Medical Manuscript

American Journal of Hospice

& Palliative Medicine®
Palliative Sedation at the End of Life: 1-5
ª The Author(s) 2015
Reprints and permission:
Patterns of Use in an Israeli Hospice sagepub.com/journalsPermissions.nav
DOI: 10.1177/1049909115572991
ajhpm.sagepub.com

Daniel Azoulay, MD1,2, Ruth Shahal-Gassner, RN, BA1,

Malka Yehezkel, MSW1, Ester Eliyahu, RN, BA1,
Nir Weingert, MD1, Eliana Ein-Mor, MA2, and
Jeremy M. Jacobs, MBBS, BSc1,2

Abstract
Palliative sedation (PS) is indicated for refractory symptoms among dying patients. This retrospective descriptive study examines
PS in an Israeli hospice. Palliative sedation was defined as PS to unconsciousness (PSU), PS proportionate to symptoms
(proportional palliative sedation [PPS]), or intermittent PS (IPS). Among 179 patients who died during 2012, PS was used
among 21.2% (n ¼ 38): (PSU 34.2%, PPS 34.2%, and IPS 31.6%), using midazolam (n ¼ 33/38), halidol (21/38), and concurrent
morphine (n ¼ 35/38). Indications included agitation (71%), pain (36.8%), and dyspnea (21%). Survival following initiation of PS
was 73 + standard deviation 54 hours. No differences in survival were observed according to who initiated the decision to
use PS (patients/medical staff/family) or type of PS (PSU/PPS/IPS). Survival following PS was longest with higher sedative doses,
an observation that may help dispel fears concerning the use of PS to hasten death.

Keywords
palliative sedation, terminal sedation, end-of-life care, palliative care, hospice, survival

Introduction sedation (PPS) may often achieve symptom relief titrated to

Palliative sedation (PS) is an increasingly common form of
treatment among dying patients, whose suffering is considered
to be refractory to other treatment modalities. Definitions of PS
include the induction of reduced level of consciousness by spe-
cific sedative medications, with the intended aim of relieving
intolerable suffering from refractory symptoms. Guidelines
commonly endorse the careful titration to reach the minimum
dose of medication necessary to achieve the cessation of symp-
toms, clear documentation of patient and/or family consent,
and the need for close supervision and guidance from experi-
enced trained medical staff.1,2
Although an important part of the continuum of palliative
care, PS is generally reserved as an ultimate tool, and used as
a last resort option among patients deemed to be close to death,
for whom all other treatments have been unsuccessful at reliev-
ing the patients’ suffering.3,4 Primary indications for PS
include poorly controlled pain, agitation, dyspnea, and vomit-
ing as well as on occasional profound psychological or existen-
tial suffering related to their close proximity to death and
dying. The PS may be intermittent (IPS), being given on an
intermittent basis (eg, just at night or for a short period of time)
or continuously.5 Similarly, the degree or depth of sedation is
generally accepted to be proportional and governed by the level
of reduced consciousness needed to alleviate 1 or more refrac-
tory symptoms. Although the use of proportional palliative
light continuous sedation, nonetheless certain extreme cases
may require the use of PS, albeit given proportionally, aimed
at reaching deep unconsciousness (PSU).4,6
The use of PS has been generally reported in the hospice set-
ting among inpatients; however, with appropriate setup, PS has
also been given to patients receiving home hospice services.
Frequency of PS varies widely in different settings, ranging
from 0% to 66.7% of patients receiving end-of-life palliative
care.3,7 The Israeli Ministry of Health recently endorsed the use
of PS among patients with an incurable disease, deemed to be
in the last 6 months of life, whose unrelieved suffering is
refractory to other treatments.8 Like many countries, the use
of PS is very poorly documented in Israel. Although PS has
been described among home hospice patients in Israel,9 the
overall frequency and patterns of use remain unclear.
1
Hadassah Mount Scopus Hospice Unit for Palliative Care
2
Department of Geriatrics and Rehabilitation, Hadassah-Hebrew University
Medical Center, and Hebrew University-Hadassah Medical School, Jerusalem,
Israel
Corresponding Author:
Daniel Azoulay, MD, Department of Geriatrics and Rehabilitation, Hadassah
Hebrew-University Medical Center, Mt Scopus, PO Box 24035, Jerusalem
91240, Israel.
Email: adaniel@hadassah.org.il

