In a prospective and parallel, randomized study, the long-term stability of epicardial defibrillation threshold was evaluated in 22 patients, using a patch-patch lead configuration at the time of implantation and generator replacement. The concomitant antiarrhythmic drug treatment consisted of either mexiletine (720 mg/day) or amiodarone (400 mg/day) and was administered to patients in a randomized and parallel manner. During a mean follow-up of 24 +/- 6 months, the defibrillation threshold increased significantly from 14.3 +/- 2.8 to 17.9 +/- 5.3 J (p < 0.05) for the entire patient group. The increase in the chronic defibrillation threshold was due to a marked increase in defibrillation energy needs in the subgroup of patients receiving amiodarone. Whereas no significant change in the defibrillation threshold was documented in the subgroup of patients receiving mexiletine, the mean defibrillation threshold increased from 14.1 +/- 3.0 to 20.9 +/- 5.4 J (p < 0.001) in those receiving amiodarone. In all patients with increased defibrillation thresholds, reevaluation showed a reduction in the defibrillation threshold after discontinuation of antiarrhythmic drug therapy. The only variable associated with an increase in the chronic defibrillation threshold was amiodarone treatment. These findings suggest that the defibrillation threshold should be measured at each generator replacement and in case of a change in antiarrhythmic drug treatment. In particular, if amiodarone treatment is initiated, it is recommended that the defibrillation threshold should be reevaluated to ensure an adequate margin of safety.