Deactivation of Implanted Cardioverter-Defibrillators at the End of Life in the Setting of an Outpatient Clinic

Dear Editor:

Indications for implantable cardioverter-defibrillators (ICD) are ever more frequent, and implanting physicians are more often faced with terminally ill patients for whom deactivation of life-prolonging therapies has to be considered. ¹ A survey of the European Heart Rhythm Association (EHRA) recently demonstrated that the majority of physicians consider ICD deactivation in patients at the end of life; nevertheless, only 6% of clinicians perform it routinely. ² Unfortunately, withdrawal of therapy is even uncommon in palliative hospice circumstances, ³ even though ethical and legal aspects have recently been discussed. ¹

We prospectively evaluated terminally ill patients where the ICD was actively deactivated between January 2009 and November 2011. For assessing the impairment in daily life activities, the Barthel Index ⁴ was used. It was noted whether the decision was requested by the patient, physician triggered, triggered by a defect such as of the system, or initiated by relatives of the patient. If the outpatient clinic was not informed about the death of the patient, a telephone contact with the family doctor was initiated once a year to gain follow-up information.

From January 2009 until November 2011, five ICDs (one female, four male; mean age 78.4±0.55 years) were deactivated in our outpatient clinic. Deactivation was performed in 0.07% (7200 contacts in three years) of patient contacts. The Barthel Index was between 0 in a patient in a vegetative state and 75 in one 78-year-old male (dialysis therapy stopped). The concomitant terminal illness was a neurological disorder in the majority (n=2 severe dementia, n=1 Parkinson's disease, n=1 cerebral bleeding); and chronic kidney disease in one patient. The deactivation of the device was done 4±2.9 years after initial ICD implantation. In three patients the deactivation was triggered by device defect or battery depletion. In one patient, the deactivation was patient requested, and in one patient the wish was expressed by the family. The pacemaker function remained unchanged in all patients. Two patients with a secondary indication for the ICD are still alive. Three patients died (two days, three months, and one year after deactivation). The time patients and their relatives spent in the outpatient clinic was 50±14.7 minutes.

The main findings of our observation are (1) deactivating ICD systems in an outpatient setting due to terminal illness is a rare event, (2) most of the ICD deactivations were triggered by device defects requiring revision, (3) neurological disorders were the predominant reason, and (4) discussion is time consuming in the setting of an outpatient clinic. These findings reflect insufficient advance care planning as recently addressed by nurses from the Mayo Clinic. ⁵ In the circumstance of a device malfunction, the operative risk has to be discussed and weighed against the potential risk of deactivation. Management of ICD patients at the end of life is a very sensitive issue and depends on the individual, ethical aspects, the family situation, and may even vary between countries. Nevertheless, ICD therapy is still focused on life prolonging, and the handling of dying with the device is disregarded.