Download Consultation table with responses from the developers

National Institute for Health and Clinical Excellence
Stroke Consultation Table
March 2008
Developer’s Response.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
SH
Association of British
Neurologists
1
Full
General
Overall, we consider the guidelines
Thank you
provide a thorough review of the
relevant literature, which has been
appropriately interpreted. The
recommendations are comprehensive
and all have our full support. The text
clearly needs editing as several typos
are still present but we have ignored
these.
We have the following comments, where
we consider that additional points,
literature or recommendations should be
considered.
All the following comments refer to the
full version, but we have included the
relevant NICE version paragraph
numbering as well.
SH
Association of British
Neurologists
2
Full
1.1.1.3
This section needs to emphasise that
the WHO definition is outdated, given
that clinicians should not be waiting 24
hours to make a diagnosis of stroke and
diagnosis these days should be based
on imaging – either an abnormal CT or
MRI if CT normal.
Similarly, there are more modern
definitions of TIA that use imaging to
distinguish between true TIA and
transient symptoms with cerebral
1
Thank you. We have kept the WHO
definition for completeness but expanded
this section to state that anyone with
continuing neurological signs at the time
of assessment should be assumed to
have had a stroke, whatever the time
since onset. We do not feel that there is
evidence to support a definition of stroke
or TIA based on imaging.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
infarction
The guideline could take a bold step and We agree that this might be helpful but
recommend revising the definition of TIA feel that redefining stroke and TIA is
and stroke in the UK – this would have
beyond the scope of the guideline.
the effect of emphasising the change in
culture underlying the National Stroke
Strategy and these guidelines.
It should be noted that brain attack is a
useful term to describe the presentation
as an emergency of stroke and TIA,
especially to lay people by analogy with
heart attack (but is not sufficient as a
diagnosis, since it includes stroke
mimics e.g. epilepsy).
The implication in the document that
Stroke and TIA have different pathways
e.g. the separate algorithms, risks
causing confusion to the public and
particularly the ambulance services
about the emergency management of
TIA. It is not necessary for the
ambulance services to try and
distinguish between TIA and stroke and
certainly not practical for them to
administer the ABCD2 score. We
therefore recommend that consideration
should be given to a section on the
emergency management of “brain
attack” in the community, which would
emphasise the need to treat the sudden
onset of neurological symptoms that
could turn out to be stroke or TIA as an
emergency by calling 999. This could be
incorporated into the TIA and Stroke
algorithms.
2
We agree that brain attack is a useful
term and have now referred to it in this
section
Thank you for your comment. The
pathways are separate because they look
at whether or not a patient has residual
symptoms. To help clarify this we have
amended the clinical introduction and
algorithms to outline the definition
between acute stroke and TIA to help
understand that these are separate
pathways.
Typ
e
Stakeholder
N
o
SH
Association of British
Neurologists
SH
Association of British
Neurologists
Docum
ent
Developer’s Response.
Section
number
Comments
3 Full
3.1.1.2
“Specialist assessment within 24 hours Thank you the section has been
of symptoms” for high risk TIA is not
amended accordingly
sufficient without access to urgent
investigations. This would be better
worded “Specialist assessment and
relevant investigations within 24 hours of
onset of symptoms”
4 Full
3.1.1.3
and
7.1.7.1
The recommendation that patients
should be “admitted directly to a
specialist acute stroke unit” would be
clearer if it specified “directly from the
community or Accident & Emergency
Department or Emergency Room”.
This recommendation assumes that
patients will receive specialist
assessment from a doctor specialising in
stroke on the Stroke Unit and the time
scale is not specified. Moreover, in
reality many patients with stroke will not
be admitted directly to stroke units (even
if this becomes hospital policy), but
instead will stay on Acute Admissions
Units until a bed becomes available on
the unit. We would therefore recommend
adding an additional recommendation
stating: “All patients with suspected or
confirmed stroke should have specialist
assessment on arrival in hospital.”
3
Thank you the section has been
amended accordingly
Thank you: we have considered this
suggestion but the view of the GDG is
that the benefits of acute stroke unit are
such that our recommendation is that
patients are directly admitted to acute
stroke units. We have clearly defined an
acute stroke unit
Typ
e
Stakeholder
N
o
Docum
ent
SH
Association of British
Neurologists
5 Full
Section
number
Comments
Developer’s Response.
3.2
We have major reservations concerning
the TIA algorithms.:
1. A large proportion of TIAs will have
recovered by the time they are seen by
health professionals. The FAST and
ROSIER screens will then be negative
(and in any case they have not been
validated for TIA). If a suspected TIA
patient has a positive FAST or ROSIER
test, then they should be on the Stroke
algorithm, not the TIA pathway. Positive
screens should either lead to the Stroke
Algorithm, or they should be removed
from the TIA algorithm. Instead there
should be a single diamond to cover
diagnosis reading “Is history compatible
with TIA?”
Thank you for your comment
Algorithm
1
2. All patients with a confirmed diagnosis
of TIA after specialised assessment
should have brain imaging, since the
pathology is always uncertain unless the
patient has previously been investigated
for the same symptoms. Those with high
ABCD2 scores definitely require imaging
to find out if they have actually had
infarction rather than ischaemia. (see
comment number 9 for further
comments re imaging after TIA).
3. The stenosis level should use the
NASCET criteria i.e. should be 50-99%,
since the evidence is based on the
combined analysis of the trials
performed by Peter Rothwell, in which
NASCET measurements were used.
More importantly, the criteria for
assessing stenosis on carotid imaging
(especially ultrasound) are based on
4
Thank you we have amended the
algorithm for further clarification.
Thank you for your comment. This area
was debated extensively by the group
and the views you put forward were put
forward by some members of the group.
These comments have been reviewed by
the developers who disagree with your
suggestion. This is a consensus based
recommendation based on little evidence.
We have amended the FETR section for
further clarification.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
NASCET criteria, not ECST criteria. The
widespread assumption that we should
use the ECST criteria of 70-99% to
select patients for treatment, means that
a substantial proportion of patients who
have ultrasounds showing 50-69%
stenosis and who would benefit from
treatment, are not being considered for
treatment by physicians and surgeons
unfamiliar with the details of the trials
and ultrasound validation studies.
The evidence discussed used NASCET
and ECST. The GDG considered your
comment but feel that their is no evidence
to state that one criteria should be
recommended over the other. We have
added in a recommendation for further
clarification that states that the criteria
used to determine stenosis level should
be reported.
4. The best medical treatment boxes
should be identical whether or not the
patient has carotid imaging.
Thank you this has been amended
5. The persons referred for carotid
endarterectomy should also have a link
to a best medical management box.
Thank you this has been amended
6. Stenting is an emerging alternative to
carotid endarterectomy and it would
therefore be appropriate to replace
“carotid endarterectomy” with “carotid
intervention”.
The GDG have found no evidence to
support the use of stenting in the acute
setting.The developers therefore do not
agree with your suggestion. For further
information please see section 6.4 of the
FULL guideline.
7. See also comment number 2 above re
incorporating brain attack into the
The term brain attack is refered to in the
algorithm.
introduction of both the FULL and the
NICE guideline. We do not wish to
incorporate this term into the algorithm as
the terms for stroke and TIA would also
need to be incorporated withi this to avoid
confusion.
SH
Association of British
Neurologists
6 Full
3.2
Algorithm
2
In the Stroke Algorithm, it would be
better to replace the text “Surgical
Intervention” with “Consider Surgical
Intervention” since not all of the patients
5
Ok
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
are candidates for surgery.
Patients with cerebellar haematoma
See amended version
should be included on the list considered
for surgical intervention.
There is a typo in “stensosis” at the
bottom right hand corner.
SH
Association of British
Neurologists
7 Full
5.2.6.2
Specialist assessment is not sufficient
Thank you for your comment we have
without access to urgent investigations. amended the guideline accordingly.
