More than 20% of all ICU deaths in the US occur during or shortly after ICU admission. At my university hospital, up to 15% die during their ICU stay, reflecting the severity of our patients' illnesses. Increasingly, decisions to withhold or withdraw life support precede death in the ICU. Critical care physicians are constantly challenged to ensure appropriate management decisions are made and effective palliative end-of-life care is provided.
Unfortunately, end-of-life care in the ICU is beset by shortcomings. Physicians frequently do not know when patients want to limit life-sustaining treatment. An alarming number of patients die with pain and dyspnea. Attempts at advance care planning are often thwarted by the inherent shortcomings of advance directives.
Widely accepted principles of bioethics, such as the duty to honour patient autonomy and to provide beneficial (and non-harmful) care, frequently come into conflict, making it difficult to guarantee ethically sound decisions. Finally, the approach to end-of-life decision-making varies considerably by physician and location, suggesting that care depends more on local practice and custom than patient preference.
DEATH FORESTALLED BY SCIENCE
The achievements of modern critical care have been truly astounding. A short list of the accomplishments includes: the advent of the modern ICU, replete with continuous pulmonary and haemodynamic monitoring; a growing number of nurses and physicians expert in critical care; the sophisticated use of complex therapeutic technologies, such as mechanical ventilation and continuous renal replacement therapy; immediate availability of highly accurate diagnostic tools, such as CT scanners; and new medications stemming from advances in biomedical research.
In Western society, the ageing of the population and the growing number of patients living with chronic illness guarantees that the number of ICU patients will grow.
The ability to forestall death in some patients has created a predicament unanticipated by previous generations of physicians, who viewed death as the inevitable consequence of ageing or poverty. Even today, for many patients with end-stage or overwhelming illness, death remains inevitable despite technology and expertise. For other patients, the burdens of acute illness may be unbearable, even if survival is possible.
As a result, today's patients, families and physicians frequently find themselves wrestling with the uniquely modern dilemma of deciding whether potentially life-sustaining treatments should be provided or withheld.
MEDICAL FUTILITY: ASSET OR DISTRACTION?
Decisions to withhold or withdraw life-sustaining therapy fall into two categories. In some cases, further therapy is considered 'futile', implying that it could not work even if provided. In contrast, 'non-futile' therapy implies that treatment could work, although the burdens may outweigh the benefits. The implications of these two distinct categories are clear. Futile care cannot achieve a desired endpoint, such as saving a patient's life, even if attempted.
Consequently, at least in theory, offering futile therapy cannot be justified because the benefits are, by definition, zero; the burdens must inevitably be excessive, no matter how small. In contrast, non-futile care requires a cost accounting on the patient's behalf, weighing the potential benefits and burdens of therapy and deciding whether starting or continuing treatment is justified.
The prevalence of futile care is uncertain. One study from California suggested that almost 5% of Medicare beneficiaries received 'potentially ineffective care'. In another study, 73% of European physicians reported that they frequently admitted patients to the ICU despite them having 'no hope of survival'.
Although pockets of controversy exist, there is ample agreement among most physicians and medical societies that futile care should not be offered even if requested. Although uncommon, legislation is starting to appear that supports the decisions of physicians and healthcare institutions to forego inappropriate medical interventions. In our hospital, a 'conscientious practice policy' has been implemented to address requests for potentially inappropriate life-sustaining care, incorporating guidelines outlined by the American Medical Association.
This practice includes ensuring an appropriate discussion with the patient or surrogate explaining the rationale for withholding interventions, offering the opportunity to transfer care to another physician or facility, providing oversight by the hospital bioethics committee, and obtaining additional medical opinions and support from the hospital administration if needed.
In practice, futility concepts are difficult to employ and, in my experience, an uncommon rationale for foregoing life-sustaining treatment. The definition of futility remains hotly contested. Some authors cite numerical definitions (such as, the likelihood of successful therapy falls below a numerical threshold); some cite qualitative definitions (for example, a minimally acceptable quality of life); and some restrict the definition to narrow physiological endpoints (such as, the ability to
achieve an acceptable blood pressure or oxygen level).
Although many authors have attempted to identify futile situations that would justify foregoing life-sustaining treatment, medical progress raises the possibility that the benefits of therapy may be underestimated, thus running the risk of denying patients potentially effective therapy.
Finally, physicians and nurses often cannot identify futile situations accurately, raising the risk that potentially salvageable patients would have life-support withdrawn inappropriately. Consequently, most decisions to limit life-sustaining therapy rely on weighing burdens and benefits rather than invoking medical futility.
The decision to forego life-sustaining therapy under non-futile circumstances imposes unique burdens on caregivers and families. Most critically ill patients are cognitively impaired, either permanently or temporarily because of their medical condition or treatment, and cannot speak on their own behalf. In the US, great effort has been devoted to promoting advance directives so patients can dictate the type of therapy they would accept should they become cognitively impaired.
Unfortunately, advance directives have numerous shortcomings: most patients do not have them; they are frequently unavailable when needed; they may be obsolete if completed in the distant past; they are often misinterpreted; and, most importantly, they often fail to address the specific questions raised by the patient's acute illness.
Because most critically ill patients cannot speak for themselves, surrogate decision-makers, most commonly close family members, are generally recruited to help make decisions. Unfortunately, family members and physicians may not be able to identify accurately a patient's wishes, a problem made worse if not discussed previously. Making critical decisions on a loved one's behalf imposes an enormous burden on family members at a time when many are experiencing severe symptoms of anxiety and depression.
Although it remains to be proven, it would seem likely that compassionate, unhurried, comprehensive meetings between patients and surrogates would promote appropriate decisions without placing an undue burden on families. Unfortunately, such meetings are not always performed optimally.
However, several strategies appear to optimise decision-making and improve family satisfaction. Highlights include allowing enough time to address the family's concerns, providing sufficient information, sharing in decision-making, supporting the family's decisions, and assuring them that their loved one will be well cared for. Even under conditions of medical futility, similar principles of excellent communication and family support apply.
PALLIATIVE CARE AND WITHDRAWAL OF LIFE SUPPORT
Symptom management and palliation are essential features of good patient care, even when death is not imminent. However, palliative care assumes central importance when life-sustaining therapy is withheld or withdrawn. Although reports indicate that too many patients die with inadequate relief of symptoms, increasing attention is being paid to promoting excellent palliative care in the ICU.
Key management principles include relief of distressing symptoms, such as pain and dyspnea, and providing emotional and spiritual support for patients’ families. When done well, outstanding end-of-life care can be an important source of family and professional satisfaction.
We live in an era of breathtaking accomplishments in critical care. An improved understanding of critical illness and the development of increasingly effective therapies have saved the lives of countless patients who could not possibly have survived a few years back.
Simultaneously, a growing number of desperately ill patients are admitted whose lives cannot be saved, either because effective treatment does not exist or because the burdens of therapy outweigh the benefits.
When death is likely, physicians and caregivers are challenged to partner closely with patients and families to make ethically sound decisions about life support. Equally importantly, it is our duty to provide effective palliative care and support to patients and their loved ones as the end draws near.