1. Classification and Definitions

1.1 Classification on the basis of clinical pattern of AF

AF pattern is defined on the basis of clinical assessment of episode persistence. These patterns have been used to characterize the severity of disease, define patient populations in clinical trials, and are used to form the basis of therapeutic recommendations regarding pharmacological and invasive arrhythmia management.^[10] Four main clinical patterns of AF have been described. Paroxysmal AF is defined as a continuous AF episode lasting longer than 30 seconds but terminating within 7 days of onset. Persistent AF is defined as a continuous AF episode lasting longer than 7 days but less than 1 year. “Longstanding” persistent AF is defined as continuous AF 1 year in duration, in patients in whom rhythm control management is being pursued. Permanent AF is defined as continuous AF for which a therapeutic decision has been made not to pursue sinus rhythm restoration. The mode of termination (spontaneous vs pharmacological/electrical cardioversion) does not influence the classification. In many patients, AF progresses from short episodes of self-terminating paroxysmal AF to more frequent exacerbations of longer-lasting persistent AF. In the event that both paroxysmal and persistent episodes are present, the classification should be defined on the basis of the predominant AF pattern. Although valuable, there are limitations to classifying AF patterns by clinical assessment. First, the distinction between paroxysmal and persistent AF is often inaccurate, because clinical assessment often underestimates the temporal persistence of AF/AF burden compared with long-term electrocardiogram (ECG) monitoring.^[11]–[13] Second, although the AF patterns have been associated with adverse outcomes (eg, heart failure [HF], stroke, and death), there remains uncertainty regarding their independent role of these clinical patterns to predict response to therapy (eg, antiarrhythmic drugs or catheter ablation).^[13]–[17]

1.2 Classification on the basis of pathophysiological pattern of AF

The likelihood of developing AF varies across physiological and pathological states. Despite similar clinical patterns the mechanisms underpinning AF vary substantially between patients. Within this context AF may be considered “primary” if the AF represents an established pathophysiological process or “secondary” if caused by a self-limited or acutely reversible precipitant (Figs. 2 and 3A).^[18] “Primary AF” should not be considered analogous to the antiquated term, “lone AF,” which previously defined AF without known cause.^[19] Common causes of secondary AF include surgery, sepsis, acute myocardial infarction (MI), thyrotoxicosis, or acute pulmonary disease. Secondary AF can be further dichotomized on the basis of the underlying cardiac substrate and risk for AF recurrence into “reversible” or “provoked” AF.^[20] Specifically, “reversible AF” defines AF that occurs solely secondary to an acute illness, with little to no abnormal underlying substrate and therefore limited future risk of AF. In contrast, “provoked AF” represents AF that is unmasked by the acute illness, occurring in patients with significant abnormal underlying substrate, and therefore ongoing risk for AF recurrence (Fig. 3, B and C). Examples of the former category include patients with hyperthyroidism or alcohol intoxication (“holiday heart”) in the absence of previous heart disease or risk factors, and the latter would include patients developing AF after mitral valve surgery or in the context of a chronic obstructive pulmonary disease (COPD) exacerbation. These concepts are explored in greater depth in section 8.3.6.

1.3 Valvular and nonvalvular AF

The term, “nonvalvular AF” (NVAF) dates back to the late 1970s and was used interchangeably with nonrheumatic AF. This early distinction was on the basis of the observation of the high risk of stroke/systemic embolism associated with severe mitral stenosis. More recently this distinction was used to define candidacy for participation in the landmark phase III trials in which non-vitamin K direct-acting oral anticoagulants (DOACs) were compared with vitamin K antagonists (VKAs) for stroke prevention in patients with NVAF.^[21]–[25] Although previous iterations of the CCS AF guidelines have considered rheumatic mitral stenosis, mitral valve repair, mechanical heart valves, and bioprosthetic heart valves to constitute valvular heart disease, the definition of “valvular AF” has continued to evolve on the basis of emerging evidence.^{[6],[26]–[28],[884]} The current definition of “valvular AF” is limited to AF in the presence of any mechanical heart valve, or in the presence of moderate to severe mitral stenosis (rheumatic or nonrheumatic).