The Uniform Determination of Death Act is Not Changing. Will Physicians Continue to Misdiagnose Brain Death?

Abstract

Efforts to revise the Uniform Determination of Death Act in order to align law with medical practice have failed. Medical practice must now align with the law. People who are not dead under the law that defines death should not be declared dead. There is no compelling reason to continue the practice of declaring legally living persons to be dead.

Keywords:

• Brain death
• determination of death by neurologic criteria
• Uniform Determination of Death Act
• American Academy of Neurology

INTRODUCTION

In the United States, a substantial proportion of patients declared dead by neurologic criteria do not meet the legal standard for death. While the Uniform Determination of Death Act (UDDA) requires the irreversible loss of all brain function, up to half of patients declared dead by neurologic criteria retain the brain function of osmoregulation, that is, hypothalamic regulation of sodium and water balance (Nair-Collins and Joffe 2021). This discrepancy between law and practice occurs because accepted diagnostic procedures for determining brain death do not assess hypothalamic function, indeed, standard guidelines explicitly consider persistence of neurohormonal function to be consistent with death (Greer et al. 2023). This “mismatch” between law and practice is well-known and has persisted for decades.

The Uniform Law Commission (ULC) is a nonprofit association supported by state governments, whose goal is to produce model legislation for states to consider adopting, with the goal of legislative and regulatory uniformity across state jurisdictions. Along with the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Biobehavioral Research (1981; henceforth “President’s Commission”), the ULC was responsible for drafting the UDDA of 1980. For the first time since then, in 2020, the ULC initiated a study committee and then drafting committee to revise the UDDA. Throughout that time, many physicians and bioethicists, along with professional societies, vigorously advocated that the law should change to fit medical practice, and not the other way around (e.g., Lewis, Bonnie, and Pope 2020; Omelianchuk et al. 2022).

The ULC suspended efforts at revising the UDDA in 2023. The uniform model law, the UDDA, remains the same: With respect to neurologic criteria for death, an individual is dead upon “irreversible cessation of all functions of the entire brain.” For the foreseeable future, the UDDA is not changing.

It appears that medical practice in the determination of brain death will not change in the foreseeable future either. Shortly after the ULC announced halting its activities, the American Academy of Neurology (AAN) released an updated set of guidelines for the determination of brain death (Greer et al. 2023). The changes across versions are superficial, with minor differences mostly revolving around ancillary tests. The core of the diagnosis of brain death remains, in broad outline, what it has been since the Harvard Committee released its report on brain death over 50 years ago (Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death 1968): in the context of a known cause of coma and with confounds ruled out, the patient must be unresponsive, apart from spinal reflexes; areflexic of pupillary, corneal, cough, gag, and vestibular reflexes; and apneic, as measured by the apnea test. Hypothalamic osmoregulation may be dismissed: “Clinicians may … determine a patient [dead by neurologic criteria] despite evidence of neuroendocrine function” (Greer et al. 2023, 1128).

Thus, some individuals who remain alive under the law will continue to be inaccurately declared legally dead. This is not due to clinician error, failure to follow accepted guidelines, or an inherently challenging diagnosis—any clinician competent in brain death diagnosis is competent to assess osmoregulation. There is no technical or diagnostic challenge here; rather, it is a choice to declare patients who do not meet the legal standard for death to be dead.

In this essay I defend the thesis that if someone is alive according to the law that defines death, they should not be declared dead under that law.

A few preliminaries: First, to say that a diagnosis is accurate or inaccurate is always relative to some condition that is being assessed. A patient might be accurately diagnosed as being in a condition characterized by the syndrome of unresponsiveness, brainstem areflexia, and apnea (i.e., in accordance with AAN guidelines), yet not accurately diagnosed as being in a condition in which all functions of the brain have irreversibly ceased (i.e., not in accordance with the legal standard). To head off confusion, when I say that brain death is misdiagnosed, or that living patients are inaccurately diagnosed as dead, I mean “relative to the legal standard stipulated in the UDDA” throughout the essay, including the title. I often add qualifiers (e.g., “legally”) as a reminder.

Below I discuss in more detail why the legal standard is important, but to provide some initial context, it is helpful to recall the significance of a legal determination of death. Being alive is a legal civil status, in which the individual—even if unconscious and no longer capable of directly expressing their preferences—is a legal living person, to whom a variety of laws and protections apply. Being dead is a separate civil status. Some protections apply to the deceased, but they are different. The rights and protections associated with the civil status of being a legally living person are separated from the civil status of being deceased precisely by the legal standard for death, a standard under the authority of an elected legislature (or in some cases an appointed judge), but not the AAN or any other medical society. Relative to that critical legal standard, patients are routinely declared to be dead in error, hence, legal rights that properly apply to legally living persons are denied as a result of following AAN guidelines which produce numerous false positives.

This does not necessarily mean that following AAN guidelines constitutes malpractice: to prevail in a malpractice suit one must show that there is a duty of care, a breach of that duty, a harm, and that the breach caused the harm. Breach of duty is almost always interpreted under the framework of negligence, that is, whether the provider followed the standard of care. The AAN guidelines are the standard of care, so I doubt that a plaintiff would prevail in a malpractice suit solely based on a provider following AAN guidelines (though malpractice is not the only legal concern, as I discuss below). The problem is with the AAN guidelines themselves.

Finally, nothing in this essay should be construed as a defense or critique of the validity of brain death itself.

