The strategy of sudden cardiac death prevention by implantable cardioverter defibrillator, in primary prevention, is mainly based on the value of ejection fraction. That means that the approach is not really patient specific. A lot of implanted patients will not receive any shock. The implantation of large categories of patients is interesting on a global therapeutical point of view but, when considering the economical aspects, it would be more useful to have better selection criteria in order to obtain a more patient-specific approach, avoiding implanting patients who will never receive shocks. The parameters commonly used to select patients for implantations have a good negative predictive value but a low positive predictive value. Concerning atrial fibrillation the approach is quite different. Antiarrhythmic drug treatment has shown many limitations. Antiarrhythmic drugs are useful and safe in atrial fibrillation patients only if the contra-indications are strictly respected. The main difficulty concerns patients with both heart failure and atrial fibrillation. The story of Dronedarone development is illustrative of the necessity of a patient-specific approach in the treatment strategy of atrial fibrillation. The ATHENA trial made with Dronedarone showed a benefit in patients with underlying heart disease but no patient with advanced cardiopathy was included in the study. On the contrary, the PALLAS trial has clearly shown that the drug is contra-indicated in patients with any type of heart failure. In atrial fibrillation, a patient-specific approach is mandatory. This review illustrates the dichotomy of the two different approaches.