In recent decades, there has been a considerable amount of scholarship surrounding the different kinds of compromised neurological states that a person may encounter following an acute severe brain injury (SBI) (see Box 1). Footnote 1,Footnote 2,Footnote 3,Footnote 4,Footnote 5,Footnote 6 More recently, Joseph T. Giacino et al. Footnote 7 ) have argued that the catalogue of such neurological states ought to be expanded to include the minimally conscious state (MCS), in which a patient exhibits some indications of consciousness. There has been research on whether such a state exists, Footnote 8 how it should be diagnosed, Footnote 9 and what treatment options may be available for individuals in MCS. Footnote 10 In this article, we approach MCS and its diagnosis from an adult clinical ethics perspective. We argue that the inclusion of MCS in the ontology of neurological states that result from an SBI in adults carries with it considerable challenges for beneficence, patient autonomy, and stewardship of resources. Consequently, we argue that the neurological community’s focus on MCS may prove to be a disservice to the very individuals who are most affected.
Box 1: Neurological states after SBI
Coma: state of unconsciousness in which a person cannot be awakened; fails to respond normally to painful stimuli, light, or sound; lacks a normal wake-sleep cycle; and does not initiate voluntary actions
Vegetative State (VS): a state in which a person is undergoing wake–sleep cycles without detectable awareness; wakefulness without awareness
Persistent VS: a VS that persists for at least 4 weeks
Permanent VS: a VS that persists for 3 months post-nontraumatic SBI or 1 year after traumatic SBI
MCS: a state of severely altered consciousness in which minimal but definite behavioral evidence of consciousness is demonstrated
To illustrate, we consider a hypothetical case like those frequently encountered in an intensive care unit (ICU) (Box 2).
Box 2: Hypothetical case
A 21-year-old male sustained a traumatic brain injury (TBI) as a result of an unrestrained motor vehicle accident. He was intubated on the scene with a Glasgow Coma Scale (GCS) score of 3, with a fixed and dilated right pupil, necessitating emergent transport to the nearest trauma center. There, a head CT scan revealed multiple calvarial fractures with underlying subdural and intraparenchymal hemorrhages with midline shift. He underwent emergent decompressive hemicraniectomy, and 1 week following his accident he remains fully supported with mechanical ventilation and artificial nutrition via a nasogastric tube.
He has intact brain stem function, intermittently blinks, does not follow commands, and has failed two attempts at extubation, suggesting need for tracheostomy. The team could perform a tracheostomy and percutaneous endoscopic gastrostomy (PEG) tube placement for long-term nutrition; however, they remain concerned he will not make a meaningful recovery. Prior to meeting with his parents, the team discussed options moving forward and believe he has the potential for minimal recovery over the ensuing months to years; that is, potentially limited ability to interact with his environment, but reliant for all activities of daily living (ADLs).
The team presents the patient’s extremely poor prognosis to the parents and expresses its support for forgoing tracheostomy and PEG placement and transitioning to comfort care. Naturally, the family is devastated by the news and expresses concern as to whether the patient would find the prospect of his condition acceptable. However, adamant that their son will recover, they research injuries like their son’s and read about patients in MCS, with some sources stating there is potential for even further recovery in the first few years.
Motivated by the promise of recovery, they reject recommendations and insist on tracheostomy and PEG tube placement. Following stabilization from surgery the patient is transferred to a long-term facility. In the years following, he remains dependent on mechanical ventilation and artificial nutrition, intermittently blinks, but is not interactive with his environment.
In this example, the family chooses for their son to receive ongoing treatment because they hope that he will make a meaningful recovery. Although the critical care team paints a bleak picture and the family doubts that their son would want to live a life dependent on others to perform ADLs or of being incapable of having meaningful interactions with his environment, they cling to the idea that he may actually be in, or transition into, MCS, and need only be coaxed out of his disordered consciousness.
Before we examine the different ethical concerns raised by this example, we provide a brief overview of MCS and consider the epistemological and phenomenological challenges inherent in its diagnosis.
