Implantable defibrillator at end of life with emphasis on deactivation and guideline compliance

Abstract

Introduction: Implantable Cardioverter Defibrillators (ICD) have been demonstrated to improve survival in cardiac patients with high risk for sudden death. The incidence of ICD implantations is increasing worldwide. The cause and nature of death in the ICD population has been insufficiently investigated. In 2010 consensus statements were published to address and highlight the management of ICD patients who were nearing end of life. The overall aim of this thesis was to study patients with implantable defibrillator in end of life as well as physicians’ knowledge about ICD treatment and compliance to guidelines concerning ICD management.

Methods and Results: Study I. An observational study exploring intracardiac electrograms from 125 deceased ICD patients. Ventricular tachyarrhythmia occurred in 35% of those patients during the last hour of their lives, and 31% received shock treatment during the last 24 hours. 52% of the patients had a do-not-resuscitate (DNR) order, but still had shock therapy active 24 hours before death in 65% of cases.

Study II. An observational analysis of 65 deceased ICD patients with DNR order. The majority (86%) of patients were treated in hospitals, mainly (63%) in university hospitals and (33%) in Cardiology wards. Patients had active ICD therapy in 51% of cases despite a DNR order, and 24% of those patients experienced shocks as consequence. Patients with active ICD therapy had a median of 4 days (IQR 1-38) from decision of DNR to death. In the 38% of the patients who had ICD therapy deactivated, the deactivation was performed two days or more after the DNR decision.

Study III. A cross-sectional comparative study with development of a questionnaire that was distributed among 432 physicians in 14 hospitals with a response rate of 99.5%. Many (83%) of the physicians said they had experience with ICD patients and 68% of physicians rated their ICD knowledge to be low. Sufficient knowledge regarding ICD therapy defined according to pre-specified criteria was observed in 41%. Physicians in Cardiology departments scored significantly higher than others. Only 30% of physicians in Internal Medicine and 19% of physicians in Geriatrics reached sufficient knowledge compared with 71% in Cardiology.

Study IV. A comparison of two cohorts of ICD patients who died in hospitals before and after the implementation of new guidelines. Almost two-thirds of ICD patients in the two groups died in wards other than Cardiology. In group 1 patients had a DNR order in 54% compared to 73% in group 2. Shock deactivation was present in 52% of the patients in group 1 compared to 67% in group 2. The difference in deactivation rate between group 1 and 2 was only significant (p=0.016) for DNR patients treated in Cardiology. A significant difference (p=0.038) was also found in deactivation within group 2 between DNR patients treated in Cardiology vs. DNR patients in Non-Cardiology.

Conclusions: Patients with ICD dies in hospitals and the majority are treated in Non-Cardiology wards. Approximately one-third of patients with an ICD have ventricular tachyarrhythmia at end of life. Many patients have a DNR order but still have shock therapy active and thus receive unnecessary shocks before death. Deactivation rates have increased but not significantly since publications of international guidelines on the management of ICD in patients at end of life. Physicians in Cardiology, Internal Medicine, and Geriatrics have a lack of basic ICD knowledge, possibly affecting their ability to manage ICD patients and may increase the risk for unnecessary suffering for these patients at end of life.