CardiologyClinical Features

Deactivation of Implantable Cardioverter Defibrillators in end-of-life situations – more progress needed

Written by Mr Paul Nolan, Clinical Lecturer & Cardiac Physiologist, Dept of Health and Nutritional Sciences, Atlantic Technological University


Implantable Cardioverter Defibrillators (ICDs) are a standard therapy to prevent sudden arrhythmic cardiac death. Implantable Cardioverter Defibrillators are proven to reduce mortality in both patients with documented ventricular arrhythmias – secondary prevention, and those at risk of such an arrhythmia – primary prevention. An Implantable Cardioverter Defibrillator monitors a patient’s heart rhythm and if they develop ventricular tachycardia, of a predetermined rate, or ventricular fibrillation, for typically 30 beats, the device will treat the arrhythmia. This is done in two ways, one a painless fast pacing therapy, called anti-tachycardia pacing (ATP), the other being defibrillation or “shock” therapy. Many of these devices are implanted in patients with heart failure and some of those patients may be implanted with a device which also improves the pumping efficiency of the heart, whilst also having the ability to defibrillate. These devices are referred to as Cardiac Resynchronisation Therapy devices or CRT.

The batteries on these devices typically last seven to ten and one of the issues with Implantable Cardioverter Defibrillators is that patient life goals may change, particularly in the setting of the diagnosis of a terminal illness. In this setting a patient may no longer wish to receive painful shock therapy, which may prolong duration, but not quality of life. Given that the number of Implantable Cardioverter Defibrillator implants in Ireland is increasing, at approximately 1000 implants per year, and the duration they may remain implanted means that the number of patients with a terminal illness and an implanted defibrillator is increasing.

Despite the fact that consensus documents from the European Heart Rhythm Association 1 and the Heart Rhythm Society 2 in the US underline the fact that it is ethical, deactivation of shock therapy in Implantable Cardioverter Defibrillators and CRTs remains an issue, with a number of barriers resulting in suboptimal care to patients in end of life situations.

Why is this still important?

Previous studies have shown that in Implantable Cardioverter Defibrillators patients in end of life situations, with a Do Not Resuscitate Order (DNR), over half still had the shock function of their device programmed on. Nearly one quarter of patients had a shock in the last hour of life with over half of these receiving multiple shocks. 3 More recent data from 2019 shows similar results with only one-third of patients having the shock function of their device deactivated and 20% of patients having at least one shock within a month of their death. 4 It would appear that the consensus documents have not had a significant impact on deactivation rates. A Swedish study 5 noted that there was no significant increase in rates of deactivation, likely related to the fact that two-thirds of deaths happened on non-cardiology wards.

Lack of discussion around Implantable Cardioverter Defibrillators deactivation and subsequent shock therapy in end of life situations is a source of frustration for family members, can be distressing for them, the patient and health care providers.

When should discussion happen?

The Heart Rhythm Society Consensus document states that discussion around the management of Implantable Cardioverter Defibrillators at end of life should be included as part of the pre-implant discussion but this is only happening in 4% of centres surveyed across Europe. 6 One of factors leading to this maybe a healthcare providers perception that such discussion might influence a patient to decline the implant, although the majority of a group of patients interviewed would prefer this to be part of a discussion. 7

What we do know is that healthcare providers within cardiology recognise the importance of having the discussion, but that how we currently do it is sub-optimal. There is no consistent approach and that such conversations either do not happen, or happen late in the patient journey. 7, 8 One might assume that advance care plans would be a tool which would naturally lead to a discussion around Implantable Cardioverter Defibrillators shock deactivation, however the evidence suggests it is not. A systematic review of the literature review showed Implantable Cardioverter Defibrillators deactivation was explicitly mentioned in only 1% of advanced directives, 9 this is sobering considering that only 12% of heart failure patients even had documented advanced directives. 10

The recent COVID-19 pandemic introduced another level of complexity to this issue but also crystalised the importance of when these discussions should happen. 11 It is clear that it is best that they happen early in the patient’s ICD pathway, preferably before implant. It should certainly happen when the patient is clinically stable, should be an ongoing in nature but there are also clear identified triggers for discussion which will be outlined below. The requirement for the conversation to be ongoing is highlighted by the fact that patients remember the benefits of ICD therapy as outlined to them pre-implant, but not issues around end-of-life care. 7

What do Healthcare Professionals know and believe?

