Of Pacemakers and Intention
Notes from a Family Meeting, Vol. 4, No. 5
Free from the ventilator, Paul had lived longer than anyone had predicted. Minutes rolled into hours which rolled into days. He didn’t wake up, and he wasn’t eating or drinking, so time was short for him, but when it became apparent it wasn’t that short, he was transferred out of the ICU to a regular ward bed.
His breathing was even and unlabored. If you paid close attention, you would see his chest remained still for twenty seconds or so at a time. His wife and his brother sat at his bedside. Concern wrinkled her forehead while his brother just looked tired.
“Did we make the right choice? Paul is still fighting,” his wife clenched and unclenched her arthritic hands, the mild pain distracting her from the grieving ache in her chest. His new doctor, whom they only met yesterday, stood near the door.
Paul’s brother scowled. “He’s got that damn pacemaker. Can y’all turn it off so he can have a peaceful death?”
The doctor cleared her throat. While it was typical for family members to decline certain medical therapies, even those that could result in death, there was something about this request that made her uneasy. Nevertheless she nodded. Paul’s wife didn’t look at all comforted, but nodded too.
My reflections here will be anchored in ideas I explored on death as a goal of care. If you feel like I’ve left something out, check out that essay. If still you think I’ve overlooked something, let me know.
People have a right not to be assaulted. That includes assault with medical interventions. If they don’t want it, and have the capacity to decide to decline it, then we can’t foist it on them. This also includes stopping therapies, and even those therapies that are helping to sustain someone’s life (e.g., a ventilator). That seems so straightforward it’s hard to believe it required development in the US legal tradition to take hold, something I’ll peek into below.
The times when we do treat someone over their objection (or without their direct consent), it’s because they lack the capacity to decline and either a surrogate decision-maker decides for them that they’ll receive it, or it’s an emergency, we’re unsure what they would have chosen. We need to implement life-saving measures to sort out the decision-making later. But whether a patient is deciding or someone is deciding for them, the clearest right regarding autonomy in healthcare is a negative one: if a patient says no, we go no further.1
Until we get to implantable cardiac devices, which stir up controversy unlike any other medical technology. These devices are pacemakers, automated implantable cardiac defibrillators (AICD), and ventricular assist devices (VAD). There are particular types of device within each of these categories (e.g., transvenous pacing vs permanent pacemakers; cardiac resynchronization therapy; anti-tachycardia pacing; LVAD vs RVAD; etc), but for the purposes of our discussion, I think these three broad categories should suffice.
Pacemakers help the heart beat at a normal rhythm and rate. They’re indicated for people whose hearts beat too slowly (or not at all) on their own. Some people depend entirely on their pacemaker; without it, their heart rate is zero. Other people only intermittently fall below a certain threshold, so their pacemaker is only pacing them a certain amount of time. These hearts will beat slowly but ineffectively without the pacemaker.
An AICD is a device that is intended to shock a heart back into a viable rhythm. People might have one of these implanted if they’ve been resuscitated from a near-lethal arrhythmia, or if they’re at high risk for such an arrhythmia due to something like heart failure. Someone might have one of these and never experience a shock; it’s there for emergencies.
A VAD is a device that takes over the pumping functions of one side (usually) of a heart that can’t pump well on its own anymore. They’re sometimes used as a final therapy for this condition, and at other times as a temporizing measure until a heart is available for transplant.
As with any other therapy, thoughts about why, when, and how you’re going to stop it should accompany the plan to start it. In the case of something like a ventilator, for example, the vast majority are stopped because people get well enough to no longer need the machine’s support. However, there are some patients for whom the ventilator isn’t beneficial because they’re dying, and they (or, more commonly, their surrogate decision-maker) opt for a plan of care that stops burdensome, unhelpful interventions.
The rationale that justifies stopping a life-sustaining therapy even when death may result is called the rule (or principle) of double effect. It’s the same rationale that justifies starting a medication even when an adverse effect might be foreseen. The rule of double effect holds that an action may have multiple good or bad effects, but is permissible if:
The action itself is not immoral.
