Introduction: This interim analysis evaluates the risk profile and incidence of atrial fibrillation (AF) in patients who underwent continuous monitoring with an implantable loop recorder (ILR) for cryptogenic (unexplained) stroke or transient ischemic attack (TIA). Methods: The observational INSIGHT XT study prospectively enrolled patients who received an ILR with dedicated diagnostics for atrial fibrillation, irrespective of the clinical indication. Of 1002 patients enrolled in the study between Aug 2008 and Jan 2012, 121 received the ILR to evaluate cryptogenic stroke or TIA. The definition of cryptogenic stroke/TIA was at the investigators' appraisal and no unified approach to patient work-up was required. This analysis includes 74 patients with cryptogenic stroke or TIA for whom at least one follow-up visit was available at the time of interim analysis. Results: The mean age was 63±12 (50% female). Stroke was the index event in 46 of 74 (62%) of patients. 61% had hypertension, 14% diabetes, and none had heart failure. The mean CHADS2 score was 3.0±0.8 and the mean CHA2DS2VASc score 4.0±1.2. Most patients (72%) had no prior symptoms or cardiac rhythm disturbances, whereas 18% had a history of prior palpitations. Sixty-seven patients were taking antiplatelet medication and four were on oral anticoagulation (OAC) at enrollment. During a median follow up of 12 months (IQR 7 to 18) AF was reported in 17 patients (23%) and two patients were started on OAC and 10 patients were converted from antiplatelets to OAC. Five patients experienced a stroke or TIA (median time to event 1.2 months), of which one patient died. Three of the patients with stroke or TIA had AF detected prior to the recurrent event. Conclusion: Continuous monitoring with an ILR in patients with cryptogenic stroke of TIA detects a high proportion of AF; this can be attributed to longer continuous monitoring in this study. These patients have high CHADS2/CHA2DS2VASc scores; documenting AF in these cases may therefore be clinically relevant in order to decide appropriate treatment.

Purpose: A syncopal event is unexplained, if its cause in still unknown after an initial evaluation according to current guidelines. The cause of syncope in such patients is often structural heart disease, and most often arrhythmias. Continuous ECG recordings captured by an Implantable Loop Recorder (ILR) are well suited to disclose the exact arrhythmia mechanism of syncope. Methods: An ILR (Reveal) was implanted in 570 patients (pts) with unexplained syncope. Pts were followed until syncope recurrence with an ECG capture or for at least one year. All devices had auto-activation of preselected arrhythmias and the option of manual activation. Both modes could be switched ON or OFF at the discretion of the investigator. Details of the device programming are not available. However, in pts with an episode of recurrent syncope, the investigator indicated whether the episode was captured automatically or whether the patient or a bystander activated the device manually. Results: Of the 218 pts with recurrence of syncope, there was a device captured ECG episode in 175 (80%), 33 of which showed no arrhythmia. In the remaining 142 pts, ECG was captured with auto-activation in 59 pts (42%), with manual activation in 52 pts (37%) and with both capture modes in 64 pts (45%). Bradyarrhythmias were documented in 43, 14 and 38 pts with auto-activation only, manual activation only, and both activation modes, respectively. Tachyarrhythmias were captured in 17, 13 and 20 pts, respectively. The investigator indicated "other arrhythmia" in 2, 2 and 8 pts, and there was no arrhythmia in the captured ECG of 4, 24 and 5 pts, respectively. Ventricular tachycardias were found by auto-activation in 2, manual activation in 2 and both activation modes in 8 pts. Conclusion: Auto-activation and manual activation contributed equally to clarify the mechanism of syncope, but in 20% of recurrences there was no ECG capture at all, implying underuse of either or both of the activation modes. Auto-activation was more often associated with bradyarrhythmias, and manual activations more often with no arrhythmias, while the most serious arrhythmias, ventricular tachycardias, were found equally often with both capture modes. Thus both activation modes are necessary for an optimal diagnostic yield.

AIMS:Little is known about the incidence of paroxysmal atrial tachycardias (PAT) in patients with heart failure (HF). The availability of cardiac resynchronization therapy (CRT) devices with extended diagnostics for AT enables continuous monitoring of PAT episodes. The aim of the study was to assess the incidence over time of PAT in HF patients treated with CRT. METHODS AND RESULTS:Consecutive patients in NYHA functional class III or IV despite optimal drug therapy, QRS duration > or = 130 ms, left ventricular ejection fraction < or = 35%, and left ventricular end-diastolic dimension > or = 55 mm were eligible for enrolment. Patients with permanent or persistent atrial fibrillation (AF) were not included in the study. The first follow-up examination was performed 2 weeks after implantation, to optimize atrial sensing and CRT. Subsequent follow-up examinations were carried out 15 and 28 weeks after implantation, to collect the telemetric data. A total of 173 patients (67 +/- 11 years, M 116) were enrolled. Complete arrhythmia monitoring data were available from 120 patients over a mean follow-up of 183 +/- 23 days. Atrial tachycardia episodes were detected through telemetry in 25 of 120 patients (21%) during at least one follow-up examination. Atrial tachycardia episodes were recorded in 29 and 17% (P = NS) of patients with and without previous history of AF, respectively. CONCLUSION:More than 20% of the overall HF patient population treated with CRT suffer PAT episodes. Paroxysmal atrial tachycardia may interfere with response to CRT. Therefore, telemetric data may be relevant to drive the appropriate therapy in each patient.

The Reveal XT subcutaneous insertable cardiac monitor detects Atrial Fibrillation (AF) from the irregularity of the ventricular rhythm. This study is a first assessment of the AF detection performance using subcutaneous signals from implanted devices. Patients implanted with a Reveal XT device were monitored by a dedicated Holter system, recording the surface ECG and uplinked device data. Device detections of AF were reconstructed from the uplinked data and classified as either true positive, false positive and false negative using the surface ECG for reference. In a total of 82 Holter recordings from 60 patients, the mean AF episode detection sensitivity was 90.6% and 98.1% for AF episodes ges2 minutes and ges4 minutes, respectively. The mean episode PPV was 55.1%. False positive detections were clustered within a few patients with irregular rhythms that were not annotated as AF.