The prevalence of implantable cardioverter-defibrillators (ICDs) is increasing. In 2003, 150000 ICDs were implanted in the United States, double the 75000 implanted in 2001.1 This is because "new guidelines dramatically increase the number of patients for whom prophylactic implantation of an ICD would be covered under Medicare."1^(p2542) Under current Medicare guidelines, patients with a left ventricular ejection fraction of less than 30% and nonischemic cardiomyopathy now qualify for these devices.1 The number of eligible patients is estimated to be around 600 000 annually, a figure that will likely increase as the population ages.1 The ICDs monitor the patient's heart rhythm and deliver electrical shocks (shock therapy) for any detected life-threatening arrhythmia2; they are not designed to maintain existing heart function. However, although shock therapy can be life saving, it can also be physically and emotionally painful. Patients describe the physical sensation of shock therapy as similar to that of being kicked in the chest by a horse or struck by lightning.3 The fear of getting a shock can also cause anxiety and psychological harm to patients and families.2

ICD Deactivation

Although ICDs can save people's lives, there are situations in which deactivating them may be appropriate. When the risk of shock therapy causes unbearable anxiety, competent patients may determine that they no longer wish to continue the use of ICDs. In addition, there are certain medical situations, such as certain critical or terminal illnesses, in which their use can be counterproductive. Sepsis, hypoxia, and electrolyte or metabolic imbalances can cause ventricular and supraventricular arrhythmias, which could prompt repetitive shocks.2 Similarly, when antiarrhythmic or vasopressor medications or other life-sustaining therapies are discontinued or decisions to forgo resuscitation are made, discussions about deactivation are warranted. Shock therapy may therefore be inappropriate for patients whose goals are focused on palliation or comfort.2 Avoiding shocks in such cases means preventing unnecessary suffering.4 In a retrospective study of patients with ICDs who received shock therapy before they died, 30% were in their last minutes of life.5 The family members of these patients "found it distressing to witness the patient being shocked at the end of life."5^(p837) On the other hand, ventricular tachycardia, a possible arrhythmia in the last minutes of life, is not considered uncomfortable.4

Several recent studies have focused on aspects of deactivating ICDs.2,4-11 Some show that both patients and clinicians have misconceptions about these devices. One study explored attitudes among 15 patients with ICDs and found that these patients did not understand how the devices worked or why they had them and that they could not conceive of circumstances where it might be appropriate to deactivate them.8 Patients also could not recall anyone ever discussing deactivation with them and were interested in receiving more information about the devices.8 Studies of physicians show that many are unaware that shock treatment can be painful and are unclear of the legal and ethical aspects of ICD deactivation.4,6,9 Among the noncardiologists in one study, 46% "either thought it was illegal or were unsure if it was legal to withdraw ICD therapy in terminally ill patients."6^(p1139)

Such misconceptions contribute to confusion over whether or when to consider ICD deactivation. As a result, deactivation is often addressed only as a patient's condition deteriorates or when the patient is very near death,5 when the potential for nonbeneficial shock therapy increases. In fact, many physicians were willing to discuss deactivation if death was more imminent. In one study, 27% of discussions took place in the last hours of patients' lives and 4% in the last minutes.5 Because 20% of deaths occur during or after an intensive care unit admission,12 decision making about ICD deactivation is often a critical care issue. When considering ICD deactivation, it is helpful to recognize the following relevant ethical principles.

Ethical Principles

Autonomy

The ICDs are a form of life-sustaining treatment, which a competent patient or surrogate has the right to refuse.2,10,13,14 In this regard, ICDs are not different from other forms of life-sustaining therapies, such as hemodialysis, mechanical ventilation, or medically supplied nutrition or hydration.10,14 Patients, or their surrogates, have the moral and legal authority to discontinue any of these treatments to allow the dying process to proceed without intervention. This is supported by the existing ethics literature.14 By the same token, patients with terminal conditions also have the right to continue these devices until death, as did all 6 cancer patients in one ICD study.11

However, deactivating ICDs makes some clinicians uneasy. The internal nature of ICDs suggests that they may fall into a different category of life support. As "biofixtures," deactivating them can feel more like the intentional ending of human life13 than withdrawing life support.13,14 This is because some people consider these devices as being part of an individual, and so deactivating them is akin to stopping the function of a natural organ.13 Ethically, however, there is no difference between internal or external life-support therapies.13,14 Discontinuing either is permissible, because doing so means removing a therapy that prolongs dying in a patient with a fatal condition.13 It is also permissible to remove such a therapy at any time if the patient or surrogate deems that its burden is greater than its benefit.

Some physicians cite concerns about the impact of withholding a treatment versus withdrawing one. "I think that people just don't think of turning off things that were already started, even though it's like all technology, even though we say ethically and legally that there's no difference between withholding and withdrawing. I think for a lot of sustaining therapies, in practice it seems like it's different."9^(p4) Although the ethics literature confirms that there is no philosophical distinction between withdrawing and withholding life-sustaining therapies,10,13-15 withdrawing can be emotionally and spiritually difficult for some clinicians.