The action is intended to achieve a good effect, not to achieve the bad effect (even if it’s foreseen)
The action doesn’t bring about the good effect by means of the bad effect.
The action weighs the proportionality between the good and bad effect (e.g., if there’s an overwhelming risk of the bad effect, but a vanishingly small possibility of the good effect, proportionality may not be in favor of acting).
Because all medical intervention involves some degree of trade-off, this rule is built into reasoning around clinical management even if most clinicians don’t know the name for it. If I plan to prescribe insulin for a patient with diabetes, we have a discussion around the burdens and benefits. Informed consent acknowledges that starting insulin itself isn’t immoral, it’s intended to help the person (not cause symptomatic or life-threatening hypoglycemia, for example, even if that’s a risk), symptomatic hypoglycemia isn’t the means through which we achieve better control of the diabetes, and the patient and I believe the benefits outweigh the burdens.
We wade into more controversial territory when we discuss withholding and withdrawing life-sustaining therapies, like a ventilator or dialysis. However, in the United States, these legal questions have long been settled: people are free to decide if the burdens of a particular intervention outweigh the benefits, even if their death is foreseen without the medical intervention.2
The controversies deepen when we come to morally contested practices. Not everyone agrees on what constitutes an “immoral action” when it comes to things like assisted suicide and euthanasia (ASE), and so some clinicians seek to discard the rule of double effect or else dilute it with so much relativism as to make it useless (what is moral after all?). That conversation is worth considering, but I just wanted to flag it here and move on to cardiac devices.3
Cardiac Confusion
Where do cardiac devices come in? At face value, they’re medical interventions, and patients can accept or decline their implantation just like any other intervention. Once they’re in and patients want them out or turned off, though, things get tricky.
Turning off the defibrillator portion of an AICD is the least controversial action. Just like a patient can decline to undergo chest compressions and external defibrillation, they can also decline to receive shocks from an AICD if they suffer an arrhythmia. There are particular clinical considerations here (e.g., some tachytherapies can restore a normal rhythm without a big shock and, even with a big shock, the heart rhythm might be restored without the need for further intervention that we typically associate with cardiac arrest - intubation, ICU admission, etc.). However, from an ethical standpoint, it’s straightforward enough: you don’t want to be shocked. Ok, we’ll turn it off for you.
Turning off a VAD can induce a great deal of moral distress in some clinicians because death will probably follow instantaneously after cessation. It feels like you’re ending the patient’s life, even if you’re not (assuming you’re relying on the rule of double effect). The burdens of a VAD can be immense, particularly as someone’s health irreversibly deteriorates. Reasoning through the rule of double effect, stopping a VAD intends the release from those burdens, even if death is foreseen but not intended.
Now, we come to the most controversial technology to stop. When functioning properly, pacemakers impose absolutely no burden. You cannot feel what they’re doing and they don’t require frequent maintenance. Pacemakers do require battery exchanges every 5-10 years, at which point someone could decline that intervention. But what about turning off a pacemaker?
Intention Matters
Daniel Sulmasy directly addresses this dilemma by drawing a line between those therapies that become “self” and those that remain “other.”4 The more closely a therapy approaches “self,” the less justification we have for withdrawing it. Consider the intuitive difference we sense between removing a ventilator, for example, and explanting well-functioning transplanted lungs. Sulmasy helps move us beyond intuition. He distinguishes between regulative and constitutive therapies. Regulative therapies are those that “coax the body back to its own homeostatic equilibrium.” Many medications do this, e.g., anti-hypertensive medications, anti-arrhythmic medications, etc. Constitutive therapies are those that “take over a function the body can no longer do for itself.” Insulin and pacemakers fall in this latter category. Sulmasy notes that regulative therapies are never “self” and therefore comfortably fall within the traditional realm of ethical decision-making regarding withdrawing and withholding medical therapies.
