Exercise testing may be used to assess symptoms occurring on effort, to search for and evaluate arrhythmias or conduction defects, antiarrhythmic drugs, pacemakers or implantable cardioverter defibrillators. Interpretation of exercise testing is difficult because of the complexity of the factors in play. Exercise itself induces changes in myocardial metabolism and the autonomic nervous system, the nature and importance of which are influenced by the underlying cardiac disease and the presence of cardiac failure or myocardial ischaemia. This is particularly true when studying the behaviour of arrhythmias on effort, which depends on many parameters, in that they may appear or disappear during exercise, irrespective of their relationship to autonomic nervous system activation. The main problem lies in the interpretation of changes in the heart rate before the onset of an arrhythmia. The sinus rhythm is both a passive indicator of the vago-sympathetic equilibrium and one of the determining factor of the arrhythmia (relationship to the rate), but it is, itself, dependent on the presence of myocardial dysfunction, a source of arrhythmias, and its changes then become difficult to interpret. These reasons explain why exercise testing is certainly a valuable tool in assessing arrhythmias but the poor reproducibility, especially in the evaluation of ventricular arrhythmias, advises prudence in the interpretation of results.