Advertisement
Mayo Clinic Proceedings Home

The Ethical Validity and Clinical Experience of Palliative Sedation

  • Paul Rousseau
    Correspondence
    Address correspondence to Paul Rousseau, MD, VA Medical Center, Phoenix, AZ 85310
    Affiliations
    VA Medical Center and RTA Hospice, Phoenix, Ariz, Arizona State University, Tempe, and Arizona College of Osteopathic Medicine, Glendale
    Search for articles by this author
      The physician's main goal in caring for a dying person is to reduce suffering, including pain, physical symptoms, and emotional, psychosocial, and spiritual distress. In refractory and intractable cases, palliative sedation offers a compassionate and humane alternative to conscious and continual suffering, both for the patient and the patient's family. Without a doubt, further studies are necessary, particularly in cases of cognitive impairment, but palliative sedation offers a valuable and efficacious intervention for interminable suffering at the end of life.
      The goals of medicine as characterized by Jonsen et al
      • Jonsen AR
      • Siegler M
      • Winslade WJ
      include the improvement of patients' functional status, education regarding their disease and prognosis, and relief of pain, symptoms, and suffering. The latter 2 precepts are cardinal to palliative care and facilitate a dignified and peaceful death for the patient and comfort and reassurance for attendant family members. However, suffering in all its domains cannot always be readily assuaged, particularly the distress of terminally ill patients. Palliative sedation (PS) is a valuable, yet controversial intervention used to manage refractory symptoms and suffering in terminal illness. It was fundamentally sanctioned by the US Supreme Court in its decision regarding physician-assisted suicide.

      Vacco v Quilt, No. 117 SCt 2293 (1997).

      Washington v Glucksberg, No. 117 SCt 2258 (1997).

      The reported incidence of PS varies from 5% to 52%,
      • Rousseau P
      Terminal sedation in the care of dying patients.
      with such variance likely attributable to diverse definitions of PS, the retrospective nature of studies, and cultural and ethnic diversity. However, while definitions are occasionally ambiguous and open to interpretation, PS can be defined objectively as the intention of purposely inducing and maintaining a sedated state, but not deliberately causing death, in specific clinical circumstances complicated by refractory symptoms.
      • Fainsinger RL
      • Landman W
      • Hoskings M
      • Bruera E
      Sedation for uncontrolled symptoms in a South African hospice.
      • Hallenbeck J
      Terminal sedation for Intractable distress: not slow euthanasia but a prompt response to suffering.
      A refractory symptom may be subjective and at times nonspecific. Cherny and Portenoy
      • Cherny NI
      • Portcnoy RK
      Sedation in the management of refractory symptoms: guidelines for evaluation and treatment.
      define a refractory symptom as one that cannot be controlled adequately despite aggressive efforts to identify a tolerable therapy that does not compromise consciousness. They further add that a symptom may be considered refractory when additional invasive and noninvasive interventions are incapable of providing adequate relief, or the therapy is associated with excessive and intolerable acute or chronic morbidity and is unlikely to provide relief within a tolerable time frame. Although the term “symptom” usually connotes a physical malady such as pain or dyspnea, Cherny and Portenoy
      • Cherny NI
      • Portcnoy RK
      Sedation in the management of refractory symptoms: guidelines for evaluation and treatment.
      also acknowledge psychological suffering and suggest that, in a small proportion of patients, therapeutic approaches, including PS, may be necessary. Cherny
      • Cherny NI
      Commentary: sedation in response to refractory existential distress: walking the fine line.
      suggests that it is much more difficult to establish the refractory nature of a psychological symptom and that such cases “skirt the distinction between assisted death and appropriate sedation for refractory symptoms at the end of life.” Palliative sedation of psychological suffering may also be more ethically and emotionally problematic for family and health care providers, as the patient is frequently awake, cognitively intact, and socially interacting prior to initiation of sedation.

