Communication between patients,

their relatives, and health care

staff is very important when

administered palliative sedation.

Photo courtesy of Lisa Scholder. Organic Walk, 16" × 24".

Palliative Sedation in Patients With Cancer Marco Maltoni, MD, and Elisabetta Setola, MD

Background: Palliative sedation involves the use of sedative medication to relieve refractory symptoms in patients by reducing their level of consciousness. Although it is considered an acceptable clinical practice from most ethical points of view, palliative sedation is still a widely debated procedure and merits better un- derstanding. Methods: The relevant medical literature pertaining to palliative sedation was analyzed and reviewed from various technical, relational, and bioethical perspectives. Results: Proportionate palliative sedation is considered to be the most clinically appropriate modality for performing palliative sedation. However, guidelines must be followed to ensure that it is performed correctly. represent the first therapeutic option and careful monitoring of dosages is essential to avoid oversedation or undersedation. Conclusions: Proportionate palliative sedation is used to manage and relieve refractory symptoms in patients with cancer during their last days or hours of life. Evidence suggests that its use has no detrimental effect on survival. A different decision-making process is used to manage the withdrawal of hydration than the process used to determine whether proportionate palliative sedation is appropriate. Communication between patients, their relatives, and the health care staff is important during this medical intervention.

Introduction sort in the final phase of a patient’s life. However, Within the context of palliative , the prac- terminal sedation is a confusing and inappropriate tice of drug-induced sedation for symptom control term because it implies that the practice is designed — the use of all possible antisymptomatic treatments to shorten life; thus, more appropriate terminology notwithstanding — is called palliative sedation. His- was needed and, hence, the term palliative sedation. torically, palliative sedation was often known as ter- Its implementation into everyday clinical practice has minal sedation because it was considered a last re- led to the use of the term in the majority of studies focusing on this topic.1 From the Unit (MM), Istituto Scientifico Romagnolo By reducing the level of consciousness, palliative per lo Studio e la Cura dei Tumori, Meldola, Italy, and the Oncology sedation uses sedative medication to control refractory Unit (ES), Saint Giuseppe Hospital, Milan, Italy. symptoms in patients with cancer.2 As this approach Submitted April 15, 2015; accepted July 15, 2015. Address correspondence to Marco Maltoni, MD, Palliative Care became more widespread and was acknowledged as Unit, Istituto Scientifico Romagnolo per lo Studio e la Cura dei an important part of palliative care, procedural princi- Tumori, Via P. Maroncelli 40, 47014 Meldola, Italy. E-mail: ples were needed to avoid malpractice. The guidelines [email protected] created clarified the definition and practice of pallia- No significant relationships exist between the authors and the companies/organizations whose products or services may be ref- tive sedation, also underlining its inherent and intrin- erenced in this article. sic difference from .3-10 Various guidelines

October 2015, Vol. 22, No. 4 Cancer Control 433 state that sedation in palliative care is a therapy mainly den procedure (see Table 1). Choice of sedation has a performed in the last days or hours of life, with an un- clinical basis and is oriented toward different clinical ambiguous aim, a precise procedure, and well-defined needs: rapid palliative sedation for catastrophic and results — all of which are in contrast to those of eutha- acute symptoms compared with proportional sedation nasia.9-12 In addition, use of palliative sedation does not for progressively worsening symptoms. Thus, imple- hasten death when conducted in accordance with stan- menting palliative sedation is based on the suddenness dard guidelines.13-15 of symptom appearance. Different methods of palliative sedation exist, Methods and Types namely, the proportional method of palliative seda- Some adjunctive characteristics of palliative sedation tion (also called proportionate palliative sedation), have been included in more detailed definitions of the and sustained, deep and continuous palliative se- practice.16 For example, the definition of palliative se- dation from the onset, regardless of the intensity of dation as the “intentional administration of sedative symptoms.17-21 The latter suggests that relational con- drugs in dosages and in combinations required to re- tinuity between patients and relatives is not neces- duce the consciousness of a terminal patient as much sarily a value, that the concept of hastening or not as necessary to adequately relieve one or more refrac- hastening death is irrelevant (given the patient’s con- tory symptoms,” considers a number of specific param- dition), and that the process of drug titration may de- eters, including short-term prognosis, proportionality of lay the clinical effect of sedation. We consider that the intervention, effectiveness of the sedation through the supporters of this view are in favor of adminis- adequate monitoring, and refractoriness of symptoms.16 tering drugs in a “standardized” way, which leads to This definition also implies that palliative sedation is not deep sedation, and interrupting all forms of hydra- a fixed intervention; rather, it is a dynamic process that tion and nutrition. Taking this concept to an extreme, can be adapted to the needs of the patient.16 many patients in this setting would be candidates for Palliative sedation can vary in terms of depth of deep continuous sedation or continuous sedation un- sedation and correlated level of unconsciousness, til death. In our opinion, those in favor of using pal- continuity, drugs used, and rapidity of implementa- liative sedation in this way do not consider that this tion (Table 1). In that sense, palliative sedation is not view is a departure from the original practice; indeed, necessarily deep continuous sedation or continuous such an approach could be counterproductive to its sedation until death, which is the last step of a pro- large-scale implementation. gressive process. By contrast, some proportional palliative seda- tion supporters view proportionality as a fundamental “Proportionality” Theme characteristic of palliative sedation and maintain that When symptom-guided, dose-titrated, and result-as- relational continuity is an important issue.6,14 They also sessed, palliative sedation can be a progressive or sud- argue that the effects of palliative sedation on refrac-