THE MISDIAGNOSIS OF BRAIN DEATH

Diabetes insipidus (DI) is a condition characterized by abnormally voluminous excretion of water through the urine, resulting either from a deficit of anti-diuretic hormone (ADH; or vasopressin), known as central DI, or from the kidney’s inability to respond to it, that is, nephrogenic DI.

Vasopressin is secreted from the posterior pituitary by neurons whose cell bodies are located in the supraoptic and paraventricular nuclei of the hypothalamus, in response to changes in fluid osmolarity. A secondary set of neurons in circumventricular areas also respond to changes in osmolarity and provide input to the primary osmoreceptors in the hypothalamus, creating an additive effect resulting in the release of vasopressin from vesicles at the axonal termina of hypothalamic neurons in the posterior pituitary (Bourque 2008). Vasopressin in blood travels through systemic circulation to reach V2 receptors in the kidney’s nephrons, triggering synthesis of channels known as aquaporins, which allow water to reabsorb back into the bloodstream, and thus ultimately controls the amount of water passed through urine. In a deficit of vasopressin (central DI), or the inability of V2 receptors to respond to it (nephrogenic DI), aquaporins are not synthesized and water is not reabsorbed into the blood, instead passing through urine, resulting in voluminous production of dilute urine and potentially severe or fatal dehydration.

In 1984, Outwater and Rockoff reported finding central DI in 14 of 16 pediatric patients diagnosed as brain dead (88%), and wrote,

It is not surprising that DI develops in patients who are brain dead. This may be due to hypothalamic-pituitary dysfunction with cessation of ADH production, or to inability to transport stored ADH because of absent cerebral blood flow. On the contrary, it is puzzling that DI does not always occur with brain death, and that DI has not been previously reported in either adult or pediatric patients as a clinical manifestation of brain death (Outwater and Rockoff 1984, 1245).

Their puzzlement was justified. Legally, the definition of death requires no brain function, so if a patient does not have central DI, they are not legally dead. Physiologically, brain death occurs as a result of a positive feedback cycle of increasing intracranial pressure and decreasing cerebral perfusion, the outcome of which is thought to be complete intracerebral circulatory arrest. Osmoreceptive neurons lie within the hypothalamus in the diencephalon, hence, if the positive feedback cycle is complete, the diencephalon would be ischemic and anoxic, resulting in complete loss of function. If osmoregulation is preserved, then some blood flow continues, at least to the supraoptic and paraventricular nuclei, and probably to the secondary osmoreceptors in circumventricular areas as well.

Many studies have since been published reporting on the presence of DI in patients meeting accepted criteria for brain death. The most comprehensive review was published in 2021, accounting for 37 studies and 2546 patients, of which 1265 (50%) were reported as having DI (Nair-Collins and Joffe 2021; this is an expansion of an earlier review, Nair-Collins, Northrup, and Olcese 2016). Due to the sample size, the Central Limit Theorem justifies application of the z-distribution to the dataset, yielding a 95% confidence interval of 47.2—52.3%. However, this is not a systematic review or meta-analysis, and the data are subject to significant limitations. Diagnostic criteria for DI were often not explicitly stated, and there are confounds in most of the studies, discussed below.

These limitations decrease confidence in any specific estimate of DI in the population of patients meeting AAN criteria for brain death, despite the calculated confidence interval. This is why I stated in the introduction that “up to” half of patients—in other words, up to but perhaps not more than half—do not meet legal criteria for death. These limitations do not, however, challenge the more general, and ethically more important claim, that some substantial proportion of patients who are declared legally dead by neurologic criteria are not, owing to preserved brain function.

It is challenging to definitively calculate the specific incidence of central DI for the entire population of people meeting accepted diagnostic criteria for brain death, not only due to limitations in the source literature, but also because it is not a uniform population. This condition can affect people of any age, and it has many etiologies, including trauma, infection, cardiac arrest, metabolic derangements, and primary neurologic pathologies. The setting in which neurologic injury occurs differs, time to and quality of both prehospital care and hospital care may be different, and general health prior to injury or insult will certainly impact the clinical course. The key point is that misdiagnosis of death relative to the legal standard is routine, and not a function of correctable clinician error or failure to follow guidelines, nor some very rare anatomic or genetic variant that no one could reasonably predict. A substantial proportion of people declared dead by neurologic criteria are legally alive, and declaring them dead anyway is a choice, not a mistake.

THE FIRST LEVEL OF RESPONSES

The UDDA requires the absence of all brain function, therefore, the persistence of any brain function is inconsistent with it. There is no argument, and no evidence, that can escape this conclusion. To deny it is to deny rudimentary deductive logic. The evidence demonstrates that a substantial proportion—up to half—of patients declared dead by neurologic criteria retain some brain function. Therefore, a sizable proportion of patients declared dead by neurologic criteria are legally alive. This kind of conclusion is rare in any academic endeavor: it is essentially certain. It is grounded in unassailable logic and clear evidence. Nevertheless, commentators have offered a wide range of responses to deny this essentially certain conclusion.

Pituitary Washout and Hypothalamic Anatomy and Blood Supply

First, we might deny that osmoregulation occurs at all: what we see is merely vasopressin passively leaking out of dead cells, sometimes described as “pituitary washout” (Varelas et al. 2023). This hypothesis posits an unlikely occurrence, commonly occurring. Vasopressin has a half-life of 15-18 minutes. As little as a 1% change in plasma osmolarity induces a change in vasopressin secretion, and the osmoregulatory system maintains plasma osmolarity within a narrow range of about 3%, as a result of minute-to-minute changes in vasopressin secretion in a tightly regulated negative feedback system (Robinson and Verbalis 2003). Without negative feedback it is unlikely that passive leakage would account for normal or close-to-normal osmolarity. Instead, there would either be insufficient amounts of vasopressin (i.e., central DI), or vasopressin would continue to leak without control, resulting in oliguria or anuria (i.e., too much vasopressin), followed by polyuria minutes after stores are depleted.

