Nova Scotia Hospice Palliative Care Association Annual
Conference 2011
PALLIATIVE SEDATION
Myth, Mercy or
Euphemism?
Dr Nigel Sykes
St Christopher's Hospice
London
A Confusion of Terms
Terms:
Palliative sedation
Terminal sedation
Early terminal sedation
Palliative sedation therapy
Primary sedation
Secondary sedation
Proportionate sedation
Controlled sedation for intractable distress
in the dying
Sudden sedation
Continuous deep sedation
A Manifold Definition of Palliative
Sedation
(Adapted from Jones, 2011)
Palliative Sedation is the use of sedatives that is either
continuous or intermittent; deep or mild; the primary or
secondary pharmacological effect; proportionate or
disproportionate to ‘refractory symptoms’ ; which
include or do not include ‘existential distress’; in a
patient who is or is not imminently dying; at the request
or not of the patient; who intends or does not intend to
be unconscious until death; with the doctor also
intending this or not; withholding or not withholding
nutrition and hydration; with or without an advance
refusal; such that this protocol does or does not actually
hasten death; and is intended or is not intended to do
so by the patient; and is intended or is not intended to
do so by the patient
This yields 4,782,969 possible definitions of
Palliative Sedation…
Is the likely clinical effect of this
Definition…
Palliative Sedation is the use of
sedatives that is intermittent;
mild; proportionate to ‘refractory
symptoms’ ; in a patient who is
imminently dying; not
withholding nutrition and
hydration
…the same as the likely clinical
effect of this Definition?
Palliative Sedation is the use of
sedatives that is continuous, deep,
disproportionate to ‘refractory
symptoms’ ; in a patient who is
not imminently dying but is
intended to be unconscious until
death; withholding nutrition and
hydration
A Euphemism for Euthanasia?
A Confusion of Purposes
Sedation can mean:
The giving of sedatives for specific
symptom control, e.g.
Seizures
Delirium in the absence of correctable
factors
A treatment for insomnia
The attempt to make a patient unaware
of a intractable symptom by reducing
their conscious level
An expert survey achieved only 40%
agreement with a single definition of
sedation (Chater, 1998)
What is an intractable
symptom?
An "intractable" symptom is one:
that does not respond to available
treatment or
for which the treatment is unacceptable to
the patient because of:
insufficiently rapid action or
excessive side effects (Cherny and Portenoy, 1994)
Sedation is used significantly more often by
doctors who predict that a symptom will be
intractable than by those who actually try all
the treatments (Morita, 2004)
Reasons for sedation
Multicentre study of 387 terminally ill
patients
Haemorrhage
Distress
Pain
Nausea and vomiting
Breathlessness
Delirium
0.8%
1.8%
1.8%
2.3%
6.5%
15.2%
(Fainsinger et al., 2000)
How often is sedation used in
Palliative Care?
Reports of the proportion of patients
who require sedation in the closing
days of life vary widely:
1% (Fainsinger, 1998)
88% (Turner et al., 1996)
This situation is not getting any better:
Prospective study of the use of all depths and
lengths of sedation in eight palliative care
units showed a rate of 7.5%
(Claessens et al, 2011)
Retrospective study from one palliative care
unit of the use specifically of continuous deep
sedation showed a rate of 43%
(Rietjens et al., 2008)
Sedation in Palliative Care
The use of sedative drugs has always been
a part of Palliative Care at the end of life:
For mental distress (but only as an adjunct to
the giving of properly attentive time)
(Saunders, 1960)
For anxiety or agitated confusion
(Saunders, 1965)
Opiates should not be used as sedatives
(Saunders, 1958)
“It should hardly ever be necessary to use
the very heavy sedation that completely
smothers the patient’s personality,
although many who see these patients
only occasionally do not believe that it is
possible to avoid this” (Saunders, 1967)
“Very heavy sedation that
completely smothers the
patient’s personality”
The crux of ethical and clinical
concern seems to be whether
sedative use:
Obliterates the patient’s personality
and destroys the possibility of further
emotional and spiritual development
Kills the patient
Sedatives can do both these
things
Sedation for intractable
symptoms
The paramount moral obligation is to
relieve suffering
“A doctor who leaves a patient to suffer
intolerably is morally more
reprehensible than the doctor who
performs euthanasia”
Twycross, 1996
Mercy
Principle of Double Effect
(The Get-out Clause)
 A harmful effect of treatment, even resulting in
death, is permissible providing that it:
 was not intended and
 arises as a side effect of a beneficial action and
 the harmful effect was not the means of achieving the
beneficial effect
 But if we need to invoke the Principle of Double
Effect does this suggest we are routinely
shortening patients’ lives by sedation?