Table 1. — Definitions of Terms Used in Palliative Sedation Term Definition Depth Superficial Patient conserves (at least partially) the ability to communicate with family or caregivers Deep Patient enters a state of total unconsciousness Continuity Intermittent Sedation is performed for a limited period of time for a specific symptom and discontinued if a reduction (albeit slight) in the distress caused by the symptom has been obtained Continuous Drugs are administered without interruption to obtain a persistent effect, which more frequently occurs near death and in severe cases Use of Drugs Primary palliative sedation Sedatives are used to lower the level of consciousness Secondary palliative sedation Dosage of drugs primarily used for symptom control (ie, morphine for pain or dyspnea) is increased to make use of an adverse event (eg, somnolence) to reduce the level of consciousness (not recommended) Rapidity of Administration and Effect Progressive Most common method in which the dose of sedative drugs is monitored and modified according to patient needs Sudden, rapid intervention method Also called “emergency” sedation; required for a catastrophic event such as massive bleeding, severe dyspnea, agitated , or acute pain

434 Cancer Control October 2015, Vol. 22, No. 4 tory symptoms must be carefully monitored to avoid been explored.5 In the event of insufficient psy- undersedation or oversedation and that the drug dos- chological support, the palliative care team may age should be individually titrated to the minimum be at risk of burn out and, thus, might perform effective dose level.6,14 Not every palliative sedation is unnecessary palliative sedation36 deep continuous sedation or continuous sedation until • Adoption of wider or narrower definition of pallia- death, and the latter approach is required in patients tive sedation may influence prevalence for whom a lower level of sedation is insufficient to However, the real frequency of palliative sedation control refractory symptoms.7,22-25 A survey carried out has also been suggested to vary: possibly between 25% by Putman et al26 on intention and practice in US physi- and 35% of all patients admitted to a or pallia- cians reported that many of these physicians preferred tive care unit.14,15 In home-care hospice programs, pal- the proportional method, because they did not wish to liative sedation may be a feasible option. Mercadante pursue full unconsciousness in patients without moni- et al37 reported that 13.2% of patients with cancer were toring the effect of such sedation. sedated in their homes. Moreover, some of the first articles published on palliative sedation were based on the use of a Symptoms and Refractoriness proportional, clinical, symptom-guided form of se- Management of refractory symptoms is the single indi- dation.5,6,11,27,28 This form of palliative sedation is ac- cation for use of palliative sedation.14,27 The definition cepted by most ethical points of view, cultural per- of a refractory symptom, which was first proposed by spectives, and procedural guidelines.8-10,28-32 ten Cherny and Portenoy,27 is still widely accepted: “Symp- Have and Welie12 used the term mission creep to de- tom for which all possible treatment has failed, or it is scribe the intention of shortening life — however estimated that no methods are available for palliation small — by using deep sedation, concluding that such within the time frame and the risk:benefit ratio that the an approach cannot be considered palliative sedation. patient can tolerate.”5,6,8 The concept of refractoriness In a clinical context, mission creep occurs when a has been associated with that of unbearable suffering, specific practice considered appropriate in a given sit- multidisciplinary evaluation, and a prognosis of short- uation is gradually extended for use with different in- term survival.7 dications, groups of patients, and different intentions. Prior to offering sedation, all other available medi- cations, procedures, and interventions should be con- Prevalence and Settings sidered and offered as appropriate. When possible, The debates about the characteristics and how to discussions should take place with patients, their rela- define palliative sedation are reflected in the wide- tives, and their caregivers so that an agreement can be ranging prevalence of the practice observed in the lit- made about the action to be taken.7-10 All dimensions erature (1%–88%), 8-10,13 which can be attributed to a of the symptom (physical, psychological, social, emo- number of factors: tional, existential/spiritual) must be taken into ac- • Different health care settings count and help offered for each component.7-10 Physical Higher prevalence in inpatient acute tertiary palli- symptoms that are most prone to refractoriness include ative care units compared with home-care hospice delirium and dyspnea; both are frequently present as programs33,34 death approaches, and their timely management is cru- • Different case mixes in similar settings cial. Pain and emesis may also become refractory, al- In a comparison between 2 Italian , a sta- beit this occurs less frequently.14,38-41 Palliative sedation tistically significant difference in prevalence of sedation initially observed in the 2 populations disappeared when the groups were corrected for age and duration of stay in hospice. This was inter- Delirium 416 (54%) Dyspnea preted as an indication that younger patients and 234 (30%) Psychological 151 (19%) those admitted for acute symptoms required more distress Pain 135 (17%) palliative sedation interventions than older patients Vomiting 37 (5%) 35 admitted to the hospice for psychosocial reasons Other (eg, itching, 30 (4%) bleeding) • Degree of adherence to palliative sedation guidelines Refractory Symptoms 0 100 200 300 400 500 or level of competence and expertise of health care Patients professionals Little experience in managing difficult symptoms Fig 1. — Main refractory symptoms requiring palliative sedation may increase the prevalence of palliative seda- in 774 patients from 10 studies with a total of 1,003 symptoms tion. The decision to initiate palliative sedation (some patients had > 1 refractory symptom). From Maltoni M, Scarpi E, Rosati M, et al. Palliative sedation in end-of-life care and survival: a systematic should be taken together with a palliative care review. J Clin Oncol. 2012;30(12):1378-1383. Reprinted with permission. specialist to ensure that all other options have © 2012. American Society of Clinical Oncology. All rights reserved.