      ETHICAL VALIDITY

      The ethical basis of PS derives from the principle of double effect, a concept developed by Roman Catholic theologians in the Middle Ages
      • Quill TF
      • Lo B
      • Brock DW
      Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia.
      • Quill TE
      • Dresser R
      • Brock DW
      The rule of double effecta critique of its role in end—of-life decision making.
      and applied to situations in which it is impossible for a person to avoid all harmful actions.
      • Quill TE
      • Dresser R
      • Brock DW
      The rule of double effecta critique of its role in end—of-life decision making.
      • Garcia ILA
      Double effect.
      The ethical principle of autonomy and the precept of informed consent are also fundamentally involved in the use of PS and directly interface with the intent of double effect.
      The traditional formulations of the doctrine of double effect emphasize 4 basic conditions.
      • Quill TE
      • Dresser R
      • Brock DW
      The rule of double effecta critique of its role in end—of-life decision making.
      • Kuhse H
      The first condition addresses the nature of the act, which must be good or morally neutral and not in a category that is absolutely prohibited and intrinsically wrong. The second condition addresses the intent of the health care provider, which must be good, and while the good effect and not the bad effect must be intended, the bad effect can be foreseen, tolerated, and permitted. The third condition addresses the demarcation between the means and effects and, in end-of-life care, relates to the fact that death must not be the means to the good effect, which is relief of suffering. The fourth condition addresses proportionality, whereby the good effect must exceed or balance the bad effect.
      • Quill TE
      • Dresser R
      • Brock DW
      The rule of double effecta critique of its role in end—of-life decision making.
      • Beauchamp TL
      • Childress J
      The use of PS favorably incorporates the 4 conditions that constitute the doctrine of double effect, although Quill et al
      • Quill TE
      • Dresser R
      • Brock DW
      The rule of double effecta critique of its role in end—of-life decision making.
      argue that the second condition, the beneficial intent of the action, is unclear when discussing PS. They propose that PS inevitably causes death, and whether death is intended or merely foreseen is ambiguous and indistinct, particularly when the patient desires death.
      • Quill TE
      • Dresser R
      • Brock DW
      The rule of double effecta critique of its role in end—of-life decision making.
      Many proponents of PS contend that such an argument is based on an erroneous interpretation and misunderstanding of the essence of double effect and that sedation is a morally and ethically acceptable strategy for the relief of intractable suffering.
      Autonomy intertwines with double effect and presupposes a patient's right to make decisions regarding treatment according to his or her own belief system, cultural and personal values, and life plans.
      • Rousseau PC
      Ethical concepts in hospice and palliative medicine.
      Obviously, autonomy is involved in the decision to use any medical intervention, including PS. However, proponents of assisted death contend that PS is analogous to physician-assisted suicide,
      • Quill TF
      • Lo B
      • Brock DW
      Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia.
      and the discriminatory use of autonomy to endorse PS is clearly no different than the use of autonomy to endorse assisted suicide. Yet PS is condoned and legal, while physician-assisted suicide is proscribed and illegal (except in Oregon, where it is legal). In terminal situations, the intent of the patient, like that of the health care provider, must also be considered, and in the use of PS, the intent must be to relieve suffering, not to hasten or cause death. Nevertheless, ethical and moral dilemmas do arise when a patient surreptitiously desiring a quick death requests PS. In such cases, ethical and psychiatric consultations are obligatory if the patient's autonomous intent is known or suspected. Moreover, ethical and moral controversy may also arise when psychological suffering precipitates a request for PS, as physical suffering is frequently perceived as more medically acceptable as a reason for PS than is psychological suffering, even though existential symptoms can be just as distressful and refractory as physical symptoms. Consequently, psychological suffering may be perceived as an invalid justification for PS, fostering questions regarding the patient's autonomy and decision-making capacity. Nevertheless, PS is a beneficial therapy for refractory psychological distress in carefully selected cases when decision-making capacity is intact.
      Finally, informed consent is intimately integrated with autonomy and allows a reasonable person to make autonomous treatment choices, assuming the patient or the patient's surrogate has decision-making capacity and the clinician provides necessary information in an understandable manner so a valid decision can be made.
      • Rousseau PC
      Ethical concepts in hospice and palliative medicine.
      Informed consent is mandatory prior to initiation of PS and should include documentation of the reason(s) for PS, the people present during the discussion, and, if required by institutional or corporate policy, a completed consent form placed in the patient's medical record.