Background
Because of advancements in intensive care technology in the mid-twentieth century, patients who sustain an acute SBI (etiologies include, but are not limited to, traumatic, anoxic, or hemorrhagic) and previously would have died almost immediately from apnea, are surviving beyond the critical stages of their injuries. Footnote 11 Once such patients emerge from coma, they often appear to be in a state in which they show no signs of awareness of self, others, or environment, and yet will open their eyes in accordance with sleep–wake cycles.
In 1972, neurosurgeon Bryan Jennett and neurologist Fred Plum first codified this state as the vegetative state (VS), describing it as “wakefuleness without awareness” Footnote 12 or “wakeful unresponsiveness.” Footnote 13 This state is characterized by the absence of consciousness due to a loss of cerebral functioning. However, unlike a coma, it is also characterized by the presence of motor functions or reflexes that are the result of a functioning brain stem Footnote 14 (see Box 1). A determination of lack of consciousness is made based on behavioral evidence, such as whether the individual lacks responsiveness to the internal or external environment or whether that individual fails to engage in purposive action. Footnote 15 Indicators of consciousness do not include reflexive responses, such as withdrawals from noxious stimuli or brief orienting to sound, as these are controlled by the brain stem and are present regardless of consciousness. From a prognostic standpoint, a VS is considered permanent if it persists for 3 months after an anoxic brain injury or 12 months after a traumatic brain injury (TBI). Footnote 16 At this point, a patient is not expected to emerge from a VS.
By the 1990s, the framework of VS had been well established. However, clinicians were making observations in vegetative patients that did not fit with the prevailing diagnostic criteria. Intermittently, they were observing what appeared to be simple command-following or sustained fixation on a stimulus. Footnote 17 This led to the delineation of MCS as a new neurological state in 2002, which in turn set off what we take to be a problematic degree of optimism with respect to neurological recovery. Footnote 18 Giacino et al. Footnote 19 sought to differentiate MCS from VS (as well as coma or locked-in state), and they defined MCS as a state in which consciousness is partially preserved, and in which “limited but clearly discernible evidence of self or environmental awareness” is demonstrable on “a reproducible or sustained” basis. Footnote 20 Moreover, as defined, this awareness must be demonstrated through one of four modes: simple command following, gestural or verbal responses, intelligible vocalizations, or purposeful behavior (e.g., pursuit of eye movement or sustained fixation on salient stimuli). Footnote 21
Diagnostic Dilemmas
The diagnostic criteria for MCS are explicitly behavioral. A person in MCS must be able to reproduce or sustain relevant behaviors. Therefore, according to the diagnostic criteria, an individual exhibiting no behavioral indication of awareness is not in MCS. Contrary to the diagnostic criteria, however, researchers soon turned to brain imaging technology to determine MCS, even for those who do not present the clinical criteria. Footnote 22 The thought was that perhaps imaging could detect something that the clinical examination had missed. This, however, had the effect of widening the de facto definition of MCS, and opening up the diagnosis to the epistemic issues inherent in imaging technology. In the case of MCS, there is no independent reason to assume that imaging technology is more revealing of consciousness than clinical examination. Such a belief may be indicative of a certain “reductive epistemology,” in which one kind of evidence is privileged over another in virtue of being closer to the physical manifestation of a phenomenon. In this case, brain imaging technology may be considered more reliable because it looks more directly at the physical object (the brain) thought to be responsible for consciousness. However, it is entirely possible that behavioral indicators are more reliably correlated with consciousness than imaged activation patterns.
We thus look at actual imaging technologies employed for the detection of consciousness. One study used positron emission topography (PET) to measure the cortical response of patients in VS and MCS, and of control subjects, on exposure to auditory clicks. Footnote 23 Melanie Boly et al. Footnote 24 found that contrary to patients in VS, patients in MCS and control subjects exhibited activation in the superior temporal gyri (Wernicke’s area) that is involved in the understanding of spoken language. They conclude that the cerebral activity observed in patients in MCS is more likely to lead to the kind of integrative processes of auditory information required for consciousness than that in patients in VS. Footnote 25 However, as Boly et al. Footnote 26 themselves acknowledge, although their imaging study suggests that patients in MCS may have activation in Wernicke’s area upon auditory clicks, this proves neither that such patients have phenomenal awareness of such stimulation, nor that such auditory clicks can be coalesced into something meaningful.
