What is the utility of a loop recorder in the evaluation of cryptogenic stroke?
The prevalence of atrial fibrillation in cryptogenic stroke is estimated to be somewhere between 5 and 25%. This study done in Denmark looked at patients with cryptogenic stroke (negative w/u for carotid artery disease, small vessel disease, cardioembolic source, and hypercoagulable disorders) and followed them with loop recorders for the development of atrial fibrillation. It's small - only 85 patients got a loop - and there is some variability in the timing to loop placement. Patients were monitored for about a year and half, and 16% developed new atrial fibrillation in that time period as detected by the loop recorder (in total 21% of the patients developed afib). It took a little over three months, on average, for the first episode:
Figure 2. Time from stroke onset to paroxysmal atrial fibrillation diagnosis by use of an implantable loop recorder
None of these patients had a stroke in that interim time period; there was one TIA. Most of the afib bursts picked up by the loop recorder were short (less than four hours), and as we know, the literature is very mixed on just how much afib you need to cross the risk threshold for stroke/systemic embolism.
This is a pretty small, straightforward study, and arguably its external validity for the Grady Clinic population is limited. But, I am struck by the prevalence of afib - higher than I thought - and the time it took to diagnose it - longer than I would have guessed. I haven't seen much loop recorder use at other hospitals where I've worked; this data helps me better understand the role for this diagnostic approach (which as you know we generally do at Grady). It's also helpful to have these numbers for when we're having conversations with patients about the utility of loop recorders. And finally, I am struck by the significant change in management that diagnosing afib entails, in terms of initiating anticoagulation, and so I do think this is an important diagnosis we should not be missing.
Christensen LM, Krieger DW, Højbergc S, et al. Paroxysmal atrial fibrillation occurs often in cryptogenic ischaemic stroke. Final results from the SURPRISE* study. European Journal of Neurology. Published online March 14, 2014. doi:https://doi.org/10.1111/ene.12400