One study combined pathology with plasma hormone levels and found that vasopressin levels rapidly dropped after brain death, in a cohort in which all (7) patients had central DI. On pathology, vasopressin positive granules were found in the posterior pituitary (Sugimoto et al. 1992). Varelas and colleagues took this to indicate hormonal leak into the periphery, thus providing evidence for the washout hypothesis (Varelas et al. 2023, 7). This interpretation is exactly the opposite of what the finding showed: Plasma hormone levels dropped quickly after brain death and all patients had central DI, indicating that any vasopressin remaining in the pituitary, even if it did leak into the periphery, did not mimic osmoregulation since, again, they all had central DI. This is consistent with the point above: osmoregulation is a highly sensitive and tightly regulated system, and it is unlikely that passive leakage would mimic this well-controlled system.

Nevertheless, the phenomenon of pituitary washout is not impossible—it cannot be ruled out—and so it might explain some cases of lack of apparent central DI. Yet, there is, and can be, no independent evidence that this phenomenon (rather than osmoregulation) occurs in an individual patient since, by hypothesis, washout mimics osmoregulation. Hence, even direct measurements of plasma vasopressin cannot distinguish washout from osmoregulation, because both would show normal levels of vasopressin in the blood. If this is correct, then pituitary washout is an absolute confound precluding the diagnosis of brain death, not a justification for determining a patient with apparent osmoregulation to be dead. It is like sedative intoxication or hypothermia, which are absolute confounds prohibiting brain death diagnosis until they are resolved. If a patient shows signs of osmoregulation, then it is either intact brain function, or it is washout that cannot be distinguished from intact brain function and therefore the determination of death under the UDDA is precluded and the patient is not legally dead. If washout truly is occurring, then central DI will set in minutes after vasopressin stores are sufficiently depleted—precisely as it did with the patients in the Sugimoto et al. (1992) study—at which point the confound is resolved.

Another response holds that the pituitary has a blood supply protected from intracranial pressure, and this explains why DI does not always occur (Russell et al. 2019). The dura mater, in addition to the cranium, is the inelastic container within which pressure rises. The posterior pituitary is supplied by the inferior hypophyseal artery, which ultimately branches off the internal carotid, and remains outside the dura (though within the cranium) until piercing the dura at the inferior aspect of the posterior pituitary, thus providing a theoretically protected blood supply to the posterior pituitary when other brain areas are subject to severe intracranial hypertension. If this were a satisfactory explanation, it would be irrelevant: The UDDA says nothing about blood supply, it says all brain function must cease. But it is not a satisfactory explanation anyway.

Osmoregulatory cell bodies are located in the diencephalon; their axons travel down the pituitary stalk and terminate in the posterior pituitary. Blood must reach cell bodies in the diencephalon to provide oxygen and glucose to sustain all parts of the cell (including the axon in the pituitary), and to provide physiologic information necessary for regulating vasopressin secretion. These cell bodies in the hypothalamus, and secondary osmoreceptors in circumventricular areas, as well as their respective arterial supply and venous drainage, are all intradural and not protected from intracranial hypertension (Nair-Collins and Joffe 2021).

Another response is to question whether the hypothalamus is part of the brain. Lewis, Bonnie, and Pope (2020), in discussing lawsuits surrounding brain death determination wrote, “this raised the question of whether the pituitary and hypothalamus are part of the ‘entire brain’” (Lewis, Bonnie, and Pope 2020, 143). They similarly wrote that perhaps the authors of the UDDA did not mean to include the hypothalamus in “all functions of the entire brain,” since the word “hypothalamus” is not found in the report (Lewis et al. 2019, 17)

There is no serious question as to whether the hypothalamus is part of the brain: it is. Furthermore, the UDDA authors did not specifically list the vast majority of brain areas or structures, but this does not mean that those areas are exempt from “all functions of the entire brain.”

An additional response is to allege that “not all cells of the brain must die” (Powner, Snyder, and Grenvik 1977). But nowhere have I stated that all cells of the brain must die; the neurons specifically responsible for osmoregulation continue to function, osmoregulation is a brain function, therefore a patient who does not have central DI does not meet the UDDA. The issue is not whether all cells have died, it is which cells have not died; namely, the ones responsible for performing an ongoing brain function.

The Incidence of Central Diabetes Insipidus in Patients Meeting AAN Criteria

A recent study (Varelas et al. 2023) challenged the claim that central DI occurs in up to half of patients meeting AAN criteria for brain death. This study examined the relationship between DI and kidney function and found, as would be expected, that as renal function worsens, the rate of apparent DI decreases. This is because decreased glomerular filtration rate (GFR) results in diminution of the amount of fluid delivered to the distal tubule, thus limiting renal water excretion independent of the action of vasopressin, and hence, renal dysfunction can potentially mask central DI if it is present by decreasing urine output, as discussed in earlier reviews (Nair-Collins, Northrup, and Olcese 2016; Nair-Collins and Joffe 2023). Examining only brain death patients with normal renal function (estimated GFR > 60 mL/min), these authors found that 78 of 101 patients (77%) exhibited central DI, that is, 23% had normal hypothalamic osmoregulatory function. They argued that this means that the true rate of DI in the general population of patients meeting AAN brain death criteria is closer to 23%, and not 50%.

