Transient ischemic attack (TIA) is a common neurologic emergency. Although the detection of atrial fibrillation (AF) has identical consequences for preventive therapy in patients with ischemic stroke and TIA, the management setting and diagnostic pathways frequently differ substantially between both manifestations. Despite these differences between stroke and TIA patients, previous studies have investigated diagnostic work-up for AF primarily in stroke patients. Thus, there is no common practice or "gold standard" of rhythm monitoring for TIA patients in most healthcare systems and the optimal method and duration of cardiac monitoring for TIA patients is currently unknown. This is likely to result in a substantial under-diagnosis of AF in TIA patients, failure to initiate appropriate secondary preventive medication (i.e. anticoagulation) and ultimately the occurrence of many otherwise preventable strokes.

The primary research question of the trial is whether prolonged ECG recording using a subcutaneously implanted event recorder increases the detection rate of paroxysmal AF (pAF) within 6 months after the event in patients with recent TIA both compared to 24-h ECG recording and compared to prolonged ECG recording using non-invasive continuous ECG recording for 28 days AND To determine whether 28-day non-invasive ECG monitoring increases the detection rate of pAF within 6 months after the event compared to 24-h ECG recording.

Transient ischemic attacks (TIA) are a common neurologic emergency. Clinical management guidelines recommend oral anticoagulation for TIA patients suffering from atrial fibrillation (AF). Therefore, a diagnosis of AF in TIA patients has a major impact on the choice of adequate secondary stroke prevention. However, detection of paroxysmal AF in patients with TIA can be challenging. AF remains undetected in a relevant proportion of stroke and TIA patients using current routine diagnostic procedures. The actual prevalence of AF in TIA patients is unknown.

Although the detection of AF has identical consequences for preventive therapy in patients with ischemic stroke and TIA, the management setting and diagnostic pathways frequently differ substantially between both manifestations. So far, only limited data exist on AF detection after TIA specifically, and the best method for diagnosis of AF has not been established. The usefulness of prolonged rhythm monitoring using event recorders or non-invasive continuous ECG in TIA patients has not been determined. While the use of an AF detection tool in TIA patients is desirable, an adequate use of resources of AF detection technologies in unselected TIA patients may be needed for this large scale health care problem. Identifying TIA patients that are at increased risk of suffering from AF using clinical and blood-based biomarkers and therefore most likely to benefit from such diagnostic procedures would be useful.

The primary research question of the trial is whether prolonged ECG recording using a subcutaneously implanted event recorder increases the detection rate of paroxysmal AF (pAF) within 6 months after the event in patients with recent TIA both compared to 24-h ECG recording and compared to prolonged ECG recording using non-invasive continuous ECG recording for 28 days AND To determine whether 28-day non-invasive ECG monitoring increases the detection rate of pAF within 6 months after the event compared to 24-h ECG recording. The ODEA-TIA trial is an investigator initiated prospective, multicentre, randomized, open study with blinded outcome assessment comparing different diagnostic methods for detection of paroxysmal AF in patients with recent TIA. The primary endpoint is the rate of AF detection during the 6 months after randomization. Approximately 40 centers in Europe (Germany and Spain) will participate in this trial. Patients with a recent TIA fulfilling the eligibility criteria (see below) will be randomized in a 1:1:1 fashion between 24 h arrhythmia monitoring (control arm) and the two procedures for prolonged ECG monitoring (interventional arms). That means we have two interventional arms, patients receiving either continuous 28d non-invasive ECG monitoring or ECG event recording using a subcutaneously implanted event recorder.

Eligibility Criteria:

Study Population Patients with a recent TIA will be enrolled during a period of approximately 24 months at participating European stroke centres. TIA patients may be enrolled after initial management as inpatients or outpatients. Consecutive screening and enrolment will be strongly encouraged and a screening log will be implemented at each site.

Inclusion Criteria

• Written informed consent by patient.
• Age ≥ 50 years.
• TIA diagnosed by a stroke physician defined as rapidly developing clinical signs of focal or global disturbances of cerebral function, lasting less than 24 hours with no apparent non-vascular cause
• 12-channel ECG available before enrolment
• Brain imaging available before enrolment (CCT or cranial MRI)
• Vascular imaging of cervical vessels performed
• Enrolment within 28 days after TIA episode. Exclusion Criteria
• Previously documented history of AF
• Ischemic stroke within the last 6 months before enrolment
• Pre-screening monitoring for cardiac arrhythmias lasting ≥72 hours
• AF lasting > 30 s on a 12 channel ECG or other ECG recording technique prior to enrolment
• Life expectancy less than 1 year.
• Significant stenosis > 50% in intracranial or extracranial vessels which, in the opinion of the investigator, is the likely cause of the patients TIA.
• Severely disabled patients (i.e. modified Rankin Score >3)
• Lack of therapeutic consequence in case of diagnosis of AF (e.g. other indication for long term anticoagulation
• Pacemaker or Implanted Cardiac Defibrillator