The ethics of unilateral implantable cardioverter defibrillators and cardiac resynchronization therapy with defibrillator deactivation: patient perspectives

Abstract

Aims

Decisions about deactivation of implantable cardioverter defibrillators (ICDs) are complicated. Unilateral do-not-resuscitate (DNR) orders (against patient/family wishes) have been ethically justified in cases of medical futility. Unilateral deactivation of ICDs may be seen as a logical extension of a unilateral DNR order. However, the ethical implications of unilateral ICD deactivation have not been explored.

Methods and results

Sixty patients who had an ICD or cardiac resynchronization therapy with defibrillator (CRT-D) were interviewed at a quaternary medical centre outpatient electrophysiology practice. Survey questions addressed whether deactivation of defibrillator function was included in advanced directives, whether deactivation constitutes physician-assisted suicide, and whether unilateral deactivation can be ethically justified. Responses were elicited to scenarios in which defibrillation function was deactivated in different contexts (including patient request to deactivate, existing DNR, and unilateral deactivation). Only 15% of respondents had thought about device deactivation if they were to develop a serious illness from which they were not expected to recover. A majority (53%) had advance directives, but only one mentioned what to do with the device. However, a majority (78%) did not consider deactivation of an ICD shocking function against patients' wishes to be ethical or moral.

Conclusion

Management of ICDs and CRT-Ds as patients near the end of their lives creates ethical dilemmas. Few patients consider device deactivation at end-of-life, although a large majority believes that unilateral deactivation is not ethical/moral, even in the setting of medical futility. Advance care planning for these patients should address device deactivation.

Introduction

Physicians and patients are increasingly confronted with questions about what to do with cardiovascular implanted electronic devices (CIEDS) in end-of-life care. Previous studies have shown that some patients and physicians relate device deactivation to physician-assisted suicide,^1,2 and that some physicians and other medical professionals are uncomfortable with device deactivation.³ These studies and others have also demonstrated that few patients have considered the advance care planning implications of having a life-sustaining implantable cardiac device.^1,4–8

Medical futility has become an increasingly important topic in the modern era of cost-conscious healthcare. Of all medical costs incurred over a lifetime, the largest percentage is spent during the last months of life for many patients.⁹ Medical providers can become frustrated by the experience of providing care they feel to be ineffectual and unwarranted, especially in the light of a perceived scarcity of expensive resources such as intensive care unit beds.¹⁰

Implantable cardioverter defibrillators (ICDs) and bi-ventricular pacemakers with defibrillation capability (CRT-D) involve a form of resuscitation. While some controversy remains concerning a patient's or authorized surrogate's decision to deactivate an ICD or CRT-D, there is an ethical and practical consensus that such deactivation is permissible.^2,11,12 A patient's or surrogate's decision to forego cardiopulmonary resuscitation (CPR) logically encompasses the choice to deactivate the resuscitative function of an ICD or CRT-D.

Decisions to forgo CPR that are made against patient or surrogate wishes in certain, defined circumstances of medical futility [sometimes termed ‘unilateral do-not-resuscitate’ (DNR) decisions] are much more controversial, yet are supported by some state laws and codified as hospital policies.^13,14 If a decision is made to forgo standard resuscitation against patient and family wishes, would such a decision logically extend to a decision to forgo resuscitation by means of an ICD or CRT-D? There has been little discussion of whether unilateral DNR orders translate into unilateral ICD or CRT-D deactivation. The objective of this study was to explore patient perspectives on ICD or CRT-D deactivation, including unilateral deactivation.

Methods

The study was approved by the local Institutional Review Board. A total of 99 patients presenting to the outpatient electrophysiology clinic at the Hospital of the University of Pennsylvania were invited to participate. Inclusion criteria consisted of having an ICD or CRT-D. Patients were approached in the waiting room and asked if they would like to participate in the study before their appointment with their physician. After excluding patients with pacemakers without defibrillation capacity and patients who declined to participate, 60 respondents' data were included in this study. The survey instrument consisted of a combination of demographic, multiple choice, and open-ended questions (see Supplementary material online for full survey). Health status was classified on a scale of 1–10 (1 being poor and 10 being excellent). Respondents were asked about advance directives and their devices, and were presented with clinical scenarios to elicit opinions regarding the ethical or moral permissibility of unilateral device deactivation.