One of the main barriers to timely
treatment of patients with TIA has been
delay in access to CT and carotid
imaging. This recommendation would be
better worded “Specialist assessment
and relevant investigations within 24
hours of symptoms”
SH
Association of British
Neurologists
8 Full
5.2.6.3
Ditto, within 1 week (Refers to their
Please see the response above (‘Thank
comment 7 ‘Specialist assessment is not you for your comment we have amended
sufficient without access to urgent
the guideline accordingly.’)
investigations. One of the main barriers
to timely treatment of patients with TIA
has been delay in access to CT and
carotid imaging. This recommendation
would be better worded “Specialist
assessment and relevant investigations
within 24 hours of symptoms” ‘)
6.2.1.1
We strongly disagree with the emphasis
in the statement in 6.2.1.1, that “not all
patients with TIA need brain scanning”.
A proportion of TIAs have subdural
haematomas, brain tumours, AVMs,
small haemorrhages and infarcts without
any symptoms other than an apparently
typical TIA. There is as much
(NICE
version1
.1.2.3)
SH
Association of British
Neurologists
9 Full
(NICE
version
1,2)
6
Thank you for your comment. This area
was debated extensively by the group
and the views you put forward were put
forward by some members of the group.
These comments have been reviewed by
the developers who disagree with your
suggestion. This is a consensus based
recommendation based on little evidence.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
6.2.6.2
and
6.2.6.3
Developer’s Response.
Comments
justification for scanning all TIAs as
there is for scanning all strokes. Some
experts in our group felt strongly that the
guidelines should not discriminate
against TIAs in this way and should
match the guideline that all strokes
should have brain imaging. We therefore
recommended having a guideline that
reads: “All TIA patients should have
brain imaging as part of their
assessment”. Not all experts in the
group supported this blanket
recommendation, pointing out that
patients with isolated retinal TIA and
some low risk TIAs may not require
imaging. A compromise
recommendation was suggested to read:
“All TIA patients should have brain
imaging considered as part of their
assessment”. All experts agreed that all
high risk TIAs require brain imaging to
exclude infarction. Hence, we suggest
The evidence and clinical consensus did
that as a minimum there should be a
not support this recommendation.
recommendation reading: “All high risk
TIA patients should have brain imaging
as part of their assessment”.
Similarly, the recommendations that only
patients with TIA “in whom vascular
territory or pathology is uncertain”
should have brain imaging implies that
the pathology is often certain. This is
misleading. Fortunately, only a small
proportion of patients have pathology
other than ischaemia on CT or MRI, and
hence those who do not scan TIAs
because they think they know the
pathology will not be wrong very often.
However, that does not mitigate the risk
7
The clinical consensus of the group was
that not all patients groups specified
within the two recommendations required
brain imaging. For clarification we
included a box to explain the population
of people who imaging may be helpful.
We did not feel that imaging should
replace clinical judgement and removing
the wording you suggest would be delay
access to imaging in those patients in
whom imaging may be required.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
of missing serious pathology. We would
therefore strongly advise rewording the
recommendations by deleting the phrase
“in whom vascular territory or pathology
is uncertain” from recommendations
6.2.6.2 and 6.2.6.3.
SH
Association of British
Neurologists
10 Full
( NICE
version
1.2.4.1)
6.4.4.1
This paragraph overstates the evidence
and misquotes the findings of reference
42 (see Table 2 in ref 42). The
comparison of patients undergoing CEA
less than one week since symptoms
only included 6 events from patients
operated early, and thus the confidence
intervals were very wide. Three out of 4
studies which compared treatment less
than 3 weeks after symptoms with those
treated more than 3 weeks after
treatment showed slightly worse
outcomes in those treated early. In those
treated either side of 4 weeks there was
no difference and it was only the 2
studies comparing less than 6 weeks
with more than 6 weeks that showed a
trend to better outcome with earlier
treatment. It should be noted that these
patients included TIAs and all had to be
neurologically stable. The paper (table
1) shows that patients with unstable
neurological symptoms did worse when
operated early.
We would therefore recommend that
paragraph 6.4.4.1 should be reworded to
read: “The systematic review reported
that there was no statistical difference
for the outcome of perioperative stroke
and death when comparing patients who
were neurological stable undergoing
8
Thank you.We have amended the text
quoting the number of events for one
week vs. greater than one week and
included the ORs and CIs. We have
added a sentence to indicate which
patients did worse if operated on early.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
CEA early (1-6 weeks) after symptoms
than those undergoing the procedure at
a later time. Patients operated early with
unstable neurological symptoms (stroke
in evolution, non-specified ‘urgent’
cases, and crescendo TIA) did worse if
they were operated in the acute phase
compared to later operation.
SH
Association of British
Neurologists
11 Full
6.4.4.2
It would be helpful to have the wording
Thank you. The wording has been made
in this paragraph clarified. The analysis clearer
of 5-year ARR refers to the delay from
symptoms to randomisation, but in
places seems to have been interpreted
as the delay between symptoms and
performance of CEA, which is not the
same and can not be calculated from the
trial data.
SH
Association of British
Neurologists
12 Full
6.4.5.1
We note the statement that “No
evidence for early carotid stenting
(within the two week period of the
guideline) was identified”. However, we
are aware of 2 publications addressing
this question. The first (Topakian et al,
Eur J Neurol 2007;14:672-678)
describes a case series of 77 patients
with symptomatic carotid stenosis
treated by stenting. Those treated less
than 2 weeks after symptoms had a
significantly higher 30 day rate of stroke
or death. The second (Groschel et al.,
Eur J Neurol 2008;15:2-5) reports a
much larger series in which treatment
less than 2 weeks from symptoms was
not associated with an increased rate of
complications compared to those treated
later.
9
This paper was published outside of the
literature review cut-off date and is a case
series. We have clarified that the papers
included were restricted to RCTs within
the two week period of the guideline.
This is a case series; although the
numbers of patients are greater in this
paper than in the previous paper quoted,
the authors of the paper concede that
there are potential biases in the selection
of patients for inclusion; for example in
the earlier part of the study only patients
thought not to be suitable for cea were
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
Carotid stenting is increasingly being
used as an alternative to carotid
endarterectomy in patients with contraindications to surgery and in patients not
willing to undergo surgery. It also
continues to be tested in a randomised
comparison with carotid endarterectomy
in symptomatic patients in the
International Carotid Stenting Study
(ICSS). The data monitoring and
steering committees of the trial have
recently encouraged investigators in the
trial to randomise and treat stable
patients as soon as possible after
symptoms.
recommended for stenting. Subsequently,
patients were offered a choice of CEA or
stenting which may have introduced
some bias. Whilst we recognise that on
occasion case series are helpful, the
GDG did not feel that the evidence fwas
reliable & hence on this occasion was
excluded from the guideline
NICE has previously issued a guideline
on carotid stenting (Interventional
procedure guidance 191, 2006). We
recommend that the NICE Acute Stroke
and TIA guideline should refer to this
earlier guideline and should include a
recommendation concerning carotid
stenting, matching the earlier NICE
Interventional procedure guidance on
the following lines: “Clinicians offering
carotid stenting as an alternative to
carotid endarterectomy should ensure
that patients understand the uncertainty
about safety and the long-term efficacy
of the procedure, and should preferably
include the patients in a randomised
clinical trial.”
10
The IP guidance was not applicable to
the management of acute stroke and is
outside the scope of this guideline. The
GDG did not find any evidence to support
a recommendation on the use of stenting
outside the remit of this guideline. There
is evidence supporting the use of stents
in long term treatment of stroke. This
timeframe falls within the intercollegiate
stroke working party guideline.
Typ
e
Stakeholder
N
o
SH
Association of British
Neurologists
SH
SH
Docum
ent
Comments
13 Full
6.4.5.1
The sentence in this paragraph reading Thank you we have amended the section
“There is less benefit from early surgery accordingly.
in patients who are medically unfit” is not
derived from the evidence. It should be
reworded to read: “There is evidence
showing that patients with unstable
neurological symptoms (stroke in
evolution, non-specified ‘urgent’ cases,
and crescendo TIA) may be harmed by
early surgery.”
