We thank Madias et al for their interest in our study. Herewith, we address their comments point by point.

First, we decided a cutoff point as 10 days according to the average of admission period and the average recovery time from LV systolic dysfunction. It is possible that our results could have been different if the cutoff point as a long-term period was more than 30 days, but unfortunately we did not have enough samples to access to that measure.

Regarding diabetes mellitus (DM), our results showed that the prevalence of DM in patients of the delayed recovery group was greater than that in patients of the early recovery group (26% vs 14%; p = 0.2), which did not reach the statistical significance. Even if the prevalence of DM was included in the multivariate logistic regression model, we have confirmed that the overall results would not become different. There were also no significant differences in the prevalence of serious co-morbidities and critical condition between patients with and without DM.

Finally, our results showed that the use of calcium channel blockers might support recovery from LV systolic dysfunction in patients with TC. We agree that calcium channel blockers might have a protective effect from calcium sequestration of the cardiomyocytes, consisting of contraction bands consistent with cytoplasmic calcium overload, in the pathophysiological cascade of TC. However, at the moment, it is just an observational study, and further studies will clarify these mechanisms.