      CLINICAL EXPERIENCE

      The literature on palliative medicine has modestly addressed the use of PS, with a small number of retrospective and prospective studies,
      • Fainsinger RL
      • Landman W
      • Hoskings M
      • Bruera E
      Sedation for uncontrolled symptoms in a South African hospice.
      • Ventafridda V
      • Ripamonti C
      • De Conno P
      • Tamburini M
      • Cassileth BR
      Symptom prevalence and control during cancer patients' last days of life.
      • Fainsinger R
      • Miller MJ
      • Bruera E
      • Hanson J
      • Maceachern T
      Symptom control during the last week of life on a palliative care unit.
      • Greene WR
      • Davis WH
      Titrated intravenous barbiturates in the control of symptoms in patients with terminal cancer.
      • Morita T
      • Inoue S
      • Chihara S
      Sedation for symptom control in Japan: the importance of intermittent use and communication with family members.
      • Stone P
      • Phillips C
      • Spruyl O
      • Waight C
      A comparison of the use of sedatives in a hospital support team and in a hospice.
      • Fainsinger RL
      Use of sedation by a hospital palliative care support team.
      • Morita T
      • Tsunoda J
      • Inoue S
      • Chihara S
      Do hospice clinicians sedate patients intending to hasten death? J Palliat Care.
      • Pcruselli C
      • Di Giulio P
      • Toscani F
      • et al.
      Home palliative care for terminal cancer patients: a survey on the final week of life.
      an article recounting 3 case reports,
      • Truog RD
      • Berde CB
      • Mitchell C
      • Grier HE
      Barbiturates in the care of the terminally ill.
      and another article detailing the results of an international mail survey
      • Chater S
      • Viola R
      • Paterson J
      • Jarvis V
      Sedation for intractable distress in the dying-a survey of experts.
      (Table 1).
      Table 1Clinical Experience With Palliative Sedation (PS)
      No. of patientsSymptoms indicating PSMedications used for PSLength of survival with PSReference
      120 (52.5% received PS)Pain, dyspnea, delirium, emesisOpioids, strong tranquilizers (diazepam, chlorpromazine, haloperidol)Median length, 25 d (23 d without PS)
      • Ventafridda V
      • Ripamonti C
      • De Conno P
      • Tamburini M
      • Cassileth BR
      Symptom prevalence and control during cancer patients' last days of life.
      100 (16% received PS)Pain, deliriumNo specific medications mentioned; morphine, diacelylmorphine, hydromor-phonc, codeine, methadone, levorphanol tartrate used for overall managementNot mentioned
      • Fainsinger R
      • Miller MJ
      • Bruera E
      • Hanson J
      • Maceachern T
      Symptom control during the last week of life on a palliative care unit.
      17 (all received PS)Bone/soft tissue pain, fecal emesis, seizures, restlessness, urinary retention due to clot formation, abdominal distentionAmobarbital, thiopental2-4 d; average, 23 h
      • Greene WR
      • Davis WH
      Titrated intravenous barbiturates in the control of symptoms in patients with terminal cancer.
      3 (all received PS)Pain and removal of mechanical ventilationThiopental, methohexital, pentobarbital, fentanylHours to 2 d
      • Truog RD
      • Berde CB
      • Mitchell C
      • Grier HE
      Barbiturates in the care of the terminally ill.
      143 (48.3% received PS)Pain, dyspnea, agitation, nausea, general malaiseMidazolam, chlorpromazinc, morphine, haloperidol, diazepam, barbiturates, bromazepam, scopolamine hydrobromideAverage, 3.9 d
      • Morita T
      • Inoue S
      • Chihara S
      Sedation for symptom control in Japan: the importance of intermittent use and communication with family members.
      115 (26% received PS)Agitated delirium, pain, mental anguish, dyspnea, “other”Midazolam, haloperidol, methotrimeprazine, chlorpromazine, phenobarbilal, “other henzodiazepine”1.3 d
      • Stone P
      • Phillips C
      • Spruyl O
      • Waight C
      A comparison of the use of sedatives in a hospital support team and in a hospice.
      278 (2 received PS, discontinued in 1 patient who went home)Agitation, painMidazolam36 h for patient who died
      • Fainsinger RL
      Use of sedation by a hospital palliative care support team.
      76 (23 received PS)Delirium, dyspneaMidazolam, chlorpromazinc, “benzodiazepines”Average, 2.5 d
      • Fainsinger RL
      • Landman W
      • Hoskings M
      • Bruera E
      Sedation for uncontrolled symptoms in a South African hospice.
      71 (all received PS)Physical restlessness with or without delirium, dyspnea, pain, nausea, myoclonus, psychological distressOpioids, midazolam, propofol, chlorpromazine, haloperidol, hydroxyzine, rectal diazepam, scopolamine hydrobromideNot mentioned
      • Morita T
      • Tsunoda J
      • Inoue S
      • Chihara S
      Do hospice clinicians sedate patients intending to hasten death? J Palliat Care.
      401 (25% received PS)Not mentioned specificallyOpioids, benzodiazepines, neurolepticsApproximately 12 h
      • Pcruselli C
      • Di Giulio P
      • Toscani F
      • et al.
      Home palliative care for terminal cancer patients: a survey on the final week of life.
