Clinical Ethics in Anesthesiology. A Case-Based Textbook

2. End-of-life issues

18. Discontinuing pacemakers, ventricular assist devices, and implanted cardioverter-defibrillators in end-of-life care

Cynthiane J. Morgenweck

The Case

Mr. K is a 59-year-old who was diagnosed with idiopathic ventricular tachycardia at age 50. He had an implantable cardioverter defibrillator (ICD) placed at that time. Within a month of placement, Mr. K experienced his first shock and described it as a ‘sledgehammer to my chest.’ He became afraid of further shocks and limited his activities in the hopes of avoiding them. The ICD was reprogrammed, his medications were readjusted and he underwent cognitive behavioral therapy to lessen his fears of the shocks. With these changes in place Mr. K was able to resume several activities. He now returns requesting that his ICD be turned off because he has been diagnosed with unresectable cancer. He had undergone some palliative chemotherapy but has decided that he does not want more chemotherapy because it makes him so ill that he cannot focus on getting his affairs in order before he dies.

Are there ethical arguments for keeping the ICD activated?

Some practitioners will not disable any cardiac devices – pacemakers, ICDs or ventricular assist devices (VADs), citing several potential ethical arguments for leaving the device intact and functioning. One argument is that the device has become part of the patient’s physiology. It is a “biofixture” and to disable it is to interfere with the patient’s current physiologic functioning in a negative manner.1 Physicians should not be a part of actions that diminish patients’ health, because it would violate the principle of nonmaleficence or “do no harm.” A second argument is that disabling the device constitutes euthanasia in that without the device, the patient will have a cardiac event that is no longer readily reversible, as it would probably be were the device active. The physician who writes the order for disabling the device writes an order that is the potential cause of the patient’s death. A third argument is that disabling a cardiac assist device could be construed as physician-assisted suicide because the order to disable the device must be written by a physician in response to a patient request. What could such an order be but an agreement to help the patient to commit suicide?

In a recent survey of physicians and their attitudes towards device deactivation, 46% of respondents believed it was illegal or were unsure if it is illegal to disable a device. Once they were reassured that it is legal, 91% of the respondents indicated that they would be willing to discuss disabling cardiac devices under certain circumstances.2

Responses to arguments against disabling cardiac devices

The biofixture analysis is adapted from legal concepts and is not a fully analogous argument. Devices are inserted and removed from patients throughout their lifetimes. For example, intraocular lenses and artificial joints are implanted in patients and explanted if not functioning properly. The device does not contain the patient’s DNA or other attributes that identify it as uniquely specific to a particular patient. Further, disabling a cardiac device does not entail its removal, so the patient is not at risk for further surgery.

When it is argued that disabling a cardiac device will harm a patient, it is important to remember that harm assessments ought to come from the patient, particularly if the therapy has been ongoing. The physician must educate the patient about the potential benefits and burdens of disabling the device, but it is the patient who decides whether or not to accept the benefits and burdens. In this case the patient initially accepted the benefits and burdens of his ICD; however, as the patient’s life circumstances changed, those benefits no longer outweighed the burdens in his view.

The writing of an order to disable a device should not be viewed as euthanasia. In general, euthanasia is considered to be an act on the part of a healthcare provider that intentionally hastens the death of a patient. A physician may believe that the lack of countershock will inevitably hasten the death of the patient; however, this may also be an overestimate of the value of the ICD. Not all delivered shocks are appropriate, there are nonfatal arrhythmias, and death may occur from other causes. Also, disabling the device is merely the removal or forgoing of a therapy that is no longer requested by a patient rather than a further intervention that is designed to hasten the death of the patient.

Disabling a device is not physician-assisted suicide. The requests for disabling cardiac devices are generally made in the context of other life-limiting diseases where the benefit of a device that causes the heart to function while the rest of the body is failing is no longer acceptable. If a practitioner is uncertain about the mental health of a patient, it is reasonable to consult with a psychiatrist or psychologist.

The doctrine of “double effect” may help in analysis of this issue. (For more about Double Effect, refer to Chapter 15). If the intention of the physician is to relieve suffering that the device is causing, and the death itself is unintended, then disabling may be ethically permissible. It is important to remember that disabling the device does not inherently cause the death of the patient – there are patients who have devices implanted who have never been shocked. Alternatively, shocking a dying patient may unnecessarily prolong death and cause further discomfort.