As emphasized above, the key point is that a sizable number of people declared to be dead are not legally dead. It is hardly a satisfactory response to say, “we are not misdiagnosing death relative to the legal standard in 1 of 2 patients, we only misdiagnose death in 1 of 5.” This is not a rejoinder to the claim that many people are misdiagnosed as dead; it is confirmation of that claim.

Furthermore, one study is not sufficient to undermine the general conclusions from the larger review. Renal dysfunction could potentially mask central DI, but hypothalamic regulation of vasopressin secretion can also continue in the context of renal dysfunction; this cannot be ruled out, as the authors noted (Varelas et al. 2023, 7). The study does not show that the true number of people with loss of hypothalamic function is in fact higher than previously reported, it shows that renal dysfunction complicates the assessment of central DI, a point both well taken and previously made.

When these and other confounds are present, the most direct means of assessing hypothalamic osmoregulation is to measure plasma vasopressin, which Varelas and colleagues did not do. This stands to reason, because it was not a prospective study designed to assess the presence of hypothalamic function in patients meeting AAN criteria for brain death. This was a retrospective chart review from a clinical database of brain death cases, and evaluating hypothalamic function is not part of clinical practice. A few studies measured vasopressin as part of a prospective design. Arita et al. (1993) diagnosed DI using rigorous methods including plasma vasopressin levels and reported that 10 of 18 patients (56%) had central DI, 7 had normal osmoregulation (39%), and 1 (6%) had the syndrome of inappropriate secretion of anti-diuretic hormone. Sugimoto et al. (1992) found that 7 of 7 (100%) had DI based on vasopressin measurement. By contrast, Hohenegger et al. (1990) found that 11 of 11 patients had plasma vasopressin levels at normal or high levels, yet several had apparent DI. Adequate decrease in urine output to administration of exogenous vasopressin indicates central rather than nephrogenic DI, however in these patients, response to exogenous vasopressin was lacking. “Consequently central diabetes insipidus can be ruled out. A renal mechanism [nephrogenic DI] has to be assumed” (Hohenegger et al. 1990, 268).

This is a messy literature, and it is difficult to draw definitive conclusions about the whole population of patients meeting AAN criteria. Renal function is obviously the main contributor to urine output and its impairment is a complication in the assessment of osmoregulation. And there are other complications, such as the role of diuretics and hypertonic saline, administration of vasopressin for hypotension, nephrogenic rather than central DI, and others.

It would be a mistake to conclude that the findings from a single retrospective chart review override, rather than add to, the findings of 37 other similar studies (yielding a new total of 1343/2647, or 51%, with DI). All are subject to limitations, and confounds run in both directions (i.e., over- and underestimating central DI). Instead, it makes better sense to be clear that the review (Nair-Collins and Joffe 2021) offers a compilation of publications in which patients were diagnosed as brain dead, and were also reported to be diagnosed with DI or not. It is not a meta-analysis of prospective trials designed to test a hypothesis. We should not have high confidence in any specific estimate of DI in the general population of patients meeting AAN criteria for brain death. But we should also not be overly concerned with that specific estimate. Whether the incidence is 1 in 5 or 1 in 2, a substantial number of people declared dead are not legally dead. Considering the accumulated literature, the appropriate general descriptor remains “up to half.”

Finally, the lesson to be drawn from reflection on the role of decreased GFR in complicating interpretation of osmoregulation is the same as that from the possibility of pituitary washout: If the clinician cannot meaningfully account for these complications, then they serve as absolute confounds to brain death testing, precluding the determination of death under the UDDA.

Function and Activity

The next series of responses revolve around the concepts of function and activity. Some argue that osmoregulation is merely an activity, but not a function (Shemie et al. 2014). This claim is ad hoc and unfounded. The regulation of sodium and water in the extracellular fluid is one of the body’s main control systems, necessary for the function of every cell. Outside this very narrow debate, no one would suggest that “antidiuretic hormone release” is a mere activity with no functional role in human physiology, as distinct from a genuine “brain function” (Shemie et al. 2014, 791). Many authors cite the President’s Commission as having distinguished function from activity, while omitting the fact that the Commission succinctly but reasonably defined “function” as activity that is “organized and directed” (President’s Commission 1981, 75). The physiologic control system that regulates extracellular fluid osmolarity is without any doubt organized and directed, and it is without any doubt a vital physiologic function.

A related claim is that a function is clinically observable by definition, while phenomena measurable with technology are mere activities (Greer et al. 2020). The idea is that brainstem reflexes, responsiveness, and breathing are clinically observable and part of the brain death diagnostic process, whereas electrophysiologic activity measured by an electroencephalogram, or hormones measured in laboratory tests, do not measure functions. This, too, is ad hoc and unfounded. If the allegation that a function can only be observed clinically is to be taken seriously, then LFTs, short for “liver function tests” (measured in blood) tell us nothing about liver function; creatinine (measured in urine) tells us nothing about kidney function; echocardiograph measurements of ejection fraction tell us nothing about heart function; and so on. Everywhere in medicine, physicians appropriately use laboratory and imaging tests to assess organ function; this is beyond dispute. It is not credible to allege that these kinds of tests only reflect activities but not functions in this one, extremely narrow context, while acknowledging that laboratory and imaging tests provide useful information about organ function in every other context.