Association of British
Neurologists
14 Full
6.4.6
In view of the above, both the first lines
of recommendations 6.4.6.1 and 6.4.6.2
should read: “People with stable
neurological symptoms from acute nondisabling stroke or TIA…”
Thank you we have amended the
wording of the recommendation
accordingly for greater clarification
Association of British
Neurologists
15 Full
6.4.6
Considerable confusion emanates from
the fact that there are 2 methods of
measuring stenosis (see comment re
algorithm 1, above) and radiologists and
ultrasound technicians rarely state what
method has been used in their reports.
It would help to avoid some of this
confusion if NICE included a
recommendation that read: “Reports on
carotid imaging should state the method
used (ECST or NASCET) to calculated
the stenosis measurements.”
The evidence based looked at 2 methods
of measuring stenosis. The GDG did not
feel that there was evidence that one
method should be recommended over the
other.
The recommendation has been inserted
to avoid confusion.
6.4.6
The final bullet point of 6.4.6.2
concerning best medical management,
should be moved to a separate
recommendation on its own, since it
applies equally to the patients requiring
carotid endarterectomy.
The mention of aspirin and dipyridamole
Thank you we have added the last bullet
of the recommendation as a third bullet of
the first recommendation. Please see
changes accordingly.
(NICE
version
1.2.4.1
And
NICE
version
1.2.4.2)
SH
Developer’s Response.
Section
number
Association of British
Neurologists
16 Full
11
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
in square brackets should be removed,
or clopidogrel added. There is some
evidence that the combination of aspirin
and clopidogrel is beneficial in short
term use prior to carotid endarterectomy
in patients with recent symptoms from a
randomised trial known as CARESS and
NICE may wish to review and
incorporate this evidence into the
guideline (Circulation. 2005 May
3;111(17):2233-40)
We have removed reference to specific
anti-platelet agents
Thank you. This paper was not included
in the evidence review because it
compares aspirin and clopidogrel
SH
Association of British
Neurologists
17 Full
8.1.1.1
The last sentence of this paragraph is
muddled. Embolism is usually the result
of thrombosis. Thrombosis and
embolism from atherosclerosis is
associated with platelet rich thrombus,
and cardiac thrombo-embolism with
thrombin rich thrombus. We suggest
deleting the last sentence.
Thank you this sentence has been
removed accordingly
SH
Association of British
Neurologists
18 Full
8.1.7.1
Although we understand that the GDG is
basing its recommendations on the IST
and CAST data, it seems to us that there
is no particular logic in delaying starting
dipyridamole until 2 weeks after onset,
or in continuing 150-300mg aspirin after
a loading dose, when we know from
other studies that 75mg is sufficient. The
one randomised trial of early
dipyridamole vs control showed no
additional harm from the combined
preparation in acute stroke and other
trials have shown that dipyridamole does
not cause cerebral haemorrhage. The
danger of delaying “definitive long-term
antithrombotic treatment until 2 weeks”
is that it will be forgotten. It is much
more effective to start long term
Thank you. We have added that patients
being discharged earlier can be started
on definitive secondary prevention (eg
aspirin and dipyridamole earlier
(NICE
version
1.4.1.1)
12
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
treatments on admission, than to plan to
start them later, given that in practice
such plans are often neglected. We
would suggest that a consensus
recommendation might be better
worded: “Patients should be given a
loading dose of aspirin 300mg;
thereafter they should receive a
combination of aspirin (50-75mg daily)
and dipyridamole MR 200mg b.d.”
SH
Association of British
Neurologists
19 Full
8.2.1.1
SH
Association of British
Neurologists
20 Full
8.2.6.1
(and
NICE
version
1.4.2.1)
SH
Association of British
Neurologists
21 Full
(and
NICE
version
1.4.4.1)
8.4.1
MR venography is not particularly
reliable as hinted at in this paragraph.
CT venography is superior and should
be mentioned here.
.
Both the randomised trials examining
the treatment of cerebral venous
thrombosis studied the use of heparin
(one with standard intravenous heparin
alone, and one a low molecular weight
heparin for 3 weeks followed by
warfarin.) The recommendation should
therefore read: “People ……should be
fully anticoagulated with heparin acutely,
followed by full-dose oral anticoagulation
(INR 2-3)….”
Thank you the guideline has been
amended accordingly
Thank you the guideline has been
amended accordingly
A distinction could be made between the Thank you for your comment. We did not
full blown antiphospholipid syndrome
review any evidence to enable us to
associated with lupus like disorders and make this distinction.
simply the association between
antiphospholipid antibodies and stroke.
Most experts would anticoagulate and
immunosuppress the former, but this
management is outside the scope of this
guideline. It would therefore be better in
paragraph 8.4.1 to say “The clinical
question remains as to whether patients
13
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
8.4.2
8.4.5.1
Developer’s Response.
Comments
with acute stroke found to have
antiphospholipid antibodies without other
major features of the syndrome should
be anticoagulated…”
The NICE researchers have missed the
pivotal study that addressed this
question, which was a nested cohort
study within a randomised comparison
Thank you. This paper does not address
of aspirin versus warfarin in stroke
the acute stroke population (within the
prevention conducted by the APASS
first 2 weeks) and hence was excluded
Investigators (JAMA 2004;291:576-584).
This showed that the risk of recurrent
stroke in those with antiphospholipid
antibodies was identical in patients
treated with aspirin or warfarin. The
authors concluded: “The presence of
aPL among patients with ischemic
stroke does not predict either increased
risk for subsequent vascular occlusive
events over 2 years or a differential
response to aspirin or warfarin therapy.
Routine screening for aPL in patients
with ischemic stroke does not appear
warranted.”
The above reference supports the
recommendation as being evidence
based and the foot note should therefore
be deleted.
SH
Association of British
Neurologists
22 Full
(and
NICE
version
1.4.7.1)
8.7.6.1.
The recommendation is that clotting
levels should be restored to normal as
soon as possible and the evidence
quoted shows that PCC acts much
quicker than other treatments. We
therefore consider that the
recommendation should add “Using
PCC”. The algorithm should also make
14
Thank you. We have amended the
guideline accordingly
Typ
e
SH
Stakeholder
Association of British
Neurologists
N
o
Docum
ent
23 Full
Section
number
9.3.6.1
(and
NICE
version
1.5.3.1)
Comments
Developer’s Response.
this recommendation and suggest a
treatment regime.
Concerning the exceptions to blood
pressure manipulation, we would add
“hypotension e.g. systolic BP less than
90mm”.
Thank you. There js no evidence
reviewed for this suggestion. People with
hypotension would be managed
according to best medical practice.
We would add a consensus
recommendation on the following lines:
“When hypertension requires
manipulation in the context of stroke,
close monitoring of blood pressure and
the effect of any agent is essential (e.g.
on an ITU) to avoid over-rapid reduction
in blood pressure or hypotension.”
The NINDS trial protocol required high
blood pressure to be lowered prior to
thrombolysis. It would therefore be
consistent with the evidence to add a
recommendation reading: “Blood
pressure reduction to 185/110 or less
should be considered in thrombolysis
candidates.” However, we are aware
that there is no other evidence to
specifically support this
recommendation.
We did not find any evidence to support
this statement specifically in acute stroke
Thank you the guideline has been
amended accordingly
SH
Association of British
Neurologists
24 Full
13.1.4.3
There is an important error concerning
the benefit of early surgery in STITCH
with haematomas less than 1cm from
the surface, in line 2: “unfavourable”
should read “favourable”.
SH
Association of British
Neurologists
25 Full
13.1.6.3
The recommendation should include
Thank you the recommendation has been
cerebellar haemorrhage as well as lobar amended for further clarification
haemorrhage, so that it reads: “…have a
lobar haemorrhage or cerebellar
haematoma with hydrocephalus or are
(and
Nice
version
15
Thank you. This has been amended
Typ
e
SH
Stakeholder
AstraZeneca UK Ltd
N
o
1
Docum
ent
Section
number
1.9.1.3
and
1.9.1.4)
13.1.6.4
Full
Comments
Developer’s Response.
deteriorating neurologically…”
Patients who present in coma from
cerebellar haematoma often make an
excellent recovery after surgical
evacuation of the haematoma.