October 2015, Vol. 22, No. 4 Cancer Control 435 is occasionally used for intractable seizures or terminal Drugs hemorrhage (Fig 1).14,42 The most widely used drug for palliative sedation in Psychological distress and existential suffering are the context of palliative care is , which complex and challenging. These terms encompass is- is a prescribed in 9 of the 11 stud- sues such as meaninglessness of life, sense of hope- ies evaluated (Fig 2).3,14 Midazolam is prescribed in lessness, perception of self as a burden to others, feel- a wide dose range; for example, in 1 study, 61% of ing dependent on others, feeling isolated, grieving, patients used doses no more than 30 mg and 8% of loss of dignity and purpose, fear of death of self, or patients used doses of at least 120 mg.47 The highest fear of the unknown. Multidimensional management final daily dose of midazolam was correlated with directed at the physical, psychological, and existential age (the younger the patient, the higher the dose) aspects is recommended, with support provided by and treatment duration (the longer the treatment, the psychologists, psychiatrists, chaplains, ethicists, or pal- higher the dose).47 Midazolam has a rapid onset of liative care specialists.3-10 action and short half-life, and both of these pharma- Psychological distress and existential suffering cokinetic features facilitate the titration procedure have many peculiarities.5,8,27 Psychological distress can during the first phase of sedation. When an ade- occur at any time during the course of a disease, and quate minimum dose (or, in some settings, a loading supportive interventions, psychological interventions, dose) has been identified for symptom control, a or both types of interventions may be ineffective and maintenance dose should be started by continuous do not completely resolve the problem; however, this intravenous or subcutaneous infusion. Because mid- does not mean that the distress is refractory.43 Rather, azolam has anticonvulsant, muscle-relaxant, hypnot- psychological distress may not have a progressive ic, and anxiolytic properties, it can be added to other course similar to that associated with physical symp- sedative classes to achieve control of these specific toms; its course is often unpredictable. For these rea- symptoms. sons, frequent meetings with the palliative care team Some studies advocate use of neuroleptics for de- are needed; the same may be true of spiritual assis- lirium (eg, haloperidol).35-37 Haloperidol has less se- tance. Intermittent or relief sedation, rather than con- dating power than midazolam, so it is typically used tinuous sedation, may also be beneficial.5,8,27 to attenuate delirium; thus, it is not the most appro- Some authors have focused on the search for early priate choice of drug for achieving continuous seda- determinants for continuous palliative sedation so that tion. Other drugs used for palliative sedation have in- patients at risk can be identified in a timely manner cluded levomepromazine, chlorpromazine (especially and their symptoms managed by other strategies.33,44-46 when profound sedation is needed for acute agitated Information on determinants that more frequently delirium), propofol, ketamine, and dronabinol.39,48-51 lead to use of palliative sedation could also be used If a patient undergoing sedation is treated with an to assess the need for advanced care planning. How- for pain or dyspnea, then the opioid must not ever, study results have suggested that implementing be interrupted. By contrast, should not be palliative sedation is more often linked to the attitudes used for sedation, because doing so would make use and beliefs of physicians and to organizational settings of their secondary effect (somnolence) and would be than to the clinical needs of patients.44 Other studies considered secondary palliative sedation, which is found a correlation between palliative sedation and not recommended.3,5,8-10 higher doses of opioids used prior to implementing palliative sedation as well as between palliative seda- tion and younger patient age.33,45,46 Midazolam 362 (49%) Haloperidol 191 (26%) Duration Chlorpromazine 107 (14%) The mean or median length of palliative sedation Morphine 79 (11%) 12 Methotrime- ranges from 0.8 to 12.6 days. One study observed 22 (3%) Drug prazine an even larger range (0–43 days), despite mean Propofol 11 (1%) and median times in line with the literature (4 and Phenobarbital 5 (1%) 2 days, respectively).13 In this study, 10.8% of pa- Other benzodiazepines 161 (22%) tients underwent palliative sedation for more than 0 100 200 300 400 13 10 days and 3.4% of patients for more than 20 days. Patients Patients requiring sedation for more than 10 days had fewer rates of delirium and dyspnea, milder and Fig 2. — Sedative drugs administered to 745 patients from 9 studies. From more frequent use of secondary palliative sedation, Maltoni M, Scarpi E, Rosati M, et al. Palliative sedation in end-of-life care and survival: a systematic review. J Clin Oncol. 2012;30(12):1378-1383. and higher rates of psychological distress than their Reprinted with permission. © 2012. American Society of Clinical Oncology. counterparts.13 All rights reserved.