Constitutive therapies require a little more thought. He admits that this hypothesis is provisional, so we need to take it with a grain of salt. He argues that constitutive therapies are either replacements or substitutes. A replacement therapy becomes a part of someone’s restored physiology (e.g., kidney transplant), whereas a substitute does not (e.g., dialysis). He offers some signs to help further distinguish these two types of therapies:
Does it change in response to the organism and environment?
Does it grow or repair itself?
Is it independent from an external energy source?
Is it independent from external control?
Is it immunologically compatible?
Is it physically integrated into the person’s body?
Relying on these signs, we discern that a porcine heart valve is a replacement therapy, whereas a pacemaker is likely a substitute therapy. Replacement therapies, by being so closely identified with “self,” cannot be so easily withdrawn. “Whereas there is no absolute standard for judging whether something is a replacement or a substitute,” Sulmasy writes, “the more clearly a technology can be classified as a replacement therapy, the greater the case for judging that its discontinuation would constitute an immoral act of killing.” Why?
“Replacement therapies become part of the restored physiology of the patient, part of the integrated unity of the patient as an intact individual organism. To discontinue such therapies is better understood as introducing a new lethal pathophysiological state rather than discontinuing a treatment that is merely substituting for a preexisting lethal pathophysiological lack of that function. The discontinuation of a replacement therapy thus becomes an act of killing.”
Therefore it’s unproblematic to discontinue dialysis, but egregious to remove a well-functioning transplanted kidney. The intention of stopping dialysis could be that one wants to avoid the burdens of the procedure, whereas the intention of removing a transplanted kidney (without subsequent plans to start dialysis or replace with another kidney, like if the transplanted kidney developed a tumor) can only be death.
Using Sulmasy’s reasoning, a pacemaker doesn’t show signs of being a replacement therapy and, in theory, could be discontinued without any ethical problem. However, elsewhere, Karches and Sulmasy note the importance of intention: "If a physician or patient surrogate intended by withholding or withdrawing a [cardiac implantable electronic device] to hasten the end of a life deemed of low quality, such an act would not be morally justified.” But intention isn’t the focus of their reasoning.
Huddle and Amos Bailey suggest there’s more here worth considering than just the self/other distinction. They argue that
“…the right to refuse treatment applies to treatments involving ongoing physician agency. This right cannot underwrite patient demands that physicians reverse the effects of treatments previously administered, in which ongoing physician agency is no longer implicated. The permanently indwelling pacemaker is best seen as such a treatment. As such, its deactivation in the pacemaker-dependent patient is best seen not as withdrawal of support but as active ending of life. ... [Demands to deactivate a permanent pacemaker] are not refusals of treatment but demands for undoing previous treatment. As such, they do not fit into the traditional analysis of legitimate physician withdrawal of support.”
It might be helpful to bring the conversation back around to the rule of double effect. “Intention” can be a confusing concept. Quill, Dresser, and Brock complain that “human intention is multilayered, ambiguous, subjective, and often contradictory.” Although that’s true, and some decisions are made with deep ambivalence, Cavanaugh brings us some clarity. How do you know whether you’re intending or foreseeing something? Deliberation. You don’t deliberate on things merely foreseen as if they were the end of your action, nor do you deliberate on what means might best accomplish something foreseen. To leverage what is foreseen to achieve what is intended is to adopt that thing as the means and thus make it part of the intention.
So, we can stop a ventilator knowing our patient will die but not intending their death. How can we make that claim? Because we haven’t deliberated on how to end their life; instead we’ve deliberated on how to free them from a burdensome medical intervention, even at the end of their life. Consider the same action, though, done so that they’ll die. The intention is starkly stated. Death is not only foreseen but intended.
Does that mean patients who request the withdrawal of a ventilator so that they’ll die should remain the ventilator over their objection? No. Not necessarily, anyway. Treatment over objection may be so logistically untenable and practically inhumane that it would be impossible to implement. Complying with the patient’s request, after a conversation or series of conversations, doesn’t necessarily mean intending their death as they intend their own death.