Another aspect of ICDs is that they have "an initial iatrogenic burden, but, over time, ... diminishing risks and burdens and continuing benefit."14^(p16) Because ICDs generally do not cause discomfort, some clinicians see no need to deactivate them. The argument that ICDs provide little burden over time, however, is less valid in patients with terminal medical conditions. In these cases, shock therapy can become potentially very burdensome when shocks administered at the time of death could be secondarily painful for the patient or family.

Informed Consent

A fundamental aspect of medical decision making is the concept of informed consent.14 This means that a competent patient should be given information on the "nature of the therapy; the purpose; the risks and consequences; the benefits; the probability that the therapy will be successful; the feasible alternatives; and the prognosis if the therapy is not given."15^(p93) Deactivation of an ICD is an option that all patients with such devices should be aware of as part of understanding the key aspects of this technology. However, discussions about ICD deactivation do not typically take place when the devices are implanted.4-6 In one study of 15 patients with ICDs, none had ever heard about the option of deactivation.8

Most physicians surveyed on this topic feel that deactivation should be a part of the informed-consent process for ICD implantation.4 Ironically, only 56% of electrophysiologists in one study felt that deactivation should be discussed at implantation, versus 79% of cardiologists and 95% of internists and geriatricians.4

At the time of insertion, the risks, benefits, and consequences should be clearly outlined, including the option or eventual need for deactivation. This does not systematically occur for ICDs for reasons having to do with both patients and physicians. For patients, many receive these devices when they are in a stable medical condition.6 At the time of insertion, most patients have not considered the possibility that despite the use of the ICD, their heart failure and arrhythmias may become irreversible and no longer amenable to intervention.8 Initially, many patients know only that these devices can save their lives and want them to do so.8 Similarly, physicians tend to focus more on the immediate medical condition when recommending an ICD.4,6,9 Almost all the physicians in one study agreed that deactivation should be discussed when the ICD is inserted, but acknowledged that they rarely did so.9 One rationale for not discussing deactivation when ICDs are inserted for some physicians was that patients would refuse the device if they could anticipate likely end-of-life scenarios.7

As time goes by, there are also patient and physician reasons for not reviewing the benefit-to-burden ratio of ICD therapy. For patients, the biofixture aspect of the ICD makes it come to seem part of their bodies and normal functioning.8,13 Some may "forget" about the device or not realize that its benefit may be diminishing as their cardiac function declines or other life-threatening issues develop.8 Among physicians, the biofixture nature of these devices makes it hard to recall that they are present in a particular patient because there is no visible evidence of them, and so it is "difficult to think of them in the same context as other management decisions at the end of the patient's life."9^(p4)

It is also difficult to know when to consider deactivation. In one study of patients in an ICD clinic, preplanning regarding ICD deactivation was possible in only 32% of patients.2 ICD deactivation is therefore not often included in routine end-of-life goals or treatment discussions.4,9 The strongest independent predictor of physician willingness to discuss ICD deactivation is a history of prior deactivation discussions.4 Because these discussions are not generally taught to medical students or residents, it is not surprising that many lack experience and expertise in discussing them.4

Moreover, there is the practical issue of who should discuss the benefits of ongoing shock therapy. Should it be the primary-care provider? The cardiologist? The electrophysiologist? A study of attitudes among these groups of physicians showed that electrophysiologists were less inclined to discuss advance directives with patients even though they were the most knowledgeable about ICD risks and benefits.4 Only a small minority (13%) of physicians surveyed would accept responsibility for having such a discussion.7 Others prefer to defer the decision,7 although lack of routine opportunities to revisit the decisions made when the device was inserted could contribute to suboptimal planning for end-of-life care.

Moreover, there is the issue of whether the physician or the patient should bring up the issue of deactivation. Patients seem to prefer that physicians raise this topic,8 but physicians cited lack of time or rapport or both with patients as additional reasons for not raising the topic.9 Interestingly, physicians do not avoid deactivation discussions out of concern for upsetting patients, as 86% of physicians in one study did not think such discussions would be upsetting to patients or families.4

One final, and surprising, physician issue was a lack of awareness of the discomfort of administered shocks. Only 59% of physicians in one study thought such shocks would be uncomfortable,6 while in another 24% thought shocks at the end of life were not distressing.4 This lack of awareness of the discomfort of shock therapy may also explain why some physicians might not consider deactivation.

Nonmaleficence/Clinician Integrity

Nonmaleficence is the ethical obligation among clinicians to prevent harm.15 Some physicians are reluctant to deactivate ICDs out of concern that doing so would result in more harm than good. To these clinicians, deactivating an ICD seems like "causing harm-even death-in the patient."14^(p15) One physician interviewed about deactivating an ICD said it is "like crossing a bridge to a certain extent .... There's a finality to turning it off."9^(p4) This is the case in terms of acknowledging that death from a fatal arrhythmia is possible after deactivation and the concern that, by deactivating the device when a shock might be imminent, death could occur quickly after deactivation. This temporal relationship between the act of turning off the device and the patient's death may fuel concerns about the ethical permissibility of doing so.