Accepting that osmoregulation occurs and is a function not a mere activity, the next step is to allege that it doesn’t matter, because this function does not count. It has been said that osmoregulation is not a clinical function, and only clinical functions count (Greer et al. 2020). The reasons why osmoregulation is allegedly a non-clinical function are the same as those mentioned above which claimed that it is not a function at all. It has also been said that osmoregulation is not a critical function, and only critical functions count (Bernat 2006).

The notion that only clinical functions count is as ad hoc as the claim that only clinically observable responses demonstrate a function, rather than mere activity. Furthermore, this relies on the prior and manifestly false claim that the diagnosis of brain death is “a clinical diagnosis” (Greer et al. 2023). It is not. An example of a clinical diagnosis would be bipolar disorder, or eczema; diagnosis of these conditions is entirely a clinical interpretation based on history, physical examination, or other clinical observation. With respect to its diagnosis, brain death is nothing like these disorders. The cause of coma must be established and confounds considered, which always includes at least one brain imaging scan and laboratory tests of kidney, liver, and endocrine function, acid-base balance, electrolytes, and sometimes toxicology screen. The apnea test requires blood gas analysis, a laboratory test. Like every other diagnosis in a modern ICU, brain death is far from a “clinical diagnosis.”

The “clinical functions” rejoinder is also subject to the same basic response as the “critical functions” claim: The UDDA makes no distinction among types of brain functions, just as it makes no distinctions by blood supply. It is irrelevant whether brain functions are classified post hoc as clinical, critical, or otherwise. The UDDA is clear and straightforward: all brain functions must cease. I am concerned here to assess whether people who do not meet the legal standard for death in the U.S. should be declared dead in the U.S. Any proposals that change the subject are not relevant. In this regard, distinguishing critical from “non-critical” brain function is irrelevant; differentiating which brain functions are “integrative of the organism as a whole” (Bernat 2006) is irrelevant, as is determining which functions are “essential clinical brain function[s]” (Greer, Lewis, and Kirschen 2024); the effort to redefine the legal standard for death in terms of the “brain-as-a-whole” standard (Bernat and Dalle Ave 2019), “neurorespiratory criteria” (Omelianchuk et al. 2022), or “permanent coma with loss of brainstem function and inability to breathe” (Greer et al. 2023) are all irrelevant. Whether a version of a higher-brain criterion (Lizza 2006) might be justified is not relevant. The idea that death is a social construction (Magnus 2018) is not relevant, since the legal definition of death most certainly is a social construction, and that is precisely what we are examining—not any proposal for “socially constructing” death as distinct from the UDDA. Nor is it relevant that other jurisdictions around the world appeal to a brainstem standard, or loss of capacity for consciousness and breathing. One might disagree with the UDDA, or prefer that it had been revised, but my focus here is precisely on the actual UDDA as it stands.

Some might also allege that determining the functional status of the osmoregulatory system is too difficult, that it is demanding a degree of epistemic certainty in the diagnosis of brain death that does not apply in any other diagnosis (Martin, Forlini, and Tumilty 2023). But death is different from any other diagnosis, and there should be as close to zero false positives as possible. This is not a contentious claim. Furthermore, assessing the patient’s endocrine and electrolyte status is already part of the diagnosis, because it is a required component of initial assessment and consideration of confounds. In fact, in any assessment for brain death in the U.S., the variables relevant for considering DI will have already been collected; nothing additional needs to be measured. The issue is not that this is a challenging diagnosis; rather, a choice is made not to consider it. There are certainly confounds, and physicians are quite competent to assess hypothalamic osmoregulation in light of them. There might be some cases in which confounds render an assessment of osmoregulation equivocal. In that case, vasopressin could be directly measured to resolve the uncertainty. If that is not possible, then confounds to brain death testing are not ruled out and the patient cannot be declared legally dead under the UDDA.

Law and Medicine

Another response holds that brain death is defined in terms of the AAN tests, not the UDDA: it is a syndrome characterized by unresponsiveness, brainstem areflexia, and apnea, in the context of a known cause of coma with confounds ruled out (Wijdicks 2010). This yet again changes the subject, and therefore is irrelevant. The AAN does not determine law, and neither does medical consensus.

Part of the underlying motivation for the above response is a belief that professional medical societies have a certain legal authority they do not, namely, to change the legal definition of death. Accordingly, some argue that the UDDA grants authority to physicians to define what death is by deferring to medicine the specific tests used to determine it. The UDDA states, “The determination of death must be made in accordance with accepted medical standards." On this point, one reviewer asserted “This statement was included in the UDDA to make the statute absolutely clear that physicians alone possessed the authority to determine the medical standards for death determination. Therefore, the UDDA’s authorization of the medical professional community to determine the standards for brain death determination can include whether hypothalamic neurosecretion counts [as] a relevant factor.”

This is a common but serious misconception. There is a difference between the diagnostic tests used to assess whether a person meets a given legal standard, and the standard itself. The statement in the UDDA refers to the tests used to determine whether the condition, irreversible cessation of all brain function, has occurred or not. The President’s Commission stated this explicitly, remarking that the meaning and intent behind “accepted medical standards” was:

to require the use of diagnostic measures and procedures that have passed the normal test of scrutiny and adoption by the biomedical community… The statutory language should eliminate wholly idiosyncratic standards or the use of experimental means of diagnosis (except in conjunction with adequate customary procedures) (President’s Commission 1981, 78-79; emphasis added).