However, that also doesn’t mean the clinician is “forcing” the patient to stay alive if they’re not being sustained by any therapy. This is sometimes an emotional appeal in favor of ASE: why are you forcing this person to live instead of prescribing them a lethal substance? A clinician mustn’t assault their patient with medical technology, even if their desire is to die, but that clinician doesn’t have a companion duty to kill a patient who wants to die.
To bring us back to pacemakers, the request to stop a functioning pacemaker on which someone is entirely dependent, if earnest and well-informed, can only be made with the intention of ending one’s life. There is no burden one is seeking to avoid by turning off a pacemaker. The patient intends their death. The clinician who turns off the device intends their death. There is no other goal and can be no other intention. Unlike a ventilator or dialysis, permanent pacemakers do not require day-to-day management; there’s no human agent that could be “assaulting” a patient day after day with the ongoing activity of a pacemaker (similar to how silly that argument would be if made about a kidney transplant).
Intention matters.
Legal Musings
Although I’m no attorney, I understand that intention is one of the major facets of criminal law. Determining mens rea is the challenge of the court. Given its relevance in determining and grading the permissibility of action, it’s strange that, in medicine, there are some who would seek to obviate or demote intention in clinical reasoning.
Susan Stefan writes in Rational Suicide, Irrational Laws about the trajectory of the US legal history around the so-called “right to die.” The discontinuation of life-sustaining therapies was equated with suicide in the pre-Quinlan era even when death wasn’t intended - something totally foreign to our modern sensibilities, testifying to the power of legal and ethical changes since then. In those days, other parties (including clinicians) had interests equal to or even overriding the constitutional interests of the patient. She cites an excerpt from the New Jersey Supreme Court opinion from Heston v. John Kennedy Memorial Hospital (1971) which gives an idea of the reasoning of the time:
“Appellant suggests there is a difference between passively submitting to death [by refusing blood] and actively seeking it. The distinction may be merely verbal, as it would be if an adult sought death by starvation instead of a drug. If the State may interrupt one mode of self-destruction, it may with equal authority interfere with the other. It is arguably different when an individual, overtaken by illness, decides to let it run a fatal course. But unless the medical option itself is laden with the risk of death or of serious infirmity, the State’s interest in sustaining life in such circumstances is hardly distinguishable from its interest in the case of suicide.”
Then the case of Karen Quinlan was brought before the same court. This tragic case provided a paradigm shift in the legal and ethical landscape of modern medicine. Quinlan’s father wanted her ventilator withdrawn. Her physicians resisted, believing this was similar to Nazi-perpetrated euthanasia. The NJ Supreme Court sided with the father: the ventilator could be removed because Karen Quinlan had a right to privacy:
“We think that the State's interest contra weakens and the individual's right to privacy grows as the degree of bodily invasion increases and the prognosis dims. Ultimately there comes a point at which the individual's rights overcome the State interest. It is for that reason that we believe Karen's choice, if she were competent to make it, would be vindicated by the law. Her prognosis is extremely poor, she will never resume cognitive life. And the bodily invasion is very great, she requires 24 hour intensive nursing care, antibiotics, the assistance of a respirator, a catheter and feeding tube.”
As an aside, Quinlan’s doctors were terribly mistaken about her prognosis. She lived for nearly ten years after her ventilator was withdrawn. If the state’s interest in preserving life diminishes as “prognosis dims,” it nevertheless should also take into consideration that its interest grows as the individual becomes more vulnerable (e.g., in the throes of a serious illness where they’re sick enough to die). This ruling links the inexact science of prognostication with notions that not just life expectancy but level of cognitive and physical functioning should inform the state’s interest in preserving life. This allows for ableist beliefs to foment within the privacy of individual psyche which will later shape culture. Nevertheless, the court’s fundamental assertion, that people have a right to be free from unwanted medical intervention, was a promising development.