Recommendation 13.1.6.4 suggesting
that patients with a GCS of less than 8
rarely require surgical intervention
should therefore not apply to cerebellar
haematoma. We therefore recommend
changing the last bullet point to read: “a
GCS of less than 8, unless due to a
cerebellar haematoma”
General
Thank you the guideline has been
amended accordingly
Thank you for your comments
AstraZeneca appreciates this
opportunity to comment on the NICE
Acute Stroke and TIA clinical guideline.
Kindly see our comments on individual
points below.
SH
AstraZeneca UK Ltd
2 Full
8.6.1.1
AstraZeneca supports the use of
Thank you we agree
Baigent et al, Lancet 2005 as evidence,
from which the GDG draws the following
conclusion: “A reduction in
concentration of LDL cholesterol by 1
mmol/L with statins over a 5 year period
reduced the relative risk of any vascular
event by 20%.” AstraZeneca believes
this may be open to the misinterpretation
that a reduction of only 1 mmol/L in LDL
cholesterol was observed in the study.
The authors of the study interpret their
findings as follows: “Statin therapy can
safely reduce the 5-year incidence of
major coronary events, coronary
revascularisation, and stroke by about
16
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
one fifth per mmol/L reduction in LDL
cholesterol.”1 To avoid misinterpretation
AstraZeneca suggests that the GDG
update their comments pertaining to this
study to read: “For each 1 mmol/L
reduction in concentration of LDL
cholesterol achieved with statins over a
5 year period, there was an associated
relative risk reduction of any vascular
event of 20%.”
Further evidence regarding the LDL
This paper was excluded in the evidence
cholesterol lowering efficacy of statins
review because it is not on acute stroke
can be found in the meta-analysis of
patients
Law et al2. This analysis involved
approximately 24,000 statin-treated
patients & rosuvastatin was shown to be
the most effective agent at reducing LDL
cholesterol.
1. Baigent C, Keech A, Kearney PM et Thank you. These papers were excluded
al. Efficacy and safety of cholesterol- in the evidence review because they are
lowering treatment: prospective
not on acute stroke patients
meta-analysis of data from 90,056
participants in 14 randomised trials
of statins. Lancet. 2005;
366(9493):1267-1278
2. M R Law, N J Wald and A R
Rudnicka, BMJ 2003;326;1423doi:10.1136/bmj.326.7404.1423
SH
AstraZeneca UK Ltd
3 Full
8.6.1.1
Typographical error at line 5:
“Observational studies have not shown
as an association…”
SH
AstraZeneca UK Ltd
4 Full
8.6.1.1
Thank you for your comment. The
guideline has been amended accordingly
Thank you for your comment. The
17
Typ
e
SH
Stakeholder
AstraZeneca UK Ltd
N
o
5
Docum
ent
Full
Section
number
Comments
Developer’s Response.
Repeated word at line 44: “The
FASTER study study planned to
assess…”
guideline has been amended accordingly
10.3.5.1
Typographical error at line 3: “with
respect to mortality, morbidity, aderse
adverse events…”
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
1
Full
General
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
2 Full
6.3.7.1
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
3 Full
7.2.6.1
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
4 Full
7.2.6.1
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
5 Full
7.2.6.2
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
6 Full
8.1.7.3
SH
Avon, Gloucestershire &
7 Full
8.1.7.4
We believe the dose of aspirin quoted
throughout the document should be
300mg not 150-300mg
We believe that this statement should
include “patients at high risk of stroke
(ABCD2 >=4 should have carotid
imaging within 48 hours” in line with the
national stroke strategy
We believe the phrasing should mirror
the National Stroke Strategy and state
that “Brain imaging are scanned in the
next scan slot within usual working
hours, and within 60 minutes of request
out-of-hours”
Thank you for your comment. The
guideline has been amended accordingly
For clarity we have changed to 300mg for
acute treatment throughout the guideline
We acknowledge that we are at variance
with the DOH strategy; however after
considering the evidence the GDG
consider this recommendation to be
underpinned by the evidence base.
We acknowledge that we are at variance
with the DOH strategy however after
considering the evidence we consider this
recommendation to be more in keeping
with the evidence available. We state
“ideally the next slot and definitely within
the hour whichever is sooner” which is
much the same.
First bullet point we question the
Thank you for your comment. Venous
indication to anti-coagulate on admission stroke, recurrent emboli and possibly
dissection may be indications to
anticoagulate on admission
We believe the phrasing should mirror
We acknowledge that we are at variance
that in the National Stroke Strategy
with the DOH strategy however after
considering the evidence we consider this
recommendation to be more evidence
based.
We feel the phrase “genuinely intolerant” Thank you for your comment. The
is ambiguous and needs to be more
wording is taken directly from the NICE
specific
TA we have included the footnote that the
TA used to define ‘genuinely intolerant’
We also feel that the word “ischaemic”
Thank you the guideline has been
18
Typ
e
SH
Stakeholder
Wiltshire Cardiac Network
Avon, Gloucestershire &
Wiltshire Cardiac Network
N
o
Docum
ent
Section
number
8 Full
8.1.7.4
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
9 Full
8.5.2.1
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
10 Full
8.6.6.1
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
11 Full
8.7.6.1
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
12 Full
8.7.6.1
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
13 Full
8.8.7.2
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
14 Full
8.8.7.4
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
Avon, Gloucestershire &
Wiltshire Cardiac Network
15 Full
9.3.6.1
16 Full
9.3.6.1
Avon, Gloucestershire &
Wiltshire Cardiac Network
17 Full
10.2.6.2
SH
SH
Comments
Developer’s Response.
should be removed from this sentence
We note that there is no explanation of
the use of anti-coagulation in AF – we
suggest a reference to section 8.8.7
should be included
We believe further clarification would be
helpful on the phrase “physicians trained
and experienced in the management of
acute stroke” to indicate that this can
include ED physicians, stroke physicians
and neurologist who are appropriately
trained and supported in line with the
interpretation of the licence from Dr
Roger Boyle (attached)
We feel that some indication of
timeframe would be helpful e.g. “until
haemorrhage has been excluded”
We feel that the phrase “as soon as
possible” should be replaced by
“immediately”
We also believe a comment suggesting
discussion with a haematologist on
diagnosis on the use of prothrombin
complex concentrate would be helpful
We feel the word “disabling” is difficult to
interpret and would suggest that further
clarification would be helpful
Whilst we agree with this statement in
principle we feel clarification on timing of
this treatment
We believe this should refer to the
manipulation of “high” blood pressure
We feel that an additional bullet point
should be added for “facilitation of
thrombolysis”
amended accordingly.
Thank you the guideline has been
amended accordingly.
Whilst we agree with this statement we
feel a definition of “early” would be
helpful. The Sentinel Audit uses 48
19
Thank you for your comment the wording
of the recommendation was quoted
verbatim from the NICE alteplase TA. We
have amended recommendation 8.5.2.2
for further clarification.
Please see the amended wording of the
recommendation
Thank you for your comment it is
impossible to do this immediately
The GDG felt this was not mandatory and
might induce delay
Thank you we have clarified this
terminology within the guideline
introduction.
We did not review timing of treatment
within the evidence.
We have amended the guideline
Please see the response to comment 23
(‘Thank you the guideline has been
amended accordingly Please see the
response to comment 23
Thank you for your comment. The
definition of early was based on clinical
consensus following the review of the
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
18 Full
10.3.6.1
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
19 Full
13.1.6.2
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
20 Full
13.1.6.5
SH
Avon, Gloucestershire &
Wiltshire Cardiac Network
21 Full
13.2.6.1
SH
Blood Pressure Association
1
NICE
Patientcentred
care
SH
Blood Pressure Association
2
Nice
1.1.2
Comments
Developer’s Response.
hours. We note that clarification is given
in a footnote but we feel this should be
added to the recommendation in this
section
We are not clear what the evidence
base for this recommendation is the
Food trial did not support early tube
feeding
evidence. The footnote has been inserted
into the recommendation (now section
10.1.6.2.