      Mail survey of 53 palliative care specialists (51 physicians, 2 nurse specialists)Pain, anguish, respiratory distress, agitation/delirium/confusion, fear/panic/terror, restlessness, seizures, nausea/vomiting, emotional/psychological/spiritual distress, “other”Midazolam, methotnmeprazinc, lorazepam, chlorpromazine, haloperidol, phenobarbilal. chlormethiazole, propofol, clonazepam, diacetylmorphineNot mentioned
      • Chater S
      • Viola R
      • Paterson J
      • Jarvis V
      Sedation for intractable distress in the dying-a survey of experts.
      Ventafridda et al
      • Ventafridda V
      • Ripamonti C
      • De Conno P
      • Tamburini M
      • Cassileth BR
      Symptom prevalence and control during cancer patients' last days of life.
      studied symptom prevalence during the last week of life in 120 terminally ill patients with metastatic cancer assisted by a home care team of the National Cancer Institute of Milan, Italy. Fifty-two percent of patients received PS an average of 2 days before death, with pain, dyspnea, delirium, and vomiting the primary symptoms indicating the need for sedation. Medications used for PS included opioids and strong tranquilizers such as diazepam, chlorpromazine, and haloperidol. The median length of survival was 25 days for patients receiving PS and 23 days for nonsedated patients, a difference that was not statistically significant.
      In a retrospective study, Fainsinger et al
      • Fainsinger R
      • Miller MJ
      • Bruera E
      • Hanson J
      • Maceachern T
      Symptom control during the last week of life on a palliative care unit.
      reviewed symptom control during the last week of life in 100 patients who died after admission to the Edmonton General Hospi­tal Palliative Care Unit for 6 days or longer during a 14-month period. Sixteen percent of patients received PS for 2 primary symptoms, pain and delirium. The authors also identified 2 other patients who died with inadequate symptom control and could have plausibly benefited from PS. No specific medications were identified for PS; drugs mentioned for overall management that may have been used for sedation included morphine, diacetylmorphine, hydromorphone, codeine, methadone, and levorphanol tartrate.
      In the study by Greene and Davis,
      • Greene WR
      • Davis WH
      Titrated intravenous barbiturates in the control of symptoms in patients with terminal cancer.
      they evaluated 17 terminally ill patients over a 14-year period receiving PS in their community-based urology practice in South Carolina. Symptoms indicating the need for PS included bone or soft tissue pain, abdominal distention, fecal emesis, protracted vomiting, seizures, urinary retention due to clot formation, and restlessness. Sedation ranged from 2 hours to 4 days, with amobarbital and thiopental the favored sedating agents, although 4 patients died without achieving total sedation. Of note in this compilation of case reports is the use of sedation for urinary retention due to clot formation, an uncommon and atypical indication for PS; however, unidentified concomitant symptoms may have contributed to the need for sedative pharmacotherapy.
      Truog et al
      • Truog RD
      • Berde CB
      • Mitchell C
      • Grier HE
      Barbiturates in the care of the terminally ill.
      described 3 patients in whom PS was used for uncontrolled pain and removal of mechanical ventilation. Medications used for PS included the barbiturates thiopental, methohexital, and pentobarbital and the opioid fentanyl. Interestingly, family members rebuffed PS in a fourth case, an 8-year-old girl with multisystem failure who was on a ventilator after an unsuccessful liver transplantation, because they feared there would be an impres­sion they were trying to “kill” their loved one.
      Among 143 inpatients in a Japanese hospice, Morita et al
      • Morita T
      • Inoue S
      • Chihara S
      Sedation for symptom control in Japan: the importance of intermittent use and communication with family members.
      found that 48.3% received PS. The symptoms indicating the need for PS included pain, dyspnea, agitation, nausea, and general malaise, with midazolam, chlorpromazine, morphine, haloperidol, diazepam, barbiturates, bromazepam, and scopolamine hydrobromide used for sedation. Patients died an average of 3.9 days after PS was initiated, with sedation usually occurring in the last few days of life. Similar to the use of sedation for urinary retention due to clot formation reported by Greene and Davis,
      • Greene WR
      • Davis WH
      Titrated intravenous barbiturates in the control of symptoms in patients with terminal cancer.
      this study described PS for general malaise, another atypical and somewhat dubious indication for sedation.
      Stone et al
      • Stone P
      • Phillips C
      • Spruyl O
      • Waight C
      A comparison of the use of sedatives in a hospital support team and in a hospice.
      retrospectively reviewed the charts of 115 patients dying in the Royal London Hospital and St Joseph's Hospice in the United Kingdom and noted 26% received PS (31% at the hospice, 21% at the hospital). No difference in total time of survival after admission between sedated and nonsedated patients was noted (18.6 vs 19.1 days, respectively), with survival after initiation of PS averaging 1.3 days. Symptoms that indicated the need for PS included agitated delirium, pain, mental anguish, dyspnea, and an amorphous category of symptoms described as “other.” Medications used to induce sedation included midazolam, haloperidol, methotrimeprazine, chlorpromazine, phenobarbital, and “other benzodiazepine.” Unfortunately, the retrospective nature of this review prohibited the authors from ascertaining whether patients were in fact sedated by the medications they received.
