Potpara et al5 show that progression to permanent AF in a subpopulation of young, otherwise healthy patients with lone AF may be an additional risk-stratification tool to identify those patients developing an increased risk for adverse cardiovascular events. This means that the transition from short self-terminating episodes of paroxysmal AF to chronic persistent and permanent AF may not simply be an “electrical” evolution but may be a marker of increasing risk, including thromboembolic risk.5 This finding may challenge the widely adopted belief that paroxysmal or intermittent AF and chronic AF carry the same risk as initially suggested by analysis of Stroke Prevention in Atrial Fibrillation (SPAF) trials.7 In daily practice, it is not easy to monitor the presence and duration of AF, but interesting observations come from studies on patients implanted with pacemakers with diagnostic capabilities that allow continuous monitoring of the AF burden.8 Some observations on patients carrying an implanted device (a subset of patients in whom arrhythmia burden is continuously monitored) suggest that the presence and duration of AF episodes have clinical implications, since they are associated with thromboembolic risk.8 In a recent study,8 the combination of CHADS2 (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, and prior stroke or transient ischemic attack) score with variable duration of AF burden (between 5 min and 24 h or >24 h) increased the prediction of thromboembolic risk achieved by CHADS2 score. In this regard, it is noteworthy to consider that in the study by Potpara et al,5 neither baseline CHADS2 score nor CHADS2 score at the time of a thromboembolic event was predictive for thromboembolism (c-statistic, 0.50). As the authors stress, the widely used CHADS2 schema may be inadequate to identify patients with AF who are truly low risk for stroke and thromboembolism. Indeed, CHADS2 score was validated in a population of patients with AF affected by underlying heart disease, with a mean age of 72 years and treated with aspirin. Therefore, the ability of CHADS2 score to properly predict thromboembolic risk in the specific setting of patients presenting with lone AF may not be the same as was observed in the validation cohort,9 in which mean age was 72 years and hypertension, heart failure, or history of stroke/transient ischemic attack were present in a substantial proportion of patients.