Limitation of life support in the ICU: Ethical issues relating to end of life care
In recent years, increasingly, intensivists have focused attention on the ethical aspects of end-of- life care. This has led to shifts in the approach from aggressive interventions to one of mitigating pain and taking into account the wishes and sensibilities of the patients' families with regard to continuing futile care. While the legal implications of this change in practice has led to the evolution of precise guidelines in the US and in Europe, in India this vital area of critical care remains largely unexplored. This review outlines the recent changes in clinical practice based on ethical principles and the legality of limiting life support in the context of futile or end-of-life care. An appraisal of the ethical issues in critical care urges us to apply intensive care with humanity and compassion. We need to respect the choices and the emotional needs of the patient and his family. Our duties must include providing information, balanced interpretation of results & counseling of the family to enable them to take rational decisions. Our strategy in end-of-life care should be unambiguous and we should ensure that there is consensus among all the physicians involved in the patient’s care. The medical community must work towards evolving legislation appropriate to Indian conditions.
Intensive care has come to be associated with high-tech, aggressive and often risk-filled medical care. Although critical care has expanded the scope of modern medicine, its mortality rate remains high at 10-36%. Aggressive treatment even in the face of hopeless prognosis has become commonplace in ICUs. While in an earlier era, the terminally ill passed away amidst their family at home, death has now become “medicalised” with the patient cut off from his loved ones in his final hour. Death often takes place in an ICU surrounded by IV lines, devices and machines.
Not surprisingly, doctors are often perceived as being insensitive to the wishes of the patient and his family when medical treatment is not beneficial in the real sense. We should therefore ask ourselves where to draw the line in applying life support methods in order that we do not unnecessarily extend the natural process of dying. Ethics do not require that we apply technology indiscriminately “to save life at all costs”, but rather to act in the best interests of the patient and his family.
In order to avoid futile medical care various practices limiting life support have been proposed [Table - 1].
ICU practices were studied prospectively by the SUPPORT group of investigator. This landmark study to understand the Prognoses and Preferences of Outcomes and Risks of Treatment showed clearly that seriously ill patients get more expensive treatment and have longer ICU stay. They often die in undesirable states such as coma and ventilatory support. It has also bought out the fact that physicians are applying treatments on patients during their end of life that they would not prefer for themselves.
Since the SUPPORT revelations the practices in ICUs are clearly changing. In an extensive survey of the end-of-life care in US ICUs, Pendergast et al found data that challenge the observations of the SUPPORT study. They found that in 48% of ICUs in the US limitation of life support is the predominant practice.
Out of 5910 deaths, 74% had been on a life-limiting care [Table - 2]. This figure is an aggregate with a wide variability in practices in different institutions. However, the median practice was to limit life support in 70% of deaths. 48% were by withdrawal or withholding of life support and only 28% received full CPR.
In contrast, in one leading hospital in India [Table - 3] only 22% had life-limiting management and probably this reflects the general trend. Do not Resuscitate (DNR) orders were written with family's consent in only 4 out of 298 deaths. LAMA (left against medical advice) is a situation peculiar to this part of the world. We have regarded these as withdrawal of treatment unilaterally by the patient's family. All of these were due to financial constraints. It tells the tragic story of the enormous financial and emotional burdens that the family has to endure in the context of critical illness.
There are misgivings about writing DNR orders or a strategy of limiting life support that the patients thus categorized may receive less medical attention and their care may be suboptimal. On the other hand, they may also be spared of the unwarranted interventions that a full-fledged management would entail.
In the study by Hall et al which compared the treatment of patients when life supporting treatment was or was not withdrawn, clear differences were evident. Those in the withdrawal or withholding group were older, received fewer interventions before death (mechanical ventilation, inotropic agents, dialysis), [Table - 4] but received more sedatives and narcotic analgesia. [Table - 5]
Total healthcare costs in the US is estimated to be 14% of the GDP and a disproportionate amount is spent on critical illness in the elderly population. Limitation of life support may therefore be expected to result in considerable fiscal efficiency.
Critical care is very expensive and on average more expenses are incurred on patients who die than on those who survive. Life sustaining treatment is often not cost-effective and data on ICU practices clearly indicate that standards of care have not been defined. Therefore, there would be plenty of room to apply cost reduction strategies in the ICU.
However, as observed by Luce and Ruberfeld, restrictions on life-sustaining treatment may not yield significant cost savings. In the US this is due to fixed costs and contractual arrangements. In the Indian context the cost benefits are real since in the most part, expenses are borne entirely by the patient. In India an increase in bed strength does not necessarily mean an increase in expenses to the hospital.
The practice of withholding or withdrawing life support in the ICU and administering palliative care is supported by the ethical principles of autonomy, beneficence and non-male fiscence.
Our understanding of the laws governing these practices in the US and elsewhere may contribute to evolving legislation to suit our own requirements.
The principle of the right to self determination led to the concept of informed consent and informed refusal. The principle holds that treatment may not be initiated without the approval of patients or their surrogates except in emergency situations and that the patients or their surrogates may refuse all therapies. American law requires that there should be clear and convincing evidence of the patient's wishes. This is achieved by Advance Directive or Will.