Multiple researchers have used functional magnetic resonance imaging (fMRI) to show that MCS patients exhibit functional activation profiles like those of healthy individuals upon hearing command-following prompts. Footnote 27,Footnote 28,Footnote 29 For example, Adrian M. Owen et al. Footnote 30 detail how they employed fMRI to demonstrate that a young woman, presumed to be in VS, displayed cortical activation in the same regions as controls when prompted to imagine herself playing tennis or walking through the rooms of her house. This was taken as evidence that the patient “understands spoken commands and can respond to them through her brain activity.” Footnote 31 Owen et al. Footnote 32 also conclude that “her decision to cooperate with the authors by imagining particular tasks when asked to do so represents a clear act of intention, which confirmed beyond any doubt that she was consciously aware of herself and her surroundings.” However, this inference does not follow. For one, the reliability of fMRI has recently been called into question. Anders Eklund et al. Footnote 33 have shown that common fMRI software packages returned an alarmingly high rate of false positives. Even if the imaging accurately represents activated brain regions, there is no way to infer from this that the patient “intentionally decided to cooperate” with the prompts.
Moreover, a technology is of limited diagnostic use if it is unable to rule out the condition in question. Owen et al. Footnote 34 argue that negative fMRI findings in patients should not be taken as evidence for lack of awareness, because false negatives are common. Owen and Coleman Footnote 35 suggest, for example, that false negatives may be the result of the patient not properly hearing or understanding the task instructions or the patient being asleep during the examination. But this means that not only can the presence of consciousness never be ruled out, but that the high probability of false positives also suggests that consciousness is often mistakenly inferred in patients where it is not present. These epistemic problems associated with the detection of consciousness through imaging technology problematically factor into ethical issues surrounding decisionmaking.
Phenomenological Issues
One vexing problem associated with MCS is that we can know only very little about what it is like to be in MCS. In the literature on VS and MCS, consciousness most often is identified as “the awareness of self and environment;” Footnote 36 however, it is not clear that “awareness of self and environment” adequately captures what we mean by consciousness. There are several conceptions of consciousness that may be implicated in MCS. Ethical decisionmaking for patients in this state warrants their delineation. If we imagine that consciousness comes in degrees, we may think of sentience as its most rudimentary form. A creature may be sentient if it experiences some basic sensation; for example, heat, cold, or pain. Undoubtedly many nonhuman animals possess (at least) this level of consciousness. More encompassing is phenomenal consciousness, by which we mean the qualitative “what-it-is-like-ness” of experiences; for example, the tartness of an apple, the whiteness of a wall, or the panginess of hunger. However, a state can also be conscious in the sense that its content is accessible to the mind for cognition, even if it lacks phenomenal properties. Ned Block Footnote 37 calls this access consciousness. This may be the kind of consciousness present in blindsight, in which individuals clearly have access to visual information and yet report that they have no phenomenal experience. Footnote 38 It is also not at all clear that complex conscious thought can do without conceptual content, meaning that there must be some ability for linguistic processing. Footnote 39,Footnote 40 Furthermore, for individuals to be self-conscious in a robust sense, they must have an explicit concept of self. We may even think that narrative consciousness is required for sense of being-over-time, in which individuals have a sense of self that extends to the past and future, and is related in appropriate ways to the present. Footnote 41 Consciousness is, therefore, not a simple concept, and consequently it is not immediately obvious what it means experientially to be in MCS.
Ethical Issues
Having laid out the epistemological issues with the MCS diagnosis, as well as the inherent difficulties of determining the phenomenological nature of being in MCS, we now turn to the ethical issues associated with decisionmaking in the critical care of affected patients.
Beneficence
Having a loved one sustain a SBI that leads to massive neurological damage and catastrophic disability is devastating. Traumatized families are forced to process and understand medically laden jargon and confusing statistics about risks and outcomes, making treatment decisions unbelievably difficult. Therefore, it is of the utmost importance that the medical field provide affected families with the most honest and helpful information and guidance possible to facilitate decisions that are most ethically appropriate.