      Fainsinger
      • Fainsinger RL
      Use of sedation by a hospital palliative care support team.
      reviewed 278 patients referred during a 1-year period to an inpatient palliative care unit at the Royal Alexandra Hospital in Edmonton, Alberta. Remarkably, only 2 patients received PS for agitation and pain (both received midazolam). One patient's condition improved with treatment, and sedation was terminated. The other patient remained sedated and subsequently died. In analyzing the paucity of sedation in the Edmonton program, Fainsinger suggested the need to recognize the potentially self-fulfilling nature of PS and questioned the indiscrimi­nate and arbitrary use of sedative palliation.
      Fainsinger et al
      • Fainsinger RL
      • Tapper M
      • Bruera E
      A perspective on the management of delirium in terminally ill patients on a palliative care unit.
      retrospectively studied the charts of 23 sedated patients admitted to a South African hospice during a 1-year period. Data were collected on diverse patient characteristics, including symptoms requiring PS and medications used in the last week of life. The 23 patients assessed as sedated included 20 with delirium, 2 with delirium and dyspnea, and 1 with dyspnea alone. Fourteen were sedated with midazolam, 7 with intermittent doses of benzodiazepines, and 2 with chlorpromazine. Patients were sedated an average of 2.5 days before death, and none received parenteral hydration. The mean equivalent dose of morphine during the last week of life demonstrated a significantly higher mean for the sedated group compared with the nonsedated group, with individuals from the PS group more likely to have had their opioid dose increased. Fainsinger et al
      • Fainsinger RL
      • Landman W
      • Hoskings M
      • Bruera E
      Sedation for uncontrolled symptoms in a South African hospice.
      noted the discrepancy in the mean equivalent dose of morphine in the sedated cohort and speculated that patients with a longer duration of admission, higher average dose of opioids, and a resultant greater exposure to accumulation of metabolites were more likely to exhibit agitated behavior resulting in sedative management.
      In a Japanese palliative care unit, Morita et al
      • Morita T
      • Tsunoda J
      • Inoue S
      • Chihara S
      Do hospice clinicians sedate patients intending to hasten death? J Palliat Care.
      prospectively evaluated the use of PS in 71 sedated patients during a 9-month period. The main symptoms indicating the need for sedation included physical restlessness with or without delirium, dyspnea, pain, nausea, myoclonus, and psychological distress. Opioids, midazolam, propofol, chlorpromazine, haloperidol, hydroxyzine, rectal diazepam, and scopolamine hydrobromide were the main sedative agents administered for PS. Morita et al noted that there were no patients for whom a physician increased the dose of sedatives despite adequate symptom palliation, which contradicts an insinuation by a select group of health care professionals of a latent physician-directed intent to hasten death.
      Peruselli et al
      • Pcruselli C
      • Di Giulio P
      • Toscani F
      • et al.
      Home palliative care for terminal cancer patients: a survey on the final week of life.
      prospectively studied 401 randomly selected patients admitted to 56 palliative care units in Italy during a 6-month period. Various data were collected, including the number of patients receiving PS. Twenty-five percent received PS; however, neither the number of symptoms nor factors such as patient or family anxiety were associated with sedation. In all but 1 case, opioids, benzodiazepines, and neuroleptics were used to induce and maintain PS, with patients surviving approximately 12 hours after PS was initiated.
      Chater et al
      • Chater S
      • Viola R
      • Paterson J
      • Jarvis V
      Sedation for intractable distress in the dying-a survey of experts.
      surveyed 53 selected palliative care specialists (51 physicians, 2 nurse specialists) by mail regarding the use of PS. Respondents were from the United Kingdom, United States, Australia, Ireland, Italy, Canada, New Zealand, and South Africa. Forty-one respondents had administered PS within the last 12 months, with the reasons for sedative management including pain, anguish, respiratory distress, agitation/delirium/confusion, fear/ panic/terror, restlessness, emotional/psychological/spiritual distress, seizures, nausea/vomiting, and “other.” Remarkably, psychological symptoms accounted for one third of the cases of PS. The most common medications used for PS included midazolam, methotrimeprazine, lorazepam, haloperidol, phenobarbital, chlormethiazole, chlorpromazine, propofol, clonazepam, and diacetylmorphine. Of considerable concern is the fact that only 50% of patients and 69% of family members were reported to have major involvement in the decision-making process for sedative therapy; however, it is not known if families were involved for patients without decision-making capacity.