Similarly, such comments may reflect concerns about how participation in acts that will ultimately lead to death will impact their perception of themselves as clinicians or the perception of patients or families. For some, it could be viewed as an expression of loss of hope,9 patient abandonment, or an unethical act. In one study, 18% of physicians thought deactivating an ICD was either unethical or possibly unethical.6 As a result, physicians often suggest additional treatments before considering deactivation.7 Again, it appears that some physicians consider these devices as being in a category different from other life-sustaining therapies, even if logically they are not.4,9 Others struggle with knowing when the burden of such devices negates their benefit.4

Doctrine of Double Effect

In certain instances, it may be ethically permissible to carry out an action even if it might have negative consequences.14 Deactivating an ICD could spare terminal patients the discomfort of shock therapy but might mean that their death could come from a fatal arrhythmia.14 Most physicians surveyed felt that unstable ventricular tachycardia or fibrillation is not generally painful.4 In a retrospective study comparing terminally ill patients who had their ICDs deactivated with those who did not, results showed no difference in time between therapy discontinuation and death.2 Those patients who were able to have their ICDs deactivated experienced fewer shocks in their final days of life than those whose ICDs were not deactivated.2

The Role of the Critical Care Nurse in ICD Deactivation

Critical care nurses have a pivotal role in the care of patients with ICDs. This role varies, depending on whether a patient has an ICD, the patient's goals of care, and code status and prognosis. Specifically

1. New ICDs: Critical care nurses caring for patients who are being recommended ICD insertion should validate patients' and families' understanding of the function of these devices and their benefits and burdens. As part of the informed consent process, it is appropriate to help patients understand that these devices can be activated and deactivated in accordance with their goals of care. Systematic assessment of preferences for life-sustaining therapies, values about quality of life, expectations of treatment, and goals of care should be routinely discussed and documented. Similarly, designation of a proxy decision maker in the event of incapacity should also be encouraged.

2. Code status: Patients coming into the critical care setting with existing ICDs should have these devices clearly noted on the chart. As with all critically ill patients, preferences for life-sustaining therapies including resuscitation should be explored and documented. If a patient chooses not to be resuscitated and a Do-Not-Resuscitate (DNR) order is written, then deactivation of the ICD should be discussed. Most physicians surveyed in one study felt "that ICD deactivation should always be discussed in the context of a DNR order."4^(p704) However, in another study, deactivation discussions occurred in fewer than 45% of patients with DNR orders.5 One way to help ensure that ICD deactivation is considered is to revise DNR forms to include this option.5

3. Changing medical conditions/goals: If the patient has an active ICD, but his or her condition deteriorates, or his or her goals of care change, discussions regarding deactivation should occur regularly. Family meetings to discuss the patient's condition and goals of care are an ideal time for critical-care nurses to ensure that the ICD is discussed and that a clear plan for handling it is agreed upon. Patients want their clinicians involved in any deactivation decisions and are often waiting for them to bring it up.8 If ICD deactivation is chosen, this can usually be accomplished easily and should be done before transferring a patient out of the critical care unit; otherwise, there is the risk of the ICD firing in a patient's last moments of life, when the agreed-upon goal was comfort. If patients want to wait to deactivate the device until they are home, that is also possible but involves coordination with the device manufacturer. The management of ICDs at the end of life, however, ultimately depends on the clinical experience of the physicians involved.4

4. Support: Deactivation of the ICD may be a new and uncomfortable concept for patients, families,8 and clinicians. Therefore, when deactivation is chosen, it is essential that critical care nurses and other clinicians are supportive of the decision. It may be too difficult for the patient or family to let go of the notion that the ICD will save the patient, no matter what.8 They are exercising their autonomy in choosing to continue or discontinue it and should be supported in whatever decision they make. ICU clinicians may also need support systems to assist in coping with their own responses to the patient's decision and their role in carrying out the decision.

Conclusion

The use of ICDs is increasing and, while they provide important benefits for treating life-threatening arrhythmias, they may not be appropriate for some critically or terminally ill patients. Deactivation may therefore become appropriate when the goals of care shift to comfort and end-of-life care. Several recent studies show patient and physician misunderstandings about these devices and deactivation. Because deactivation is typically not discussed at implantation, this issue may arise only when the patient is in a critical care setting. Deactivation is medically, legally, and ethically appropriate when a patient decides to forgo life-saving or sustaining treatment. Making a decision about deactivation involves the ethical principles of autonomy, informed consent, nonmaleficence, integrity, and the doctrine of double effect. The critical-care nurse can play an important role in helping the team, patient, and family consider these issues and make the right choice for the patient.

REFERENCES