Here we see that the questions surrounding phenomenology in MCS are not merely academic. In conjunction with multiple societies, the Society of Critical Care Medicine (SCCM) states that an ICU intervention should be considered inappropriate “when there is no reasonable expectation that the patient’s neurologic function will improve sufficiently to allow the patient to perceive the benefits of treatment.” Footnote 42 This requires not only that an individual have some degree of sentience or phenomenal consciousness; that is, that there is some perception occurring, but also that that individual be capable of conceiving of that perception as beneficial, which Alexander Kon et al. explicitly state. Footnote 43 Therefore, this kind of cognition requires, at minimum, some rudimentary, probably conceptual, capacity to grasp the idea of “beneficial,” along with the ability to predicate that property of a given conscious state. This implies access consciousness. This also means that an intervention for an individual in MCS that results in some phenomenal experience; for example, a visual or tactile perception, but for whom this experience is not accompanied by a recognition of its beneficialness, is ineffective, according to the SCCM guidelines, and, therefore, should not be provided. A fortiori, any intervention that may result in a perception that the individual experiences as being wholly burdensome is unwarranted.
When initially caring for a patient with a devastating neurologic injury, intensivists recognize there are variable phenotypic manifestations and clinical trajectories based on the age of the patient, specific etiology, location of injury, and level of neurodegeneration post-injury that affect the patient’s potential for recovery. Footnote 44 There are prognostic models to guide intensivists for some forms of SBI (e.g. TBI), Footnote 45 with other etiologies less delineated. However, although it is true that intensivists cannot know with certainty what the outcome for a particular patient will be, this does not preclude them from making reasonable predictions based on a combination of indicators, algorithms, and experience, which allows them to prognosticate the patient’s likely trajectory. The codification of MCS has contributed to reifying a neurological category for which the literature focusses on prospects of “recovering” to MCS or “emerging” from MCS. However, what is less emphasized is that regardless of underlying neurological condition (i.e. VS or MCS), the vast majority of patients are left with severe neurologic disabilities.
When a patient is admitted to an ICU with an SBI, extensive invasive critical care interventions are performed to ensure oxygenation, adequate circulation, and reduction in intracranial pressure. Usually this involves intubation, placement of central intravenous catheters, and craniotomy or extraventricular drain placement to relieve and monitor intracranial pressure. Generally, patients stabilize within days to weeks (if there is not progression to brain death) and the immediate extent of the neurological damage becomes apparent. At this point, the family has to decide whether to withdraw life-sustaining treatments (LST) and keep their loved one comfortable until they pass away, or whether to transfer their loved one to a long-term care facility in the hopes that he or she makes a recovery that he or she would consider meaningful.
Certain clinical presentations (i.e., global hypoxic ischemic encephalopathy) clearly indicate the high likelihood of a poor outcome, in which the patient would not make a meaningful recovery and would either remain in VS or MCS over the long term, or would live with severe disability. This, coupled with the difficulties in diagnosing MCS, means that the prospect of any benefit from long-term treatment is highly uncertain. In addition, we must consider the burdens imposed by the continued treatment of patients in prolonged states of VS or MCS. Beyond the invasive interventions in the acute stabilization phase, a prolonged period in VS or MCS brings with it considerable hardships and risks. Patients may be dependent on a ventilator, requiring a tracheostomy, for the remainder of their lives (if unable to breathe on their own), which can precipitate lung trauma caused by mechanical ventilation, increased risk for pneumothorax, and ventilator-associated pneumonias (VAPs), in turn worsening antimicrobial resistance. All patients, even those able to breathe without a ventilator, require the placement of a percutaneous endoscopic gastrostomy (PEG) tube. Even patients who make some recovery most often depend on extensive nursing support with ADLs, including attention to nutrition, bowel and bladder care, and skin care to prevent breakdown, along with positioning and range of motion exercises to avoid contractures. Footnote 46
In addition to these physiological burdens, a patient in MCS may experience considerable psychic burdens. If patients are sentient, they may experience pain, both from the original insult and from the continued treatments they undergo. Nearly all medical professionals surveyed in a study by Athena Demertzi et al. believed patients in MCS feel pain. Footnote 47 (Surprisingly, half to two thirds of medical professionals also thought that patients in VS experienced pain. Footnote 48 ) Additionally, Boly et al. Footnote 49 demonstrated that noxious stimulation activated areas in the brain that were identical to those of controls and are thought to make up the pain matrix, suggesting the capacity for pain perception in patients in MCS. If patients in MCS retain some degree of self-consciousness and access consciousness, they may be able to piece together some inkling of their condition. It is hard to know what kind of psychic suffering may accompany such a state, but it is likely that a recognition of one’s condition in this case would be devastating to most. For someone with partial consciousness, we can only imagine that such a state may very well be terrifying and confusing. Perhaps patients in this condition have no capacity for narrative consciousness, unable to carry forth their sense of self through time, and consequently are horrified anew every time they find themselves in this situation. If they have only a tentative grasp of conceptual content, they may not be able to coalesce sensations and emotions into meaningful thoughts, and exist in a state of constant frustration at their inability to understand and process their existence.
