High performance of an implantable Holter monitor in the detection of concealed paroxysmal atrial fibrillation in patients with cryptogenic stroke and a suspected embolic mechanism

Introduction. Implantable loop recorders (ILR) may allow detection of occult paroxysmal atrial fibrillation (PAF) in patients with cryptogenic ischemic stroke. However, optimal selection algorithm and ideal duration of monitoring remain unclear. Aim. To determine the incidence and time-profile of PAF in patients with cryptogenic ischemic stroke studied with Reveal XT (R) ILR, who were selected based on a high suspicion of cerebral embolism. Patients and methods. Selection criteria: absence of stroke etiology after complete study including vascular imaging, transesophageal echocardiography and at least 24 hours of cardiac rhythm monitoring, and confirmation of acute embolic occlusion of intracranial artery by transcranial duplex or characteristics of acute ischemic lesion on neuroimaging suggesting embolic mechanism of ischemia. After implanting Reveal XT ILR, patients were trained to perform transmissions monthly or when symptoms occurred. We reviewed the information online each month and patients underwent clinical visits at 3rd and 6th month and then every six months. Results. We included 101 patients with cryptogenic ischemic stroke and at least one month of follow-up after ILR implant. Mean age was 67 years, 54 women (53.5%). Mean follow-up after implantation was 281 +/- 212 days. Occult PAF was detected in 34 patients (33.7%). Frequency of false positives: 22.8%. Median time from implant to arrhythmia detection was 102 days (range: 26-240 days). 24 (70%) patients with PAF had several arrhythmic episodes detected with ILR. The majority of events (75%) were detected during the first six months of monitoring. Conclusions. In our patients with probably embolic cryptogenic ischemic stroke, PAF was detected by Reveal XT ILR in 33.7%. One in four events occurred after the first six months of monitoring.