Paroxysmal atrial fibrillation detection in patients with acute ischemic stroke through prolonged Holter: prospective...

The detection of paroxysmal atrial fibrillation (PAF) is crucial in the etiological study of an acute ischemic stroke (AIS), although this type of arrhythmia is often under-diagnosed. This prospective study aims to (1) evaluate the new-onset PAF detection rate among elderly patients with AIS in an acute setting, and (2) to assess the applicability of the STAF score (Score for the Targeting of Atrial Fibrillation) for such patients.

An observational and prospective study was performed over a period of 11 months. Patients with acute ischemic stroke who had been admitted to the stroke unit were included. Exclusion criteria included prior AF, AF readings on ECG during admission, or stroke mimic. The patients were monitored with a bedside ECG monitor throughout the first 48 h, then simultaneously with an extended Holter recorder for a maximum of 6 days. The occurrence and duration of PAF events was evaluated. The STAF risk score for PAF was calculated and compared to the detection rate of PAF for each patient.

The sample population consisted of 67 patients, whose median age was 76 years (IQR 71-84) and who were 57% males. The median duration of Holter recording was 3.5 days per patient (IQR 2-5). We detected new PAF cases in 16% of patients, and a total of 31% had supraventricular arrhythmia episodes lasting 10 s or longer. Among the patients in which this arrhythmia was detected, 10% had it detected by the bedside ECG monitor in the first 48 h. The median risk score was 5 (IQR 4-6). There was no association between the risk score used and PAF.

In elderly patients with AIS, the use of an extended Holter recorder led to the detection of more cases of PAF, offering them the chance to initiate hypocoagulant therapy. Among this population, the STAF score was non-discriminative and did not prove useful in predicting PAF.