Not only was the President’s Commission explicit that “accepted medical standards” refers to diagnostic measures and procedures, not the legal standard itself, but furthermore, if deference on diagnostic tests entailed authority to change the criteria to be identified by those tests, then the first sentence of the UDDA, which provides the legal standard for death, would be moot. Indeed, the entire Act would be self-defeating, since the purpose of the Act is to provide a legal standard for determination of death. If medical societies could simply change the legal standard, then what is the point of specifying a particular standard in the statute? Why not just write, “a patient is dead when they are determined to be dead according to accepted medical standards”? But the statute does not say that, the first sentence is not moot, and the UDDA is not self-defeating.

Further still, it makes no sense to appeal to the UDDA as the legal foundation for the determination of death by neurologic criteria (e.g., Greer et al. 2023, 1113) while at the same time asserting that the medical community has authority to change or not follow the legal standard for death explicitly articulated in that very same law.

In short, the UDDA does not authorize professional medical societies to modify or determine the legal standard for death, nor does it grant medical societies or individual physicians authority to declare someone legally dead if they do not meet the legal standard for death. The UDDA only authorizes the medical community to develop medical standards, that is, “diagnostic measures and procedures that have passed the normal test of scrutiny and adoption… [in order to] eliminate wholly idiosyncratic [diagnostic] standards” (President’s Commission 1981, 78). Medical standards are diagnostic tests and procedures; they are not on par with, exchangeable for, nor superordinate to, the legal standard explicitly articulated in the UDDA.

As a legal boundary that separates the civil statuses of living legal person from deceased person, it is the law, not the AAN or any other medical society, that defines the legal standard for death. The physician’s role is to evaluate a patient and certify whether they do or do not fulfill the legal standard, and that is the extent of medical authority with respect to the law and death.

Although this essay is focused on the determination of death by neurologic criteria, the UDDA provides a disjunctive definition of the legal standard for death, with the first disjunct defining an individual as dead upon “irreversible cessation of circulatory and respiratory functions.” This standard is used to declare an organ donor to be dead in a “non-heart-beating” or donation after circulatory death (DCD) protocol. In this circumstance, the patient is declared dead 2-5 minutes after cardiac arrest. The purpose of the wait-time is to ensure that the heart will not restart spontaneously, known as autoresuscitation. So long as autoresuscitation will not occur, and the clinicians standing by will not attempt to resuscitate the patient (among other reasons, because to do so would violate the patient’s valid DNR), then the cessation of function will not reverse; cessation of function is permanent. This “permanent” cessation of circulatory and respiratory function is judged to satisfy the UDDA’s standard of “irreversible cessation” because the word “irreversible” has come to be interpreted to mean “permanent.”

Some might argue that this is analogous to considering neuroendocrine function to be consistent with brain death. In both cases, medical practice does not actually comport with the law as written: In some cases (not all), return of spontaneous circulation can be achieved after 5 minutes of cardiac arrest. In other words, the cessation of circulatory and respiratory function is not known to be irreversible at 5 minutes of cardiac arrest. But medical professionals interpret “irreversible” to mean “permanent.” By analogy, the law requires absence of all brain function to determine death by neurologic criteria, but medical professionals interpret preserved neuroendocrine function to be consistent with “irreversible cessation of all functions of the entire brain.” Thus, if it is justified to reinterpret the first component of the UDDA from irreversible to permanent, then why insist on strict adherence with respect to neurologic criteria?

The problem with this argument is that it is not justified to reinterpret—that is, change—the meaning of the first part of the UDDA any more than it is justified to change the second. “Permanent” and “irreversible” are neither synonymous nor coextensive, and a patient whose circulatory and respiratory functioning has ceased, but not yet irreversibly, is plainly not dead under the UDDA.

Notably, this does not imply that every patient must be subjected to CPR prior to declaring them dead in order to assess whether function is irreversible or not. If CPR is not appropriate for whatever reason, most commonly because a valid DNR is in place, then it is not appropriate and should not be done. Whether the patient is, at that very moment, in a condition in which circulatory and respiratory functions have ceased but are reversible (i.e., dying but not yet dead) or has passed into irreversibility (death), they will pass into irreversibility in a few minutes, and death should be legally declared in the medical record in a few minutes. There was never a need to rush the declaration of death until it became relevant to DCD.

Furthermore, there is an important disanalogy between the issues of irreversibility and permanence in determination of death by circulatory criteria, and the preservation of neuroendocrine function in determination of death by neurologic criteria. The issue with irreversibility and permanence is that at a given time, we do not know whether the patient is in a state of irreversible loss of function. But the opposite holds in brain death: We do know that they do not meet the legal standard for death, because some brain function is preserved, and this is readily apparent through urine output and common laboratory values that will have been measured in every patient anyway (e.g., electrolytes, plasma and urine osmolarity). Uncertainty about reversibility is one issue, and only applies specifically to the DCD setting, not the normal clinical setting. But declaring someone legally dead when they manifestly do not meet the legal standard for death is a very different issue.

Ultimately, DCD and questions about irreversibility and permanence are different topics than the question of this essay. If DCD is ethically and conceptually unimpeachable, or if it is ethically and conceptually unsalvageable, neither circumstance bears on the question of whether it is permissible to declare someone legally dead by neurologic criteria when they manifestly do not meet the legal standard for death. These issues constitute another red herring.

Finally, some might simply reply, “but the AAN, and the World Brain Death Project, and other authoritative medical bodies have explicitly stated that neuroendocrine function is consistent with death.” They have indeed.

