February 17, 2016
At an invited symposium at the ISC, attendees were treated to a thorough discussion regarding extended monitoring for AF, biomarkers for AF, and atrial cardiopathy.

First, Dr. Bernstein reviewed recent trials demonstrating the importance and yield of prolonged cardiac monitoring to detect AF in patients with cryptogenic stroke. As we know, approximately 30% of stroke remains cryptogenic after a standard stroke work-up. In EMBRACE, patients with cryptogenic stroke or TIA were randomized to standard of care of 30 days of monitoring. Prolonged monitoring doubled the rate of AF diagnosis (14.8% at 4 weeks as compared to 2.2% after 24 hours and 7.4% after 1 week) and led to an increase in appropriate therapy. In CRYSTAL AF, a similar population was randomized to standard of care and an insertable cardiac monitor. AF diagnosis increased in a time-dependent manner. At ISC 2016, FIND-AF was presented, which randomized all-comer stroke patients to standard of care or 3 10-day monitoring episodes. 4.5% of patients in the control were found to have AF, versus 13.5% at 6 months with up to 3 10-day sessions.

These trials all suggest the importance of prolonged cardiac monitoring. Dr. Bernstein astutely conveyed the opportunities inherent to this data: we may be able to determine the dose-dependence and thresholds of toxicity with regards to AF and stroke.

Second, we were treated to a discussion of structural and serum biomarkers of AF. The goal of biomarker development is to guide early detection of AF in order to initiate early therapy to prevent early recurrent cerebrovascular events. Noteworthy potential biomarkers include cardiac MRI (delayed enhancement reveals atrial fibrosis), atrial blood flow dynamics, and BNP and ANP, for which there are a range of potential cut-offs.

Third, Hooman Kamel provided an overview of electrocardiographic markers of atrial dysfunction that may be independent of AF. He argues that it is important to search for occult AF because 70% of patients with cryptogenic stroke (e.g. in EMBRACE) do not have AF even after prolonged study, suggesting that other mechanisms of disease may have a role. He notes that endothelial dysfunction, fibrosis, impaired myocyte function, and chamber dilation are noted in patients with AF. He conducted multiple studies in various, large cohorts (MESA, Cardiovascular Health Study, ARIC, NOMAS) to demonstrate that the P-wave terminal force on EKG (lead V1) is associated with ischemic stroke and, more specifically, cryptogenic and non-lacunar stroke. This association suggests that atrial cardiopathy may independently be a risk for cardioembolic stroke, with AF perhaps indicating more severe atrial cardiopathy.

Last, Shadhi Yaghi reviewed the therapeutic implications of the atrial cardiopathy model. He has also discussed some of these views in an article in Neurology (2015). A major advantage of the atrial cardiopathy model is that it shrinks the cryptogenic stroke population, which allows more targeted, individualized therapy for secondary stroke prevention. Studies to evaluate the effectiveness of anti-coagulation for secondary prevention in patients with stroke or TIA and markers of atrial dysfunction should be considered. Additionally, this model may allow for prediction of AF, which could, prompt a search for therapies to halt or reverse atrial cardiopathy and also prompt trials for primary prevention of stroke.

In summary, patients with cryptogenic stroke and TIA should undergo prolonged cardiac monitoring to prompt appropriate secondary prevention (anti-coagulation). Structural and serum biomarkers of AF may help us identify patients who can be initiated on anti-coagulation earlier after a stroke, before AF is conclusively diagnosed. AF may be a marker of severe atrial cardiopathy, which may be independently associated with ischemic stroke; emerging data will inform approaches to primary and secondary stroke prevention in patients with risk for cardioembolic stroke in the absence of documented AF.

– Neal S. Parikh, MD