Downloaded from ajh.sagepub.com at University of Sussex Library on October 13, 2015

2 American Journal of Hospice & Palliative Medicine®
The subject of PS often rouses emotive debate, and ethical
considerations have typically focused upon the ‘‘primacy of
intention’’ (the aim to reduce suffering), despite the possible
‘‘double effect’’ (the potential of hastening death). Indeed fears
that PS has the potential to hasten death have led critics to asso-
ciate PS with the debate concerning patient-assisted suicide,
and euthanasia.5,6,7 However, to date, available evidence does
not support the belief that PS, when administered correctly,
actually hastens death.3,10
The goal of this study is to widen knowledge concerning PS
in Israel. In order to achieve this end, we performed a retro-
spective descriptive study, collecting data from all patients who
died as inpatients during 2012 in the Hadassah Hospice Unit,
Mount Scopus, Jerusalem.
Specifically, we aimed to describe the frequency and type of
PS, the indications, associated medical diagnoses, the extent of
patient and family involvement in the decision to initiate PS,
and the improvement in symptoms following PS. The study
also describes the time to death among different subgroups fol-
lowing initiation of PS.
Methods
Basic Methodology
This was a retrospective, descriptive observational study of all
patients cared for from January 1, 2012, to December 31, 2012,
in the Hospice Unit of the Hadassah Hebrew University Med-
ical Center, situated on Mount Scopus in Jerusalem, Israel.
Patients
All patients who were admitted to the hospice during 2012
were included in the study. Patients eligible for hospice care
were referred by 1 of the 4 local Health Maintenance Organi-
zations (HMO). All Israeli citizens by law belong to one of the
HMOs, which are obliged to provide comprehensive health
coverage, including palliative care at the end of life. Patients
are referred to the hospice from a variety of settings, including
hospitalized inpatients, oncology day care units, primary care
physicians as well as from home hospice services which exist
in Jerusalem and the surrounding area. The majority of patients
have advanced cancer, and the average duration of admission is
about 3 weeks.
Setting
The Hadassah Hospice Unit is a 14-bed unit, founded in 1986,
which serves the local Jerusalem and surrounding population.
The hospice is a dedicated unit, providing multidisciplinary
care, which includes a team of physicians, nurses, social work-
ers as well as physiotherapists, volunteers, and chaplaincy.
Palliative Sedation
Proportionality indeed underlines the use of all PS, with the end
point of PS for all patients being ‘‘to reduce the consciousness
Downloaded from ajh.sagepub.com at University of Sussex Library on October 13, 2015
of a terminal patient as much as necessary to adequately relieve
one or more refractory symptoms.’’11 Nonetheless, as outlined
by Quill et al,6,12 PS aimed at inducing deep unconsciousness
from the outset will be needed in certain, most severe, immedi-
ate, and compelling cases. Accordingly, we identified the use
of (1) proportional palliative sedation (PPS), which was usually
started as a mild continuous sedation, (2) Palliative sedation to
unconsciousness (PSU), which involved the use of deep PS,
albeit given proportionally, in certain extreme circumstances,
and (3) intermittent use of palliative sedation (IPS), whereby
PS was continued up until the time of death, but sedation was
not continuous, rather intermittent throughout different parts of
the day.
The decision to initiate PS was routinely documented in the
medical file, following discussion with the patient, family, and
staff. Informed consent by the patient was a necessary prerequi-
site, without which PS was not initiated. In keeping with guide-
lines laid down in the Israeli Law of the Dying Patient,13 in
cases where the patient was physically and mentally unable
to give informed consent, then consent was given by people
legally designated as holding power of attorney for medical
decisions concerning the patient, or in the case that no such
legal power of attorney or guardianship existed, then the closest
significant family was consulted.
There is a growing awareness concerning palliative care in
general and PS in particular among the Israeli public, and it was
not uncommon for patients or family members to initiate the
request to initiate PS. Accordingly, we documented who was
most proactive in the decision to initiate PS-the patient, the
family, or the treating staff members.
Choices of medication were based upon individual patient
characteristics and tailored to most effectively address the
patient’s symptoms. Route of PS drug administration was sub-
cutaneous, with delivery via calibrated pumps as well as addi-
tional injections as needed. Dosages were titrated according to
the patient response. Some patients received transdermal fenta-
nyl, and when urgently necessary, opiates were also available
via intravenous administration. The majority of patients in the
hospice receive morphine or similar opioids in the last week of
their lives in order to alleviate a range of symptoms, irrespec-
tive of whether or not they also receive PS. In the patients
reported in this article, morphine was used largely for allevia-
tion of pain and dyspnea, and while it was not directly given for
its sedative effects, nonetheless a possible sedative effect may
have existed among the 35 of the 38 patients who received PS.
We have therefore included the information concerning the
concurrent use of morphine among the PS patients.
Data Collection
Data were collected retrospectively during 2013 from medical,
nursing, and drug records, by the author (DA), and included
sociodemographic variables; medical diagnosis; the symptoms
for which PS was indicated; medications used for PS including
route of administration and daily doses used (total dose given in
24-hour period, for each of the 7 days prior to death); type of PS
Azoulay et al 3

