIENT-REPORTED OUTCOMES

There is increasing recognition of the important effects of AF on patient-reported outcomes (PROs), which encompass quantitative assessments of symptoms, functional status, and quality of life.^{2,3,289–291} While early evidence suggests that stressor-associated AF may negatively impact PROs, these outcomes remain inadequately characterized, particularly given the high burden of comorbid illness in individuals affected by stressor-associated AF (including the acute stressor itself).²⁹² A more comprehensive understanding of how stressor-associated AF affects functional capacity and mental well-being is important for several reasons. First, data from cardiac surgery populations suggest that, even when baseline characteristics are similar with respect to comorbid disease burden, individuals who develop stressor-associated AF experience marked reductions in physical and social function and report greater postoperative pain than those without AF.²⁹² These findings support the potential value of individualized rehabilitation and symptom management strategies following discharge. Second, evidence from the care of primary AF illustrates the clinical utility of PROs in informing treatment decisions, as reflected in consistent evidence supporting catheter ablation as a preferred rhythm control strategy with respect to functional status and quality of life.^2,3,293 Building on these insights, future studies, such as those evaluating surveillance and longitudinal rhythm control following stressor-associated AF, should formally incorporate PROs alongside clinical endpoints to identify preventive and therapeutic strategies that improve health, functional status, and psychological well-being.

SUMMARY OF FUTURE DIRECTIONS

Over the past decades, extensive research has elevated stressor-associated AF from a condition of minimal clinical interest to an important focus of modern cardiovascular care. Although we have gained a number of insights into common stressors, their likely pathophysiologic mechanisms, clinical implications, and optimal management strategies for certain forms of stressor-associated AF, several critical areas of future research remain (Central Illustration). We have identified key knowledge gaps throughout this narrative review, and succinctly highlight the following critical unmet needs:

1. Developing more effective prevention: Stressor-associated AF in the postcardiac surgery setting has provided a blueprint for translating pathophysiological insights into effective clinical interventions (eg, posterior pericardiotomy to prevent stressor-associated AF episodes). Our understanding of strategies to prevent stressor-associated AF, and manage it optimally when it does occur, is comparably lacking in other settings. Identifying patients at high risk of stressor-associated AF in noncardiac surgery and medical settings may facilitate further discovery and guide the use of evidence-based prophylaxis.

2. Building precision methods for longitudinal surveillance and treatment: Better methods of stratifying individual risk of AF recurrence and AF-related adverse events following a stressor-associated AF episode are essential to prioritize individuals for resource-intensive but more accurate monitoring technologies (eg, implantable loop recorders). Additionally, such methods could guide preventive interventions focused on reducing AF-related heart failure (which are yet unknown but may include early rhythm control or even certain guideline-directed medical therapies) and stroke (eg, empiric OAC or left atrial appendage closure).²⁹⁴ A better understanding of the natural course of stressor-associated AF and the potential interplay between AF burden and adverse outcomes could further improve prognostication and establish a foundation for shared decision making that incorporates patient-level risks and benefits. Given established limitations in the predictive utility of clinical factors, AI-enabled analysis and other novel biomarkers (eg, imaging, genetic scores) may represent a promising avenue. Further work including the collaborative development of rich data sets comprising ethnically, racially, socioeconomically, and geographically diverse patients is needed to advance future efforts to achieve precision management of stressor-associated AF. Harmonizing definitions of stressor-associated AF (eg, establishing consensus diagnostic criteria and potentially introducing unique billing codes) could substantially facilitate research, reduce study heterogeneity, and enhance replicability of findings.

3. Utilizing a patient-oriented approach: Integrating patient-centered outcomes and socially informed approaches into the management of stressor-associated AF may alleviate symptom burden and address comorbid anxiety and depression. Further investigations into the critical role of social determinants in stressor-associated AF onset, short- and long-term outcomes, and longitudinal management are needed to promote high-quality equitable care for all patients with stressor-associated AF.

CONCLUSIONS

Once considered a transient and benign phenomenon, mounting evidence highlights that stressor-associated AF carries a substantial risk of AF recurrence and AF-related adverse events, comparable to primary AF. Advances in our understanding of the epidemiology, risk factors, and pathophysiology of stressor-associated AF have refined approaches to prevention and management, particularly in the postcardiac surgery setting. Stressor-associated AF likely represents a spectrum spanning variable relative contributions of stressor and underlying predisposition, with location on the spectrum representing a useful framework for conceptualizing risk of recurrent AF and AF-related adverse events. Nevertheless, many key knowledge gaps remain, including the need for better risk stratification methods, more precise strategies to monitor for recurrence, and evidence-based approaches to prevention, longitudinal arrhythmia management, stroke and heart failure prophylaxis. Although clinical factors possess limited utility for prognostication, emerging AI-based methods and other novel biomarkers (eg, laboratory assays, imaging, omics-based risk) represent promising avenues for individualized risk assessment and management. Incorporating PROs into routine care following stressor-associated AF may enhance quality of life and long-term cardiovascular health. Future research should focus on refining risk prediction tools, rigorously investigating individualized management strategies, and integrating patient-centered approaches to optimize outcomes after stressor-associated AF.

ABBREVIATIONS AND ACRONYMS

AI

artificial intelligence

CABG

coronary artery bypass grafting

CRP

C-reactive protein

IL

interleukin

LAAO

left atrial appendage occlusion

MI

myocardial infarction

OAC

oral anticoagulation

PRO

patient-reported outcome

PRS

polygenic risk score

APPENDIX

For a supplemental table and figure, please see the online version of this paper.