13th International Congress
THE "NEW FRONTIERS"
OF ARRHYTHMIAS 1998

January 24-31, 1998
Marilleva, Trento, Italy

RT-41

Analysis of the non-linear relationship between cycle length and QT interval in healthy subjects

Gabriella Malfatto, Antonio Zaza*, Gastone Leonetti, Mario Facchini.
Divisione di Cardiologia, Istituto Scientifico Ospedale San Luca, Istituto Auxologico Italiano IRCCS, *Dipartimento di Fisiologia e Biochimica Generali, Universitą di Milano, Milano, Italy

Heart rate is the major determinant of action potential duration at the cellular level and of QT interval on the ECG, and attempt have been made in the last decades to define their relationship. In the clinical setting, the Bazett's formula has gained a widespread use for its simplicity, but the influence of heart rate on repolarization and QT is far more complex¹. Recently, a linear relationship has been used to describe the rate dependence of QT interval in normal subjects and in patients with long QT syndromes^2,3, but this is an obvious oversimplification. Indeed, in the experimental setting, it has been demonstrated that heart rate, or the cycle length (CL) of stimulation of a cardiac cell, may bear various kinds of relationship with the duration of action potential, depending on whether the system was in dynamic conditions or in steady-state. In the first case, i.e. during restitution, the APD/CL relationship is exponential⁴. In the second case, i.e. during constant rate, the relationship is hyperbolic^4-6: APD=APDmax*CL/(CL*CL₅₀). In this equation, APDmax (APD extrapolated at infinite CL) is a rate-independent measure of APD^5,6.

The validity of this latter steady-state relationship has never been verified in the clinical setting so far. We therefore decided to use the hyperbolic fitting to test whether it could describe the relationship between RR intervals (i.e. cycle length) and QT (i.e. action potential duration) in humans within a physiological range of heart rates, in a situation quite different from the extremes low and high rates used experimentally. In 14 normal subjects (age 37 ± 13 years, 5 females), undergoing a bicycle stress test (steps of 25 W every 2 min) as a routine evaluation for sports practice, the steady-state heart rate at the various loads (expressed as RR interval in ms) and the QT duration in lead V₅ were measured at 50 mm/s and 20 mV/cm. The QT/CL relationship was well fitted for each patient by the hyperbolic equation (R > 0.95, Fig. 1). In this healthy population, QT_max was 426 ± 43 ms, women had longer QT_max than men (472 ± 66 vs 409 ± 35 ms, p < 0.05). In an additional subject treated with 200 mg/die of amiodarone for recurrent lone atrial fibrillation, QTmax was 717 ms, pointing to the increase of the intrinsic duration of repolarization as a consequence of I_K blockade. The values of QT_max were of the same order of magnitude, even if slightly lower, of those obtained from right ventricular monophasic action potentials in four patients undergoing an electrophysiological study, in whom APD_max was 487 ± 26 ms. Moreover, they are consistent with the values from QT/CL and APD/CL relationships estimated in the experimental setting from cardiac tissues^4-6. The shorter QT_max obtained in the stress test with respect to that found in the catheterization laboratory and in vitro may be the consequence of the attendant adrenergic activation during exercise: catecholamines do indeed shorten APD_max⁵.

Fig. 1: Example of a QT/RR relationship in a healthy women 28 years old. The hyperbolic fitting (R 0.94) is shown together with its confidence limits: note that this fitting takes care of the tendency of QT intervals to plateau at longer RR intervals.

In conclusion, a non-linear, hyperbolic relationship between ventricular repolarization and cycle length may be demonstrated - even in the narrow physiological range of heart rates during a routine stress test - in normal subjects, simply measuring RR and QT intervals. The parameter QT_max indicates the intrinsic characteristics of repolarization without any confounding rate influence^4-6. As it has been already demonstrated, women and patients under class III agents had longer repolarization. Moreover, sympathetic activation seems to shorten repolarization also in the clinical setting. The analysis of QT_max can therefore help the understanding of repolarization in its physiological and pathophysiological aspects, such as during drug therapy, or during diseases like myocardial infarction or long QT syndromes.