[Experimental arrhythmia models. Critical study of correlations with arrhythmias observed in clinical practice]

A number of different in vitro or in vivo experimental tests or models are now available for the demonstration of antiarrhythmic properties. In vitro electrophysiological studies on cells or healthy tissues usually establish the properties of antiarrhythmic agents, rather than their actual antiarrhythmic properties. New information has recently been provided by studies on abnormal tissues or tissues which have been made abnormal experimentally. The methods used to cause this dysfunction are discussed to determine their possible clinical relevance. There are over 50 in vivo arrhythmia models but only 2 or 3 are in common usage: aconitine or digitalis intoxication and experimental myocardial infarction (Harris' method and variants). The study of thresholds to ventricular fibrillation is more a study of an electrophysiological parameter than a study of an arrhythmia. Statistical studies require the uniformity of an experimental model and its stability (or relative stability) over a sufficiently long period from one animal or preparation to another. These conditions widen the gap between some experimental models and clinical reality. Present experimental models are essentially models of acute arrhythmias as experimental animals with chronic degenerative heart disease are not usually available (the problem of cardiomyopathic hamsters and Japanese woodcocks is discussed). The failure to appreciate the differences between species, the direct action of the drug under study (particularly on the autonomic nervous system), the hemodynamic changes induced by the experimental conditions, and the pharmacokinetics of the drug, often make extrapolations to clinical practice very difficult. Although the recent advances in our understanding of the mechanisms of arrhythmias derived from these models have not lead to any significant changes in the clinicians' attitude to the choice of antiarrhythmic therapy or to the introduction of new criteria of antiarrhythmic efficacity, none of the active antiarrhythmic agents used in clinical practice have been submitted to these antiarrhythmic screening tests. These models even provide an "honourable" classification of the relative efficacity of antiarrhythmic drugs.