Is Pacemaker Deactivation at the End of Life Unique?: A Case Study and Ethical Analysis

Determining pacemaker dependency

Since the first pacemaker was implanted in 1958, pacemakers have become a mainstay of treatment for patients with bradyarrhythmias and end-stage heart failure.^6,7 Some of these patients have an inadequate or absent underlying cardiac rhythm and will develop severe symptoms of heart failure, syncope, or even cardiac arrest following pacemaker deactivation. These patients may be considered “pacemaker dependent,” although there is no consensus in the literature regarding how to define or determine pacemaker dependency. ⁸ Some cardiologists consider a patient to be pacemaker dependent if the ventricle is constantly paced by the device, although this could vary substantially depending on how the pacemaker is programmed. ⁹ Others label patients as pacemaker dependent only if pacemaker deactivation results in asystole in the absence of an underlying escape rhythm. ¹⁰ There is also uncertainty as to how the presence of a native heart rhythm should be determined. Some experts have recommended decreasing the pacing rhythm abruptly to 30 or 40 beats per minute to allow an underlying escape rhythm to be observed. Others favor gradually decreasing the rate of the pacemaker. There has been some evidence to suggest that artificial pacing can inhibit the native automaticity of the heart ¹⁰ ; as the underlying escape rhythm may initially be suppressed, patients can temporarily become asystolic if the pacemaker is suddenly turned off.^9,11

Attitudes toward the deactivation of pacemakers versus automated implantable cardioverter defibrillators

Much of the available literature on deactivating cardiac devices near the end of life focuses on automated implantable cardioverter defibrillators (AICD).^12–19 The majority of these ethical discussions conclude that the deactivation of an AICD at the end of life is a morally permissible act.^{12,13,17,19–21} In contrast, there has been more debate regarding the deactivation of pacemakers in terminally ill patients.^21,23–28 A recent survey of physicians, nurses, and industry representatives revealed significant differences in attitudes regarding pacemaker versus AICD deactivation. Clinicians are less comfortable disabling a pacemaker, less likely to grant requests for pacemaker deactivation, and more likely to consider pacemaker deactivation euthanasia. ²⁹

There is some evidence to suggest that some clinicians' discomfort with deactivating pacemakers may be related to a lack of experience in dealing with these requests. In 2003, Mueller et al. published one of the first ethical analyses of pacemaker withdrawal at the end of life. ³⁰ Over the course of 6 years, they identified six terminally ill patients or family members who had requested cardiac device withdrawal near the end of life. This included five requests for pacemaker deactivation and one request for AICD deactivation. In the earliest cases, cardiologists refused to deactivate pacemakers, sometimes even after an ethics consult had concluded that the act was morally permissible. Over time, the cardiologists seemed to become more comfortable with deactivating pacemakers at the end of life, perhaps because of more familiarity with this issue after discussion with the ethics consultation service.

The concept of burden

The differences in attitudes regarding deactivation of pacemakers versus other implantable cardiac devices may partly stem from the perceived burden of the device on the patient. Defibrillators can cause physical harm to the patient. One study found that 27% of patients received a shock within the last month of their life, and 8% of patients received a shock within minutes of death. ¹⁴ In addition to physical pain and discomfort associated with shocks themselves, other studies have exposed a set of psychological harms that can accompany AICD placement. Moderate to severe panic disorder, agoraphobia, and other anxiety disorders have been documented in patients with no pre-existing and only mild psychiatric illness.^31–34 Family members may also experience significant distress from watching their loved one being shocked so close to death. ¹⁶

It may be more personally acceptable to some providers to discontinue therapies that can cause physical or psychological distress, such as an AICD, or have an ongoing physical burden associated with them, such as a left ventricular assist device (LVAD). In these examples, withdrawal has the benefit of ending the physical burden associated with the device itself. Some have contended that such a case cannot be made for pacemakers, which have little burden after the initial implantation. ²³ Rhymes et al. have argued that pacemaker withdrawal does not ease any burden associated with the device itself; but rather acts only to alleviate the burden of continued life by changing the timing or cause of death. ²³ The idea that death is the “intended means” to achieve the alleviation of suffering may not be psychologically or morally acceptable to some providers. Some clinicians may not be comfortable with carrying out an act that will likely hasten death in order to relieve the suffering associated with life itself. However, there are many other life-sustaining treatments considered low burden (such as intravenous fluids or antibiotics) that clinicians are willing to discontinue at the request of a patient or family. Therefore, the issue of burden alone cannot be the sole reason for the psychological and moral distress caused by pacemaker deactivation.