The rationale for disabling the device are based on a fundamental notion of patient autonomy, informed by patient awareness and acceptance of the consequences of the choices made. This set of conditions is analogous to the withdrawal of other medical therapies such as ventilatory support or dialysis. A debate about disabling cardiac devices has arisen in part because these devices are relatively new, because they literally support the symbolic organ of life (the beating heart), and because true to the technological imperative, health care providers rush to support those with debilitating diseases with less thought to the limits and burdens of the technology.

Distinguishing characteristics of current cardiac devices

Each of the three types of currently used cardiac devices has different functions and a different mix of harms and benefits. These differences may impact the decision to disable the device. Each device may be disabled by simply “deprogramming,” or by battery pack disconnection.


Pacemakers are the oldest of the devices, having been used since the early 1960s. There are risks with implantation. There can be infection, lead fractures, and inability to place the leads in a satisfactory manner with a resultant inability to pace. If the patient is able to have a pacer placed successfully, there appears to be minimal downside. It is a small device with little disfigurement of the patient and it performs its function without patient awareness. By maintaining an adequate cardiac rhythm, a pacemaker can help a patient to have a nearly normal life with few restrictions.

When a patient is dying, continued pacing may be seen as a burden. The pacing can be viewed as a mandatory continuation of one physiological system that is out of sync with the overall condition of the (dying) patient. But exactly what happens if a pacemaker is discontinued is of concern. A slower heart rate may cause congestive heart failure with shortness of breath, pulmonary edema, and an inability of the patient to enjoy the last days of life. These symptoms may be ameliorated with good palliative care. It might also be that the patient’s intrinsic rate is sufficient to maintain the limited activities of a bedridden, dying patient.

There are few stories available for review of what happens to patients when their pacemakers are disabled and so it is hard to predict what will happen. If the device is disabled within the context of multiple lethal diseases, there is little reason to fix blame for all of the patient discomfort on that action alone. It is important to remember that good palliative care can ameliorate the symptoms that the patient might experience during the dying process when the pacer is disabled. If the patient is completely pacer dependent – i.e., there is no underlying cardiac rhythm – death will likely occur shortly after the pacer is disabled. Obviously, specific plans should be made regarding do-not-resuscitate orders prior to disabling such a device.

Implantable cardioverter defibrillator (ICD)

Today, pacemaker technology is usually enveloped in the ICD so that one device functions as a pacemaker or ICD or both. This section will focus on the ICD function of the device. The initial placement of the device has the risks of inability to place, infection, and lead fracture. Recovery time from the placement is short, if the device is functioning well. It is a relatively small device with minimal disfigurement. However, once the ICD is in place and functioning, it may not be as benign as a simple pacemaker. Patients are frequently aware of the firing of the device and describe the shock as traumatic. These experiences create anxiety, depression, and panic states in some patients. They may self-impose activity limitations in the hope of avoiding being shocked. Thus, patients may have a greater interest in ICD removal than with simple pacemakers in which awareness of function is not an issue. If there are other disease states that are not reversible or controllable, the patient may decide that the burden of ICD shocks added to the incurable illness creates a quality of life that is unbearable and ask for disabling of the device.

Disabling an ICD in a patient with a terminal diagnosis creates the potential for a more peaceful death. With no ICD function, death will occur either as the result of a fatal arrhythmia or as a result of the terminal illness. Shocks that would occur if the device were not disabled might prevent a peaceful death that is being sought by many patients. Patients who accept their imminent death may desire comfort care only and might actually prefer a sudden cardiac death to a more prolonged dying with its attendant diminution of functioning.

When discussing ICD disabling with a patient (and/or surrogates), it is necessary to describe the possibility of arrhythmias that do not cause the death of the patient, but do create increased disability. Further, it is important to point out that disabling the device does not guarantee a sudden cardiac death. This may also be the time to initiate a do-not-resuscitate (DNR) order. (See Chapter 2.)

