Background : It is well-known that the QT interval and the RR change in the same direction on electrocardiograms (ECGs). This change appears to be modified by various factors such as sex, age, and race. Accordingly, it is necessary to investigate the joint distribution of QT and RR in resting ECGs of healthy young Japanese men, before undertaking trials to test the effect of drugs on QT. To eliminate the effects of RR on QT and obtain new values that are independent of the RR intervals, many formulas have been proposed. Among them, Bazett's and Fridericia's corrections are the most widely used. We compared these two correction methods by means of correlation coefficient and determined which one was best fit to determine the independence between QTc and RR.
Methods : By using all cases in the subject population, the QTc values corrected according to Bazett (QTcB) and Fridericia (QTcF) were calculated. The correlation coefficients between QTcB and RR, and between QTcF and RR were calculated. In doing such a comparison, the usefulness of these two corrections was assessed.
Results : The mean of the conditional QT distribution, given linear RR, increased along with the increase of RR. We found that the correlation coefficient between QTcF and RR was almost null, whereas that between QTcB and RR was -0.5 and it was statistically significant.
Conclusion : We showed that the mean of the conditional distribution of QT given RR was closely approximated by Fridericia's correction formula. We also transformed QT to QTcF by means of Fridericia's correction formula and succeeded in obtaining QTcF values that were statistically independent of the RR. From these results, we conclude that the application of Fridericia's correction is more suitable for the evaluation of QT in resting ECGs of healthy young Japanese men.
View full abstract