Other cases took intention into account. Elizabeth Bouvia, contesting her right to stop tube feedings over her objection, won on appeal. The California court, remarking on the lower court’s opinion, ruled:
“…the trial court seriously erred by basing its decision on the ‘motives’ behind Elizabeth Bouvia's decision to exercise her rights. If a right exists, it matters not what ‘motivates’ its exercise. We find nothing in the law to suggest the right to refuse medical treatment may be exercised only if the patient's motives meet someone else's approval. It certainly is not illegal or immoral to prefer a natural, albeit sooner, death than a drugged life attached to a mechanical device.”
This perspective presents a clinical challenge, however, as motive is a major facet of determining decision-making capacity embedded in whether one appreciates one’s current circumstances and the various medical options available. If a man opts against surgery because of a paranoid delusion, that is a relevant, important clinical and ethical concern. Nowhere in the USA, including in California, is it treated as an untouchable motive shielded behind the right to privacy.
The judicial challenges continued throughout the 1980s and 1990s. There were other major cases, but the one perhaps most relevant to our discussion about intent was Vacco v. Quill, argued before the US Supreme Court. When discussing the difference between the withdrawal of life-sustaining therapies and suicide, the Court grounded their reasoning on the “fundamental legal principles of causation and intent.” The cause of death of someone who declines a life-sustaining therapy is their disease, whereas it’s the substance they ingest in (a medically assisted) suicide. Furthermore, a physician intends to “only respect his patient’s wishes and ‘to cease doing useless and futile or degrading things to the patient when [the patient] no longer stands to benefit from them” when that physician withdraws or withholds a life-sustaining therapy at the patient’s request. “Similarly, a patient who commits suicide with a doctor's aid necessarily has the specific intent to end his or her own life, while a patient who refuses or discontinues treatment might not.”
A lot hangs on that “might not,” and unfortunately the Court didn’t clarify the recourse clinicians have if they suspect a patient is wishing to withdraw or withhold a life-sustaining therapy because of suicidal intent. They also didn’t warn clinicians against intending a patient’s death via withdrawal or withholding life-sustaining therapies.
In his concurring opinion in Cruzan v. Director, Missouri Department of Health, Justice Scalia argues that suicide could entail action or inaction, and leaves room for the possibility that one could die by suicide if one intends their own death by withdrawing or withholding a medical therapy. The counter-balancing consideration is that clinicians may have to go to inhumane, logistically impossible lengths in order to sustain life, such that a person may die by suicide even if clinicians have the means to avert it.
Who Cares?
Why all this rigmarole about intent? Well, intention frames responsibility. If a tree falls over and kills someone, we don’t say the tree intended for them to die, and it would be awkward language to claim the tree was “responsible” for the person’s death, even if it caused it. If a physician mistakenly prescribes the wrong medication in an electronic medical record, a number of things may have made such an error more likely, and even if they didn’t intend for it to happen, they’re responsible for the error, because humans have agency and can bear responsibility. Their lack of intent lessens the consequence (if any) they’ll face, for if they intended to prescribe the wrong medication, they’d receive much heavier consequences.
Those who argue that intention in medical ethics is too unreliable or irrelevant seem to view clinicians (either intentionally or unintentionally - I don’t know!) as mechanisms in a bureaucratic machine that delivers a service. Their agency doesn’t matter, only the widget they produce. Peter Berger wrote that this was one characteristic of the frame of mind embodied in technological production:
“A strategic element in the cognitive style in question is componentiality. The components of reality are self-contained units which can be brought into relation with other such units. … This apprehension of reality in terms of components is essential to the reproducibility of the production process as well as to the correlation of men and machines.”
He goes on:
“A further implication of this is the separability of means and ends. Since reality is apprehended in terms of components which can be assembled in different ways, there is no necessary relationship between a particular sequence of componential actions and the ultimate end of these actions. To take an obvious example, a particular assemblage of cogs produced in a highly specific production sequence may eventually go into a passenger automobile or a nuclear weapon. Regardless of whether the worker involved in this particular production process approves or even knows about its intended end, he is able to perform the actions that are technologically necessary to bring it about.”