It is highly questionable whether being minimally conscious is desirable, and consequently whether a diagnosis of MCS is preferable over one of VS. The risks and burdens of prolonged treatment of individuals in MCS are grave, and should not be dismissed. Moreover, epistemic concerns about the MCS diagnosis mean that expectations of benefits need to be tempered. Given this equation, it is likely that in many cases, prolonging treatment for individuals in MCS causes more harm than benefit.
In response to the previous argument, one might think that it is precisely the uncertainty in the diagnosis that warrants continuing treatments, even when such treatments constitute severe burdens for the patient, because we cannot know who among the hundreds of thousands with severe brain damage will make a meaningful recovery. However, this seems to place an unjust responsibility on the preponderance of individuals who do not reap any benefits from prolonging treatment and merely suffer. To argue that it is permissible or obligatory to keep anyone suspected of being in or progressing to MCS on LST means that one accepts that the vast majority of affected individuals will have burdens without outweighing benefits.
Autonomy
It is also unclear that prolonged treatment with a very low likelihood of meaningful recovery respects a patient’s wishes. Jennett reported that 90% of people consider VS a state worse than death. Footnote 50 Other surveys have shown that 82% of healthcare professionals would not want to remain in VS and would want treatment withdrawn. Footnote 51 Demertzi et al. Footnote 52 also showed that two thirds of healthcare professionals in MCS would want the withdrawal of treatment for themselves. However, it is telling that a much smaller percentage (28%) of healthcare professionals feel that they would withdraw treatment from their patients in MCS.
The prospect of their loved one being the one rare individual who not only survives severe brain damage, but also emerges from MCS to a state that that individual finds worth living, may cause families to err on the side of extending treatment, beyond what the patient would have wanted. As we have noted, the prognosis for individuals with severe brain damage, regardless of whether in VS or MCS, is very poor. Demanding that such a patient tolerate pain, suffering, and low quality of life that we ourselves would not, does not respect what we should infer are his or her wishes. Permitting a loved one with a bleak prognosis to pass away by forgoing a tracheostomy or artificial nutrition may cause him or her less harm and respect their preferences more than withdrawing such therapies months or years after injury.
Stewardship
A further issue that must be considered with respect to MCS is that of responsible stewardship of resources. Although systematic numbers about the prevalence of VS and MCS are scarce and divergent, values for individuals in VS in the United States are estimated between 14,000 and 35,000, and estimated at 280,000 for individuals in MCS. Footnote 53,Footnote 54,Footnote 55 These figures are bound to vary considerably regionally and internationally. As the diagnosis of MCS becomes more well known, it is likely that families of patients who have sustained an SBI will be encouraged and increasingly request prolonged treatment for their loved ones.
At the same time, skilled nursing facilities are facing historic shortages. The number of nursing facilities nationwide has remained mostly stagnant in the past 10 years, Footnote 56 and yet demand for skilled nursing beds is predicted to increase greatly in part as a result of an aging population and healthcare innovations. Footnote 57 By 2030, in California alone, the number of seniors expected to require skilled nursing facility placement could increase by 32,000, which would increase the senior demand for nursing home care by 40%. Footnote 58 Skilled nursing facility demand in the San Francisco Bay Area, where we are located, is expected to outstrip supply by 2020. Footnote 59 Moreover, multiple studies have found that as a result of being greatly impacted, many United States nursing homes are already operating with a poor quality of care. Footnote 60 Furthermore, the United States is facing a deficit of approximately 1,000,000 registered nurses by the year 2030. Footnote 61 It is of concern that if patients cannot be discharged from an ICU to a skilled nursing facility because of lack of supply, they tend to remain in the ICU longer. Reduced ICU bed availability in turn leads to increased rates of ICU readmissions as well as worse outcomes on the floor. Footnote 62 With ICUs highly impacted across the nation, triage decisions are inevitably made. Footnote 63 This may prevent elective procedures such as cardiac surgery, which may be cancelled or postponed without an ICU bed available.