436 Cancer Control October 2015, Vol. 22, No. 4 Monitoring of palliative sedation on quality of life and participant Guidelines for palliative sedation recommend close well-being.15 None of the 14 studies included in the re- surveillance of patients with respect to the manage- view took any of the above issues into consideration.15 ment and relief symptoms and suffering, depth of se- The impact on symptoms was partially reported and dation (level of consciousness), and potential adverse with different methods, thus making data pooling im- events of sedation.3-10 Family members of patients and possible.15 Furthermore, the studies reviewed had nu- their health care team should be monitored for psycho- merous biases, with the authors concluding that the logical and spiritual distress.3,5,6,8 Vital signs should be data are insufficient and the evidence is of poor qual- assessed in nonimminently dying patients undergoing ity with regard to the qualitative effectiveness of pal- short-term, intermittent, or light sedation. liative sedation.15 They also recommended that studies Evaluating symptom relief or relief of suffering on palliative sedation focus on qualitative, rather than in unconscious patients can be difficult. Some scores quantitative, end points (ie, length of survival from the monitoring verbal or facial expression, body move- start of palliative sedation).15 ments, and response to nonpainful stimuli may be useful when deep sedation is performed, and some au- Impact on Survival thors have suggested using the level of sedation as a For some, the impact of palliative sedation on survival proxy for evaluating symptoms.3,25,31 is not an issue that merits much attention, given that Depth of sedation is a common theme in studies the priority of palliative sedation in the end-of-life set- and the tools used to assess it can vary. For example, ting is quality of life; thus, even a detrimental effect guidelines from the European Association for Pallia- on survival may not be considered of particular rele- tive Care recommend the Critical-Care Pain Observa- vance.59 By contrast, some physicians may consider it tion Tool or the Richmond Agitation-Sedation Scale fundamental to know whether the sedation proposed (RASS).52,53 A prospective study performed by Arevalo has a high, medium, or low risk of hastening the death et al54 evaluated the validity and reliability of 4 scales: of their patient.14 However, this question cannot be the Minnesota Sedation Assessment Tool, the RASS, studied with the highest evidence-based methodology. the Vancouver Interaction and Calmness Scale, and Although the authors of the Cochrane review were un- a sedation score proposed by the Royal Dutch Medi- able to pool data on qualitative outcomes of palliative cal Association. The RASS was as reliable as the score sedation, they did provide reasonably good, quantita- proposed by the Royal Dutch Medical Association tive evidence that palliative sedation does not have a and study results claimed that the RASS was the least detrimental impact on survival.15 The authors looked time consuming, clearest, and easiest to use of the at 14 studies involving 4,167 adults — nearly one-third 4 tools.54,55 Another study tested a modified form of of whom were sedated.15 Survival time measured from RASS (RASS-PAL) for use in patients in the palliative admission or referral to death was not statistically sig- care unit, observing that the tool was useful for assess- nificant between those who were sedated and those ing sedation; however, further validation studies are who were not.15 A prospective study of patients admit- needed to confirm these results.56 ted to hospices matched patients who were sedated More objective methods to evaluate depth of se- and those who were not for sex, age, reason for admis- dation have been proposed because some sedated pa- sion, performance status, and prognostic score.13 Sur- tients may continue to experience symptoms without vival from admission to death was compared between being able to communicate their distress.57,58 Deschep- the 2 groups and no detrimental effect of palliative se- per57 proposed a mixed-evaluation method that com- dation on survival was found.13 Of note, even intensive bined use of a scale, a subjective assessment by profes- procedures, deep continuous sedation or continuous sionals, as well as neuroimaging, electrophysiological sedation until death, did not hasten death when per- techniques, or both. However, such an approach is formed in a proportional, step-by-step manner.13 A impractical in end-of-life care and may be more suit- subsequent systematic review also identified 11 studies ed to a research setting.57 A systematic review of the totalling approximately 2,000 patients.14 No difference literature revealed that 5 clinical studies and 1 guide- in survival was seen between patients receiving hos- line-based article included use of a validated scale to pice care who underwent sedation (median, 7–27 days) evaluate the results of palliative sedation.58 During the and those who did not (median, 4–40 days).14 titration phase of sedation, clinical parameters should Such results confirm that palliative sedation can be evaluated every 15 to 30 minutes, but the exact in- be considered a legitimate clinical intervention from terval should be calculated on the basis of the drugs, most ethical viewpoints and that the principle of dou- doses used, and clinical conditions of the patient; once ble effect is unnecessary to justify its practice. This an appropriate level of sedation has been achieved, the ethical criterion is appropriate in a small percentage frequency of assessment can be reduced to once every of patients receiving sedation in whom respiratory or 24 hours.3 One Cochrane review evaluated the impact circulatory function depression is recorded following