Thank you for your comment. The FOOD
trial did not support nor argue against
early tube feeding. The rationale for
recommendation is explicit in the from
evidence to recommendation section
We agree with this statement but feel
Thank you for your comment. It is beyond
that it should also be recommended that the remit of a clinical guideline to name
these patients “are discussed with a
specific professions.
neurologist”
We feel an additional bullet point should Thank you for your comment. The
be added for patients with “posterior
guideline has been amended accordingly.
fossa haemorrhage”
Whilst we agree that rapid treatment is
Thank you the timescales are based on
necessary for these patients we feel the the trial for hemicraniectomy
timescales indicated here are too long
i.e. referred and treated within 24 hours
We welcome the commitment to patient- Thank you for your comment. As part of
centred care in the introductory sections, the publication of the NICE guideline
but we are concerned that this
“understanding NICE guidance” (UNG)” a
commitment is not carried through in the booklet is produced to help patients &
main body of the report. We
carers to understand the
acknowledge that established systems
recommendations made within the
of patient and carer engagement may
guidance. It also includes helpful
appear contrary to the urgency of the
questions for patients or carers to engage
acute setting, but we would argue that
with healthcare professionals. Patient
this makes it all the more important to
centred care is covered in much more
institute clear routes for engagement in detail in the Royal College of Physicians
this document. In general, we feel there stroke guidelines.
is a risk of patient-centred care being
accepted in principle but not in practice,
and of achieving only a limited role
compared to so-called “real medicine”.
We are dismayed that the assessment
Thank you for your comment secondary
of blood pressure is not given greater
prevention is covered within the RCP
prominence in this section. TIA is clearly ICSWP guideline
20
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Blood Pressure Association
3
Nice
1.1.2
SH
Blood Pressure Association
4
Nice
1.2.4.2
Developer’s Response.
Comments
the single greatest risk factor for stroke,
but it does not work in isolation.
Hypertension confers up to a sevenfold
increase in the risk of stroke, and is
thought to account for up to 60% of
strokes worldwide. Although blood
pressure is measured in the ABCD test,
it remains the single feature of the test
that can be altered. The early treatment
of TIA should include the assessment
and treatment of hypertension more
prominently, to acknowledge its
importance in the prevention of stroke
post-TIA. Inclusion in the
commencement of secondary prevention
does not achieve this. It may also be
sensible to differentiate between blood
pressure management after TIA, and
blood pressure management after stroke
(which is covered to a degree in section
1.5.3 of the Nice version).
People with TIA should be given
opportunities to discuss their health and
future health risks with an informed
health professional. Although the place
for this is more likely in the longer-term
follow-up post-TIA, there is a role for
information provision at the point of
assessment. An assessment such as
the ABCD can provide a picture of
current risk of stroke, but the person’s
health behaviours in the longer term
period will also impact significantly on
their stroke risk. Information is needed at
this point to educate the person and
their family as to the required measures
they can take to manage their health
and their risk of future problems.
We are pleased that blood pressure
21
Thank you we have added “including
discussion of individual risk factors” into
the guideline. Information provision is
outside the scope of the guideline.
Thank you for your comment. Your
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Blood Pressure Association
5
Nice
1.5.3
SH
Blood Pressure Association
6
Nice
1.5.3
SH
Blood Pressure Association
7
Nice
General
Comments
Developer’s Response.
control is included in “best medical
treatment”. We would like to see it
included elsewhere in the document, for
example in section 1.1.2 above. We
would further wish to see information on
prevention of stroke or stroke recurrence
being provided at this point, and at other
points in the document. Once again, we
would stress that the role of information
is obscured in this document, and needs
to be given greater clarity.
We accept the recommendation that
blood pressure manipulation in acute
stroke may not be good practice.
However, we would refer to section
1.4.6.3 of the Nice guidance regarding
statin treatment post-discharge. We
would argue that, as a major risk factor
for stroke, assessment and treatment of
hypertension should be considered in
the context of discharge after stroke,
and would wish to see a
recommendation that information and
treatment be considered in hypertensive
people at this point.
We would question whether the
inclusion of blood pressure control in this
section (as opposed to section 1.4
where statin treatment is considered)
downgrades the relative importance of
blood pressure and hypertension to TIA,
stroke and secondary prevention of
stroke
We are concerned that this guidance
makes no mention of treatment and
secondary prevention. While the acute
setting is not where the bulk of this work
will take place, we would argue that the
initiation of rehabilitation and secondary
suggestion is outside the remit of our
guideline.
22
Thank you for your comment this
suggestion is outside the remit of the
acute stroke guideline. This information
would be included as part of secondary
prevention. Please refer to the royal
college of physicians intercollegiate
stroke working party which includes
secondary prevention
Blood pressure control is not usually part
of the immediate management of a
patient with stroke which is considered in
1.4
Agreed: comment added to 1.1.2
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Blood Pressure Association
8
Nice
4.5
SH
Boehringer Ingelheim Ltd
1
Full
Table 16;
page 162
SH
Boston Scientific Limited
1
Full
General
Developer’s Response.
Comments
prevention should take place at the
earliest possible stage. The lack of any
sense of follow-on from the acute setting
is troubling, as it does not suggest a
seamless transition into longer-term
treatment.
We welcome the research
recommendation concerning blood
pressure manipulation in acute stroke. In
particular, the SCAST trial will provide
some insight into the use of angiotensin
receptor blockers. However, we would
wish for a more detailed
recommendation including the various
other blood pressure treatments which
have not been considered at this point,
in particular ACE inhibitors. More
prominence should also be given to the
long-term efficacy of blood pressure
treatment in reducing mortality and
vascular outcomes.
We welcome this document as a fair and
balanced appraisal of the evidence for
the use of antiplatelet agents in
secondary prevention of stroke.
However, for accuracy all the trial
evidence for the use of dipyridamole in
stroke relates to the “modified release
formulation” of dipyridamole. We
therefore recommend that any reference
made in the document to the use of
dipyridamole is corrected to “modified
release dipyridamole”.
•
The NICE clinical guideline is an
extremely timely follow-up to the Stroke
Strategy. The chapter ‘Time is Brain’ is
clear: immediate diagnosis and
management of stroke and TIA can save
lives, avoid subsequent strokes and
23
This relates to secondary prevention
which is covered within the ICSWP
Thank you
We have amended the guideline
accordingly.
Thank you. We did not review any RCT
evidence for carotid stenting within the
‘acute’ two week period of the guideline
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
improve outcomes of stroke survivors
•
The role of carotid stenting in
the patient pathway is unclear in the
draft guideline and should be
systematically mentioned as an
alternative to endarterectomy
•
The evidence base used for
carotid artery stenting is incomplete and
should include the several randomized
controlled trials between stenting and
surgery (see below)
•
If both treatments are clinically
indicated, patients should be offered the
choice of a minimally-invasive option
SH
Boston Scientific Limited
2
NICE
1.2.4
and
1.2.4.1
SH
Boston Scientific Limited
3
Full
Algorithm
The title of 1.2.4 refers to carotid
endarterectomy and carotid stenting.
However the details in 1.2.4.1 only offers
one option: patients should ‘be assessed
and referred for carotid endarterectomy
within 1week of onset of symptoms’
The option of carotid stenting should be
offered in the details of 1.2.4.1
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.
Please see reponse to comment 61
(‘Thank you. We did not review any RCT
evidence for carotid stenting within the
‘acute’ two week period of the guideline
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.’)
Please see response to 61
Carotid stenting is absent from the
Please see response to 61 Please see
treatment pathway algorithm and should reponse to comment 61(‘ Thank you. We
be added
did not review any RCT evidence for
carotid stenting within the ‘acute’ two
week period of the guideline
24
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.’)
