RESUMEN
AIM: Atrial fibrillation (AF), which can lead to cardioembolic stroke, is often not properly diagnosed in hospital outpatient departments or medical clinics. We therefore used a pulse analysis to screen patients for AF, and examined the benefits of using this method in screening. METHODS: We performed screening of the hospital's first-visit and ambulatory patients during the afternoon in 2014 (total number, 50,875; true number, 16,356), mainly targeting patients older than 65 years of age. Among the true number of outpatients, the device was used on 5,013 patients, 8,656 times. We independently developed a pulse analysis software application which analyzed the pulse interval variation. We assessed the accuracy of this analytical method in the detection of AF. RESULTS: AF was detected in 56 patients, who were considered for or introduced to anticoagulation treatment. In their cases, the method was considered useful for detecting undiagnosed or untreated AF. This figure amounts to 0.34% of all outpatients and 1.1% of the patients who were screened in 2014. The average age was 76.9±7.7 years, 67.9% of the patients had a CHADS2 score of more than 2, half had a history of arrhythmia in the past, and 37.5% were first-visit patients. The sensitivity of the device used was 89.7%. CONCLUSIONS: Using the method described in this study, we detected asymptomatic AF in numerous patients, and demonstrated that this method is potentially useful in screening outpatients for asymptomatic AF.