      DISCUSSION

      The process of dying is a ubiquitous milestone that allows patients and family members a time of reconciliation, growth, and spiritual enrichment as life enters its final chapter.
      • Rousseau P
      Dying and terminal sedation [editorial].
      • Cherny NI
      • Coyle N
      • Foley KM
      Suffering in the advanced cancer patient: a definition and taxonomy.
      Lamentably, it can also be a time of consider­ able suffering, abolishing hope of a serene death, precipitating physical and emotional anguish, and fomenting requests for physician-assisted death. When palliative interventions have been exhausted and symptoms remain refractory, many palliative care clinicians consider PS a valuable therapeutic adjunct that contributes to a peaceful, comfortable, and dignified death.
      • Rousseau P
      Dying and terminal sedation [editorial].
      Unfortunately, data regarding the use of PS are incomplete, controversial, and disparate, in part secondary to various definitions of PS and in part secondary to the retrospective nature of studies and anecdotal case reviews. Moreover, the use of sedative pharmacotherapy in nonmalignant diseases such as chronic obstructive lung disease and congestive heart failure have not been adequately studied and recounted, further confounding prevalence data on the incidence and use of PS.
      The ethical validity of PS has been an arguable issue with Quill et a1
      • Quill TF
      • Lo B
      • Brock DW
      Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia.
      ; however, the basis of sedative therapy is the intent of the intervention, which is to relieve suffering. Although PS may hasten death, its use is generally reserved for patients near death, except in select cases of psychological suffering. Psychological suffering is certainly a tenuous area regarding sedation but, as mentioned previously, can be just as distressful and refractory as physical suffering.
      Quill et al
      • Quill TF
      • Lo B
      • Brock DW
      Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia.
      in effect support PS when they affirm that the time between the initiation of PS and death permits second-guessing and reassessment by health care providers and family members. They also argue that the notion that PS is “letting nature take its course” is contentious, because patients often die of dehydration and the withholding of fluids. On the other hand, most patients who receive PS have already stopped eating or drinking, thus negating the argument of PS-induced death due to fluid deprivation and/or starvation. Although the issue of discontinuing tube feedings prior to PS can be complex and should be discussed with the patient or the patient's surrogate, if such feedings are requested, they can continue during administration of sedation, as the intent of PS is not to hasten death but rather to relieve suffering. Many patients receiving enteral nutrition have had tube feedings withdrawn prior to initiation of PS. Quill et al
      • Quill TF
      • Lo B
      • Brock DW
      Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia.
      also argue that PS could be initiated without explicit discussions with alert patients who appear to be suffering intolerably and that PS is closer to physician-assisted suicide and euthanasia than is ordinarily acknowledged. As with any therapeutic intervention, paternalistic nuances must be precluded and informed consent obtained, especially with the use of PS. The suggestion of Quill et al
      • Quill TF
      • Lo B
      • Brock DW
      Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia.
      that PS could be instituted without overt discussion with patients or their surrogates is obviously a possibility that can occur with any medical intervention, but such an action is contrary to ethical and moral medical practice and should be prosecuted if and when discovered. Nevertheless, the concern of Quill et al
      • Quill TF
      • Lo B
      • Brock DW
      Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia.
      regarding the absence of informed consent is valid. Such potential should not encumber the use of PS to alleviate interminable suffering and afford a tranquil death.
      Finally, in the case Vacco v Quill,

      Vacco v Quilt, No. 117 SCt 2293 (1997).