In addition to a shortage of long-term beds and nursing staff for patients in VS or MCS, such patients may require expensive and highly limited technology, both in the diagnosis of their state and in proposed treatment protocols. For example, Owen et al. Footnote 64 have suggested that it may be necessary to communicate via fMRI with patients in MCS. But medical centers have a very limited number of fMRI machines—skilled nursing facilities have none at all—and these scanners are used for everything from early stroke detection to identifying tumor progression. Employing this scarce resource to monitor the brain of an individual who is not exhibiting clinical signs of consciousness would likely prevent the use of this technology in the diagnosis and treatment of many conditions with known treatments.
Burden on Families
In addition to issues of beneficence, autonomy, and stewardship, we should also consider the burdens on those who love and care for someone in MCS. Families of patients in long-standing VS or MCS are conflicted with an emotional paradox; they are unable to mourn the loss, as the person is still alive. Footnote 65 Paola Chiambretto et al. Footnote 66 looked at caregiver reactions to a loved one in VS and found a great degree of emotional distress, which increased with the duration of the patient’s condition. Furthermore, caregivers experienced anxiety surrounding their future death and revealed significant depressive symptoms. Many family members of patients in MCS were adamant that if they were ever in a similar condition, they would not want to be kept alive, which contributed to remorse surrounding decisions to place the person they loved on artificial life-sustaining measures. Footnote 67
Conclusion
We now return to our hypothetical case of the 21-year-old who sustained a devastating TBI. The patient is young and his family is understandably unprepared for the grim prognosis that the critical care team presents. However, 1 week after the accident, he remains fully dependent on the ventilator, and although he has emerged from his coma (he is blinking), he exhibits no signs of awareness of his self or environment, or any indication of purposive action. Although there is a small chance that he may make some recovery with long-term treatment, there is a much greater chance that he will make no, or only very moderate, recovery. Perhaps he will be able to experience some glimpse of consciousness. Perhaps he will be able to even express preferences through eye movements or through particular thought patterns detectable via fMRI. But he will certainly not return to a state comparable to the one he was in before the accident. His family does not believe that he would find such a life worth living, but they cling to the hope, precipitated by the promise of MCS, that he will beat the odds and make a miraculous recovery. However, the price for this small chance of a meaningful recovery is immense. Our patient faces the overwhelmingly large chance of years of invasive interventions, mechanical ventilation, and artificial hydration and nutrition, and of years of physiological and psychological pain. In his case, the fleetingly small chance of an outcome he himself would find desirable does not justify the near certainty of an outcome that he would not.
In this article, we have argued that the delineation of MCS as a distinct neurological state, along with research that has been conducted in order to expand how MCS is diagnosed, may pose ethical difficulties for the decisionmaking families and providers must perform. The prospect of an MCS diagnosis for their loved one who has sustained an SBI may lead families to be more resistant to the decision to withdraw life-sustaining treatment, even when such a decision is medically and ethically appropriate. This is problematic on the one hand, because there are inherent epistemological issues with the diagnosis of MCS, as well as phenomenological issues with the nature of MCS. At the same time, it is highly likely that a patient who presents a grim clinical picture after a reasonable time in the ICU will not make a recovery that that patient would find acceptable, regardless of whether the patient is in VS or MCS. This means that opting for prolonged treatments may be objectionable on the basis of patient beneficence, respect for a patient’s likely preferences, stewardship of resources, and burdens on family. It is understandable that MCS has sparked excitement among researchers eager to explore a novel neurological state. However, it is imperative that such excitement not obscure well-established principles of decisionmaking for patients with devastating neurological injuries.
