Intraoperative ICD-testing is traditionally performed in many hospitals in order to ensure reliable sensing, detection, and defibrillation of induced ventricular fibrillation. The technical progress of defibrillators allows rapid detection and delivery of high energy shocks which defibrillates effectively in the vast majority all patients at implant. This review describes arguments pro and contra of systematic testing of the defibrillation threshold in all patients. Many reasons argue against testing in all patients: experimental considerations, patients' specific and nonspecific factors, e.g., underlying severity of cardiac disease, ischemia, and medication, as well as factors specific to the ICD system, e.g., implanted type and location of electrodes and active cans. Finally, the testing method is very important, since it bears the risk of false negative test results because the a priori probability of a positive test result is >95%. Therefore, data from prospective randomized studies are necessary in order to abandon the tradition of ICD-testing on an evidence-based background.