was classified as (1) PS to unconsciousness (PSU), where the Table 1. Characteristics of Patients.
decision was made from the outset to achieve rapidly deep
P
sedation, (2) proportionate PS (PPS), where the patient’s level PS No PS value
of sedation was titrated according to relief of symptoms, with-
out necessarily the initial aim of rapidly inducing deep seda- Total N ¼ 38 N ¼ 141
tion, or (3) IPS, where PS was given intermittently; we also (21.2%) (78.8%)
noted who initiated more actively the decision for PS, whether Mean length of hospice 22.92 + 33.8 16.6 + 29.1 .25
admission, days
it was from the patient, family, or the hospice staff. Overall
Gender
symptom control was assessed (from the nurse and physician Male 29% 46.80% .048
follow-up records) as a subjective global measure of whether Female 71% 53.20%
an improvement was noted following PS initiation. Duration Age, years 73.3 + 12.1 75.9 + 15.2 .21
of PS was measured in hours from the first dose of PS to the Married 50% 55.30% .49
time of death. Tumor
The Hadassah-Hebrew University Medical Center institu- Brain 2.60% 4.30% .763
Breast 15.80% 7.10%
tional review board approved the study.
Lungs 21.10% 15.70%
Pancreas 13.20% 16.60%
Statistical Analyses Colon 13.20% 16.40%
Stomach 2.60% 5.70%
Descriptive statistics were performed using chi-square tests for Prostate 0 1.40%
categorical variables, and t test and analysis of variance for Hematologic 5.30% 7.10%
duration of PS until death. Data storage and analysis were per- Other 23.70% 24.30%
formed using SAS 9.1e package (SAS Institute Inc, Cary, North Metastases 92.10% 83.10% .17
Carolina). All P values were 2 tailed, and P < .05 was consid- Subcutaneous fluids 78.40% 71.70% .65
Antibiotics 33.30% 29.70% .67
ered significant.
Abbreviation: PS, palliative sedation.

