Occult Paroxysmal Atrial Fibrillation in Non-Cryptogenic Ischemic Stroke

Recent studies suggest that their is a relatively high incidence of silent paroxysmal atrial fibrillation (PAF) in cryptogenic ischemic stroke patients. Detection of silent PAF in this population, however, requires prolonged cardiac monitoring. On the other hand, there is no understanding of the frequency of silent PAF in patients with a "known" or presumed etiology for their ischemic stroke other than AF. If this incidence is also found to be significant, a dramatic change in stroke evaluation and management would likely occur. This study intends to find the incidence of PAF in a population of recent (within 7 days) ischemic stroke patients (or MRI positive "transient ischemic attack (TIA)" patients) who have presumed etiology for their stroke other than atrial fibrillation or other high risk cardiac lesion. Stroke etiology will be determined through a standard post-stroke evaluation that includes: * Brain MRI * 12-lead electrocardiogram (ECG) for AF detection * 24-h ECG monitoring for AF detection (eg, Holter monitor or cardiac telemetry) * Transthoracic and/or Transesophageal echocardiogram * CT or MR angiography of the head and neck * Hypercoagulable blood panel for patients less than 55 years-old. Preliminary hypercoagulable work-up within 7 days will include the antiphospholipid antibody syndrome results, which will be needed to determine stroke management. All other results, which could take longer to return, including genetic tests of hypercoagulability, rarely change stroke management. To find occult PAF in this population, participants will have an insertable cardiac monitor (Reveal LINQ ICM) implanted within 7 days of the incident stroke and will be monitored for at least 1 year.