A New Clinical Construct for Strokes: Not So “Cryptogenic” Anymore?

April 17, 2014

If common mechanisms underlying cryptogenic stroke can be clarified, we will need to reevaluate the current approach to secondary prevention. Here, elucidation of "emoblic stroke of undetermined source."

Twenty-five percent of strokes are “cryptogenic,”1 that is, they cannot be explicitly ascribed to the “traditional” causes of strokes (eg, large-vessel carotid disease, small-vessel lacunar strokes, and atrial fibrillation). They form a robust subset of strokes, totaling 300,000 per year in North America and Europe.1

Now the Cryptogenic Stroke International Working Group wants to take the “cryptogenic” out of our stroke vocabulary and replace it with a new clinical construct for this heretofore elusive category. This group believes that these specific strokes are also embolic and should now be identified as embolic stroke of undetermined source (ESUS).

Common Cardiac Abnormalities

The authors in the working group observe that many patients in the cryptogenic category have common cardiac abnormalities associated with strokes (such as patent foramen ovale) and have these lesions more frequently than the general population. Other putative lesions that might contribute to ESUS (also found more commonly in persons with cryptogenic strokes) are systolic dysfunction, left ventricular regional wall motion abnormalities, and “covert” atrial fibrillation. Even a short run of reversible atrial fibrillation can be bad for your brain health. In fact, in patients experiencing cryptogenic strokes, extended cardiac monitoring suggests that 10% to 20% have paroxysmal, not fixed, atrial fibrillation.1

Moving away from the heart to the vasculature, high-resolution MRI scans have demonstrated that persons with cryptogenic strokes also have complicated, nonstenotic plaques that may be the source of brain emboli and ischemic strokes.1 There is more to vascular risk than high-grade carotid stenosis.

If all this makes sense, the next thing to do is diagnose strokes in this particular cohort and then identify the ESUS pathology as different from traditional risks. The authors propose a step-wise approach.

First, neuroimaging would be used to confirm the diagnosis of an ischemic stroke. Step 2 would be to rule out traditional risks and look for the purported abnormalities associated with ESUS. Holter monitoring would uncover cardioembolic sources (covert atrial fibrillation), or echocardiography would identify patent foramen ovale or ventricular clots. If the exact pathology is still not pinned down, vascular imaging would focus on proximal vascular atherosclerosis (as opposed to high-grade carotid disease). With this approach, ESUS would be proved to be responsible for those strokes previously labeled cryptogenic.

Aggressive Anti-Embolic Approach

Why all the effort to clarify mechanisms in the cryptogenic cohort? If the strokes in this group are embolic, worthwhile secondary prevention will not be achieved with aspirin and cholesterol lowering. An aggressive anti-embolic approach will be necessary.

Future trials will have to consider this cohort’s unique risk factors and determine whether warfarin or the newer anticoagulants (novel oral anticoagulants, or NOACs) are the missing piece in the puzzle of prevention. If cryptogenic strokes are a misnomer, and ESUS is the real pathology, randomization for future studies will have to compare aspirin with warfarin or an NOAC.

Yes, this will be a new and potentially fruitful clinical construct relevant to 300,000 persons each year. It seems this working group is really onto something.


  • Hart RG, Diener HC, Coutts SB, et al. Embolic strokes of undetermined source: the case for a new clinical construct. Lancet Neurol. 2014;13:429-438. (Abstract)