Results
Table 2. Sedation.
During 2012, a total of 179 patients were admitted to the hos-
pice, among whom 21.2% (n ¼ 38) received PS. Among N ¼ 38, %
patients receiving PS, 71% were women, while among patients Indication for sedation
not receiving PS women formed 53.2% (P ¼ .048). All PS Agitation 71
patients had advanced cancer. No significant differences were Pain 36.80
observed among patients with or without PS for age, marital Dyspnea 21
status, diagnosis, and presence of metastases. Patients treated Vomiting 7.90
with PS tended to have longer admissions in the hospice com- Other 2.60
pared to patients who did not receive PS (22.92 + 33.8 vs 16.6 Number of indications
1 55.30
+ 29.1 days, P ¼ .25; Table 1). 2 39.50
Among the reasons for PS, agitation was the most common 3 5.30
(71%), followed by pain (36.8%), and dyspnea (21%). In all, Type of sedation
55.3% had 1 indication for PS and 39.5% had 2 reasons. The PSU Palliative sedation to unconsciousness (PSU) 34.20
was used for 34.2%, PPS for 34.2%, and IPS for 31.6% of patients. Proportional palliative sedation (PPS) 34.20
The decision to start PS was initiated by the patient in 18.4%, by Intermittent palliative sedation (IPS) 31.60
the family in 26.3%, and by the hospice staff in 55.3% of the cases. Decision to initiate PS
Patient request 18.40
Following PS, there was documentation of a global symptomatic Family 26.30
improvement in 73.7% of the patients (Table 2). The medications Staff 55.30
used were midazolam and morphine, which were often used in Improvement following PS initiation
combination with haloperidol (Table 3). Yes 73.70
Overall time to death following the initiation of PS was Not documented 26.30
73.0 + 54.4 hours. No difference in survival was observed Abbreviations PS, palliative sedation.
among patients who themselves initiated the request for PS
compared to medical staff initiation compared to family initia-
tion (70.13 + standard deviation [SD] 56.1 hours vs 81.5 + following initiation of PS according to average daily dose and
SD 62.81 hours, vs 57.2 + 39.6, P value ¼ .51). Similarly, mean daily dose of PS medications is shown in Table 5. In all
no significant differences in duration were observed for type 3 drug categories, a similar pattern was observed, whereby
of PS (PSU/PPS/IPS): 68.6 + 46.0 vs 83.8 + 70.7 vs 66.0 + patients who received highest dosages of midazolam, morphine,
44.8, P ¼ .68, as shown in Table 4. An analysis of time to death or haloperidol actually survived the longest (Table 5).

Downloaded from ajh.sagepub.com at University of Sussex Library on October 13, 2015

