Bear tracks hypothesis: from atrial fibrillation to atrial fibrosis syndrome in stroke risk assessment

We read the articles by Thanigaimani et al. [1] entitled ‘Molecular mechanisms of atrial fibrosis: implications for the clinic’ and by O’Neill et al. [2] entitled ‘Clinical, electrophysiological and imaging predictors of atrial fibrillation ablation outcome’ with great interest.

Strokes related to atrial fibrillation (AF) are generally more severe and more often fatal or disabling than non-AF strokes [3] but can be effectively prevented with oral anticoagulant therapy [4,5]. Therefore, the documentation of AF has been the cornerstone of primary (before the first stroke occurs) and secondary prevention of embolic and cryptogenic stroke.

Since AF may be asymptomatic in up to 30–40% of the patients [6], long-term monitoring using implanted recorders improves detection in patients with cryptogenic stroke. The Cryptogenic Stroke and Underlying AF (CRYSTAL AF) trial used implantable monitors for AF diagnosis after cryptogenic stroke and demonstrated superiority in documenting a half-minute episode of AF more than sixfold compared to the external conventional monitors [7]. While this is a significant improvement in the ascertainment of the arrhythmia, it is a rather surprising finding that AF did not always precede thrombo-embolization in 70% of patients with permanent pacemakers who also experienced thromboembolic stroke [7]. REVEAL AF trial [8] evaluated the incidence of subclinical AF that lasted ≥6 min in high-risk patients who were previously undiagnosed with AF and found that the incidence of ≥6 min episodes of AF in high-risk patients for AF and stroke was essentially the same as it was in patients with cryptogenic stroke. The median time from device insertion to detection of the first AF episode was 120 days. A sub-analysis from the Asymptomatic AF and Stroke Evaluation in Pacemaker Patients and the AF Reduction Atrial Pacing Trial (ASSERT) also showed that only 15% of patients with an embolic event had evidence of subclinical AF ≥6 min in the previous month. But a more recent exploratory analysis of ASSERT demonstrated that whereas only AF episodes lasting longer than 24 h were associated with an increased risk of ischemic stroke, this risk was not significantly different in patients with asymptomatic AF between 6  min and 24  h and patients without asymptomatic AF [9]. The IMPACT study reported that stopping and starting anticoagulation based on presence of device detected AF had no significant effect on ischemic stroke [10]. Moreover, the majority of subclinical AF events occurring prior to the embolic event were far shorter than 48 h, which is commonly believed to be the minimum duration required for thrombus formation in the left atrial appendage. All these results showed no temporal relation between ischemic stroke and AF paroxysms and raised cause and effect questions about the role of AF, the arrhythmia, in thrombus formation and embolization. Indeed, AF may be related to embolic events via indirect mechanism or could only be a risk factor or marker for other comorbidities that increase the risk of stroke, rather than the sole or primary etiology. The phenotypic classification of AF into paroxysmal, persistent, long-standing persistent, and permanent AF1 is clinically useful but does not reflect the underlying pathophysiology and substrate characteristics of the atria [11]. Moreover, in risk estimation scores; such as the congestive heart failure, hypertension, age, diabetes and previous stroke or transient ischemic attack, vascular disease and female sex category (CHA₂DS₂-VASc) [12], the mechanism relating their individual components to thrombogenesis is unknown. Given these limitations, a more comprehensive and mechanistic evaluation of atrial disease is needed to better identify patients at risk for stroke and AF.

While electrical dysfunction is widely linked to arrhythmogenic triggers, other hallmark features of human AF are inflammation and atrial fibrosis [1]. In one case report of AF as the initial presentation of cardiac sarcoidosis, the patient had recurrent AF after pulmonary vein isolation whereas AF burden decreased after immunosuppression [13]. Indeed, most of components of CHA₂DS₂-VASc are also predictive of atrial fibrosis, and might not directly reflects AF but also underlying atrial fibrosis more accurately. Indeed, the proposed associations need to be rigorously tested to establish a link between atrial fibrosis and thromboembolism, independent of the presence of AF. Recently, biomarkers of atrial cardiomyopathy have been associated with embolic stroke risk even in the absence of diagnosed AF [11,14], suggesting that the presence of AF is not required for left atrial (LA) thromboemboli to occur. In the literature, there are some different classifications of AF such as ‘primary versus secondary persistent AF’ [15], ‘atrial cardiopathy/cardiomyopathies’ [14] or ‘fibrotic atrial cardiomyopathy (FACM)’ [11]. However, whether fibrosis is causally related to AF or an epiphenomenon, the precise mechanisms underlying atrial fibrosis are not fully elucidated [12]. There are AF patients in whom AF is secondary to atrial fibrosis syndrome (AFS) and there are AF patients in whom AF is a manifestation of primary electrical triggers; however, the lack of this mechanistic information in AF ablation trials should be kept in mind. Ghanbari et al. reported that the presence of sinus rhythm at follow-up after ablation was associated with a significantly lower risk of cardiac mortality but not a reduction in stroke risk [16]. In selected patients after apparent successful AF ablation, the risk of stroke might be different depending on the underlying AF mechanism regardless of the CHA₂DS₂-VASc score even in the absence of oral anticoagulation. In patients with pure focal AF, triggers isolation can be considered a curative treatment option and may eliminate the AF-related stroke risk in the absent of overlapping atrial fibrosis; however, it would not be enough for AFS-related stroke risk without changing the lifestyle modifications and/or atrial fibrosis based ablation strategy.

In the future, we need a personalized approach to identify which patients are likely to benefit more from appropriate ablation and anticoagulation to reduce thromboembolic risk; and we would use the more mechanistic scores predicting AFS and stroke risk. The LEGACY, DECAAF and SORT-AF trials have stressed the importance of the risk-factor treatment in AF patients with or without ablation therapy, probably via treating not only AF itself but also underlying atrial fibrosis by these modifiable lifestyle changes. Since most modifiable risk factors such as hypertension, alcohol, smoking, overweight and obesity, sleep apnea, smoking, and diabetes mellitus contribute significantly to the degree of atrial fibrosis and AF burden, targeting the substrate for AF with cardiovascular risk factor management has gained significant momentum with studies demonstrating improvement in AF burden and severity, together with reverse atrial remodeling/fibrosis and improved outcomes post catheter ablation. Therefore, it may be time to consider AF as only one more factor of adverse atrial/endothelial structural remodeling rather than the main contributor to stroke [17].

Without treating AFS, treating AF by ablation may not be enough to prevent embolic strokes. In many cases, effective treatment of AF is severely hindered by a lack of mechanistic understanding relating features of fibrotic remodeling to dynamics of reentrant arrhythmia. Moreover, after multiple extensive AF ablations, active left atrial (LA) function is significantly impaired by ablation-related LA scar burden [18,19]. Therefore, it may not be safe to stop anticoagulants for patients with a high-risk score even in the absence of AF recurrence in patients undergoing ablation not for trigger-based AF but for AFS-related AF. The decision about long-term oral anticoagulation strategy could be based on baseline clinical risk scores predicting underlying atrial fibrosis rather than the status of AF recurrence [5,20]. All these might be explained by the bear track hypothesis: ‘If you see bear tracks in the forest, this means that there would be a risk of attack by a bear. However, to erase these tracks does not eliminate this risk’.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This paper was not funded.