Does pulmonary vein isolation prolong QT-interval?- Authors' reply.

We would like to thank you for the opportunity to respond to the issues raised in Dr Chikata’s letter titled ‘Does pulmonary vein isolation prolong QT-interval’. In their letter to the editor, they raise three concerns regarding our study which we will further discuss in this response: (1) the use of an automated algorithm to measure the QT-interval; (2) the use of antiarrhythmic drugs (AAD) in our study population, and (3) the heart rate correction methods for QT-intervals.

Indeed, long QT syndrome experts have advocated manual QT-interval assessment because of erroneous QT-interval measurements determined by available algorithms. This was the reason why we developed and validated our own QT-interval algorithm which was also used in this study. We validated our algorithm by comparing the QT-intervals with the mean QT-interval measured by three independent observers. This resulted in an intra-class coefficient of 0.981 and a mean difference in a Bland–Altman analysis of −0.38 ms with limits of agreement of −25.41:24.65 ms. The inter-observer variability of the three observers showed very similar results. Based on these results we concluded that ouralgorithm is as accurate in determining QT-intervals as manual observers are and, hence, we believe that the use of our algorithm to study QT-interval changes after pulmonary vein isolation (PVI) is justified and that the use of our methodology does not explain differences in raw RR- and QT-interval between their and our study.

We do agree with Dr Chikata et al. that AAD might affect both the RR- and QT-interval in our study and could be an explanation for the differences in raw RR- and QT-intervals between their and our study. However, as we already mentioned in the discussion of our original paper, we believe that our study population better reflects the real-world population of patients undergoing pulmonary vein isolation since these patients are frequently treated with AAD before ablation. Therefore, we believe we induce a bias if we would only include patients without AAD.

Finally, Dr Chikata et al. suggest that Bazett’s formula is superior in detecting small QTc changes. In our opinion, however, our reported QTc prolongation 1 day after PVI when using Bazett’s formula is most likely caused by the well-known over- and underestimation of QTc at RR <1000 ms and >1000 ms, respectively. Furthermore, we would like to highlight that in the original paper by Chikata et al. the average QTc Bazett increased only mildly from 404.9 ± 25.2 ms at baseline to 420.1 ± 21.8 ms at 3 months post-PVI. Therefore, although the increase in QTc Bazett was significant, this does not automatically imply it is clinically relevant since the average QTc Bazett remains well within the normal range.

Recently, Chikata et al. published an interesting case report of Torsade de Pointes due to QT prolongation after PVI. Clearly, we have different experiences and therefore endorse their suggestion to perform multi-centre studies with large number of patients and many time courses to clarify the influence of PVI on the QT-interval. However, we do not agree that only patients off antiarrhythmic drugs should be included but think the inclusion criteria should meet daily practice.

Conflict of interest: none declared.