Background: Use of implantable cardioverter defibrillators is increasingly common. As patients approach the end of life, it is appropriate to deactivate the shock function. Aim: To assess the prevalence of implantable cardioverter defibrillator reprogramming to deactivate the shock function at the end of life and the prevalence of advance directives among this population. Design: Following a previously established protocol available in PROSPERO, we performed a narrative synthesis of our findings and used the logit transformation method to perform our quantitative synthesis. Data sources: We searched seven bibliographic databases (Embase, Cochrane Central register of controlled Trials, Medline-Ovid, Web-of-Science, Scopus, PsychInfo, and CINAHL) and additional sources until April 2019. Results: Of the references we identified, 14 were included. We found a pooled prevalence of implantable cardioverter defibrillator reprogramming at the end of life of 28% (95% confidence interval, 22%–36%) with higher reprogramming rates after the recommendations for managing the device at the end of life were published. Among patients with advance directives, the pooled prevalence of advance directives that explicitly mentioned the device was 1% (95% confidence interval, 1%–3%). Conclusions: The prevalence of implantable cardioverter defibrillator reprogramming and advance directives that explicitly mentioned the device was very low. Study data suggested reprogramming decisions were made very late, after the patient experienced multiple shocks. Patient suffering could be ameliorated if physicians and other healthcare professionals adhere to clinical guidelines for the good management of the device at the end of life and include deactivating the shock function in the discussion that leads to the advance directive.