Ventricular assist device (VAD)

VADs are the most recently developed cardiac assist devices, with clinical trials ongoing to determine the best design and most appropriate candidates for VADs. VAD placement is a more invasive surgical procedure with a longer in-hospital recovery time. VADs are more noticeable than pacers or ICDs. VADs were initially viewed as “bridging devices, ” that is, placed until the patient was able to either undergo definitive therapy (cardiac transplantation) or recover from the cardiac illness. Currently, VADs are being used as destination devices, i.e., they are implanted with the understanding that they will remain in place as long as the patient is alive. Patients who receive a destination VAD are very ill – they have refractory Class IV heart failure and systolic dysfunction. Even with a properly functioning VAD, a patient may decide after living with the device that the quality of life improvement is not sufficient to continue. The patient may experience neurologic complications, repeat infections, thromboembolic complications, and bleeding problems with recurrent hospitalizations. When a VAD is disabled, patient death is likely to occur within a short period of time, although this is sometimes hard to predict. Currently most VADs have external battery packs, and the VAD can be disabled by simply disconnecting the battery. Again, plans should be made regarding do-not-resuscitate orders prior to disabling the device.

“Preventive” ethics

The decision to disable a cardiac device can be wrenching for the patient, for family members and for healthcare providers. Preventive ethics may ease some of the emotions that are experienced when such a request is made.3 Physician self-knowledge, communication with patients at the time of placement, and anticipation of points of dispute can diminish some of the anxiety about such decisions.

Physicians should reflect on their own responses to such requests before they are made.4 These responses must be communicated to the patient. If the physician would never honor such a request, the patient ought to be informed at the beginning of the patient/physician relationship. If the physician is willing to consider patient-initiated requests for disabling the device, a great opportunity for conversation is when the device is placed. Discussion should include the circumstances under which the physician would recommend disabling or when the patient might request that the device be disabled. Patients should be encouraged to fill out advance directives if they have not already done so.

Sometimes the decision to disable a cardiac device may be advocated by all parties involved in the healthcare of the patient. At other times, there may be disputes that will have to be resolved so that the most appropriate care of the patient can be rendered with the best possible understanding of the choices made in the treatment plan.

In the ICU environment, it is perhaps more difficult, but still important to develop as complete an understanding of the patient’s goals and values as possible. Knowing the patient’s goals and values will enable the physician to craft a plan of care that is respectful of the patient’s life views. It is nearly always helpful to involve the patient’s family to the extent that the patient will permit such involvement. A family member usually provides a broader understanding of the patient’s life and can become a reflective sounding board for decisions. This may minimize disputes.

What to do if a request is made to disable a device

When a request to disable a cardiac device is made, it is important to assure the patient that the request is being taken seriously, but also to inform the patient that disabling the device will not occur immediately, because the potential irreversibility of the disabling necessitates assessment of the request. Conversation with the patient, exploring the reasons for the request is the starting point for the evaluation the request. The dialogue may not be long, since there may be obvious reasons that need little explanation. In other cases, review of the patient’s overall disease state with an emphasis on the prognosis may help to determine if disabling of the device will meet the patient’s real goals.

It is important to determine if the patient has “decisional capacity” (the capacity to make decisions). Elements of “decisionality” include the patient’s abilities to comprehend information, to consider the information in the context of the patient’s preferences, and to communicate a decision that is consistent over time. It is imperative that the physician recognize and accept that decisional patients may refuse any and all therapies, even ones that have been instituted years ago. If there are unanswered questions about the patient’s decisionality, formal psychiatric or psychologic evaluation will be beneficial.

The physician should also consider the request in the context of prior knowledge of the patient. If this request seems consistent with the patient’s previous choices in healthcare, then the request is likely to be valid. If the patient has an advance directive, the request can be evaluated against the choices made in the document and in conversation with the appointed agent.

If the reasons do not appear sound, then discussion with colleagues as well as the patient’s family and friends (assuming patient permission) is reasonable. Ethics consultation may also be of value. If the physician, as a moral agent, finds herself incapable of honoring a valid request, the physician must assist in the transfer of care of the patient to another physician who is willing to honor the request.

A palliative care consultation may also be helpful when a patient requests that their cardiac device be disabled. The focus of this consult will be to anticipate and alleviate uncomfortable symptoms that may occur when the device is disabled, to describe the timeline of the patient’s death as well as the possibility of continued life, and to prepare the patient, family, and friends for the death.