October 2015, Vol. 22, No. 4 Cancer Control 437 palliative sedation (3%–4%) and hypothesized as hav- terruption of artificial nutrition and hydration — both ing potentially hastened death.60 In that sense, it may of which are considered medically assisted treatments be appropriate to consider this effect a serious adverse or as vital support interventions, at least in this patient event, which could occur after any medical or surgical setting. However, medically assisted hydration is a top- procedure. ic more often discussed, given that artificial nutrition may be interrupted prior to palliative sedation due to Ethical Considerations its proven lack of impact on the duration and quality of If a survival impact of palliative sedation was ever life of patients with a terminal disease.61-66 The role of demonstrated, then palliative sedation might have hydration in end-of-life care is subject to different per- been equated with “slow” or “soft” physician-assisted spectives and is widely discussed in the literature.64 In suicide; however, absence of a detrimental impact on certain situations, excess hydration can lead to water survival is sufficient to distinguish palliative sedation retention and exacerbation of pleural or peritoneal ef- from physician-. The European Associ- fusion, among other problems; by contrast, interrupt- ation for Palliative Care has identified 3 areas in which ing liquids may provoke thirst and an increase in drug palliative sedation differs from euthanasia11: metabolites, worsening delirium, and agitation, thus • The intention of palliative sedation is to provide making it difficult for patients to communicate dis- relief from unbearable suffering caused by a re- tress.65 Thus, maintenance or withdrawal of hydration fractory symptom, whereas euthanasia is de- must be individually evaluated and managed. signed to end the life of someone who is suffering. Results from a large epidemiological study on • Palliative sedation is a proportionate and symp- 20,480 questionnaires completed by physicians expe- tom intensity–guided procedure using the mini- rienced in deep continuous sedation or continuous se- mum useful dose of a sedative drug (eg, a ben- dation until death showed that, even in this “extreme” zodiazepine); it is individually oriented and form of sedation, up to two-thirds of patients contin- monitored. Conversely, euthanasia uses neuro- ued hydration for clinical reasons.66 The take-home muscular relaxants and barbiturates and is not message is that use of palliative sedation and evalua- monitored on the basis of the symptom relief ob- tion of whether or not to continue hydration involve tained.61,62 2 different decision-making processes. Thus, hydration • Success of palliative sedation is measured in terms does not have to be automatically stopped because pal- of the relief it provides from distress. liative sedation has begun. Although palliative sedation does not play a role in hastening death, its use prevents many patients from Communication and Decision-Making maintaining verbal contact with their relatives, which Process makes it unique among medical interventions.63 Inju- Talking about death with patients and their families is dicious uses of palliative sedation include inadequate a delicate and challenging aspect of palliative care. Ide- assessment of potentially reversible causes of severe ally, discussions of this kind should be initiated when symptoms, lack of involvement of specialists in the a patient’s prognosis is years to months and then re- condition underlying the presenting symptoms, exces- evaluated when the patient’s life expectancy decreas- sive use by overwhelmed health care professionals, es to months to weeks.9 The conversation should be a performing the procedure at the request of the rela- balance between maintaining hope and realistic and tives, and untimely start (ie, too early, too late).6,12,60 achievable goals.6,67 Health care professionals must Although some may view palliative sedation, deep plan these end-of-life interventions by taking into ac- continuous sedation, or continuous sedation until count the goals, values, needs, and wishes of the pa- death as bringing patients to a sort of “living dead-like” tient. Patients should be asked where they prefer state in which they are no longer considered persons, to die so that adequate plans can be made to respect it is our opinion that patients who are sedated can be their request. All decisions should be documented in talked to, wanted, cared for, and loved by their rela- the patient’s medical records.9 When a health care pro- tives, sometimes “answering” with their presence in fessional recognizes the imminence of suffering, he surprising and unexpected ways.22,28 The “innate” core or she should discuss the possibility of sedation with of personhood (“individual component”) has been sug- the patient, outlining the aims, risks, and benefits of gested to persist even after relational and societal com- the procedure. The patient may not want to broach the ponents are severely reduced.29 subject and may authorize the health care professional to make decisions on his or her behalf or delegate a Nutrition and Hydration caregiver to do so. If the patient is no longer capable of Those who believe that palliative sedation plays a role making an autonomous decision, then the health care in hastening death may place most of the responsibil- professional may approach a family member or a sur- ity for this “secondary effect” on the simultaneous in- rogate decision maker to clarify the patient’s wishes.6