Responses were recorded on an audiotape device, and interviews were later transcribed. Statistical significance between respondent demographic groups and answers was tested using cross tabulations. A P-value of <0.05 was considered statistically significant. Qualitative answers were analysed using NVivo software, where answers were coded to identify consistent themes.

The average age of respondents was 59 (range 23–89), and 70.0% (n = 42) were male. Of the respondents, 35.0% (n = 21) had ICDs, and 65.0% (n = 39) had CRT-Ds. From the total, 81.7% (n = 49) of respondents were Caucasian, 15.0% (n = 9) were African American, and 3.3% (n = 2) were Hispanic. The sample had varied levels of education and income. Sixty-five per cent (n = 39) were married, and 78.3% (n = 47) had children (Table 1). The average health classification was 6.8 on a scale of 1–10. Respondents had had their devices for an average of 6.74 years (range 0.11–25) (Table 2).

Results

Advance directives and devices

A small majority, 53% (n = 32), of the respondents reported that they had an advanced directive. Of those who did have an advanced directive, only one respondent's advance directive mentioned what they would like done with their ICD or CRT-D at the end of his/her life. This respondent reported that their advanced directive said to ‘turn it off’ in an end-of-life situation. Only 15% (n = 9) of respondents reported that they had thought about what they would like done with their ICD or CRT-D if they were to lose decision-making capacity and develop a serious illness from which they were not expected to recover. Even fewer respondents, 3% (n = 2), had ever discussed with a physician what should be done with their device (Table 3).

Clinical scenario responses

Scenario 1 described deactivation of ICD shocking function in accordance with a patient's wishes (Figure 1). A majority of respondents (76.7%, n = 46) stated that this was not a form of physician-assisted suicide. One patient reasoned that it was not about suicide, and that God made the decision for the patient to die. Of those who answered that this was a form of physician-assisted suicide, one respondent stated that turning off the ICD was ‘giving somebody a directive to shut them down’. Another said that in shutting of the device, ‘you are almost guaranteeing the patient dies’.

Scenario 2 described deactivation of ICD shocking function in a patient with a pre-existing DNR order (Figure 2). A majority (65%, n = 39) indicated that the physician had made the right decision, and that the device should be turned off. Two of the respondents who indicated the device should be turned off stated that the decision was made respecting the patient's wishes. Two others stated that the doctor was ‘following protocol’ by extending the patient's DNR to the functions of the defibrillator. Respondents who said that the device should be left on stated that they were uncertain whether the patient's DNR should extend to their device unless explicitly stated in an advance directive. Two respondents commented that, with this ambiguity, the decision should be left to the patient's family instead of the physician.

Scenario 3 described deactivation against a patient's wishes by two physicians who believed treatment to be futile (Figure 3). A majority (60%, n = 36) indicated that they thought the physicians should not unilaterally deactivate the device. One patient commented that the physicians ‘overstepped their bounds’. Multiple respondents said that there may be legal issues associated with unilateral device deactivation, with one respondent saying ‘I would sue the doctor’, and another saying ‘Doctors have no right to do that unless I give the order or my power of attorney does’. One respondent stated that this scenario was ‘Obviously a little more than physician-assisted suicide. That's physician-assisted homicide’. Of those who responded that the device should be turned off, four commented ‘the doctor knows best’. One respondent said ‘I think that the doctor should have the last say in it’.

Respondents were asked directly whether it is ethical or moral for the physicians to deactivate a defibrillator against the patient's wishes. A large majority (78.3%, n = 47) responded that it was not, and 6.6% (n = 4) respondents said that their answer to this question would depend on the patient's age, prognosis, decision-making capacity, and/or quality of life (Table 3).

