Background: Decision-making in primary prevention is not always trivial and many clinical scenarios are not reflected in current guidelines. To help evaluate a patient’s individual risk, a new score to predict the benefit of an implantable defibrillator (ICD) for primary prevention, the MADIT-ICD benefit score, has recently been proposed. The score tries to predict occurrence of ventricular arrhythmias and non-arrhythmic death based on data from four previous MADIT trials. We aimed at examining its usefulness in a large single-center register of S-ICD patients with various underlying cardiomyopathies. Methods and results: All S-ICD patients with a primary preventive indication for ICD implantation from our large single-center database were included in the analysis (n = 173). During a follow-up of 1227 ± 978 days, 27 patients developed sustained ventricular arrhythmias, while 6 patients died for non-arrhythmic reasons. There was a significant correlation for patients with ischemic cardiomyopathy (ICM) (n = 29, p = 0.04) to the occurrence of ventricular arrhythmia. However, the occurrence of ventricular arrhythmias could not sufficiently be predicted by the MADIT-ICD VT/VF score (p = 0.3) in patients with (n = 142, p = 0.19) as well as patients without structural heart disease (n = 31, p = 0.88) and patients with LV-EF < 35%. Of the risk factors included in the risk score calculation, only non-sustained ventricular tachycardias were significantly associated with sustained ventricular arrhythmias (p = 0.02). Of note, non-arrhythmic death could effectively be predicted by the proposed non-arrhythmic mortality score as part of the benefit score (p = 0.001, r = 0.3) also mainly driven by ICM patients. Age, diabetes mellitus, and a BMI < 23 kg/m2 were key predictors of non-arrhythmic death implemented in the score. Conclusion: The MADIT-ICD benefit score adds a new option to evaluate expected benefit of ICD implantation for primary prevention. In a large S-ICD cohort of primary prevention, the value of the score was limited to patients with ischemic cardiomyopathy. Future research should evaluate the performance of the score in different subgroups and compare it to other risk scores to assess its value for daily clinical practice