Abstract
Patients at short-term risk of paroxysmal atrial fibrillation (PAF) often exhibit increased RR interval variability during sinus rhythm. We studied if RR dynamic analysis, applied in the first hours after stroke unit (SU) admission, identified acute ischemic stroke patients at higher risk for subsequent PAF episodes detected within the SU hospitalization. Acute ischemic stroke patients underwent continuous cardiac monitoring (CCM) using standard bedside monitors immediately after SU admission. The CCM tracks from the first 48 h were analyzed using a telemedicine service (SRA clinic, Apoplex Medical, Germany). Based on the RR dynamics, the stroke risk analysis (SRA) algorithm stratified the risk for PAF as follows: low risk for PAF, high risk for PAF, presence of manifest AF. The subsequent presence/absence of PAF during the whole SU hospitalization was ruled out using all available CCMs, standard ECGs, or 24-h Holter ECGs. Two hundred patients (40% females, mean age 71 ± 16 years) were included. According to the initial SRA analysis, 111 patients (56%) were considered as low risk for PAF, 52 (26%) as high risk while 37 patients (18%) had manifest AF. A low-risk level SRA was associated with a reduced probability for subsequent PAF detection (1/111, 0.9%, 95% CI 0-4.3%) while a high-risk level SRA predicted an increased probability (20/52, 38.5% (95% CI 25-52%). RR dynamic analysis performed in the first hours after ischemic stroke may stratify patients into categories at low or high risk for forthcoming paroxysmal AF episodes detected within the SU hospitalization.