Infarct Topography and Detection of Atrial Fibrillation in Cryptogenic Stroke: Results from CRYSTAL AF

Richard A. Bernstein, Vincenzo Di Lazzaro, Marilyn M. Rymer, Rod S. Passman, Johannes Brachmann, Carlos A. Morillo, Tommaso Sanna, Vincent Thijs, Tyson Rogers, Shufeng Liu, Paul D. Ziegler, Hans Christoph Diener

Research output: Contribution to journalArticlepeer-review

54 Scopus citations

Abstract

Background: Insertable cardiac monitors (ICM) have been shown to detect atrial fibrillation (AF) at a higher rate than routine monitoring methods in patients with cryptogenic stroke (CS). However, it is unknown whether there are topographic patterns of brain infarction in patients with CS that are particularly associated with underlying AF. If such patterns exist, these could be used to help decide whether or not CS patients would benefit from long-term monitoring with an ICM. Methods: In this retrospective analysis, a neuro-radiologist blinded to clinical details reviewed brain images from 212 patients with CS who were enrolled in the ICM arm of the CRYptogenic STroke And underLying AF (CRYSTAL AF) trial. Kaplan-Meier estimates were used to describe rates of AF detection at 12 months in patients with and without pre-specified imaging characteristics. Hazard ratios (HRs), 95% confidence intervals (CIs), and p values were calculated using Cox regression. Results: We did not find any pattern of acute brain infarction that was significantly associated with AF detection after CS. However, the presence of chronic brain infarctions (15.8 vs. 7.0%, HR 2.84, 95% CI 1.13-7.15, p = 0.02) or leukoaraiosis (18.2 vs. 7.9%, HR 2.94, 95% CI 1.28-6.71, p < 0.01) was associated with AF detection. There was a borderline significant association of AF detection with the presence of chronic territorial (defined as within the territory of a first or second degree branch of the circle of Willis) infarcts (20.9 vs. 10.0%, HR 2.37, 95% CI 0.98-5.72, p = 0.05). Conclusions: We found no evidence for an association between brain infarction pattern and AF detection using an ICM in patients with CS, although patients with coexisting chronic, as well as acute, brain infarcts had a higher rate of AF detection. Acute brain infarction topography does not reliably predict or exclude detection of underlying AF in patients with CS and should not be used to select patients for ICM after cryptogenic stroke.

Original languageEnglish (US)
Pages (from-to)91-96
Number of pages6
JournalCerebrovascular Diseases
Volume40
Issue number1-2
DOIs
StatePublished - Aug 20 2015

Keywords

  • Atrial fibrillation
  • Brain infarction
  • Cryptogenic stroke

ASJC Scopus subject areas

  • Neurology
  • Clinical Neurology
  • Cardiology and Cardiovascular Medicine

Fingerprint

Dive into the research topics of 'Infarct Topography and Detection of Atrial Fibrillation in Cryptogenic Stroke: Results from CRYSTAL AF'. Together they form a unique fingerprint.

Cite this