In short, the intentions of individual clinicians threaten the standardization of clinical practice. That practice is oriented toward a particular end: the autonomous use of the clinician as a tool for whatever purposes the patient sees fit. Health becomes “whatever I want.” The customer is always right. I had considered before what happens when we try to supplant health with respect for patients’ autonomy as the end toward which healthcare aims. Nothing has changed my worry there.
So, if a clinician turns off a patient’s pacemaker so that they’ll die, they may claim they’re “just doing their job.” The only end they serve is to realize the patient’s request. In so doing, they feed a way of practicing medicine that views clinicians as instruments. You’re there to carry out the patient’s will, not to pursue their health (which may, at times, conflict). You should only deny a patient’s request when it exceeds your expertise (find someone else to do it then) or they lack the capacity to request it (find someone else to decide then). This is a significant step toward refining the clinician’s self-image as one of machine. They are to become like pacemakers.
The alternative is living medicine: a relationship among human beings, true moral agents, who each have duties, rights, and evaluations. This means intention is going to matter. This is non-controversial in wide swaths of medical care. A clinician might prescribe an anti-hypertensive to lower someone’s blood pressure, a proximal intention, but their ultimate intention is to reduce their patient’s risk of a major health problems like a stroke, heart attack, or even death. It would be an awkward framing to claim the clinician prescribed the medication merely and only because the patient requested it. If that were the case, then I suspect we would have seen many more prescriptions of ivermectin for COVID.
In such a routine practice of medicine we have the seeds for properly orienting the care of patients with serious illness who will die, sometimes as a result of the things we do or don’t do. In all that turmoil, let’s not lose sight of why we do what we do.
Trajectories
Following a meandering reading-path, sharing some brief commentary along the way.
“Patient values: Three important questions - Tell me more? Why? What else?”
A helpful guide to using three tools (questions) to explore patients’ values and goals.
“Core roles and responsibilities of physicians in hospice care”
An important document outlying what constitutes high-quality care from physicians in hospice care.
“Love your patient as yourself: On reviving the broken heart of American medical ethics”
Tyler Tate and Joseph Clair write about the importance of attention and virtue in medical care and ethics, aligning well with explorations here in Notes, particularly here and here.
“What should we do if our compassion runs out?”
Eric McLaughlin, a physician working in Burundi, reflects on his practice of medicine when so many of his patients die. Where is there hope? How can he sustain compassion in the face of such enormous suffering?
Closing Thoughts
“What is needed, then, is a different, broader understanding of restoration. Could it be that restoration has less to do with repairing the body (as important as it is) and more to do with returning those who are ill to a more homelike way of being-in-the-world? In other words, ought the practice of medicine attempt to make our falling out of health less acute and less lonely? If so, it might mean that in addition to treating the biological body, an effort is made to make patients feel acknowledged and heard, to make the "they” of illness less alienating, to create a space where the articulation of a person’s illness experience can be expressed and heard alongside conversations about the objectified body and disease processes.”
Nicole Piemonte, Existential Medicine
Well, that’s not entirely true. I’d continue the conversation, ask why, encourage them to tell me more about their thinking. But we’re not going to force something on someone who has the decision-making capacity to decline the intervention.
I included this in a footnote because it’s a big can of worms. I don’t want it to derail the main discussion, but there are certain situations where clinicians will “cheerlead” (to put it lightly) patients through a required period of life-sustaining therapies. For example, in the days and weeks following a transplant, the clinical course can be tumultuous. Patients may want to stop; they’re tired, demoralized, hurting, and scared. It’s too much, they say. Clinicians will work together to both simultaneously encourage the patient and also avoid honoring their request (except in the most extreme of situations). They’ll also exert a “gift of life” philosophy to foster adherence with post-transplant medical recommendations.
The rule of double effect doesn’t tell us when a clinician is obligated to act; it only guides us about what is permissible or impermissible. There’s a bigger conversation about duty that I can’t address here. The rule of double effect, though, is a useful tool for handling those things you’re permitted to do that could have bad effects.
The Heart Rhythm Society grounds its consensus statement in large part on Sulmasy’s argument.