Discussion

Expert consensus statements by the Heart Rhythm Society and the European Heart Rhythm Association consider withdrawal of pacemakers and ICDs to be legal and ethical, as long as this is consistent with patients' end-of-life goals and values. The consensus statements classify these devices as life-sustaining therapies: interventions that prolong life but may or may not reverse the underlying disease. Patients therefore have the right to refuse such treatments, and the act of deactivation does not constitute either euthanasia or physician-assisted suicide/physician aid in dying. However, these statements do not consider unilateral deactivation.^11,12

Prior studies have investigated physician,^2,3,15 other medical provider,^16,17 and patient^18–21 perspectives on withdrawal of rhythm management device therapies. Our findings agree with these past studies that show that a majority of patients do not consider deactivation of an ICD to constitute physician-assisted suicide/physician aid in dying. However, the topic remains controversial, as patient requests to deactivate rhythm management devices when healthcare providers disagree can lead to moral distress. As noted in prior work,^15,18–21 we found that a majority of patients saw no ethical distinction between ‘external’ defibrillation during a code and ‘internal’ defibrillation by ICD or CRT-D. Another study of 384 electrophysiology practitioners found that a majority did not think that it is ethical/moral for a physician to deactivate a device against a patient's wishes.¹⁵

This preliminary work constitutes the first empirical study of patients that extends the exploration of the ethical issues surrounding deactivation to unilateral deactivation. While most respondents in our sample did not believe deactivation in accordance with patient wishes to be ethically problematic, a similar majority rejected the notion that unilateral deactivation was ethically or morally appropriate.

Unilateral do-not-resuscitate

There is legal precedent for upholding unilateral DNRs. The Patient Self Determination Act of 1990 firmly established medical decision-making, and focuses on patients' rights to refuse therapies, but does not dictate that healthcare providers must honour demands for therapies.²² Building on long-standing precedent, however, the American Medical Association (AMA) took the position over 20 years ago that providers need not provide medical interventions that they consider to be futile. The AMA included CPR, and stated ‘in the unusual circumstance when efforts to resuscitate a patient are judged by the treating physician to be futile, even if previously requested by the patient, CPR may be withheld’.²³ The American Heart Association (AHA) states that patients should be given resuscitative care unless there is clear and quantitative evidence of medical futility.²⁴

Some state legislatures and health systems have officially adopted provisions that allow unilateral DNRs—the withholding of resuscitation against patient and/or family wishes. In 1999, Texas introduced the ‘Texas Advanced Directives Act’. This act states that if a physician believes further treatment to be futile for a given patient who wishes it continued, he or she is required to engage an ethics committee to make a decision. Patients and surrogates are given time to seek transfer of care to another institution, but the withdrawal or withholding order goes into effect after that time period. Futile treatments in these cases may include CPR.¹³ Individual hospitals have adopted policies that allow physicians to order a unilateral DNR if the physician believes that CPR is ‘unlikely to be successful’. This policy was enacted after the hospital utilized the Houston Model to assess medical futility. While the Houston Model was helpful in defining the process through which determinations of futility should be made, it did not clearly outline how various actors' opinions, such as physicians, family, or other ‘third parties’, should be balanced.¹⁴

There has been considerable debate in the ethics literature about what exactly is meant by ‘futile’, as some definitions are thought to be too vague in their meaning and clinical application.^25,26 Waisel and Truog²⁷ argue that only a physiological definition of futility should be used, as definitions which rely on qualitative (‘life not worth living’) or quantitative (‘less than 1 in 10 chance of working’) aspects inappropriately involve clinicians' values. The Hastings Center defines ‘medical’ or ‘physiological’ futility, which is measureable quantitatively, and argues that if a treatment will not achieve its physiological objective or provide a physiological benefit, it does not need to be performed.²⁸ This definition aims to eliminate any bias or value-ridden judgements from decisions to end (or not perform) treatment, circumventing paternalistic decision-making by the physician for the patient.^26,29

Unilateral deactivation

Regardless of the definition of futility, is unilateral ICD or CRT-D deactivation a logical extension of unilateral DNR? There is no legal precedent for this specific type of unilateral decision-making, and these authors are not aware of any specific cases.