438 Cancer Control October 2015, Vol. 22, No. 4 A multidisciplinary team of oncologists, palliative patients, a Portuguese case study determined that the care specialists, psychologists, psychiatrists, nurses, decision to begin sedation was made by health care chaplains, or spiritual advisors, among others, may professionals alone in 21 cases.38 The authors attrib- also form the decision-making group, together with uted this result to the fact that 76% of study patients the patient and his or her family.5 In particular, such required urgent and nonpostponable sedation.38 a team can help ensure that all conditions have been Nurses also play an essential role in the decision- met for palliative sedation, that informed consent making process. A survey of 576 nurses from The is obtained, that the type of intervention has been Netherlands showed that most responders had cared planned, and then modify any plans, when and if nec- for at least 1 patient who received palliative sedation.73 essary, during the course of sedation. Further thera- Nearly all of the nurses indicated that they had been peutic interventions may also be performed if other involved in the decision to perform palliative sedation conditions arise (eg, urinary retention, constipation, and were present when palliative sedation was started, myoclonus, “death rattle”). unlike the physicians who were present in fewer than Keeping an open dialog with relatives and the 50% of cases.73 A study of Flemish nurses also revealed patient’s loved ones throughout the palliative seda- that the majority of the nurses supported use of pal- tion process is fundamental so that they understand liative sedation for refractory symptoms.74 Patel et al75 the characteristics of this medical intervention (ie, it studied the attitudes and perspectives of nurses work- is performed in the absence of other means to man- ing with patients receiving palliative sedation in 3 dif- age refractory and unbearable symptoms; it neither ferent settings (oncology, intensive care, and hospice), hastens nor postpones death; it reduces or eliminates highlighting their ability to define palliative sedation, the possibility of verbal communication with patients; their skill set for administering palliative sedation, pol- clinical conditions permitting, it is reversible in na- icy, and procedural guidelines, and their education on ture; it is carefully monitored to avoid oversedation palliative sedation and end-of-life care. or undersedation; it is performed for the well-being of patients).68,69 Guidelines Although some relatives experience psychological Many clinical associations and international agencies distress, Bruinsma et al69 reported that most are com- have produced procedural guidelines for use of pallia- fortable with palliative sedation if they are appropriate- tive sedation so as to provide health care professionals ly informed about the procedure and actively involved with a framework on which to base the clinical deci- in the decision-making process. A study of Swiss care- sion-making process.3-10 Key issues in palliative seda- givers of patients who died while undergoing palliative tion were also addressed in a 10-item framework cre- sedation found that most caregivers agreed about the ated by the European Association for Palliative Care timing of the initiation of palliative sedation and con- (Table 2).5,6 An audit performed in a single center to as- firmed that it led to a substantial improvement in re- sess adherence to guidelines on palliative sedation in fractory symptoms.70 In a small study of Dutch relatives of patients receiving palliative sedation until death, Table 2. — European Association for Palliative Care many of the relatives acknowledged that palliative se- 10-Item Framework for Guidelines in Palliative Sedation dation was indicated, but they felt that communication between the health care professionals and the relatives Need for preemptive discussion of potential role of sedation in end-of-life care and contingency planning was poor.71 Although an open dialog between the health care Description of when sedation is indicated team, the patient, and his or her family members and Description of necessary evaluation and consultation procedures loved ones is important, we do not feel that such open Indications for informed consent requirements communication is always respected. In an Italian study Indications for need to discuss decision-making process with of communicative behavior in 2 different hospice set- patient’s family tings, family involvement in the decision-making pro- Indications for selecting method of sedation cess with regard to palliative sedation was 100% in Indications for dose titration, patient monitoring, and care both hospices, but the rate of patient participation in Indications for decisions regarding hydration, nutrition, and such decision-making varied from 24% to 59%.35 An- concomitant medications other study of Dutch and US physicians revealed that Indications for needs of the patient’s family open discussion among physicians, patients, and fami- Support for medical and nursing staff lies was more diffuse in The Netherlands than in the United States.72 However, the study involved too few From Cherny NI; ESMO Guidelines Working Group. ESMO clinical practice guidelines for the management of refractory symptoms cases to conclude that an international difference ex- at the end of life and the use of palliative sedation. Ann Oncol. ists when communication about sedation due to dif- 2014;25(suppl 3):iii143-11152. Reprinted by permission of Oxford ferent cultural approaches.72 In a group of 27 sedated University Press.

