How to correct the QT interval for the effects of heart rate in clinical studies

Abstract

Much inter- and intra-subject variability in the QT interval in health and disease is accounted for by differences in heart rate, leading to difficulties when determining the effects of disease and drugs on the QT interval. Traditionally, heart rate correction formulae have been used to overcome this problem in man. However, the commonly used Bazett's heart rate correction formulae (QT=QT C√RR interval) does not remove the effect of heart rate; indeed, it overcorrects at high heart rates. Fredericia's formula (QT=QT C× 3√RR interval) does remove the effects of heart rate; this is the preferable formula, if one is to be used. However, all formulae make assumptions about the nature of the QT–heart rate relationship, assumptions that may not apply to those with disease or on drugs. A more intellectually rigorous approach to QT interval–heart rate correction is to determine the QT–heart relationship for each individual, using data obtained from exercise tests or 24-h Holter tapes. The best mathematical relationship (linear, exponential, etc.) is obtained from analysis of this data, and is used to determine the QT interval at a heart rate of 60 bpm, the QT 60. The QT 60 measure makes no assumptions about the nature of the QT interval–heart rate relationship, removes the dependence of QT interval on heart rate, and maintains genuine biological differences in the QT interval. It should become the standard in QT interval–heart rate correction.