The vegetative state
- Correspondence to: Professor B Jennett, 83 Hughenden Lane, Glasgow G12 9XN, UK;
The definition, diagnosis, prognosis and pathology of this state are discussed, together with the legal implications
In the 30 years since this state was first described and named1 it has provoked intense debate not only among clinical scientists and health professionals but also among moral philosophers and lawyers. Considering its relative rarity there is also, courtesy of the media, an unusual degree of public awareness of the condition. What attracts attention and curiosity is the dissociation between arousal and awareness—the combination of periods of wakeful eye opening with lack of any evidence of a working mind either receiving or projecting information. The advantage of the term “vegetative state” is that it simply describes observed behaviour, without implying specific structural pathology. However, since the realisation that this state is frequently temporary, the original term persistent vegetative state is potentially misleading as it suggests irreversibility. After a certain length of time it may none the less be reasonable to describe this state as permanent.2
Before specific diagnostic criteria were agreed, some reports had used less rigorous definitions.3 The most widely accepted criteria are those in the 1994 report from the US Task Force.4 These are essentially negative—the lack of evidence of awareness of the self or the environment, of interaction with others, or of comprehension or expression of language. By implication, external stimuli do not evoke purposeful or voluntary behavioural responses that are sustained and reproducible. There are two problems in applying these criteria. One is the wide range of reflex responsiveness in some vegetative patients, some of which can give rise to suspicion of meaningful mental activity. The other is the limited range of voluntary responses available to severely brain damaged patients, which can make the detection of awareness difficult.5 The recent publication of detailed practical tests for detecting that a patient is in a minimally conscious state rather than vegetative should help to reduce the risk of misdiagnosis.6,7
The simplistic assertion that vegetative patients can be defined as showing no sign of cerebral cortical function has fuelled debate as to whether certain behaviours entail any cortical activity. More important, however, is whether there is evidence of sufficient cortical integrity to indicate awareness—which is the crux of the definition. Such integrity is not required for sudden light or sound to stimulate a startle response or an orienting reflex with the head and eyes turning briefly towards the stimulus. Whether blink to visual threat need imply limited cortical activity is debatable, but this response is not itself evidence of awareness. Because visual pursuit is usually one of the first observable signs of recovery in patients who go on to regain consciousness, some consider that patients who show this phenomenon are no longer vegetative. However, several authorities maintain that occasional vegetative patients do regain some capacity for visual pursuit without developing any other behavioural evidence of consciousness during the following months or years. The Task Force concluded: “One should be extremely cautious in making a diagnosis of the vegetative state when there is any degree of sustained visual pursuit or consistent and reproducible visual fixation.”4 The Royal College of Physicians criteria in 1996, which seemed to be designed to exclude cortical activity rather than showing lack of awareness, list “no nystagmus to caloric testing, no visual fixation, tracking, or response to menace” under the heading `Other criteria’.2 In several cases since then the English High Court has chosen to accept expert opinion that particular patients in whom one or more of these eye responses were active were indeed unaware and therefore vegetative.3
The spastic limbs may move in a non-purposive way, including groping and grasping. Noxious stimuli may provoke limb withdrawal that may be associated with facial grimacing and a rise in respiratory and pulse rates and in blood pressure; these do not in themselves indicate conscious experience of pain. Some vegetative patients may occasionally smile or frown, and less often laugh or weep, but these emotional behaviours show no consistent relation to an appropriate stimulus. Intermittent episodes of complex behaviour have been reported in three American patients who were vegetative by all other criteria, and whose global brain metabolism level was below 50% of normal.8 One patient, 20 years vegetative, uttered single words from a vocabulary of four or five once every two or three days. Another, seven years vegetative, responded to loud noise or attempts at nursing care by clenched teeth, rigid extremities, and high pitched screaming that abated in response to soothing voices or music. All three patients had limited areas of cortex that showed metabolic levels that were higher than elsewhere although still well below normal. The conclusion was that these patients had isolated segregated cortical networks that retained connectivity and partial functional integrity. In two other vegetative patients without unusual behaviours this team observed some preserved cortical metabolism; in one case this was virtually normal throughout the brain but there was severe bilateral thalamic damage. Others have recorded evidence of residual cortical activity in vegetative patients,9 and there has been an exchange of views between these two laboratories about the significance of their findings.10 Some patients diagnosed as permanently vegetative have had repeated episodes of tonic-clonic epilepsy, further evidence of residual cortical activity.
The dominant lesions in the brains of many vegetative patients are thalamic and subcortical white matter damage.11 Of 35 traumatic cases 80% had thalamic damage and 71% severe diffuse axonal injury; in seven the cerebral cortex was normal and in 21 there were only minor cortical contusions. By contrast study of the brains of 30 cases severely disabled after head injury found that half had neither severe diffuse axonal injury nor thalamic damage, a negative combination found in not a single traumatic vegetative case.12 Of 14 non-traumatic vegetative cases only 64% had diffuse cortical damage and one had a completely normal cortex, but every case had severe thalamic damage.11 The conclusion is that consciousness depends on the integrity of sufficient thalamocortical and intercortical connections, and that isolated neural activity involving parts of the cortex, even when associated with minimal stereotyped behavioural expression, need not indicate even minimal consciousness. However, much remains to be understood regarding the mechanisms underlying the vegetative state.
