In reply

I thank Lugassy et al for their interest in our study. However, they provide poor reasoning for citalopram overdoses requiring prolonged monitoring. Although the metabolite didesmethylcitalopram has been associated with cardiac toxicity in dogs, this has not been shown in humans and Lugassy et al confirm this in the case they reference where only citalopram and desmethylcitalopram were detected. In a full pharmacokinetic-pharmacodynamic analysis of citalopram in overdose there was a delay in peak QT prolongation compared to peak drug concentrations. However, simulations based on the model showed that after 13 hours the chance of an abnormal QT occurring in a patient with a normal QT was less than 1%. In the single case reported by Tarabar et al2 the patient had QT prolongation on admission which does not support the case for delayed QT prolongation, as it is quite possible that she simply survived with QT prolongation for over 24 hours prior to presentation