Different voltage dependence of I blockade in nonselective I blockers causes their opposite effects on early afterdepolarization in drug-induced arrhythmia.

Several false-positive results in the human ether-à-gogo-related gene test suggest that blockers of the rapid component of delayed rectifier K current (I) do not necessarily produce drug-induced arrhythmias. Specifically, the occurrence of early afterdepolarization (EAD) differs among I blockers, even if the prolonged action potential duration is in the same range. To predict EAD in drug-induced arrhythmias, we proposed a prediction method based on the mechanisms underlying the difference in frequency of EAD among nonselective I blockers. The mechanisms were elucidated by examining how different blockade kinetics of L-type Ca current (I) affect the frequency of EAD, using mathematical models of human ventricular myocytes. Addition of voltage-independent I blockade resulted in the suppression of EAD. However, when voltage-dependent I blockade kinetics of amiodarone, bepridil, and terfenadine were incorporated into I in the model, bepridil and terfenadine induced EAD more than the voltage-independent I blockade, while amiodarone suppressed EAD more effectively. Opposite effects were accounted for by the difference in I blockade at negatively polarized potential. EAD occurrence was found to be associated with I blockade measured at -20 mV. These results suggest that voltage dependence of I blockade may be useful in predicting the different risks of nonselective I blockers.