Abstract
Dear Editor:
In their Letter to the Editor, Rady and Verheijde 1 take issue with the consensus view 2 that deactivating cardiovascular implanted electronic devices (CIEDs) is equivalent to withdrawal of any other potentially life-prolonging intervention. Instead they argue that deactivating pacemakers (PMs), implanted cardioverter-defibrillators (ICDs), and ventricular assist devices (VADs) constitutes physician-assisted death, and that therefore deactivation of any CIED should be prohibited.
If their recommendation were adopted, clinicians would have to turn back the clock on what has become a widely accepted standard of practice. In a recent survey, 97% of hospice agencies reported that they enroll patients with ICDs. Fifty-eight percent of these patients had received shocks in the past year, and in 42% the shocking function was deactivated. 3 In a survey of 787 physicians, nurses, and representatives of two device manufacturers caring for patients with CIEDs, 86.8% of respondents reported involvement in ICD deactivation, and 77.6% in PM deactivation. 4
The authors' argument boils down to two assertions. First, they say that because CIEDs replace native cardiac functions, discontinuing them determines both the timing and the mode of death. This implies that clinicians who knowingly deactivate these devices intend to cause the patient to die. Second, in order for death to be “natural” after a CIED has been implanted, it must be due to “new lethal pathophysiology,” e.g., infection, hemorrhage with circulatory collapse, or respiratory failure. Deactivation-related death resulting from the pre-existing cardiac dysfunction that led to device implantation, they say, is by definition “unnatural,” and by implication unethical and perhaps illegal.
There are problems with each of these arguments. First, arguing that CIED deactivation should be illegal goes against widely accepted standards established over a third of a century of case law. In 1976, the New Jersey Supreme Court ruled that Karen Ann Quinlan's ventilator could be discontinued because her personal interest exceeded that of the state. Nine years later, the same court decided that any medical intervention could be withdrawn if the patient wished. The US Supreme Court ruled similarly in the Cruzan case 5 years later. Most authorities now believe that legally and ethically, CIEDs do not differ from feeding tubes, mechanical ventilation, and dialysis, which replace native digestive, respiratory, and renal function respectively.
Furthermore, the authors' argument that device deactivation directly causes death is not supported by experience or evidence. After Quinlan's ventilator was withdrawn, she did not immediately succumb to respiratory failure as expected. In fact, she lived for 9 more years before finally dying of pneumonia. Similarly, except for VADs that provide complete circulatory support and PMs implanted in pacemaker-dependent patients, CIED deactivation does not necessarily cause death.
The authors lump all CIEDs together, condemning deactivation of any of them. In practice, however, benefits and burdens differ from device to device. ICDs, for example, do not invariably prolong life. In major ICD trials, no survival benefit was documented until after the first year of implantation. 5 Because their effectiveness decreases as illness advances, ICD implantation is not recommended for patients with end-stage heart failure (HF), the very cohort in which deactivation would most likely be contemplated. By the same token, in non-PM–dependent patients, pacemakers prevent bradycardic episodes that may cause discomfort, but they do not necessarily prevent death.
So much for the benefits. The burdens are another story. ICD shocks are painful to the patient and upsetting to those around them, 6 and adverse effects are common. “ICD storm,” defined as the delivery of two or more shocks in 24 hours, occurs in 10%–20% of ICD recipients and is associated with long-lasting physical and psychological trauma in both patients and caregivers. 7 Normally functioning PMs, on the other hand, are not associated with significant morbidity.
Taken together, these findings imply that deactivating a PM near the end of life may cause discomfort, because the resulting bradycardia can reduce cardiac output below levels that sustain physiologic organ perfusion. Turning off an ICD, on the other hand, may prevent significant discomfort caused by shocks from the device itself; deactivating an ICD does not cause pain or any other symptom. Not surprisingly, a survey of 658 patients, as well as professionals in medicine and law, showed that overwhelming majorities in each group favored deactivating ICDs at the end of life, whereas fewer (although still majorities) agreed with turning off a pacemaker. 8 Deactivation is not a black and white issue. A thorough consideration of burdens versus benefits of continued CIED operation should take place when implanted patients progress to advanced illness. Beyond that, even patients who are completely device-dependent, such as those on destination VAD therapy, should not be prohibited from discontinuing it, any more than they currently are from, say, discontinuing a ventilator.
In a survey referenced by the authors, some physicians felt that the inherent nature of CIEDs made their deactivation difficult to discuss with implanted patients, even though these same physicians also thought that the topic of ICD deactivation should be included in advance care planning conversations. 9 The generalists surveyed in this study may have been less familiar with CIEDs than with older technologies where controversies over withdrawal have long been settled. Electrophysiologists, who have more experience in the implantation, management, and deactivation of CIEDs, do not express the same reservations. 10 This controversy will likely resolve over time, as it has with other life-prolonging interventions.
Finally, the authors contend that in order to die “naturally,” patients with CIEDs must die of non-cardiac causes. This argument is troublesome. Because CIED implantation rates are increasing, and because the final common pathway of coronary disease is progressive heart failure, we are only now seeing the first phase of an inevitable increase in the number of CIED-implanted patients with end-stage HF. Prohibiting deactivation of these devices would condemn this growing cohort to suffering caused by the continued operation of CIEDs. Moreover, it would compel them to trade a quick and relatively merciful death from cardiac arrest, which many patients say they would prefer, 11 for a lingering death from pump failure. For these reasons, most authorities now agree that CIED deactivation, like withdrawal of any other life-prolonging medical intervention, should be at the discretion of the patient.
The authors' criticism, which extends beyond CIED deactivation, may not enjoy widespread support. One of their articles opens with this statement: “There are growing concerns in the medical community about conflating physician-assisted death (PAD) with palliative care in end-of-life (EOL) care.” 12 These “growing concerns,” however, do not refer to expert opinion in peer-reviewed literature. They are the authors' own views from one of their previous articles. 13
Regardless of the authors' beliefs, CIEDs do not possess special characteristics that convey unique moral status relative to ventilators, feeding tubes, or dialysis. 14 Arguments to the contrary might be based not on objective factors, but on more subjective principles, e.g., those based on emotion or religious belief. In fact, the authors have invoked these principles explicitly in prior published statements questioning the morality of organ sourcing for transplants 15 and palliative sedation. 13 If they are pursuing a similar agenda in this letter, they should acknowledge it. This might promote a more open discussion, not just about clinical decision making in difficult cases, but also about important cultural issues. It might also prevent unnecessary suffering for some of our most vulnerable patients.