The prevalence of atrial fibrillation increases with age, with rates of 2-5% among people over the age of 60 years. Patients may be highly symptomatic or may suffer from hemodynamic compromise or thromboembolic complications. However, antiarrhythmic drug treatment implies problems like the choice of the suitable drug, the individual benefit/risk profile, and alternative treatment strategies. Experimental and clinical data support the concept that atrial fibrillation in the clinical setting in most cases is due to multiple reentrant wavelets. A critical number of three to six simultaneously circulating reentrant wavelets seems to be necessary for the maintenance of atrial fibrillation. Consequently, antiarrhythmic drugs may terminate or prevent atrial fibrillation by prolonging the refractory period or slowing conduction velocity, thereby leading to conduction block. In clinical practice, antiarrhythmic therapy may act by slowing of the ventricular rate due to depression of atrioventricular nodal conduction or by termination and/or prevention of atrial fibrillation. Digitalis is commonly used for the control of the ventricular rate. Betablocking drugs and verapamil are effective in this respect during exercise performance. For antiarrhythmic conversion and prophylaxis of recurrences of atrial fibrillation, class Ia (e.g., quinidine), Ic (e.g., flecainide and propafenone), and class III (e.g., amiodarone and sotalol) drugs of the Vaughan Williams classification are useful. Presently, no general concept exists whether medical or electrical cardioversion should be used as a first line approach for termination of atrial fibrillation. In the individual patient with atrial fibrillation, the potential benefit of restoring sinus rhythm must be weighed against the morbidity and mortality of the arrhythmia and the morbidity and mortality of the antiarrhythmic agents used.(ABSTRACT TRUNCATED AT 250 WORDS)