It is also important to note that an intervention deemed without burden by the clinical team may still cause significant distress and suffering on the part of the patient and family. ⁵ A patient with decision-making capacity has the right to assess the relative burdens and benefits of a medical treatment for him/herself. As noted in the HRS consensus guidelines, even in the absence of a direct physical burden, there may still be substantial indirect burdens of continued pacemaker therapy, including loss of control and dignity, fears about a prolonged dying process, and substantial psychological and existential suffering. ⁵ No distinction should be made between a low-burden intervention such as taking a pill and an invasive, potentially uncomfortable intervention such as endotracheal mechanical ventilation. A patient has the autonomous right to decide what quality of life is acceptable and how much suffering and physical invasion to endure. ¹

Implanted cardiac devices and the patient's self

Some have likened pacemaker deactivation with euthanasia, claiming that pacemakers in pacemaker-dependent patients have become “an integral part of the heart.” ²¹ Similar to a mechanical heart valve, it has been argued that implanted cardiac devices become “biofixtures” considered intrinsic to the patient and part of the patient's “self.”^25,35 Using this reasoning, some have concluded that pacemakers are distinct from external life-sustaining treatments, such as ventilatory support, antibiotics, or intravenous hydration. ²⁵ It has also been suggested that pacemakers may be distinguished from AICDs in this respect because the latter only act intermittently upon the activity of the heart or not at all. ²¹

As technology progresses and medical interventions become less burdensome, yet more efficient and more integrated into the patient's body, it may become more challenging for clinicians to decide whether it is ethically permissible to withdraw certain interventions. Ethicist Daniel Sulmasy differentiates between those interventions that seem “indistinguishable from the patient's self,” such as a renal transplant, and those interventions that substitute for impaired bodily function, but remain distinct, such as peritoneal dialysis. ³⁶ In this analysis, a transplanted kidney is considered to be a “replacement therapy,” which has become part of the patient's restored physiology and replaces a bodily function in the same manner in which the patient's body was able to perform the function when healthy. In contrast, peritoneal dialysis may be considered a “substitutive therapy,” which regulates bodily functions by helping the body return to normal homeostasis. Even when internalized, a substitutive therapy remains distinct from the patient's body and extrinsic to its function. Sulmasy also suggests other criteria for determining whether an intervention is a replacement therapy, including the ability to grow and perform self-repair, as well as independence from external energy sources and control. ³⁶

A recent ethical analysis by Kay and Bittner argues that a permanent pacemaker may be considered a replacement therapy and that it is morally acceptable to withdraw a substitutive therapy, but not a replacement one. ²⁷ For pacemaker-dependent patients, the authors maintain that a pacemaker functions as the normal conduction system and therefore, withdrawing pacemaker function is akin to resecting a transplanted organ. Conversely, an AICD does not replace the normal heart rhythm, but rather intervenes on an abnormal rhythm by shocking the heart back to normal sinus rhythm. ²⁷ Kay and Bittner maintain that an AICD functions as a risk-reducing substitutive therapy making it ethically permissible to withdraw the device near the end of life. However, this analysis disregards the other key criteria suggested by Sulmasy, as outlined above. A pacemaker is a mechanical device incapable of growth or self-repair with a limited power source (battery) that must eventually be replaced. Pacemakers are also easily controlled by a computer and often need manipulation by expert clinicians. Based on these criteria, an ethical distinction cannot be made between the withdrawal of a pacemaker versus an AICD.

Pacemaker dependency and timing of death

The timing of a patient's death following cessation of a life-sustaining therapy may also have a significant impact on physician comfort with withdrawal. In contrast to discontinuing an AICD, which may result in death in days or weeks, the case of pacemaker deactivation in a pacemaker-dependent patient is notable because of the expected immediacy of the outcome (the patient's death) to the action (turning off the pacemaker). However, there is no ethical or legal distinction between withdrawing a therapy that results in a patient's demise in days or hours or seconds. Deactivation of the pacemaker would allow an underlying disease process to take its natural course, but would not introduce an external mechanism of death (such as administering a medication to make the patient asystolic).

Because of this temporal proximity, it may be psychologically difficult for the clinician participating in the withdrawal to separate the action from the outcome. If a patient is alive in one moment and passes away the next and the only intervening act was the deactivation of a cardiac pacemaker, then it is both predictable and reasonable that the agent who carried out this act would feel responsible for the outcome. This is akin to the discomfort some clinicians and family members may feel when withdrawing, rather than withholding life-sustaining therapies. Similarly, it may be psychologically more acceptable to respect a patient's right to refuse the initial placement of a pacemaker than it would be to later turn off the device.

Timing of death and perceived burden

In our case, a clinician was expected to stop a patient's heart by pushing a few buttons on a computer. The physician expected that he would be able to speak and interact with a living, breathing, coherent person in one moment and then watch her die almost instantly as a result of his intervention. It is certainly understandable that this clinician would question whether this is a beneficent action promoting patient well-being (by relieving suffering) or rather a maleficent one causing the patient harm (by causing death). The combination of temporal proximity, as well as the perception of a treatment without burden, may elicit substantial moral angst in a clinician. This can be contrasted with the typical view of ventilator withdrawal, an event after which death can result within seconds to minutes, but prior to which there appears to be a high burden to the patient. Conversely, intravenous antibiotics may be considered low burden, but death is usually temporally separated from the act of withdrawal by days. The presence of an intrinsic heart rhythm in a non-pacemaker-dependent patient may lessen the moral anguish experienced by the clinician upon disabling a pacemaker because this action does not typically result in instantaneous death. Ironically, although an underlying rhythm may allow the clinician to feel more comfortable discontinuing the device, the patient may suffer more at the end of life due to worsening heart failure and prolongation of the dying process. ²²