October 2015, Vol. 22, No. 4 Cancer Control 439 5. Cherny N, Radbruch L. The Board of the European Association for a palliative care unit confirmed an adherence rate of Palliative Care. European Association for Palliative Care (EAPC) recom- 100%.76 Conversely, a survey of Dutch general practi- mended framework for the use of sedation in palliative care. Palliat Med. 2009;23(7):581-593. tioners revealed a high resistance to following guide- 6. Cherny NI; ESMO Guidelines Working Group. ESMO clinical practice lines, and this was particularly true with regard to the guidelines for the management of refractory symptoms at the end of life and the use of palliative sedation. Ann Oncol. 2014;25(suppl 3):iii143-11152. inadvisability of using palliative sedation without pre- 7. Dean MM, Cellarius V, Henry B, et al. Framework for continuous vious expert consultation and to the recommendation palliative sedation therapy in Canada. J Palliat Med. 2012;15(8):870-879. 8. Kirk TW, Mahon MM; Palliative Sedation Task Force of the National of the involvement of a multidisciplinary palliative care Hospice and Palliative Care Organization Ethics Committee. National team.5,77 Without such an approach, misconceptions Hospice and Palliative Care Organization (NHPCO) position statement and commentary on the use of palliative sedation in imminently dying ter- about palliative sedation may be further misconstrued minally ill patients. J Pain Symptom Manage. 2010;39(5):914-923. and perpetuated. 9. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: Palliative Care. v2.2015. http://www.nccn.org/ professionals/physician_gls/PDF/palliative.pdf. Accessed November 4, 2015. Conclusions 10. American Academy of Hospice and Palliative Medicine. Palliative sedation position statement. Published December 5, 2014. http://aahpm. Some patients with cancer experience refractory symp- org/positions/palliative-sedation. Accessed November 4, 2015. toms during the last hours or days of their life and are 11. Materstvedt LJ, Clark D, Ellershaw J, et al. Euthanasia and physician assisted suicide: a view from an EAPC Ethics Task Force. Palliat Med. no longer responsive to antisymptomatic treatments. 2003;17(2):97-101. Managing such symptoms is the objective of palliative 12. ten Have H, Welie JV. Palliative sedation versus euthanasia: an ethical assessment. J Pain Symptom Manage. 2014;47(1):123-136. sedation, a proportional medical approach that must 13. Maltoni M, Pittureri C, Scarpi E, et al. Palliative sedation therapy be individually tailored to each patient and closely does not hasten death: results from a prospective multicenter study. Ann 3-10 Oncol. 2009;20(7):1163-1169. monitored. Undersedation and oversedation can be 14. Maltoni M, Scarpi E, Rosati M, et al. Palliative sedation in end-of-life avoided by titrating the level of sedating drugs to the care and survival: a systematic review. J Clin Oncol. 2012;30(12):1378-1383. 15. Beller EM, van Driel ML, McGregor L, et al. Palliative pharmaco- minimum useful dose (termed proportionate palliative logical sedation for terminally ill adults. Cochrane Database Syst Rev. sedation). A different approach involving standardized 2015;1:CD010206. 16. Broeckaert B, Nunez Olarte J. Sedation in palliative care. Facts and drug dosages and deep continuous sedation or contin- concepts. In: ten Have H, Clark D, eds. The Ethics of Palliative Care: European uous sedation until death risks being performed with Perspectives. Buckingham, UK: Open University Press; 2002:166-180. 3-10 17. Berger JT. The proportionate value of proportionality in palliative the total or partial intention to hasten death. How- sedation. J Clin Ethics. 2014;25(3):219-221. ever, palliative sedation should only be used in a pro- 18. Rys S, Deschepper R, Mortier F, et al. Continuous sedation until death with or without the intention to hasten death--a nationwide study in nursing portional way; such a view has been accepted by many homes in Flanders, Belgium. J Am Med Dir Assoc. 2014;15(8):570-575. bioethical viewpoints. 19. Seale C, Raus K, Bruinsma S, et al. The language of sedation in end-of-life care: the ethical reasoning of care providers in three countries. Although evidence suggests that proportionate Health (London). 2015;19(4):339-3354. palliative sedation does not have a detrimental impact 20. Anquinet L, Rietjens JA, Seale C, et al. The practice of continuous deep sedation until death in Flanders (Belgium), the Netherlands, and the on survival, further research is needed to monitor the UK: a comparative study. J Pain Symptom Manage. 2012;44(1):33-43. qualitative results of the procedure.14,15 Benzodiaze- 21. Swart SJ, van der Heide A, van Zuylen L, et al. Considerations of physicians about the depth of palliative sedation at the end of life. Can Med pines — in particular, midazolam — are the first choice Ass J. 2012;184(7):E360-E366. of drugs for this type of sedation. The decision-making 22. Maltoni M, Scarpi E, Nanni O. Palliative sedation for intolerable suffering. Curr Opin Oncol. 2014;26(4):389-394. process to initiate proportionate palliative sedation is 23. Broeckaert B. Palliative sedation, physician-assisted suicide, and complex and should involve patients, their relatives euthanasia: ‘same, same but different’? Am J Bioeth. 2011;11(6):62-64. 24. Claessens P, Menten J, Schotsmans P, et al. Palliative sedation, not and loved ones, and the health care team. This pro- slow euthanasia: a prospective, longitudinal study of sedation in Flemish cess varies from those used for other clinical decisions palliative care units. J Pain Symptom Manage. 2011;41(1):14-24. 25. Raus K, de Laat M, Mortier E, et al. The ethical and clinical impor- such as maintenance or withdrawal of artificial nutri- tance of measuring consciousness in continuously sedated patients. J Clin tion and hydration. Proportionate palliative sedation is Ethics. 2014;25(3):207-218. 26. Putman MS, Yoon JD, Rasinski KA, et al. Intentional sedation to a universally accepted medical intervention that differs unconsciousness at the end of life: findings from a national physician survey. from physician-assisted suicide in terms of intention, J Pain Symptom Manage. 2013;46(3):326-334. 27. Cherny NI, Portenoy RK. Sedation in the management of refrac- procedure, and results. tory symptoms: guidelines for evaluation and treatment. J Palliat Care. Acknowledgments: The authors thank Gráinne 1994;10(2):31-38. 28. Materstvedt JL, Bosshard G. Deep and continuous palliative seda- Tierney and Ursula Elbling for editorial assistance. tion (terminal sedation): clinical-ethical and philosophical aspects. Lancet Oncol. 2009;10(6):622-627. References 29. Krishna LK, Kwek SY. The changing face of personhood at the end of life: the ring theory of personhood. Palliat Support Care. 2014;3:1-7. 1. Hasselaar JG, Verhagen SC, Vissers KC. When cancer symptoms 30. Menaca A, Evans N, Andrew E, et al. End-of-life care across south- cannot be controlled: the role of palliative sedation. Curr Opin Support ern Europe: a critical review of cultural similarities and differences between Palliat Care. 2009;3(1):14-23. Italy, Spain and Portugal. Crit Rev Oncol Hematol. 2012;82(39):387-401. 2. De Graeff A, Dean M. Palliative sedation therapy in the last weeks of 31. Gu X, Cheng W, Chen M, et al. Palliative sedation for terminally ill life: a literature review and recommendations for standards. J Palliat Med. cancer patients in a tertiary cancer center in Shanghai, China. BMC Palliat 2007;10(1):67-85. Care. 2015;14:5. 3. Schildmann EK, Schildmann J, Kiesewetter I. Medication and 32. Sterckx S, Raus K, Mortier F. Continuous Sedation at the End-of-Life: monitoring in palliative sedation therapy: a systematic review and qual- Ethical, Clinical, and Legal Perspectives. Cambridge, UK: Cambridge Univer- ity assessment of published guidelines. J Pain Symptom Manage. sity Press; 2013. 2015;49(4):734-746. 33. Caraceni A, Zecca E, Martini C, et al. Palliative sedation at the end of 4. Schildmann EK, Schildmann J. Palliative sedation therapy: a sys- life at a tertiary cancer center. Support Care Cancer. 2012;208(6):1299-1307. tematic literature review and critical appraisal of available guidance on in- 34. Mercadante S, Porzio G, Valle A, et al; Home Care Italy Group. dication and decision making. J Palliat Med. 2014;17(5):601-611. Palliative sedation in patients with advanced cancer followed at home: a