      Quill's argument that the freedom to refuse or withdraw life-sustaining therapy is analogous to physician-assisted suicide places considerable substance on the outcome of therapy, completely discounting the value of intent. The Supreme Court, in a majority opinion, disagreed with such an argument and reaffirmed the importance of intent. In fact, the Court advocated the use of terminal sedation to palliate refractory symptoms and expectantly preclude a request for physician-assisted suicide.
      The concern and trepidation that PS is hastily prescribed without adequate evaluation and treatment of underlying conditions, particularly as it relates to terminal agitation and delirium, merit further discussion. Fainsinger et al
      • Fainsinger RL
      • Tapper M
      • Bruera E
      A perspective on the management of delirium in terminally ill patients on a palliative care unit.
      have argued for a thorough evaluation of cognitive impairment and agitation in terminally ill patients and have reduced the use of PS to 3% of inpatients on their palliative care unit secondary to disciplined assessments for causes of cognitive deterioration, treatment of reversible causes, and the use of hydration and opioid rotation. Moreover, Fainsinger et aL
      • Fainsinger RL
      • Tapper M
      • Bruera E
      A perspective on the management of delirium in terminally ill patients on a palliative care unit.
      note that a retrospective review of 117 and 162 patient charts before and after implementing a standard practice of assessment and management of cognitive impairment demonstrated a decline in the prevalence of agitated and impaired mental status from 26% to 10%, with an associated reduction in frequency and dose of neuroleptic and benzodiazepine medication use. In addition, another study of patients in the same palliative care unit
      • Pereira J
      • Hanson J
      • Bruera F
      The frequency and clinical course of cognitive impairment in patients with terminal cancer.
      revealed that 25 of 240 patients admitted with cognitive failure were found to have reversible causes, with a subsequent improvement in findings on the Mini-Mental State Examination. Clearly, such data are conspicuous and warrant further studies of cognitive impairment and agitation, since agitated behavior, variously described as delirium, terminal restlessness, mental anguish, and agitation,
      • Fainsinger RL
      • Landman W
      • Hoskings M
      • Bruera E
      Sedation for uncontrolled symptoms in a South African hospice.
      is a recurring symptom that may indicate the need for sedation.

      CONCLUSION

      To care for a dying person is an honor and privilege afforded few in our society. As physicians, our main goal is to reduce suffering, including pain, physical symptoms, and emotional, psychosocial, and spiritual distress. In refractory and intractable cases, PS offers a compassionate and humane alternative to conscious and continual suffering, both for the patient and the patient's family.
      • Rousseau P
      Dying and terminal sedation [editorial].
      Without a doubt, further studies are necessary, particularly in cognitive impairment, but PS is a valuable and efficacious intervention for interminable suffering at the end of life. And to those who question the medical, ethical, and moral validity of PS, the following quotation from Henry Fielding's Amelia should cause personal reflection on caring for the dying: “It hath often been said that it is not death, but dying which is terrible.”
      I will tell him, the doctor, that he must think of something else. It is impossible, to go on like this.
      Leo Tolstoy, The Death of Ivan Illych

      REFERENCES

        • Jonsen AR
        • Siegler M
        • Winslade WJ
        Clinical Ethics: A Practical Approach to Ethical Decisions in Clinical Medicine. 4lh ed. McGraw-Hill, New York, NY1998
      1. Vacco v Quilt, No. 117 SCt 2293 (1997).