SH
Boston Scientific Limited
4 Full
6.4.2.1
p.43
See the presentation of CAS vs CEA
randomized trials below
Thank you. We did not review any RCT
evidence for carotid stenting within the
‘acute’ two week period of the guideline
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.’)
SH
Boston Scientific Limited
5 Full
NICE
6.4.5.1
p.45
1.2.4.1
p.10
The guideline indicate that patients
should ideally have a carotid intervention
within 2 weeks of the onset of
symptoms. It is derived from a pooled
analysis of NASCET and ESCT and
rated as the highest level of evidence.
Considering the evidence presented in
comment 4 and the equivalence (or noninferiority) between stenting and surgery,
the recommendation on time to
intervention should be extended to
stenting. The NICE guideline should
state that
Paragraph 1.2.4.1
People with acute non-disabling stroke
25
Thank you. We did not review any RCT
evidence for carotid stenting within the
‘acute’ two week period of the guideline.
We have amended the FETR to help
clarify the rationale behind the GDG
decision for this.
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.’)
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
or TIA who have symptomatic carotid
stenosis of 50–99% according to the
NASCETcriteria, or 70–99% according
to the ECST criteria, should:
• be assessed and referred for carotid
intervention (endarterectomy or
stenting) within 1week of onset of
symptoms
• receive treatment within a maximum
of 2 weeks of onset ofsymptoms.
The abstract below corroborates this
assumption:
Reference: Gröschel K, Knauth M,
Ernemann U, et al. Early treatment after
a symptomatic event is not associated
with an increased risk of stroke in
patients undergoing carotid stenting. Eur
J Neurol. 2008 Jan;15(1):2-5. Epub
2007 Nov 14.
A recently symptomatic carotid artery
stenosis carries a high risk of
subsequent ischaemic events and thus
requires rapid treatment. We
investigated the influence of the time
delay between the last symptomatic
event of a carotid stenosis and
subsequent carotid artery stenting (CAS)
with respect to the combined 30-day
outcome of stroke and death. In a group
of 320 patients undergoing CAS the
median delay before the intervention
was 19 days (interquartile range 10-36)
and the combined 30-day complication
rate was 8.4%. Time delay was not
significantly associated with periprocedural complications, regardless of
26
This paper is a case series and therefore
was not considered by GDG
Thank you. This paper is outside of the
cut off date period for literature search.
In addition only RCTs were appraised.
This has been clarified in the text.
Typ
e
Stakeholder
SH
British and Irish Orthoptic
Society
SH
British and Irish Orthoptic
Society
SH
British and Irish Orthoptic
Society
SH
British Association of Stroke
Physicians
N
o
1
Docum
ent
Full
2 Full
3
1
Section
number
General
6.3.4.1
Full
11.1.6.1
Full
general
Comments
Developer’s Response.
whether this variable was dichotomized
(<14 days and > or =14 days),
separated into interquartile ranges or
analysed as a continuous variable. Our
results indicate that early CAS is not
associated with an increased
complication rate in patients with a
recently symptomatic carotid stenosis.
Thus, if CAS has been selected as the
treatment modality for a patient, it should
be performed as soon as possible to
maximize the benefit of the intervention
in reducing the risk of stroke.
Thank you. We did not review any RCT
evidence for carotid stenting within the
‘acute’ two week period of the guideline
The guidelines are helpful in identifying
the areas of good practice and areas
where there are substantial gaps in the
knowledge base.
We agree that amaurosis fugax should
be considered an indicator for carotid
artery disease. Please note the correct
spelling of this condition.
We recommend that assessment of
visual status be specified. This should
include assessment of eye movement in
addition to assessment of visual acuity,
visual field and perceptual problems
such as inattention. It is well reported in
the literature that visual impairment can
have a substantial impact on mobility,
dependency, increased risks of falls,
depression and other quality of life
measures. For example, see Jones &
Shinton, Age and Ageing 2006; 35: 560565
We welcome this comprehensive
document on the management of the
Thank you
27
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.’)
Thank you the guideline has been
amended accordingly.
Thank you. We have given falls risk and
sitting balance as two examples. We
acknowledge visual status is important
we are unable to include this in the
recommendations. The assessment of
visual status was not included as part of
the scope and hence was not part of the
evidence review.
Thank you for your comment. We have
amended and edited the guideline.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
acute stroke patient. The layout of the
full guidance was at times a little
confusing with a number of headings
and subheadings. There were quite a
few typing and spelling errors which
need to be corrected and some
statements which do not make sense
(see below). There was a certain
amount of repetition in the document eg
statements in the “From evidence to
recommendations” which really was a
repeat of what was in the Clinical
evidence statements.
It is important that when the guidance is
published it is made clear how it relates
to pre – exisiting guidance ie National
Clinical Guidelines for Stroke. This will
help avoid confusion or indeed
differences in opinion.
SH
British Association of Stroke
Physicians
Full
5.1.6
Full
5.2.6
2
SH
British Association of Stroke
Physicians
3
Whilst we agree in principle with the
recommendations it would be useful to
explain that the FAST test should be
used outside the hospital setting to
screen patients with an acute
neurological event.
While we agree that a scoring system
will help identify high risk patients, we
suspect that 24 hour assessment by a
specialist service will not be practicable
in a large number of hospitals. The
default position will be the use of
medical assessment units to provide this
assessment. These units may not be set
up to deal adequately with this group of
patients. Secondly, while patients who
have a low ABCD2 score are more likely
28
Thank you we have amended the main
clinical introduction for further
clarification.
We disagree, there is no evidence and no
reason why FAST could not be used in
some circumstance (eg pt having stroke
on medical ward) in a hospital setting to
screen for stroke
Thank you. Service provision is beyond
guideline scope
Typ
e
SH
Stakeholder
N
o
British Association of Stroke
Physicians
Docum
ent
Section
number
Full
6.2.5.1
Full
6.4.6
Full
6.4.6
Full
7.1.2.1
Full
7.1.2.2
Full
7.1.7
4.
SH
British Association of Stroke
Physicians
5.
SH
British Association of Stroke
Physicians
6.
SH
SH
SH
British Association of Stroke
Physicians
7.
British Association of Stroke
Physicians
8.
British Association of Stroke
Physicians
9.
Developer’s Response.
Comments
to have a non – vascular cause of their
symptoms. Finally, evidence for the
management of patients who have
transient ischaemic events on
anticoagulants or recurrent events on
antiplatelet therapy is not addressed.
Further discussion is required over the
accuracy of the clinical assessment of
TIA patients. Studies have suggested
that less than 50% of referrals to a
neurovascular clinic may be due to a
vascular event.
The EXPRESS and SOS studies
suggest that aspirin and clopidogrel
were the two antiplatelet drugs of choice
in prevention of secondary events. The
use of aspirin and dipyridamole is based
on studies which often randomised
patients out of the time limits suggested
by NICE.
No mention is made of the management
of patients with bilateral severe internal
carotid artery disease. Control of blood
pressure in these patients may do more
harm than good. Blood pressure
reduction should only be considered
following endarterectomy.
We thought the phrase “including brain
imaging should be added after
diagnostic tests.
The term “A comprehensive ward” is
confusing. This could also mean an
acute and rehabilitation stroke unit.
“all people with suspected stroke should
be admitted directly to a specialised
stroke unit”. We agree with the
sentiments in this statement. However,
as mentioned in section 7.1.6 “there is a
need for a randomised trial comparing
29
Stroke mimics are discussed in this
section
Reference to specific antiplatelet drugs
has been removed
No evidence was identified for this group
The text has been amended
This has been clarified
This is a consensus recommendation.
Please see section 14.
Typ
e
SH
Stakeholder
N
o
British Association of Stroke
Physicians
Docum
ent
Section
number
Full
8.1.4.1
Full.
8.1.5.2
British Association of Stroke
Physicians
British Association of Stroke
Physicians
12 Full.
.
Full.
13
.