440 Cancer Control October 2015, Vol. 22, No. 4 systematic review. J Pain Symptom Manage. 2011;41(4):754-760. euthanasia and lethal injection. J Pain Palliat Care Pharmacother. 35. Maltoni M, Miccinesi G, Morino P, et al. Prospective observational 2012;26(1):30-39. Italian study on palliative sedation in two hospice settings: differences in 63. Janssens R, van Delden JJ, Widdershoven GA. Palliative sedation: case mixes and clinical care. Support Care Cancer. 2012;20(11):2829-2836. not just normal medical practice. Ethical reflections on the Royal Dutch 36. Morita T, Akechi T, Uchitomi Y, et al. Practices and attitudes of Medical Association’s guideline on palliative sedation. J Med Ethics. Japanese oncologists and palliative care physicians concerning terminal 2012;38(11):664-668. sedation: a nationwide survey. J Clin Oncol. 2002;20(3):758-764. 64. Bruera E, Hui D, Dalal S, et al. Parenteral hydration in patients with 37. Mercadante S, Porzio G, Valle A, et al; Home Care–Italy Group. advanced cancer: a multicenter, double-blind, placebo-controlled random- Palliative sedation in advanced cancer patients followed at home: a retro- ized trial. J Clin Oncol. 2013;31(1):111-118. spective analysis. J Pain Symptom Manage. 2012;43(6):1126-1130. 65. Beland P. Ethical issues around continuous deep sedation without 38. Goncalves F, Cordero A, Almeida A, et al. A survey of the seda- hydration. Nurs Times. 2012;108(38):22-25. tion practice of Portuguese palliative care teams. Support Care Cancer. 66. Miccinesi G, Rietjens JA, Deliens L, et al. Continuous deep seda- 2012;20(12):3123-3127. tion: physicians’ experiences in six European countries. J Pain Symptom 39. Jaspers B, Nauck F, Lindena G, et al. Palliative sedation in Ger- Manage. 2006;31(2):122-129. many: how much do we know? A prospective survey. J Palliat Med. 67. Harrison JD, Young JM, Price MA, et al. What are the unmet sup- 2012;15(6):672-680. portive care needs of people with cancer? A systematic review. Support 40. Kang JH, Shin SH, Bruera E. Comprehensive approaches to Care Cancer. 2009;17(8):1117-1128. managing delirium in patients with advanced cancer. Cancer Treat Rev. 68. Maltoni M, Scarpi E, Nanni O. Palliative sedation in end-of-life care. 2013;39(1):105-112. Curr Opin Oncol. 2013;25(4):360-367. 41. Lawlor PG, Bush SH. Delirium in patients with cancer: assessment, 69. Bruinsma SM, Rietjens JA, Seymour JE, et al. The experiences of impact, mechanisms and management. Nat Rev Clin Oncol. 2015;12(2):77-92. relatives with the practice of palliative sedation: a systematic review. J Pain 42. Teunissen SC, Wesker W, Kruitwagen C, et al. Symptom prev- Symptom Manage. 2012;44(3):431-445. alence in patients with incurable cancer: a systematic review. J Pain 70. Vayne-Bossert P, Zulian GB. Palliative sedation: from the family Symptom Manage. 2007;34(1): 94-104. perspective. Am J Hosp Palliat Care. 2013;30(8):786-790. 43. Sadler K. Palliative sedation to alleviate existential suffering 71. Bruinsma S, RietjensJ van der Heide A. Palliative sedation: a focus at end-of-life: insight into a controversial practice. Can Oncol Nurs J. group study on the experiences of relatives. J Palliat Med. 2013;16(4):1-7. 2012;22(3):195-199. 72. Rietjens JA, Voorhees JR, van der Heide A, et al. Approaches to 44. van Deijck RH, Hasselaar JG, Verhagen S, et al. Determinants of suffering at the end of life: the use of sedation in the USA and Netherlands. the administration of continuous palliative sedation: a systematic review. J Med Ethics. 2014;40(4):235-240. J Palliat Med. 2013;16(12):1624-1632. 73. Arevalo JJ, Rietjens JA, Swart SJ, et al. Day-to-day care in pal- 45. Oosten AW, Oldenmenger WH, van Zuylen C, et al. Higher doses liative sedation: survey of nurses’ experiences with decision-making and of opioids in patients who need palliative sedation prior to death: cause or performance. Int J Nurs Stud. 2013;50(5):613-621. consequence? Eur J Cancer. 2011;47(15):2341-2346. 74. Gielen J, Van den Branden S, et al. Flemish palliative-care 46. Alonso-Babarro A, Varela-Cerdeira M, Torres-Vigil I, et al. At- nurses’ attitudes to palliative sedation: a quantitative study. Nurs Ethics. home palliative sedation for end-of-life cancer patients. Palliat Med. 2012;19(5):692-704. 2010;24(5):486-492. 75. Patel B, Gorawara-Bhat R, Levine S, et al. Nurses’ attitudes and 47. Morita T, Tei Y, Inoue S. Correlation of the dose of midazolam for experiences surrounding palliative sedation: components for developing symptom control with administration periods: the possibility of tolerance. policy for nursing professionals. J Palliat Med. 2012;15(4):432-437. J Pain Symptom Manage. 2003;25(4):369-375. 76. Benitez-Rosario MA, Castillo-Padrós M, Garrido-Bernet B, et al. 48. Shlamovitz GZ, Elsayem A, Todd KH. Ketamine for palliative Quality of care in palliative sedation: audit and compliance monitoring of a sedation in the emergency department. J Emerg Med. 2013;44(2):355-357. clinical protocol. J Pain Symptom Manage. 2012;44(4):532-541. 49. Hernandez SL, Sheyner I, Stover KT, Stewart JT. Dronabinol treat- 77. Koper I, van der Heide A, Janssens R, et al. Consultation with ment of refractory nausea and vomiting related to peritoneal carcinomatosis. specialist palliative care services in palliative sedation: considerations of Am J Hosp Palliat Care. 2015;32(1):5-7. Dutch physicians. Support Care Cancer. 2014;22(1):225-231. 50. Dietz I, Schmitz A, Lampey I, Schulz C. Evidence for the use of Levomepromazine for symptom control in the palliative care setting: a systematic review. BMC Palliat Care. 2013;12:2. 51. McWilliams K, Keeley PW, Waterhouse ET. Propofol for terminal se- dation in palliative care: a systematic review. J Palliat Med. 2010;13(1):73-76. 52. Gélinas C, Fillion L, Puntillo KA, et al. Validation of the critical-care pain observation tool in adult patients. Am J Crit Care. 2006;15(4):420-427. 53. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation- Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002;166(10):1338-1344. 54. Arevalo JJ, Brinkkemper T, van der Heide A, et al. Palliative seda- tion: reliability and validity of sedation scales. J Pain Symptom Manage. 2012;44(5):704-714. 55. Benítez-Rosario MA, Castillo-Padrós M, Garrido-Bernet B, et al. Appropriateness and reliability testing of the modified Richmond Agitation- Sedation Scale in Spanish patients with advanced cancer. J Pain Symptom Manage. 2013;45(6):1112-1119. 56. Bush SH, Grassau PA, Yarmo MN, et al. The Richmond Agita- tion-Sedation Scale modified for palliative care inpatients (RASS-PAL): a pilot study exploring validity and feasibility in clinical practice. BMC Palliat Care. 2014;13(1):17. 57. Deschepper R, Laureys S, Hachimi-Idrissi S, et al. Palliative se- dation: why we should be more concerned about the risks that patients experience an uncomfortable death. Pain. 2013;154(9):1505-1150. 58. Brinkkemper T, van Norel AM, Szadek KM, et al. The use of ob- servational scales to monitor symptom control and depth of sedation in patients requiring palliative sedation: a systematic review. Palliat Med. 2013;27(1):54-67. 59. Rys S, Deschepper R, Mortier F, et al. The moral difference or equiva- lence between continuous sedation until death and physician-assisted death: word games or war games? A qualitative content analysis of opinion pieces in the indexed medical and nursing literature. J Bioeth Inq. 2012;9(2):171-183. 60. Morita T, Chinone Y, Ikenaga M, et al. Efficacy and safety of pal- liative sedation therapy: a multicenter, prospective, observational study conducted on specialized palliative care units in Japan. J Pain Symptom Manage. 2005;30(4):320-328. 61. Rietjens JA, van Delden JJ, van der Heide A, et al. Terminal seda- tion and euthanasia: a comparison of clinical practices. Arch Intern Med. 2006;166(7):749-753. 62. Hahn MP. Review of palliative sedation and its distinction from

October 2015, Vol. 22, No. 4 Cancer Control 441