Estimates of the frequency of the vegetative state vary widely because cases are difficult to locate in many different sites of care, and because of variations in how long patients need to have been vegetative for inclusion in a survey. This can vary from one to six months, one month being specified by the Task Force. American prevalence figures, including those resulting from chronic conditions, vary from 64 to140 per million population (PMP), about one third of them children.13 The annual incidence from acute causes varies in different countries from 14 to 67 PMP at one month after the insult but only 5 to 25 PMP at six months, with the UK figures being the lowest.3 This reflects the lower incidence of severe head injuries in the UK, as trauma accounts for 40%–50% of vegetative survivors after acute insults.
The best data available on the prognosis of vegetative patients are from the analysis of 754 published (acute) cases by the Task Force.4 Of those vegetative at one month 43% had regained consciousness by one year, 34% were dead, and only 23% were still vegetative. The longer the vegetative state had lasted the fewer recovered—after six months only 13% regained consciousness. Trauma cases did better (table 1), and head injured children did marginally better than adults. It is because of the attrition by death and recovery that the incidence and prevalence rates reduce as time passes since the insult. There are few systematic data on cases followed up for more than a year but the Task Force decided that it was reasonable to declare the vegetative state permanent after one year in traumatic cases and after three months in the non-traumatic. However, the Royal College of Physicians report recommended six months for the latter cause.2 There are occasional reports of late recoveries, more in the public media than are medically verified. It is, however, important to realise that many patients who regain consciousness after several months remain speechless and tube fed with very limited ability to communicate. To state that these patients have “recovered” may seem to some a rather optimistic claim. Some of them are in the minimally conscious state,6 showing some responses that indicate a limited degree of awareness. However, the wisdom of attempting to define such a state has been questioned by disability rights campaigners in America.14
There is a high mortality during the first year and the Task Force reported a mean survival of only two to five years for patients vegetative at one month. What is of more interest, however, is the expectation of life for those who have already survived in a vegetative state for a year or more. The continuing annual mortality then reduces year by year, and there have been occasional survivals of 20 or more years. A systematic study in California drew attention to the difference between mean and median survival, the latter being shorter because not distorted by the occasional very long survival.15 For a young adult vegetative after one year this study calculated a mean survival of 10.5 more years, the median being 5.2 years. For one still vegetative four years after the insult these figures increase to 12.2 and 7 more years respectively.
In the early months it is important that everything is done to maintain the patient in the best possible general condition so that the most can be made of any spontaneous neurological recovery that may occur.16 Good nutrition is essential to maintain body weight, usually via a percutaneous endoscopic gastrostomy, and joint mobility should be preserved to minimise contractures. Regimens of sensory stimulation (so called coma arousal programmes) have been advocated but there is no good evidence that these promote the recovery of consciousness.17
The ethical and legal debates focus on attitudes to patients declared permanently vegetative. There have been many declarations that survival in a permanent vegetative state is not a benefit to the patient, some regarding it as a fate worse than death. Numerous surveys of the attitudes of patients, doctors, nurses, and ethicists have confirmed that this is a widespread view, with many respondents indicating that they would not want life prolonging treatment if in this state.3 At the same time there is increasing concern in medical ethics for respecting patient autonomy when making decisions about treatment. Competent patients have an absolute right to refuse the initiation or continuation of any treatment, even when this is life prolonging, if they regard it as bringing more burdens than benefits. The problem is that vegetative patients are not competent to refuse continued treatment, and there is concern about how best to protect them from treatment that they would probably refuse if they could. Few patients have made an advance directive and there has been much debate about appropriate surrogate decision makers. In the US families are allowed to assume this role but in the UK only a doctor may make such a decision about an incompetent adult patient, albeit after discussion with the family. While it is easy enough to decide to withhold cardiopulmonary resuscitation or antibiotics for acute infections, many vegetative patients survive for years after a decision to limit these treatments. The question then arises of withdrawing artificial nutrition and hydration—regarded by most authorities as a form of medical treatment. There is no moral or legal obligation for a doctor to provide a treatment that is not in the best interests of his patient. Although some object that such withdrawal might seem to condone euthanasia this has been challenged.18,19 A recent review of the outstanding ethical issues in the UK notes the need to consider also the question of justice in allocating scarce resources to the indefinite support of vegetative patients.20 In many common law jurisdictions it is now agreed that withdrawal of life sustaining treatment is lawful,3,21 and an English judge recently ruled that it does not infringe the European Convention on Human Rights.22 Only in the UK, however, is High Court approval still required before taking such action, and some lawyers do not consider that the legal situation has yet been satisfactorily resolved.23,24
The definition, diagnosis, prognosis and pathology of this state are discussed, together with the legal implications
Competing interest: BJ has appeared as an expert witness in the High Court when a declaration was sought that withdrawal of treatment from a number of specific patients in the permanent vegetative state would be lawful.