Truth or Myth?
Classification of end of life care
sedation
(Broeckaert, 2000)
Mild
Intermittent
Acute
Deep
Continuous
Non-acute
Classification of end of life care
sedation
(Morita, Tsuneto and Shima, 2001)
Mild
Intermittent
Primary
Pain
No organ
failure
Deep
Continuous
Secondary
Psychological
distress
Organ
failure
How is depth of sedation
assessed?
 Glasgow Coma Scale (Teasdale and Jennett, 1974)
 Communication Capacity Scale (Morita et al., 2001)
 Consciousness Scale for Palliative Care
 Physicians’ unsubstantiated report
(Goncalves et al.,2008)
 Assessment of the deepest sedation requires
infliction of pain:
 Supra-orbital pressure (GCS)
 Pain (unspecified method) or change in position (CCS)
 Trapezius pinch (CSPC)
 How willing are palliative care staff to carry out
these assessments routinely?
Depth of Sedation
It has been suggested that the depth of
sedation tends to increase as death
approaches
45% of patients originally given ‘mild’
sedation had ‘deep continuous’ sedation by
two days before death (Claessens et al., 2011)
 But this is based on only 9 patients and it is not
clear how the sedative doses changed in the
interim
How different is this from the natural
trajectory of dying?
50% of Palliative Care not receiving
sedatives are unable to manage complex
communication five days before death
(Morita et al., 2003)
Comparison of percentage of patients
spontaneously awake in sedated and nonsedated groups during the last week of life
(n=23) (from Fainsinger et al. 1998)
100
%
80
60
Sedated
40
Non-sedated
20
0
6
5
4
3
2
Days before death
1
0
%
Comparison of percentage of patients
spontaneously awake in sedated and nonsedated groups during the last week of life
(n=124)
(from Kohara et al. 2005)
100
90
80
70
60
50
40
30
20
10
0
Sedated
Non-sedated
6
5
1
2
3
4
Days before death
0
Comparison of percentage of unrousable
patients in sedated and non-sedated
groups during the last week of life (n=23)
(Fainsinger et al, 1998)
100
%
80
60
Sedated
Non-sedated
40
20
0
6
5
4
3
2
1
Days before death
0
%
Comparison of percentage of unrousable
patients in sedated and non-sedated groups
during the last week of life (n=124)
(Kohara et al, 2005)
80
70
60
50
40
30
20
10
0
Sedated
Non-sedated
6
5
4
3
2
1
Days before death
0
What is in the Name?
The root of the word sedation is the
Latin sedatio meaning ‘soothing’ or
‘allaying’
The clinical purpose of sedative drugs in
palliation is the reduction of irritability
or agitation, i.e. the relief of distress
Sleep is not the intention but may occur
either:
if a high enough sedative dose is required
to relieve the distress or
If a tired, ill patient is enabled to be
comfortable and relaxed
Continuousness
 Sedation is a response to a symptom
 Continuous symptoms need continuous relief
 See use of regular morphine in chronic pain for
details
 30% of patients receiving sedatives do so only
on an ‘as required’ basis
 Median 2.5mg midazolam on a median of 2
occasions (Dunn et al., 2008)
 Liverpool Care Pathway Guidance suggests use
of a continuous subcutaneous infusion if two
or more ‘as required’ doses of sedative have
been given in 24h (NCPC, 2006)
 So ‘as required’ rapidly becomes ‘continuous’
Continuous sedative administration is neither
rare nor necessarily sinister
Proportionality
 There is a growing consensus that the essence
of sedative use in Palliative Care is
proportionality
 Morita, Tsuneto and Shima, 2002
 De Graeff and Dean, 2007
 Brockaert and Claessens, 2009
 Cherny and Radbruch, 2009
 Hasselaar, Verhagen and Vissers, 2009
 Quill et al., 2009
 But not everywhere. The Dutch National
Guideline on Palliative Sedation speaks of
proportionality but assumes:
 The aim is to reduce consciousness
 The patient should be within 2 weeks of dying
 Administration of fluids should be stopped
 A doctor should be present at initiation of sedation
(KNMG 2005/2009)
Proportionate Responses are key
to Palliative Care Practice
The mode of use of sedatives is
analogous to that of other symptom
control measures, such as opioids for
pain:
A low initial dose is titrated higher against
the response until distress is relieved, i.e.
the dose used is proportional to severity of
distress
Relief of distress is the end-point, not a
particular level of consciousness
What is a Proportionate Dose of
Sedative?