4 American Journal of Hospice & Palliative Medicine®

Table 3. Palliative Sedation Medications. midazolam, with concurrent use of morphine, frequently with
haloperidol. Patients receiving PS tended to live longer with
Drugs
Midazolam 33/38
increased length of admission in the hospice when compared
Halidol 21/38 to those who did not receive PS. Among patients receiving
Concurrent use of morphinea 35/38 PS, survival time following initiation of PS was not influenced
Drugs combinations by the type of PS or by the person who initiated the decision for
Midazolam þ halidol 3/38 PS. The finding that 1 in 5 patients received PS is in keeping
Midazolam þ morphine SC 15/38 with comparative literature from other countries, which show
Halidol þ morphine SC 3/38 a wide variation in prevalence of PS among hospice patients
Midazolam þ halidol þ morphine SC 15/38
at the end of life.3,7,10
Abbreviation: SC, subcutaneous. Interestingly, the highest doses of PS medication were actu-
a
Includes 2 patients who received only morphine on as as needed basis and not ally observed in patients who lived the longest following the
via syringe driver.
initiation of PS. This observation is in keeping with recent
opinion that PS does not per se shorten survival.10 Nonetheless,
Table 4. Palliative Sedation and survival. our findings are purely descriptive and thus cannot be used to
draw conclusions of causality between survival time and PS
Survival following PS, hours treatment dosage. There is accumulating evidence that pallia-
tive care may actually prolong survival, whether initiated early
Mean Median
on at the time of cancer diagnosis in conjunction with standard
All patients 73.02 + 54.4 58.7 of care oncology treatment14 or when delivered at much more
Decision to initiate PS P ¼ .514 advanced stages of illness. Similarly, there is growing evidence
Patient 70.12 + 46.1 66.0 that rising doses of opiates used to achieve optimal symptom
Family 57.19 + 39.56 47.4 are actually associated with longer survival.15
Staff 81.53 + 62.8 62.8
Type of PS P ¼ .684
Barriers to appropriate care are numerous, ranging from lack
PSU 68.6 + 46.0 66.0 of technical knowledge and professional experience to more
PPS 83.8 + 70.7 70.5 challenging issues of local policy and attitudes among health-
IPS 66.1 + 44.8 48.0 care professionals concerning the use of PS in the context of
refractory symptoms at the end of life. Similarly, attitudes of
Abbreviations: IPS, intermittent PS; PPS, proportional palliative sedation; PS,
palliative sedation; PSU, palliative sedation to unconsciousness. patients and their family toward PS are likely to play a crucial
role in the decision concerning the initiation of PS. Concerns
surrounding the use of PS to hasten death are recognized. In
Table 5. Drug Dosage and Survival. this study, death was clearly not immediate, with a median time
to death being 58.7 hours and a mean of 73.0 + 54.4 hours.
Survival time following start of PS, hours
Furthermore, the observation that higher doses of PS medica-
Average daily dose Mean Median tions were actually observed with longer duration of PS treat-
ment, and thus a longer time before death, may be useful in
Midazolam, mg/d arriving at an informed decision and emphasizing the clear dif-
0-5 54.3 + 38.3 (16) 44.8
ference between euthanasia and PS initiated for intractable
6-10 70.0 + 41.7 (8) 65.4
11-20 90.3 + 78.8 (8) 68.3 symptoms at the end of life.
>20 90.4 + 49.0(4) 79.0 Although ethical concerns are to be considered, no less
Morphine, mg/d important are concerns of safety, adequate supervision and con-
0-20 77.7 + 57.4 (26) 66.6 trol, and the containment of PS to guarantee its use only in
20-99 65.3 + 46.9 (6) 58.7 unequivocally indicated circumstances. Recent guidelines from
>100 80 + 56.5 (4) 64.5 the Israeli Ministry of Health, which outline the criteria for PS
Haloperidol, mg/d
usage, will need to contend with these barriers in delivery of
0-2 65.7 + 44.0 (23) 50.0
>2 94.0 + 67.3 (13) 79.0 care. Since the concerns and barriers surrounding PS are cultu-
rally sensitive, providing local data are important. Our data will
Abbreviation: PS, palliative sedation. serve a useful purpose in providing a first-time snapshot of the
usage and, by implication, the degree of need for PS among
hospice patients in Israel. Similarly, local data emphasizing the
Discussion time course of treatment, the observation of increased survival
This study of 179 patients who received end-of-life care in a despite increasing doses, and the distinctive lack of similarity
hospice setting describes the use of PS among 38 (21.2%) between PS and active euthanasia, will be helpful in the local,
patients in the last week of their life. No significant differences culturally sensitive debate concerning the use of PS in Israel.
in baseline characteristics were observed other than increased The descriptive nature of our data is a limitation worthy of
use of PS among females. The most commonly used drugs were mention. Conclusions cannot and should not be drawn

Downloaded from ajh.sagepub.com at University of Sussex Library on October 13, 2015

