Risk Stratification for Stroke in Atrial Fibrillation: A Critique

Ammar M. Killu; Christopher B. Granger; Bernard J. Gersh


Eur Heart J. 2019;40(16):1294-1302. 

In This Article

Future Directions

In spite of the immense library of literature, many unanswered questions remain. For example, despite the significant implications of AF it is unclear whether systematic population-based screening, and by what methods, is worthwhile though studies are ongoing. In addition, what one should do for patients with high CHA2DS2-VASc and either very short episodes or even no documented AF is unclear. Patients with elevated CHA2DS2-VASc are at increased risk of stroke even in the absence of AF. Therefore, it remains unknown whether these individuals should undergo screening to detect subclinical AF or whether they might warrant anticoagulation independent of whether AF has been documented. Indeed, individuals with elevated CHA2DS2-VASc are more frequently found to have AF on prolonged ambulatory monitoring. As mentioned above, in the ASSERT-II study patients ≥65 years with CHA2DS2-VASc ≥2 had a 34.4% incidence of AF per patient year.[76] In the REVEAL-AF study, 29.3% of patients with a CHADS2 ≥3 (or 2 with ≥1 additional risk factor) had AF detected.[86] In the PREDATE-AF study, 22.4% of patients with CHA2DS2-VASc ≥2 had AF detected at 15 months follow-up.[87] One common feature of these studies is that they implemented implantable recorders. In the REHEARSE-AF study, however, a twice weekly acquisition of a single-lead electrocardiogram (ECG) using the AliveCor Kardia monitor demonstrated a 3.8% incidence of AF detection at 12 months compared with 1.8% in those undergoing usual care.[88]

If AF is detected, another dilemma relates to the duration at which it is deemed to be significant. The ASSERT study showed that AF episodes as short as 6 min were associated with an increased risk of stroke.[77] However, the timing of AF did not correlate with the occurrence of stroke and the annual rate of stroke only became significant when 18 h duration was reached. An analysis from the MOST study showed that ≥5 min was significant, while the TRENDS study suggested a cut off of >5.5 h duration.[89,90] This was in contrast to a prospective study that argued only >24 h duration is significant.[91] Therefore, what duration of AF should be documented before anticoagulation is controversial. The Apixaban for the Reduction of Thrombo-Embolism in Patients With Device-Detected Sub-Clinical AF (ARTESiA, NCT01938248), Non-vitamin K Antagonist Oral Anticoagulants in Patients With Atrial High Rate Episodes (NOAH, NCT02618577) and the AF Detected by Continuous ECG Monitoring (LOOP, NCT02036450) studies should provide important data to address this issue. Excellent commentaries have been published on AF burden and stroke.[92,93]

While anticoagulation is the preferred stroke reduction therapy in patients with AF, not all patients are suitable. Percutaneous procedures aimed at isolating the LAA have gained popularity as the LAA is the usual source of embolization. At present, appendage exclusion procedures are approved only in patients with a contraindication to chronic anticoagulation therapy, but can tolerate short durations to protect against thrombosis as the device endothelializes. It remains to be seen whether they will have a broader indication. However, LAA occlusion is unlikely to completely mitigate anticoagulation requirement given that AF appears to be sequelae of a more widespread vascular disease.