The present study provides supportive empirical evidence that the majority of patients do not consider unilateral deactivation to be ethical or moral, even in the setting of futility, although a minority either deferred to physician expertise and knowledge over patient preferences or indicated that other factors must be considered. Interestingly, fewer respondents indicated that unilateral deactivation was ethically/morally problematic in Scenario 3, in which doctors had determined defibrillation was futile than when asked directly about unilateral deactivation without mention of futility. The views on the issue of futility determined by doctors suggest some degree of ethical ambiguity in the permissibility of unilateral deactivation. Further study of this topic is needed to build upon these findings.

In addition to ethical and moral concerns over clinician bias and loss of patient autonomy, unilateral ICD or CRT-D deactivation is also problematic when considered in the legal context. In the 1914 trial of Schloendorff v. Society of New York Hospital, Judge Cardozo stated ‘every human being … has the right to determine what shall be done with his own body’.³⁰ In criminal law, battery is defined as ‘a physical act that results in harmful or offensive contact with another's person without that person's consent’, and thus deactivation could fall under this definition.³¹ These legal precedents and definitions suggest that if a device is considered part of the body or even property of the patient, unilateral deactivation would be legally problematic. Forced deactivation against patient or surrogate wishes, regardless of futility, requires a physician or medical provider to perform an action to a patient's body. Such an action could be considered battery, as physical touching is required to turn off an ICD or CRT-D with a programmer, even though this action is not invasive or painful. Wireless deactivation, performed unilaterally, could complicate strict legal notions of battery, although it may still be defined as an act of battery if the device is thought of as a part of the body.

Limitations

This preliminary, single academic, urban centre study with a relatively small and ethnically homogenous sample size may have limited generalizability, in addition to compromising the ability to detect statistically significant associations between respondents' demographics and survey answers. Generalizability may also have been compromised by recruitment of respondents from the outpatient clinic setting. On the other hand, the average age of the patient's interviewed in this study was 59 (with a range from 23 to 89). An older patient cohort, presumably closer to end-of-life, may have given different answers than the population of this sample that varied in age. At the same time, the questions asked device patients to consider issues related to ICD and CRT-D deactivation in a general sense, not specifically applied to themselves. In addition, the limited scope of the survey could not fully explore the beliefs and experiences underlying each patient's individual perspective on unilateral deactivation, which likely influenced their answers. In particular, we did not address the issue of the unconscious vs. conscious patient, and whether unconsciousness was likely to be permanent. Withdrawal or deactivation of therapies in an unconscious patient may have very different emotional consequences for decision-makers (both family members and physicians), and while outside the scope of our survey, play a role in different decisions. Although religious background and beliefs may play a significant role in decisions about end-of-life care of devices, we did not perform a detailed exploration of each patient's spiritual or faith tradition/adherence (or lack thereof), as we believed it would have added to the length of the survey and was beyond the scope of the current project.

Conclusion

In an era of cost-consciousness and scrutiny of resources, management of ICDs and CRT-Ds as patients near the end of their lives creates ethical dilemmas. While the extension of the concept of unilateral DNRs to unilateral ICD and CRT-D deactivation may be attractive in the setting of cost consciousness, scarce resources, and frustration over provision of futile treatment, our study found that patients raise serious ethical and moral objections to this practice. A substantial majority of respondents believe that unilateral deactivation is not ethical or moral, even in the setting of medical futility. However, few patients had previously considered device deactivation at end-of-life, or discussed their preferences with their physicians. Advance care planning for these patients should address device deactivation, and future studies need to address institutional policies and communication between patients and providers about DNR and device deactivation.

Supplementary material

Supplementary material is available at Europace online.

Acknowledgements

The authors thank Miss Tiffany Young for her help with data collection.

Conflict of interest: none declared.

References