To conclude this section, there are no good responses. The UDDA requires no brain function, yet some people declared dead have some brain function. The conclusion is rock solid and unavoidable: Death is routinely misdiagnosed relative to the legal standard, where legally living people are declared dead, and not by accident but by standard practice, following accepted guidelines.

But why? A general intransigence is not characteristic of medicine, so why not simply update diagnostic guidelines in light of evidence? Clearly, there are deeper issues here.

DEEPER REASONS

Heart-beating donors are the primary source of transplantable thoracic and visceral organs. If the UDDA were followed, the number of patients declared dead by neurologic criteria would decrease substantially, potentially by half, and the number of potential heart-beating donors would decrease. (Many patients would presumably choose to donate under a DCD protocol in the envisioned circumstance. But I will simply assume that all erstwhile heart-beating donors would not be organ donors at all, if the determination of death by neurologic criteria proceeded in accordance with law). Is accurate determination of death under “a strict interpretation” of the UDDA really worth this cost?

This rhetorical question relies on two underlying premises. First, the dead donor rule (DDR), which prohibits causing death by organ donation, must be followed. Second, the DDR is only to be followed nominally but not substantively; it does not really need to be followed if doing so causes outcomes we are not prepared to accept. People without central DI are legally alive, and they constitute up to half of the population of patients declared dead by neurologic criteria. If we wish to remove their organs through heart-beating organ procurement anyway, then we are advocating abandoning the DDR, but surreptitiously, using inaccurate death determination as a workaround because, as this objection goes, “is accurate determination of death really that important?.”

The DDR is a deontic constraint on medical practice. It does not facilitate organ donation; it does the opposite. If we support adhering to the DDR, then we have to pay its price. The price is that legally living people cannot be heart-beating donors, and that means fewer organs. If, on the other hand, we find the outcome of accurately determining death to be unacceptable—namely, fewer organs available for heart-beating procurement—then it is with the DDR that we have a quarrel. Simply put: If we want organs from people who are not legally dead, then we must advocate for abandoning the DDR. Otherwise, we must pay its price.

The foundation of the organ transplantation system is public trust: since organ donors opt-in, proactively or through surrogate consent, anything that calls into question the reliability of brain death diagnosis can threaten the transplantation enterprise. And I am certainly calling into question the reliability of brain death diagnosis. This concern about trust generates another motivation to deny the obvious fact that some people declared dead are legally alive.

This also relies on two premises: First, that trust in brain death diagnosis and organ transplantation is of paramount importance, and second, that the diagnosis of brain death and the transplant enterprise being worthy of trust are not important, if acting in ways deserving of trust would lower organ donation rates. There is a clear legal standard for death in the U.S. Whether it is a good standard or not, it is our governing legal standard. Choosing to routinely declare people who do not meet that standard to be dead anyway, regardless of the reason, is not compatible with trustworthiness.

Other concerns surround medical futility, just use of resources, and the role of consciousness, personhood, or quality of life. The patient with profound neurological injury, who is comatose, brainstem areflexic, and apneic, yet retains hypothalamic osmoregulation, has a dismal prognosis. For some, cardiovascular collapse will occur in a week or two despite continued treatment. Others will survive this acute phase and stabilize, able to be discharged to a long-term care facility or to home on long-term ventilation. There will almost certainly be permanent profound neurologic injury, and the likelihood of return to normal conscious functioning approaches zero. This is a paradigm case of “medical futility,” both in its quantitative and qualitative senses. It would seem that further treatment is futile; and therefore, it should be stopped regardless of the family’s wishes.

Though I do not endorse this role for futility judgments, here I assume this analysis is adequate. This would show that treatment ought to be stopped on grounds of medical futility. It would not show that it would be justified to falsely certify that the patient meets a clear legal standard they do not meet. Consider any paradigm case of futility, such as a patient with multiple organ failure, intractable hypotension, and sepsis. It may be the case that treatment should be stopped over objection, but it does not follow that it would be permissible to inaccurately certify that they meet a legal standard for death they do not meet. The same is true of the patient with unresponsiveness, brainstem areflexia, apnea, and intact hypothalamic function. Finally, if treatment withdrawal over objection is justified on grounds of futility, that case needs to be made on its own grounds, not by falsely certifying a living patient to be dead.

It is common to express concerns about distributive justice, where treating patients with profound brain injury is considered a tremendous, and unjust, use of resources. But distributive justice is a different issue from whether it is permissible to misdiagnose death. In a situation of severe resource shortage, it can be just to deny a patient access to an ICU bed, or a ventilator, or downgrade a patient from the ICU to a floor, and perhaps even to remove a ventilator, and the considerations on which these decisions should be made (efficacy, prognosis, resource intensity) apply to everyone. But that has nothing to do with the permissibility of certifying that someone meets UDDA criteria when they do not.

It is also important to temper concerns about the resource impact of accurately diagnosing death with a dose of realism. Brain death is rare. In the U.S. it accounts for 1-2% of deaths (Seifi, Lacci, and Godoy 2020). About half (let us assume) do not meet the legal standard and would not be declared dead. But most people do not want continued treatment in this kind of condition, whether the hypothalamus functions or not. Of those who do want extended treatment, some will not long survive even with treatment. The process of increasing intracranial pressure itself causes systemic complications which can damage several organ systems. While it is not true that every patient who meets AAN diagnostic guidelines will suffer cardiovascular collapse, many will despite treatment, decreasing the number of patients by another factor. Thus, were death no longer misdiagnosed the additional patients we might expect would be a percentage of a percentage of a percentage of a rare incidence to begin with. (And if we include consideration of DCD, those who donate through this route would decrease the total by another factor.) It is safe to assume that the viability (or not) of the U.S. hospital system does not turn on misdiagnosing brain death. More importantly, the just use of resources is a very different question from whether it is permissible to misdiagnose death.