It is vital to document the conversation and the plan developed for disabling the device. Sometimes, an outside technician is needed to disable the device. They may seek greater understanding of the physicians order for disabling the device before carrying out such an order. A well-written note will prevent miscommunications. This is also true for all other healthcare providers who may be cross-covering or coming on service shortly after the decision was made.

Key points

• Refusal of standard medical care, even if life threatening, is within a patient’s right to personal autonomy.

• Pacemakers, ICDs, and VADS represent forms of cardiac-supportive therapies that patients may legitimately forgo or discontinue as part of end-of-life decision-making.

• In situations where the benefit vs. burden balance of medical care does not seem appropriate to medical personnel, additional psychological or ethics team consultation may be helpful.

• Discontinuing cardiac device therapy should be accompanied by thoughtful discussion, discernment of patient goals, and plans for appropriate palliative care.


1* Paola, F.A. and Walker, R.M. (2000). Deactivating the implantable cardioverter-defibrillator: a biofixture analysis. So Med J93, 20–3.

2* Sherazi, S.Daubert, J.P.Block, al. (2008). Physicians’ preferences and attitudes about end-of-life care in patients with an implantable cardioverter-defibrillator. Mayo Clin Proc83, 1139–41.

3* Wiegand, D.L. and Kalowes, P.G. (2007). Withdrawal of cardiac medications and devices. AACN Adv Crit Care18, 415–25.

4* Mueller, P.S.Jenkins, S.M.Bamstedt, K.A., and Hayes, D.L. (2008). Deactivating implanted cardiac devices in terminally ill patients: practices and attitudes. Pacing Clin Electrophysiol31, 560–8.

Further reading

Ballentine, J.M. (2005). Pacemaker and defibrillator deactivation in competent hospice patients: An ethical consideration. Am J Hosp and Pall Med22, 14–19.

Berger, J.T.Gorski, M., and Cohen, T. (2006). Advance health planning and treatment preferences among recipients of implantable cardioverter defibrillators: an exploratory study. J Clin Ethics17, 72–8.

Dudsinski, D.M. . (2006). Ethics guidelines for destination therapy. Ann Thorac Surg81, 1185–8.

England, R.England, T., and Coggon, J. (2007). The ethical and legal implications of deactivating an implantable cardioverter-defibrillator in a patient with terminal cancer. J Med Ethics33, 538–40.

Goldstein, N.Carlson, M.Livote, E., and Kutner, J.S. (2010). Brief communication: management of implantable cardioverter-defibrillators in hospice: a nationwide survey. Ann Intern Med152(5), 296–9.

Goldstein, N.E.Lampert, R.Bradley, al. (2005). Management of implantable cardioverter defibrillators in end-of-life care. Ann Int Med141, 835–8.

Goldstein, N.E.Mehta, D.Siddiqui, al. (2008). “That’s like an act of suicide.” Patients’ attitudes toward deactivation of implantable defibrillators. J Gen Intern Med23 Suppl 1, 7–12.

Kirkpatrick, J.N.Fedson, S.E., and Verdino, R. (2007). Ethical dilemmas in device treatment for advanced heart failure. Current Opin Support Palliat Care1, 267–73.

Lewis, W.R.Luebke, D.LJohnson, al. (2006). Withdrawing implantable defibrillator shock therapy in terminally ill patients. Am J Med119, 892–6.

Manganello, T.D. (2000). Disabling the pacemaker: the heart-rending decision every competent patient has a right to make. Health Care Law MonJan, 3–15.

Rizzieri, A.G.Verheijde, JL.Rady, M.Y., and McGregor, J.L. (2008). Ethical challenges with the left ventricular assist device as a destination therapy. Phil, Ethics, Humanit Med3, 20.

Zellner, R.A.Aulisio, M.P., and Lewis W.R. (2009). Should implantable cardioverter-defibrillators and permanent pacemakers in patients with terminal illness be deactivated? Deactivating permanent pacemaker in patients with terminal illness. Patient autonomy is paramount. Circ Arrhythm Electrophysiol2(3):340–4.