Abstract 10764: A Comparison of Atrial Fibrillation Detection Strategies After Ischemic Stroke- A Retrospective Study

Abstract

Listen

Translate article

Introduction: In 20-30% cases, the cause of stroke remained unexplained which has led to coining of the term, Cryptogenic Stroke (CS). Similarly, about 48% cases of transient ischemic attacks (TIAs) had no identifiable cause after standard diagnostic workup. Undiagnosed Atrial Fibrillation (AF) is a prime suspect in CS but guidelines do not recommend initiation of anticoagulation unless AF has formally been detected. Methods: In a IRB approved retrospective study we included patients with at least one episode of ischemic stroke or TIA without identifiable cause and was monitored with either 48-hour Holter Monitor (HM), 30-day Event Monitor (EM) or Implantable loop recorder (ILR) to diagnose any undiscovered AF. All patients had at least 1 year, and up to 3 years, of follow-up after device placement. SAS Version 9.4 was used for statistical analyses. Results: Out of a total of 531 patients, 150 patients were monitored by ILR, 286 by EM and 95 by HM. Primary Outcome- detection of AF. The diagnosis of AF within 1 month of the stroke was 5.59% (16/286), 6.32% (6/95) and 9.33% (14/150) in the EM, HM and ILR cohorts, respectively (p=0.33). At 6, 12 and 24 months, ILRs detected AF in 15.33% (23/150), 16% (24/150) and 20% (30/150) of patients respectively (p=.0017, .0008 and .0001, respectively). Hence the Chi-Squared analysis showed no statistically significant difference among 3 devices for the detection of AF within 1 month of the index stroke but a significant difference in AF detection was observed at 6, 12 and 24 months. Similarly, the multivariable logistic regression model demonstrated no significant difference in capturing AF between HM, EM and ILR within 1 month (p=0.29) but showed a significant difference in AF detection when ILR was compared to HM and EM at 6,12 and 24 months (p=0.0027, 0.0012 and <0.0001 respectively). Secondary Outcome- Kaplan Meier estimator analysis and Cox proportional Hazard model showed device type had no significant impact on secondary outcomes of the study: 1) Subsequent ischemic stroke or TIA 2) Initiation of anticoagulation 3) Mortality 4) Incidence of major bleeding. Conclusion: In conclusion, despite the superiority of AF detection by ILR, it is not superior to HM or EM in lowering the risk of subsequent stroke or TIA, or in reducing mortality.