I read with interest the recent article by Kirkpatrick et al regarding deactivation of implantable cardioverter–defibrillators (ICDs) in end-of-life care. The authors surveyed 278 patients regarding their attitude to ICDs at the end of life. They specifically asked about advance directives and patient opinion on whether subjects defined ICD deactivation as physician-assisted suicide. Half of patients had some form of advance directive, but only 3 had included a plan for their ICD at the end of life. 44% of patients in 2009 and 12% of patients in 2010 considered ICD deactivation to be physician-assisted suicide. This was a worrying statistic because physician-assisted suicide is illegal in most parts of the world and most states in the United States. Patients interviewed in 2010 were more likely to want their ICD deactivated at the end of life than those interviewed in 2009 (49% vs 0%).

One point that was not specifically addressed within the study was patients’ understanding of the ICD at the end of life. Of the 139 patients who had an advance directive, only 3 had included a plan for their ICD. Our data in the United Kingdom suggest that patients’ understanding of the function of the device may be poor, creating a barrier in end-of-life discussions. Doctors often assume that patients understand the implications of their device and what “deactivation” means. It may be that some patients who considered ICD deactivation to be analogous to physician-assisted suicide were of the impression (as one of our patients was) that ICD deactivation would immediately lead to death because the ICD was the only thing “keeping his heart alive.”

Increased awareness of the specific issues an ICD creates at the end of life and formal guidance on this such as the American College of Cardiology/American Heart Association and Heart Rhythm Society consensus statement has encouraged physicians to discuss end-of-life and advance directives with patients with an ICD and this may explain the observed differences in opinion between patients interviewed in 2009 and 2010. However, for these discussions to be effective, it is important to ensure that patients understand the function of their device and what deactivation means (and does not mean) in an end-of-life setting.