부산대학교 도서관

Atrial fibrillation (AF) is the most common cardiac arrhythmia in adult and old people and represents a risk factor for stroke. Correct AF individuation bears strong relevance in primary and secondary stroke prevention. Our goal was to evaluate the reliability of a low-cost, non-invasive technology in detecting AF in acute stroke patients. AFib model BP3MQ1-2D (Microlife USA, Dunedin, FL) showed good accuracy in diagnosing AF in a general cardiologic outpatient population. We carried out an observational study in patients with recent stroke. We studied 207 subjects, 103 men, 104 women, mean age (±SD) 77.7 ± 11.34 years, who underwent a test by AFib device with indication of AF or lack of it. The golden standard was a 12-lead EKG done immediately and evaluated by a certified cardiologist. We computed estimates of Sensitivity and Specificity and their 95 % confidence intervals (CI). AF was present in 38 subjects from the sample of 207 (18.4 %). AFib correctly demonstrated AF in 34 and failed diagnosing AF in 4 cases; on the other hand, AFib correctly excluded AF in 167 and caused an erroneous diagnosis of AF in 2 cases. The Sensitivity was 0.895 (95 % CI 0.7597-0.958) and the Specificity was 0.988 (95 % CI 0.958-0.997). The AFib device global accuracy was 0.971 (95 % CI 0.938-0.987). This device was able to detect AF with high specificity and a good sensitivity. This device may be considered as an accurate tool in detecting AF in stroke patients.