A 2009 survey of over five hundred US based cardiologists, electrophysiologists, internists and geriatricians, presented with 5 end of life case vignettes, found significantly more would discuss a DNR, rather than ICD deactivation, in 4 out of five cases being 20-30% higher. Despite the 2010 guidance, a 2021 Canadian survey showed that 24% of nurses, 10% of fellows and 7% of cardiologists still believed shock deactivation was unethical. 13 Predictors of such beliefs were sociocultural, including profession, region of training and spiritual beliefs.

Despite clinicians stating that they would discuss ICD shock deactivation in certain circumstances, such as when signing a DNR, there is also an admission amongst noncardiology professionals that lack of knowledge can be a barrier. 8 Other reasons quoted which result in clinicians rarely explicitly discussing ICD shock deactivation includes personal discomfort, lack of experience, time constraints and lack of guidance documentation. 14

Given that 11% of the population over 50 years of age in Ireland suffer from two or more chronic diseases and many of our patients see different teams, and multiple members within the one team, as well as their own General Practitioner, the question of who should partake in discussions around ICD shock deactivation can be a difficult one. With potentially fragmented care, there can be a feeling that the responsibility lies elsewhere. 8 In the UK, data would suggest that Cardiologists often focus on getting the patient better, and this perhaps results in the conversation not being had early enough. Cardiac Physiologists, who follow-up patients with devices, have opportunity to discuss but are reluctant because of lack of knowledge or confidence. In general terms, Heart Failure Nurses, feel they are already having those end of life conversations but perhaps are not explicitly talking about ICD deactivation. 7

The truth is that all teams, involved in the care of the patient, particularly at end of life, need to be aware of the issue of ICD shock deactivation and have the knowledge to initiate the discussion and empower the patient decision.

What do our patients know and believe?

For patients to be part of an informed discussion, they need a good understanding of their device and the end of life issues. A 2014 study showed that over 25% of patients believed deactivation equated to euthanasia, the device had to be surgically removed or that it could not be reactivated if the patient improved15. Disappointingly, five years later, the percentage of patients with those misconceptions had increased to nearly 40%. 16

Unsurprisingly the data around patient perceptions and attitudes is varied, some feeling that such a discussions would give them a sense of control, but others feeling it would raise concerns. 7 Data would suggest that up to 40% of our ICD patients do not wish to have a discussion around end of life issues, with no significant difference between those with and without heart failure. 17 However it also informs us that those who have experienced shocks, have high anxiety levels or concerns about their device are more likely to have these discussions. It has also been shown that past discussion around disease progression were a predictor of openness to discuss ICD deactivation. 18 The data is also telling us that 60% of our patients do want to have such conversations.

A really interesting paper from Stoevelaar in 2020 showed the nuances of patients’ knowledge and attitudes to the issues. Some were unaware it was an option, other were uncertain as to when it would be appropriate and that relatives would play an important role in the conversation. Most would want to get information from their healthcare provider, but not necessarily their cardiologist, feeling others might have more time. Some felt documenting their wishes would help but others had concerns about changing their mind as people adapt to their situations. 19

What the evidence tells us is that we should have ongoing conversations with our patients, tailoring the information to their needs, their current situation and we should include close relatives.

What else can we do better?

