In response to the letter by Dr. Salinas and colleagues, we agree that the conclusions of our study (1) regarding the relationship between new-onset atrial fibrillation (AF) and embolic events are limited by the low number of events, and this has been clearly stated in the study limitations section and conclusions of our manuscript. In these sections, we also emphasized the need for further studies in larger series of patients. However, we believe that the point questioning the link between the occurrence of new-onset AF and embolic events lacks a solid scientific rationale. Atrial fibrillation is well recognized as 1 of the main causes of cerebrovascular embolic events, and several studies have shown an association between new-onset AF and cerebrovascular events in many circumstances, including cardiothoracic surgery, myocardial infarction, and sepsis. Indeed, the occurrence of short, silent episodes of AF has recently been associated with a higher rate of cerebrovascular events (2). Dr. Salinas and colleagues argue that no plausible relationship could be established in those patients having an embolic event several days after an AF episode. To this effect, the elegant work of Glotzer et al. (3) clearly shows an increased risk of embolic events up to 30 days after short episodes of AF. Furthermore, several studies demonstrated a significant delay in the recovery of atrial function leading to an increased risk for embolic events after AF cardioversion. This delay was even longer in the presence of left ventricular hypertrophy and in patients with cardioembolic risk factors leading to a high CHADS2 score (4- 5), which was the case in most patients included in our study. Also, we showed a much higher rate of embolic events (up to 40%) in patients with short AF episodes who did not receive anticoagulation therapy, and that is another argument indirectly supporting a potential relationship between atrial arrhythmias and embolic events. The occurrence of stroke is a major issue in the field of transcatheter aortic valve implantation (TAVI), and knowledge of the mechanisms involved in such events will allow us to implement the appropriate preventive strategies. Our work represents a very first step in this direction, and it will be followed by a large multicenter study (ongoing) evaluating the factors associated with this important complication.