In the absence of Advance Directive the surrogates (family members) may act on his behalf. The problem of Will is that it is rarely available and the patient may change his mind or may not have visualized his present situation. In France where the State does not recognize the right of surrogates to represent the incompetent patient, the doctor is given the responsibility of acting in his best interests. Thus in France the principle of beneficence takes precedence over that of autonomy. This is often dubbed as “paternalism”.
In the case of Quinlan (1976) the new Jersey supreme court held that the patient had the right to refuse mechanical ventilation and, because she was vegetative and could not exercise that right directly, her parents could act as surrogates for her.
In the case of Barber (1983) the California Court held that physicians charged with murder had not committed an unlawful act, when, with the permission of the family, they removed nutrition and hydration from a comatose patient
The issue of withholding and withdrawal of life support was first addressed by the US court in the Cruzan case (1990). The Cruzans wanted a feeding tube removed from their vegetative daughter. In Missouri, where they lived, the court required specific evidence that an incompetent person would want treatment withdrawn. The court required “clear and convincing” evidence of the patient's wishes, which limits the role of surrogates in case the patient is incompetent. The reliability of surrogates as representing truly the patient's wishes has been questioned in the UK and in France it is acceptable practice for the physician to act unilaterally in a “paternalistic” manner. Nevertheless, in the Cruzan case later the court acknowledged the right of a competent patient to forgo treatment even if it meant that this would result in his death. In most states of the US, surrogates' requests are honored despite the Cruzan judgement.
Physicians often cite futility of continuing therapy as justification to withhold or withdraw it. Since the perceptions of the patients or their surrogates may be different from the physicians' these have been modified to indicate the so-called “physiologic futility'. Even with this term value judgements are inevitable as few things are of absolute certainty in clinical practice. For eg., in writing DNR orders it would be impossible to be sure of an immediate as opposed to eventual physiological futility of CPR except in the circumstances such as cardiac rupture or severe outflow obstruction. Using this principle physicians have been found to act unilaterally. The position of the US courts has been rather ambiguous in this regard.
In the Baby K Case (1994) when the court was asked to approve in advance a physician's decision to withhold life-sustaining treatment, the Court refused to do so. On the other hand, in Gilguun (1995) the courts were reluctant to punish doctors after they issued a DNR order and removed the ventilator from a patient over the objections of the patient's daughter. In the Wendland case, the wife wanted to remove the feeding tube from her husband who had been comatose for 2 years. Her husband's mother, however, went to court to block the plan. The court thus recognizes a hierarchy of proxy where its ruling supports one who is in favour of protecting life. It also points to the difficulties in asking the surrogates to take decisions, as the dilemma of disagreement within the family members may arise.
The trend in futility cases is that while courts do not permit life support limitation prospectively on appeal from doctors, they tend to defend decisions to limit life sustaining therapy when they are made within acceptable professional standards.
Despite the outcome of these cases we must eschew the practice of taking unilateral decisions on what is evidently a sensitive matter. This trend has become evident in France as well where there are less legal obligations to do so. Decisions are to be made jointly with the patients or surrogates. This respects their right of self determination, and avoids misunderstanding and conflicts that may lead to litigation. The physician may in his discussion with the family play the role of an informer or a counsellor. He should allow time for the family to accept the realities and for a consensus decision to emerge through a cooperative process. The physicians should clear the air of wrong notions, prejudices and misinformation. Often a period of de-escalation of aggressive treatment enables the family to come to terms with the realities and agree to a formal consent for withdrawal or withholding of life support.
In the cases of Glucksberg (1997) and Vacco (1997) the US Supreme Court confirmed its approval of forgoing life sustaining treatment and provided guidelines for administering palliative care. These cases dealt with the constitutionality of the laws prohibiting physician-assisted suicide in the states of Washington and New York. Since there are laws prohibiting suicide, and there are legitimate fears of misuse, physician-assisted suicide continues to be illegal. However the Court acknowledged the right of patients to receive adequate pain relief during the dying process.
The patients' access to palliative care may be limited by the principle of nonmalefiscence as it does not permit physicians to intend potentially harmful effects. The US supreme Court distinguished assisted suicide from palliative care by accepting the principle of “double effect”. Quill and associates described it and distinguished between intended and unforseen effects. Giving of analgesics and sedatives are justifiable if their bad effects are not intended, and there are no less risky ways of relieving suffering.
The US Supreme Court also sanctioned the practice of terminal sedation, in which patients are rendered comatose and then may have nutrition and hydration withdrawn. Though this issue is contentious, the Court allows the practice if it is based on informed consent.
In physician assisted suicide (PAS) the actual killing drug is self administered with help from a physician and in euthanasia the physician himself administers the lethal dose. The much-debated practice has received legal sanction in very few regions of the world. The first legal sanction of euthanasia (the Northern Territory rights of the terminally ill Act 1995, Australia) was repealed in 1997. The Netherlands never had legal sanctions but because of a “tolerant Judiciary” the practice of euthanasia is common among Dutch physicians. In the state of Oregon in the US, death with Dignity Act was passed allowing PAS. Even in this state the practice is not widespread.
