Reliable identification of "truly low" thromboembolic risk in patients initially diagnosed with "lone" atrial fibrillation: the Belgrade atrial fibrillation study.

BACKGROUND The CHA(2)DS(2)-VASc (Congestive heart failure, Hypertension, Age ≥75 years, Diabetes mellitus, previous Stroke/transient ischemic attack [TIA], Vascular disease, Age 65-74 years, and Sex category [female gender]) schema recently has been introduced to complement the CHADS(2) (Congestive heart failure, Hypertension, Age >75 years, Diabetes mellitus, and previous stroke or TIA) score and improve the identification of atrial fibrillation (AF) patients at "truly low risk" for thromboembolism. We tested the predictive ability of the CHA(2)DS(2)-VASc, CHADS(2), and van Walraven risk stratification schemes in a cohort of "lone" AF patients with a 12-year follow-up. METHODS AND RESULTS We conducted a registry-based, observational cohort study of 345 patients initially diagnosed with "lone" AF between 1992 and 2007. At baseline, all patients had the CHADS(2) and van Walraven scores of 0, and 262 (75.9%) had a CHA(2)DS(2)-VASc score of 0. During follow-up (or within a year prior to stroke), 228 (66.1%), 234 (67.8%), and 150 patients (43.5%) retained the CHADS(2), van Walraven, and CHA(2)DS(2)-VASc scores of 0, respectively. The overall rate of ischemic stroke was 0.19 (95% CI: 0.18-0.20) per 100 patient years. In the multivariable analysis, only the CHA(2)DS(2)-VASc score of 0 was significantly related to the absence of stroke (odds ratio 5.1, 95% CI: 1.5-16.8, P=0.008). Only the CHA(2)DS(2)-VASc score had a significant prediction ability (c-statistic 0.72 [0.61-0.84], P=0.031). CONCLUSIONS The CHA(2)DS(2)-VASc score reliably identified the "lone" AF patients who were at "truly low risk" for thromboembolism, and was the only tested risk stratification scheme with a significant predictive ability for thromboembolism among lone AF patients.