The mix of patients’ views on ICD therapy at the end of life supports the need for more open communication and ongoing education on this topic for both patients and providers. A number of projects have aimed at improving either deactivation rates or documentation of patients wishes regarding same. These have all included staff education, changes to end-of-life care and/or DNR documentation and other tools such as reminder posters for staff. 20, 21 In one setting in the UK, education was targeted at all specialities and also on wards where missed deactivation opportunities had been identified by audit. 22

What are the practicalities?

It is clear that discussion with patients need to include education around their device and end of life issues and that these need to be ongoing and developing, aligned to patients’ current disease trajectory. Despite that, there are some clear events which should trigger such a discussion. These would include, significant disease progression, repeated hospitalisations, refractory symptoms, deterioration in quality of life or if a patient or relative requests same.

There are a number of issues to explore prior to discussion, in an effort to overcome the “guardian angel” view of the ICD. The evidence suggests it is desireable to include close relatives as part of the discussion if the patient consents. These include patients experience of their device and therapies, the meaning of the device to them, their understanding of the option regarding their device at end of life in an effort to assess their readiness to have the discussion (Fig 1).

When the discussion moves specifically to deactivation of shock therapy, a number of points must be emphasised, which address some of the misconceptions highlighted in the literature. We first must be careful about language and avoid using phrases such as “switching off the ICD”, rather we should state that we are turning off potentially painful shocks. We should ensure our patients and their close relations have a clear understanding of the deactivation procedure. Firstly that this does no hasten death but simply allows for a peaceful, natural death. Also important is the fact that deactivation is painless, similar to the procedure for a routine device check. It should also be made clear that shock therapy can be turned back on if the patients condition improves or they change their mind, and that the care they are receiving won’t change (Fig 2).

Once the decision has been made, contact should be made with the patient’s current follow-up centre. Most centres in Ireland have a policy and procedure for deactivating an ICD in end of life settings and will endeavour to facilitate same. This will generally involve some form of documentation, such as clear evidence of a shared decision in the notes, and/or a completed deactivation request form. Whilst it is preferable for this to be carried out in a planned manner, during a hospital admission or visit, this is not always possible, and whilst challenging, centres endeavour to support deactivation in the community. This may require the GP or a member of the primary team, for example palliative care, to meet a Cardiac Physiologist at the patient’s home. In emergency situations where the device cannot be programmed in a timely manner a strong donut shaped magnet can be placed over the device. This disables shock therapy for as long as the magnet is over the device.


The consensus documents from the Heart Rhythm Society and the European Heart Rhythm Society are clear that deactivation of the shock function of the ICD is appropriate and ethical in end of life situations. What is clear is that we are falling short in delivering this important aspect of care to our patients.

We need to increase awareness of the issues and the practicalities around management of ICDs in end of life situations to all our colleagues, across all relevant specialities. We need to ensure that patients are educated around their device function and their potential options in an end of life situation. We need the conversation with patients, and their close relatives, to be ongoing, ideally prior to implant, tailored to our patients needs at that time and should develop over time.

In an Irish setting we need to develop an education programme for all relevant healthcare providers, especially those outside of cardiology, with clear referral pathways for timely deactivation. The Irish Heart Foundation had, prior to COVID, put a working group together to progress these issues and one hopes that this process can be reinvigorated. There is good data that local quality improvement projects can also have a significant impact on appropriate discussions, documentation and deactivation rates in end of life settings. It is over ten years since the consensus documents but as a healthcare community we have some progress to make on this issue.

About the author

Paul Nolan is a Clinical Lecturer on the new MSc in Clinical Measurement Physiology in Atlantic Technological University and is a Cardiac Physiologist. He is a Certified Cardiac Device Specialist by the Heart Rhythm Society and the European Heart Rhythm Association.

References available on request

Read the latest edition of Hospital Professional News: August HPN

Catch up on our Clinical Features

Leave a Reply

Your email address will not be published. Required fields are marked *

Please Confirm

This website is only for the eyes of medical professionals. Are you a medical professional?