The dominance of the principle of autonomy over that of beneficence often leads to inappropriate treatment. The physician's role is also hampered by his inability to exercise his judgement in the best interests of his patients. Rejecting the idea of futility, the patient's family often impose unreasonable demands on him. The doctor often finds himself in a moral dilemma without adequate legal safeguards against misinformed decisions by the families. In this part of the world his problems are compounded by the need to ration resources and the moral obligation to protect families from financial ruin. He is also subjected to societal pressures that erodes his self esteem as a professional.
Unlike in the US, in India legal opinion and legislation relating to end-of-life care is scarce. It is rather disturbing that there is no clearly stated legal opinion regarding discontinuation of life support systems even in brain dead patients.
In India Article 21 provides the right to life. However, the concept of autonomy is still weak and even the right to live a dignified life is not adequately explored for the purpose of end-of-life care.
The Transplantation of Human Organ Act (1994) defines brain death. There have been a paucity of cases dealt with by Indian Courts in the matter of terminal illness. In the P. Rathinam Vs Union of India, 1994, the Supreme Court conceded that in the case of terminal illness, attempts to hasten death may be viewed as acceleration of a dying process already started. The Court acknowledged that “a person cannot be forced to enjoy the right to life to his detriment, disadvantage or dislike.”
However, in the Gian Kaur Vs State of Punjab the Court categorically stated that the right to live cannot be interpreted to include the right to die an unnatural death curtailing the natural span. The judge declared that the plea to support the view of permitting termination of life in cases of a dying man, who is terminally ill or in persistent vegetative state to reduce the period of suffering during the process of certain natural death is not available to interpret Art 21.
Our religious texts glorify the act of suicide as “mahaprasthana” (great departure) Voluntary starvation was considered the befitting conclusion of a hermit's life and this was prevalent among the Jain ascetics as well. However, the position of the apex court in this regard is different. The natural corollary of the judgement in the Gian Kaur case is that it disallows the concepts of euthanasia as it violates article 21 of the constitution. It also reaffirms that abetment to suicide is a punishable offence while attempted suicide is viewed more compassionately.
The Indian Penal Code declares attempt to suicide or abetting to suicide as illegal. The sale of morphine sulfate is also illegal in most States.
Lakhs of rupees of compensation can today be claimed without a penny being tendered as court fee in consumer disputes redressal fora. The deterrence of an ad valorem court fee is no longer there and consequently an increasing number of professionals are being involved in interminable cases. The problem becomes more complicated in grey areas where there is a legal penumbra. Without proper legal definitions of Rights often the doctor is called upon to take on-the-spot decisions on legal conundrums which have required decades of deliberations in the Courts.
The predominant factor that impacts decision-making in the ICU is the unbearable financial burden that it entails. As a consequence the family goes through complex emotional upheaval. The doctor is often a witness to the feelings of guilt, fatalism and denial that the family would experience. The family tends to nurture false hopes even in the face of hopeless prognosis that is supported by misinformation. There is also the unwanted intrusion of alternative systems of medicine that only helps to reaffirm entrenched prejudices.
Above all, the legal process is too slow to allow legal interventions prospectively in a patient's management where required. Redressal has been expedited by consumer courts, but they have not been used to address issues related to end-of-life care. Generally, Indian sensibilities rarely allow withdrawal of support and withholding may be more acceptable.
Providing emotional support to the patient's family is an essential part of the responsibilities of the treating team. In a multicentre study conducted in France, the needs of the ICU patients' families were evaluated. The important predictors of family satisfaction included absence of contradiction in information given by caregivers, information provided by a junior physician, nurse to patient ratio < 3, help from the patient's family doctor and sufficient time spent giving information.
In another study looking at parental perspectives on end-of-life care in the pediatric ICU, parents were found to place the highest priority on the perception of their child's pain, the quality of life and the likelihood of getting better when considering withdrawal of life support. A questionnaire revealed that parents depended mainly on critical care practitioners for decision-making and emotional support.
An appraisal of the ethical issues in critical care clearly urges us to apply intensive care with humanity and compassion. It is evident that setting appropriate goals and strategies in end-of-life care and in patients with poor prognoses is as important as other skills in critical care. We need to respect the choices and the emotional needs of the patient and his family. Our duties must include providing information, balanced interpretation of results & counseling of the family to enable them to take rational decisions. In consonance with the principle of autonomy, we should always take informed consent from the patient or his family for withholding or withdrawing support including DNR orders. Our strategy in end-of-life care should be unambiguous and we should ensure that there is consensus among all the physicians involved in the patient's care. Our duties are not only towards the patient but also to the families who look to us for emotional support and rational decisions to avoid unnecessary financial and emotional burdens. Finally, the medical community must work towards evolving legislation appropriate to Indian conditions.