Low Yield of Mobile Cardiac Outpatient Telemetry after Cryptogenic Stroke in Patients with Extensive Cardiac Imaging

Background Detection of paroxysmal atrial fibrillation (AF) after cryptogenic stroke (CS) or transient ischemic attack ranges from 5% to 24%, but previous studies have had varying definitions of both. We aimed to evaluate the yield of up to 30 days of mobile cardiac outpatient telemetry (MCOT) for this dysrhythmia in CS patients who had undergone extensive cardiac imaging before monitoring. Methods We reviewed data from our center on patients with CS who completed MCOT within 3 months of the cerebrovascular event from May 2009 to January 2014; 14-30 days of monitoring was performed using one of 3 approved devices after cardiac imaging did not demonstrate a clear embolic source. We estimated the prevalence and 95% confidence intervals of AF. Results Eighty-five patients met the study criteria; 89.4% underwent transthoracic echocardiogram, 68.2% underwent transesophageal echocardiography, and 38.8% completed cardiac magnetic resonance imaging. We found 4 (4.7%, 95% confidence interval 1.5% to 11.9%) patients with AF by MCOT. There were no univariate predictors of AF. Conclusions The diagnostic yield of cardiac rhythm monitoring for up to 30 days in CS patients may be lower than previously reported. This may be because of the routine use of cardiac imaging to identify a likely source of embolism, resulting in a lower incidence of occult AF in patients who are labeled as "cryptogenic." Longer monitoring may be needed to detect this dysrhythmia in high-risk patients who have already undergone extensive cardiac imaging.