8.1.5.3
British Association of Stroke
Physicians
Full.
8.1.7.1
Full
8.1.7.4
16 Full
8.2.6.1
10
.
SH
SH
SH
SH
British Association of Stroke
Physicians
11
.
8.1.6.5
14
.
SH
British Association of Stroke
Physicians
15
.
SH
British Association of Stroke
Comments
Developer’s Response.
direct admission to an acute stroke unit
versus admission to a medical ward, at
least while the latter remains standard
clinical practice”. Therefore this
recommendation may be challenged in
the absence of high quality randomised
trial evidence.
The word “died” should be changed to
“dead”. It would be useful to give
numbers needed to treat to prevent one
death or dependent person per 1000
patients treated to give an indication of
the size of the benefit.
The last sentence needs rewritten as it
does not make sense as it currently
stands.
Thank you for your comment. The
wording has corrected accordingly
Thank you. We have now corrected this
sentence. 8.2.5.2. corrected to “more
effective than aspirin and less expensive
than clopidogrel”
Was confusing to read and should be
Thank you. We have now simplified this
simplified.
section.
Should be clarified that the use of proton The section clarifies aspirin intolerance
pump inhibitors and aspirin is for those
using the NICE vascular TA definition.
patients who have GI intolerance to
aspirin.
It is recommended that patients should
.WThe evidence around long term
be continued on aspirin 150 – 300mg
management of antiplatelets has been
per day for two weeks. In clinical
considered by the ICSWP. The long term
practice a number of stroke patients will management is outside the scope of this
be discharged before that time.
guideline.
Therefore we would recommend that on
discharge patients are put on 75 mg /
day. This will avoid patients being left on
high dose aspirin until reviewed in clinic.
While anticoagulants are not used
Thank you the text has been altered.
routinely for the treatment of the acute
ischaemic stroke patient. It would be
useful to specify circumstances when
anticoagulants should be considered eg
venous stroke.
It should be made clear how long
Please see section 8.3.5.1. for further
30
Typ
e
SH
SH
Stakeholder
N
o
Physicians
.
British Association of Stroke
Physicians
British Association of Stroke
Physicians
Docum
ent
17 Full
.
Full.
Section
number
8.5.2.2
8.6.6
18
.
SH
British Association of Stroke
Physicians
Full.
8.8.7.3
British Association of Stroke
Physicians
British Association of Stroke
Physicians
20 Full.
.
Full.
21
.
9.2.2.2
British Association of Stroke
Physicians
Full.
19
.
SH
SH
SH
22
.
9.3.2.2
and
9.3.2.3
10
Comments
Developer’s Response.
anticoagulation is recommended in
venous stroke disease.
Mention is made of level 1 and level 2
nursing staff. This should be explained.
The question of delaying statin treatment
in patients with ischaemic stroke is poor.
If this recommendation is based on the
SPARCL study, patients were
randomised in the non – acute phase
(TIA / iscahemic stroke in preceeding 16 months). Secondly, there is no
mention of statin use in patients with
haemorrhagic stroke. We presume, that
as there is a risk of haemorrhagic
expansion with statin treatment that
patients with a haemorrhagic stroke
should not be started on a statin.
However, this needs to be formally
documented.
Not all DVTs carry the same risk. It may
not be appropriate to anticoagulate a
patient with a below knee DVT whereas
a patient with a large femoral DVT will
always be anticoagulated.
clarification.
Hypoglycaemia should read
hyperglycaemia.
It is not clear whether these statements
relate to patients with hypertension or
relate to patients with normal and high
blood pressure.
There is very little in the way of
recommendations relating specifically to
hydration or the type of fluid used for
rehydrating patients.
The text has been amended
31
Thank you a reference has been added
for further clarification.
Thank you. This is stated in the text
Thank you the text has been altered.
Unfortunately, this information is not
reported in either of the reviews
Thank you the guideline has been
amended accordingly. See 10.3.6.4. The
type of fluid used for hydrating patients
was not included in the search. The
developers feel that it is not appropriate
to recommend types of fluid and feel that
this will depend on a patients
condition/co-morbidities.
Typ
e
Stakeholder
SH
British Association of Stroke
Physicians
SH
SH
British Association of Stroke
Physicians
British Association of Stroke
Physicians
Docum
ent
Section
number
Comments
Developer’s Response.
Full.
12.1.1.1
24 Full.
.
Full.
12.1.2.2
Line 10 states that “little is known about
the safety of water by mouth”. This is
contradictory to what is stated in the rest
of the paragraph.
Line should read at the level of vocal
folds (not focal folds).
The expert panel was unable to come up
with any recommendations although it is
commented in the first chapter on the
virtues of common sense. Dysphagic
patients do aspirate and measures are
needed to prevent this happening. The
use of modified diets are well tolerated
in properly selected patients, although
this is not based on a randomised
controlled trial.
Overall comment: The evidence base for
this section is very weak. Early
mobilization has not been defined.
Inappropriate references have been
adduced (not relating to early
mobilization). The recommendations are
likely to lead to confusion. I suggest that
the BASP SDQ committee should
recommend revision of the chapter
taking the comments below into account.
Results form the recently published
AVERT trial should be included.
Bernhardt J, Dewey H, Thrift A, Collier J,
Donnan G. A very early rehabilitation
trial for stroke (AVERT): phase II safety
and feasibility. Stroke. 2008
Feb;39(2):390-6. Epub 2008 Jan 3.
We disagree, patients are producing
large volumes of saliva – placing a
patient ‘nil by mouth’ does not stop saliva
production.
Thank you. The guideline has been
corrected.
Thank you we have inserted a
recommendation
The term early mobilization needs to be
defined. Is mobilization within a week of
the stroke really early? Current
understanding of early mobilization
Early mobilisation is defined as ‘as soon
as possible’.To help clarify this we have
added into the recomendation “when
clinical condition permits”
N
o
23
.
12.1.6
25
.
SH
British Association of Stroke
Physicians
Full.
11.
26
.
SH
British Association of Stroke
Physicians
Full.
27
.
11.1.2
32
No studies were identified for very early
mobilisation. Very early mobilisation is
defined in 11.1.2.2. Please see section
11.1.5.1. for the rationale of the group.
This is phase II trial and in addition is
outside of the literature search cut-off
date period
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
relates to the first 24 hours, while
mobilization within one week is standard
therapy.
SH
British Association of Stroke
Physicians
Full.
11.1.4.4
Given the evidence from 11.1.4.4. that
We have amended the wording of this
stitting out of bed is associated with
recommendation for greater clarification.
better oxygenation than mobilization
within bed a further statement should be
added to the recommendations: “
Wherever possible stroke patients
should be sat out of bed.”
Full.
11.1.6
The guidance states that people with
Agreed. The wording has been
acute stroke should be mobilized as
amended accordingly.
soon as possible following an
assessment (e.g. sitting balance and
falls risk) by an appropriately trained
healthcare professional with access to
appropriate equipment. ‘Appropriately
trained’ and ‘appropriate equipment’ are
not defined. This statement is not based
on evidence. This is likely to lead to
confusion amongst stroke service staff
who are left wondering whether they are
appropriately trained and appropriately
equipped to mobilize the patient. This
sentence should be rephrased as:
“People with acute stroke should be
mobilized as soon as possible as part of
an active management programme on a
specialist stroke unit. “
Full.
13.1.1.1.
Lines 6 – 8 attribute increased risk of
intracerebral bleeding to the use of
aspirin. There is no evidence to support
this statement. Although aspirin use
does increase the risk of intracerebral
bleeding, the risk of bleeding does not
appear to be greater in the elderly. He
28
.
SH
British Association of Stroke
Physicians
29
.
SH
British Association of Stroke
Physicians
30
.
33
Thank you the wording has been
changed
Typ
e
SH
Stakeholder
N
o
British Association of Stroke
Physicians
Docum
ent
Full
31
.
SH
SH
British Association of Stroke
Physicians
British Association of Stroke
Physicians
Full.
32
.