Azoulay et al
concerning the influence of PS on survival. Only an interven-
tional study might address this very sensitive issue; however,
performing such a study is itself fraught with numerous meth-
odological and ethical issues among dying patients. The
patients in our study were referred by the local HMO for
end-of-life care. Since alternative care may also be offered
by the HMO in less specialized geriatric long term care facility
(LTCF), it is possible that the sample of patients who were
admitted to the hospice was biased toward those demanding
more complex management issues, with a greater need for
PS. Clearly, the hospice setting has the expertise and experi-
ence, the option of PS is accepted as standard practice, and
in general the therapeutic environment is minded to the option
of PS as a last resort option. Thus the use of PS is likely to have
been higher than other less specialized settings, which may care
for patients at the end of life. A comparison of symptom man-
agement alongside the availability and frequency of PS use
among patients being cared for at the end of life in different
health-care environments would shed light on this issue.
In conclusion, our study documents the usage of PS among
an Israeli hospice and shows similar rates of use in comparison
to other countries. The PS was used for 1 in 5 patients, with an
average time to death of 3 days. No significant differences were
observed according to whose decision it was to initiate PS or
the type of PS. The observation that survival was longest
among those patients who actually received the largest doses
of PS medication may help dispel fears concerning the use of
PS to hasten death.
Authors’ Note
The sponsors had no role in the design and conduct of the study;
collection, management, analysis and interpretation of the data; or
preparation, review, or approval of the article. Dr Azoulay had full
access to all of the data in the study and takes responsibility for the
integrity of the data and the accuracy of the data analysis. These funds
were used exclusively to support the research effort, primarily as
salaries to ancillary staff. No research funds were received by any
author of this article.
Acknowledgments
We wish to acknowledge all members of the Hospice team.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the
research, authorship, and/or publication of this article: This work was
supported by funds from The Minerva Center for the Interdisciplinary
Study of the End of Life, at Tel Aviv University.
Downloaded from ajh.sagepub.com at University of Sussex Library on October 13, 2015
5
References
1. de Graeff A, Dean M. Palliative sedation therapy in the last weeks
of life: a literature review and recommendations for standards.
J Palliat Med. 2007;10(1):67-85. http://dx.doi.org/10.1089/jpm.
2006.0139.
2. Claessens P, Menten J, Schotsman P, Broeckaert B. Palliative
sedation: a review of the research literature. J Pain Symptom
Manage. 2008;36(3):310-333 http://dx.doi.org/10.1016/j.jpain-
symman.2007.10.004.
3. 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 http://dx.doi.org/10.1200/JCO.2011.37.3795.
4. 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.
CMAJ. 2012;184(7):E360-E366. http://dx.doi.org/10.1503/cmaj.
110847.
5. Lo B, Rubenfeld G. Palliative sedation in dying patients: ‘‘we turn
to it when everything else hasn’t worked’’. JAMA. 2005;294(14):
1810-1816. http://dx.doi.org/10.1001/jama.294.14.1810.
6. Quill TE, Lo B, Brock DW, Meisal A. Last resort options for pal-
liative sedation. Ann Intern Med. 2009;151(6):421-424. http://dx.
doi.org/10.7326/0003-4819-151-6-200909150-00007.
7. Quill TE, 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 eut-
hanasia. JAMA. 1997;278(23):2099-2014. http://dx.doi.org/10.
1001/jama.1997.03550230075041.
8. Guidelines for Palliative Sedation. Medical Statement. Israel:
Medical Division, Israel Ministry of Health; 2014.
9. Rosengarten OS, Lemed Y, Zisling T, Feygin A, Jacobs JM.
Palliative sedation at home. J Palliat Care. 2009;25(1):5-11.
10. Morita T, Chinine Y, Ikenaga M, et al. Efficacy and safety of
palliative sedation therapy: a multicenter, prospective, observa-
tional study conducted on specialized palliative care units in
Japan. J Pain Symptom Manage. 2005;30(4):320-328. http://dx.
doi.org/10.1016/j.jpainsymman.2005.03.017.
11. Broeckaert B, Leuven KU. Palliative sedation, physician-assisted
suicide, and euthanasia: ‘‘same, same but different’’? Am J Bioeth.
2011;11(6):62-64.
12. Quill TE, Brock D, Lo B, Meisel A. Justifying different levels of
palliative sedation. Author response. Ann Intern Med. 2010;1(5):
332-333.
13. Steinberg A, Sprung CL. The dying patient act, 2005: Israeli inno-
vative legislation. Isr Med Assoc J. 2007;9(7):550-552.
14. Temel JR, Greer JA, Muzikansky A, et al. Early palliative care for
patients with metastatic non-small-cell lung cancer. N Engl J
Med. 2010;363(8):733-743. http://dx.doi.org/10.1056/NEJMoa
1000678.
15. Azoulay D, Jacobs JM, Cialic R, Ein Mor E, Stessman J. Opiods,
survival, and advanced cancer in the hospice setting. J Am Dir
Assoc. 2011;12(2):129-134. http://dx.doi.org/10.1016/j.jamda.
2010.07.012.