 Midazolam is the most commonly used sedative in
Palliative Care (Sykes and Thorns, 2003a)
 Mean midazolam doses reported range from 22 to
70mg/24h (Mercadante et al., 2009)
 But individually as high as 240mg/24h
 In our study of 238 patients:
 Overall mean midazolam dose was 25.7 mg/24 h
 Mean midazolam dose for patients receiving sedation
throughout the last week of life
was 54.5 mg/24 h (Sykes and Thorns, 2003b)
Effect of Sedation on Palliative
Care patients’ Survival
Study
Stone, 1997
(UK)
With
sedation
18.6 days
Ventafridda, 1990 25 days
(Italy)
Without
sedation
19.1 days
23 days
Chiu, 2001
(Taiwan)
28.5 days
24.7 days
Sykes, 2003b
(UK)
38.6 days
14.2 days
Kohara, 2005
(Japan)
28.9 days
39.5 days
Duration of Sedation
Mean duration of sedation
estimated to be 2.5 days (range
1.3-3.9)
Based on ten studies, totalling
1,900 patients (Porta Sales, 2001 updated)
Suggests that sedation is generally a
response to symptoms associated
with the onset of dying
Midazolam use at St
Christopher’s
In a random recent month:
 55 patients died
 51 (93%) had at least one dose
 35 (64%) had a continuous s.c. infusion
 14 (40%) of infusions started within 48h of death
 14 (40%) of infusions started 3 to 7 days before
death
 All had either already stopped eating or ate until 3 to 5
days before death
 7 (20%) infusions lasted between one week and
one month
 Of these patients five continued to eat until 3 to 5 days
before death
 The other two had gastrostomy feeding
Was our sedation rate 93% or zero?
Case History 1
58 year old man with astrocytoma
General condition noted to be
deteriorating
Developed an acute onset of violent
agitation and paranoia
Midazolam 20mg given i.m. stat
followed by 55mg/24h by s.c.
infusion
Died 55 hours later
Case History 2
70 year old woman with lung cancer and a
previous history of schizophrenia
 Admitted because of general deterioration
 Developed delusions and progressive agitation
unresponsive to haloperidol doses up to 12mg
per day
 Over 24h she received 125 mg
levomepromazine and 60 mg midazolam by
s.c. infusion, but also another 60 mg
midazolam and 200 mg levomepromazine in
s.c stat doses for continuing agitation
 At the end of this period her breathing was
noted to be noisy. 200 mg phenobarbital was
given s.c. and the patient died 6 h later.
And yet some will ask:
Is sedation used to cover up potentially
remediable delirium?
73% of delirium in palliative care is irreversible
Life expectancy of patients with irreversible
delirium is under 17 days (Leonard et al., 2008)
What about provision of hydration and
nutrition?
This is a separate decision, but the great
majority of patients who receive sedatives
already have minimal oral intake
What about sedation for existential
distress?
Does not correlate with physical deterioration
Use of sedatives in existential or
psychological distress
 Hard to tell if such distress is really intractable
 Level of distress can be variable and idiosyncratic
 Standard treatments have low morbidity
 Intractability can only be decided by a
multiprofessional clinical team skilled in psychological
care who know both patient and family and have made
repeated assessments
 Team access to psychiatry, chaplaincy and ethics is
required
(Cherny and Radbruch, 2009)
 Some sedative use may be helpful, as may
respite sedation to provide periods of ‘time out’
 But the induction of sleep for extended periods
should be a truly exceptional occurrence
Conclusions
 ‘Sedation’ continues to mean different things to
different people
 In specialist palliative care units use of
sedatives in the last days of life is not
associated with shortened survival overall
 Most use of sedatives is for the management
of restlessness and confusion occurring as part
of the process of dying
 Impaired consciousness is common at the end
of life with or without sedatives
 The aim of sedative use is to relieve distress,
not to induce sleep
 The key to ethical use of sedatives is
proportionality, whatever the indication
If Palliative Sedation is
approached properly…
It will be an act of Mercy for our
patients whose distress cannot be
relieved by other means
It will be a Myth that it shortens
patients’ lives
And so
It will not be a Euphemism for
Euthanasia