Response to letter regarding article, "Embolic stroke of undetermined source in the athens stroke registry: a descriptive analysis".

As it is mentioned in our article, 142 (51.6%) patients with Embolic Stroke of Undetermined Source (ESUS) had a 24-hour ambulatory Holter monitoring, whereas 195 (70.9%) ESUS patients had continuous electrocardiographic (ECG) monitoring for 1 week or until discharge from the stroke unit. For the latter patients, continuous ECG monitoring was observed by the trained nurse personnel of the acute stroke unit and intermittently analyzed by the treating physician. Mahagne et al point to the group of our patients who had only continuous ECG monitoring (but not an ambulatory 24-hour Holter monitoring) and comment that “it is not possible to classify patients in ESUS group if these minimal cardiac investigations are not performed.” We disagree with this opinion. It was previously shown by the Heidelberg group that the use of Holter (for a period of 24 hours) does not provide any additional benefit compared with continuous monitoring with intermittent analysis by trained staff alone. In particular, among the entire population of 370 patients with ischemic stroke or transient ischemic attack, no patient with atrial fibrillation (AF) would have been overseen performing only the continuous ECG monitoring. What is even more striking in this study is that compared with the 24-hour Holter monitoring, the prolonged (ie, during entire hospitalization because it was also the case in our study) continuous ECG monitoring lead to significantly higher detection rate of AF. In particular, ≈85% of patients with a newly diagnosed AF detected by prolonged continuous ECG monitoring would have been overseen by performing only the 24-hour Holter monitoring. Therefore, we think that our strategy for the detection of covert AF, as described in the Methods section of our recent article, is not suboptimal compared with the minimal requirement of a 24-hour ambulatory Holter monitoring but rather perhaps superior. This is also supported by our data where the rate of detection of covert AF during the postdischarge follow-up was numerically higher in ESUS patients who had only prolonged continuous ECG monitoring during hospitalization compared with ESUS patients who had only an ambulatory 24-hour ambulatory Holter monitoring (27.5% versus 30.8%; P=0.64).