      2. Washington v Glucksberg, No. 117 SCt 2258 (1997).

        • Rousseau P
        Terminal sedation in the care of dying patients.
        Arch intern Med. 1996; 156: 1785-1786
        • Fainsinger RL
        • Landman W
        • Hoskings M
        • Bruera E
        Sedation for uncontrolled symptoms in a South African hospice.
        J Pain Symptom Manage. 1998; 16: 145-152
        • Hallenbeck J
        Terminal sedation for Intractable distress: not slow euthanasia but a prompt response to suffering.
        West J Med. 1999; 171: 222-223
        • Cherny NI
        • Portcnoy RK
        Sedation in the management of refractory symptoms: guidelines for evaluation and treatment.
        J Palliat Care, Summer. 1994; 10: 31-38
        • Cherny NI
        Commentary: sedation in response to refractory existential distress: walking the fine line.
        J Pain Symptom Manage. 1998; 16: 404-406
        • Quill TF
        • Lo B
        • Brock DW
        Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia.
        JMAM. 1997; 278: 2099-2104
        • Quill TE
        • Dresser R
        • Brock DW
        The rule of double effecta critique of its role in end—of-life decision making.
        N Engt J Med. 1997; 337: 1768-1771
        • Garcia ILA
        Double effect.
        in: Reich WT Rev ed. Encyclopedia of Bioethics. Vol. 2. Simon & Schuster Mac-millan, New York, NY1995: 636-641
        • Kuhse H
        Tie Sanctity-of-Life Doctrine in Medicine: A Critique. Garenden Press, Oxford, Kngland1987
        • Beauchamp TL
        • Childress J
        Principles of Biomedical Ethics. 3rd ed. Oxford University Press, New York, NY1989
        • Rousseau PC
        Ethical concepts in hospice and palliative medicine.
        in: Schonwelter RS Hawke W Core Curriculum in Hospice and Palliative Medicine. Kendall/Hunt Publishing, Dubuque, Iowa1999: 153-158
        • Ventafridda V
        • Ripamonti C
        • De Conno P
        • Tamburini M
        • Cassileth BR
        Symptom prevalence and control during cancer patients' last days of life.
        J Palliar Care. Autumn 1990; 6: 7-11
        • Fainsinger R
        • Miller MJ
        • Bruera E
        • Hanson J
        • Maceachern T
        Symptom control during the last week of life on a palliative care unit.
        J Palliai Care. Spring 1991; 7: 5-11
        • Greene WR
        • Davis WH
        Titrated intravenous barbiturates in the control of symptoms in patients with terminal cancer.
        South Med J. 1991; 84: 332-337
        • Morita T
        • Inoue S
        • Chihara S
        Sedation for symptom control in Japan: the importance of intermittent use and communication with family members.
        J Pain Symptom Manage. 1996; 12: 32-38
        • Stone P
        • Phillips C
        • Spruyl O
        • Waight C
        A comparison of the use of sedatives in a hospital support team and in a hospice.
        Palliat Med. 1997; 11: 140-144
        • Fainsinger RL
        Use of sedation by a hospital palliative care support team.
        J Palliat Care. 1998; 14 (Spring): 51-54
        • Morita T
        • Tsunoda J
        • Inoue S
        • Chihara S
        Do hospice clinicians sedate patients intending to hasten death? J Palliat Care.
        Autumn. 1999; 15: 20-23
        • Pcruselli C
        • Di Giulio P
        • Toscani F
        • et al.
        Home palliative care for terminal cancer patients: a survey on the final week of life.
        Palliat Med. 1999; 13: 233-241
        • Truog RD
        • Berde CB
        • Mitchell C
        • Grier HE
        Barbiturates in the care of the terminally ill.
        N Engl J Med. 1992; 27: 1678-1682
        • Chater S
        • Viola R
        • Paterson J
        • Jarvis V
        Sedation for intractable distress in the dying-a survey of experts.
        Palliat Med. 1998; 12: 255-269
        • Fainsinger RL
        • Tapper M
        • Bruera E
        A perspective on the management of delirium in terminally ill patients on a palliative care unit.
        J Palliai Care Autumn. 1993; 9: 4-8
        • Rousseau P
        Dying and terminal sedation [editorial].
        Clin Geriatr. January 1999; 7: 19-20
        • Cherny NI
        • Coyle N
        • Foley KM
        Suffering in the advanced cancer patient: a definition and taxonomy.
        J Palliat Care Summer. 1994; 10: 57-70
        • Pereira J
        • Hanson J
        • Bruera F
        The frequency and clinical course of cognitive impairment in patients with terminal cancer.
        Cancer. 1997; 79: 835-842