Hypothalamic osmoregulation, on its own, makes no difference to consciousness. A patient who is otherwise irreversibly and completely unconscious would not become conscious if osmoregulation were restored. Thus, one might think, what difference does this make? The ability to concentrate urine in a comatose body is not what patients or families care about, and it is not what physicians aim to preserve or restore. As a result of loss of the capacity for consciousness, many would argue that the person is gone; their moral status is lost. They have no quality of life and never will again. With all that said, why should anyone care about vasopressin?

Each of the above comments changes the subject, yet again. I ask a simple question: Is it permissible to declare a patient dead in the U.S. if they do not meet legal criteria for death in the U.S.? I do not contest the fact that patients meeting AAN diagnostic guidelines for brain death are unconscious, and I think it reasonable to suppose that the majority of people in the U.S. would consider life in the absence of consciousness to be not worth living and would want treatment withdrawn. But that is relevant only insofar as it tells us something about the number of people wanting continued treatment, namely, a small minority. It’s got nothing to do with the UDDA, which does not define death in terms of loss of the capacity for consciousness and more importantly, it’s got nothing to do with the question of this essay: Is it permissible to routinely misdiagnose death according to the legal standard?

THE FINAL RESPONSE

Once we dig deep enough, through the superficial layers of obfuscation and the underlying motivations which are understandable but do not justify misdiagnosing death according to the law, we arrive at what I suspect is the core source of ambivalence about this issue for many people. Namely: why does it matter? This is the way brain death has been determined for a half-century, most people truly don’t care about vasopressin in the context of irreversible coma, and it would be disruptive to change. Wouldn’t it be better to simply keep things the way they are, and perhaps continue efforts to get the law changed, either by another round of ULC committees or pursuing federal legislation?

At least this response does not attempt to deny the obvious, nor wander off into irrelevancies. It simply says, “you are right, but this issue is not important enough to change death determination practices.”

This reply presumes that the sentence “people who are legally alive should not be declared legally dead” is not already self-justifying. Instead, it presumes that some other reason is required to justify the demand for accurate death determination practices from physicians. I do not accept this idea, and no one else should, either. The medical profession’s role in determining death is to certify whether an individual meets the legal standard or not; medicine has no further authority to de facto change the law through practice guidelines. It is right to demand that physicians accurately determine death according to the legal standard, without conceding that any other reason is needed.

Nevertheless, there is another, very important, reason. A critically ill patient lacking decision-making capacity has certain entitlements that a dead body does not. The individual is considered a legal person (conscious or not), and not a corpse. Being a living patient reflects a particular civil status, which is not dependent on judgments of quality of life, nor assessment of their moral status or ontological personhood status. The Patient Self-Determination Act, the Uniform Health Care Decisions Act, requirements of accreditation by the Joint Commission on Accreditation of Healthcare Organizations, and relevant state laws all apply. Broadly speaking, the aforementioned affirm that the individual retains a legal right, even when incompetent, to direct medical decision-making through advance directives or surrogates. There are also standard practices in every hospital for resolving disputes about treatment, often revolving around medical futility. This usually takes the form of a clinical ethics consultation or appeal to an ethics committee and might also include an attempt to transfer the patient to another physician or hospital. Even when treatment is removed over objection, there is due process. The patient has a right to a patient advocate or legal representation. Insurance companies continue to provide coverage for a living patient.

All the above are nullified when the patient is declared dead. A physician certifies that a particular civil status no longer applies, because the patient (allegedly) meets a legal standard distinguishing one civil status from another. Declaring the patient dead is a performative utterance, legally transforming the patient’s legal civil status from person with rights to corpse with a different status. If the patient is inaccurately certified to meet the legal standard that separates these two civil statuses, they are unjustly denied access to all the rights associated with being a living legal person. This is a violation of civil rights, namely, the patient’s right of self-determination through advance directives or a surrogate; the right to be treated similarly to other patients with respect to standard end-of-life practices and due process for treatment disputes; access to rights enshrined in the Patient Self-Determination Act and others mentioned above; and in short, the right to equal protection under the law.

There is an ongoing debate about whether people should have the right to opt out of being determined dead by neurologic criteria, but that debate has nothing to do with people who do not meet legal criteria for death. There is no sense in asking whether people have a right to “opt out” of a legal standard they do not meet, nor should there be any question as to whether one legally living patient has the same rights as any other living patient. To falsely declare a person legally dead is to violate their civil rights and in particular their right to equal treatment under the law.

CONCLUSION

There have been vigorous efforts to revise the UDDA and, for better or worse, those efforts failed. The UDDA is not changing in the foreseeable future. It has long been common practice to declare some patients dead by neurologic criteria even though they do not meet the legal standard for death, and the AAN has made clear that they have no intention of changing their guidelines. Thus, legally living people will continue to be declared dead, not because of a mistake, but because of a choice.

The decision to continue misdiagnosing death according to the law will create routine violations of civil rights, will continue to violate the DDR that allegedly is such an important red line for organ transplantation, and will contribute to a well-deserved mistrust in the determination of death. Most importantly, the demand that physicians accurately determine death according to the legal standard is rightful and transparently correct, with no further justification needed.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.