CI-452768-1 IMPACT OF NEW-ONSET ATRIAL FIBRILLATION DETECTED BY CARDIAC IMPLANTABLE ELECTRONIC DEVICES ON CARDIOEMBOLIC STROKE EVENTS

Abstract

The association between atrial fibrillation (AF) detected by implanted cardiac implantable electronic devices (CIEDs) and the incidence of cardioembolic stroke events has been reported. However, the cut-off value of new-onset AF duration detected by CIEDs for cardioembolic stroke is still unknown. In previous studies regarding the cut-off value of AF duration, patients who had a history of AF or had received anticoagulant therapies were also included. Therefore, the solo effect of AF on the risk of cardioembolic stroke cannot be reliably evaluated on these studies. The purpose of this study is to identify the incidence of cardioembolic stroke events in Japanese patients with a CIED and examine the cut-off value of AF duration detected by CIEDs as a risk factor for cardioembolic stroke events. We retrospectively analyzed the database of our CIEDs clinic. Every 6 months, AF events were checked by CIEDs. Patients who had a history of AF or had received anticoagulant therapies before a CIED implantation and who prescribed anticoagulants after a CIED implantation were also excluded. We examined the characteristics and incidence of cardioembolic stroke events and investigated the relationship between new-onset AF detected by CIEDs and the incidence of cardioembolic stroke events. We enrolled 710 consecutive patients who were followed-up at our CIED clinic. We excluded 295 patients who had a history of AF or had received anticoagulant therapies before a CIED implantation. Moreover, we excluded 66 patients who were founded to have been prescribed anticoagulants after a CIED implantation. Finally, 349 patients (follow-up period, 75±58 months; age, 70±16 years; male sex, 64%; defibrillator, 51%) were included in this study. The mean CHA2DS2-VASc and HAS-BLED scores were 2.8±1.5 points and 1.7±1.6 points, respectively. During the follow-up period, 24 patients (6.9%) had cardioembolic stroke events. Receiver-operating characteristic curve analysis confirmed that 48 seconds was the best cut-off value of AF duration detected by CIEDs for cardioembolic stroke events (AUC = 0.72). Multivariate analysis demonstrated that independent predictors for cardioembolic stroke were left atrial diameter ≥40mm (odd ratio [OR] 3.8, 95% confidence interval [CI] 1.1-6.2, p=0.019), new-onset AF ≥ 48 seconds (OR 5.4, 95% CI 2.1-14 P = 0.0008). AF detected by CIEDs that lasted for ≥ 48 seconds was risk factor of cardioembolic stroke events in this population.