Full.
33
.
SH
British Dietetic Association
1
Full
SH
British Dietetic Association
2
NICE
Section
number
Developer’s Response.
Comments
et al, published a meta – analysis on 16
trials from 55,462 patients [JAMA
1998;280: 1930 – 1935]. The study
suggests that aspirin treatment was
associated with an absolute risk
increase in haemorrhagic stroke of 12
events per 10,000 persons (95% CI: 5 –
20; p< 0.001). This risk did not differ by
participant or study design
characteristic.
13.2.6
Patients with malignant MCA occlusion
aged up to 60 years should be referred
for hemicraniectomy. Yet in preceeding
paragraph 13.2.5.1 there was a
significant increase in morbidity in
patients over the age of 50 years. It
would be useful for NICE to explain how
they have extrapolated from the trial
data to include patients aged 50 – 60
years in their recommendation.
Alogorithm (i) Must ensure that FAST is used in
for
appropriate setting ie paramedics.
suspected (ii) No mention of atrial fibrillation and
stroke.
secondary prevention.
We refer to the pooled analysis data. We
have amended the from evidence to
recommendation section of the guideline
to aid clarification
Thank you we do not feel that FAST
should be limited to the paramedic setting
and it is outside the scope of the
guideline to deal with issues pertaining to
secondary prevention.
Alogorithm Use of <70% - 99% is confusing. Should Thank you please see the amended
for
read either < 70% or >99%.
algorithm
suspected
TIA.
General
Thank you for giving The British Dietetic Thank you for your comment
Association the opportunity to comment
on this guidance. We fully support the
guidelines and have the following
comments, where we consider that
additional points or recommendations
should be considered.
1.6.1.2
There are screening tools that are in use Thank you for your comment. This
that are not based on BMI, %
recommendation was taken from the
34
Typ
e
SH
Stakeholder
British Dietetic Association
N
o
3
Docum
ent
NICE
Section
number
General
p.6
Comments
Developer’s Response.
unintentional weight loss etc. It is not
always appropriate to weight/measure
someones height on patients with large
strokes/frail etc. While in agreement
that these measurements are valuable
markers in nutritional assessment,
surely the use of any appropriate
validated screening tool is a far more
important point to make.
NICE nutrition support (NS) clinical
guideline. The MUST tool does offer
alternative measurements,
considerations, and subjective criteria in
circumstances where it is difficult to
measure height and weight.
Under communication, the word
Thank you this has been amended
dysphasia or aphasia could be included, accordingly
as this is common after stroke and a
specific problem for clear
communication of information.
SH
British Dietetic Association
4
NICE
1.6.1.2
It is appreciated that the
recommendation is taken directly from
NICE guideline for Nutritional support in
adults, but feel that some comment
should be included about assessing
specific nutritional risk factors to patients
after stroke, e.g. dysphagia, oral state
(mouth shape changes, dentures etc)
and the ability to self-feed.
SH
British Dietetic Association
5
NICE
1.2
Should section be titled ‘nutrition and
Thank you this has been amended
hydration’, instead of ‘hydration and
accordingly.
nutrition’ as the nutrition is covered first?
SH
British Dietetic Association
6
NICE
1.6.3
Why is jejunostomy included in the title – Thank you title amended as suggested
it is not the routine practice and the title
may suggest that it should be
considered. We would suggest that
jejunostomy be removed from the title.
The title would be better phrased as
‘timing of enteral nutrition’ or ‘timing of
non-oral nutrition’ or ‘when to start
enteral nutrition?’
35
Thank you. This recommendation was
taken from NICE Nutritional Support
guideline we did not review any additional
evidence and are unable to change the
wording for this guideline
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
SH
British Dietetic Association
7
NICE
1.6.3.1
This statement doesn’t appear to take
into account the clearly dying patient?
Thank you the introduction has been
altered accordingly
SH
British Dietetic Association
8
NICE
1.6.3.1
Would like to see added in ‘and be
referred to an appropriately trained
healthcare professional for detailed
nutritional assessment, individualised
advice and monitoring’.
Thank you the guideline has been
amended accordingly
SH
British Dietetic Association
9
Full
General
Should say The British Dietetic
Association but is written incorrectly as
British Association of Dietetics.
Thank you the guideline has been
amended accordingly
SH
British Dietetic Association
10 Full
GDG
member
page 8
10
SH
British Dietetic Association
11 Full
General
Should there be a general/over-arching See comment 75 (‘Thank you the
paragraph highlighting that these
introduction has been altered
guidelines are for patients having active accordingly’) See comment 75
treatment and that palliative care
pathways should be followed for the
dying patient? This is however included
under nutrition section in full document.
SH
British Dietetic Association
12 Full
10.1.5.1
Double full stop in line 2.
SH
British Dietetic Association
13 Full
10.1.6.2
It is appreciated that the
recommendation is taken directly from
NICE guideline for Nutritional support in
adults, but feel that some comment
should be included about assessing
specific nutritional risk factors to patients
after stroke, e.g. dysphagia, oral state
(mouth shape changes, dentures etc)
and the ability to self-feed.
Should section be titled ‘nutrition and
Thank you the guideline has been
hydration’, instead of ‘hydration and
amended accordingly
nutrition’ as the nutrition is covered first?
36
Thank you the guideline has been
corrected.
Thank you the guideline has been
amended
Typ
e
Stakeholder
N
o
SH
British Dietetic Association
SH
Docum
ent
Developer’s Response.
Section
number
Comments
14 Full
10.1.6.4
We understand that 10.1 relates to
The evidence from the FOOD trial relates
patients ‘not identified as being
to patients who were not considered to be
malnourished’. However, we are
malnourished on admission
concerned that if the recommendations
are read alone, without the background,
this recommendation could be
interpreted that even if the patient was
subsequently identified as being
malnourished, they should not receive
nutritional supplementation as they were
‘adequately nourished on admission. We
suggest that the statement would be
more logical if the last two words ‘on
admission’ were omitted.
British Dietetic Association
15 Full
10.3
Why is jejunostomy included in the title – Thank you we have altered the title
it is not the routine practice and the title accordingly
may suggest that it should be
considered. I would suggest that
jejunostomy be removed from the title. I
also wondered if the title would be better
phrased as ‘timing of enteral nutrition’ or
‘timing of non-oral nutrition’ or ‘when to
start enteral nutrition?’
SH
British Dietetic Association
16 Full
10.3.1.1
See comments above about
jejunostomy/PEJ – feeding into the
stomach has less risks associated than
the jejunum due to the acid barrier. Jej
should not be a first choice method
without clinical rational and is therefore
not relevant to this section.
SH
British Dietetic Association
17 Full
10.3.5.2
Could a comment be included about the We do not feel it is appropriate to pick out
non significant trend towards reduced
and discuss non significant trends in one
mortality in early feeding group be made particular group / RCTs
here? We understand that this was
discussed at length at the GDG, and felt
to be especially important to the patient
37
We agree but there are some
gastroenterological indications for PEJ
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
group reps. Also could it be mentioned
that the study compared early with
delayed feeding, but that in reality
people could be randomised up to 3
days post event and then took 1-2 days
to start feeding, so this may have
masked the possible benefit/harm of
very early feeding.
We have included this in section 10.2.2.1
SH
British Dietetic Association
18 Full
10.3.5.3
With the morbidity and mortality
Nasal bridles are now included in the
associated with PEG placement, we are evidence review and recommendation
concerned that within the first few weeks
‘PEG should be the intervention of
choice if it is impractical to use an NG
tube’.
We agree with the guidelines that
feeding via an NG tube should be used
initially, however, there are alternative
techniques to PEG that we believe
should be used prior to resorting to PEG
in the early days post stroke, such as
nasal bridles, which can secure the
position of the NG tube, if the tube is
frequently being removed inadvertently
or otherwise. We understand that some
units also use mitts, although there are
issues around consent / restraining of
patients.
SH
British Dietetic Association
19 Full
10.3.6.1
We are concerned that ‘early tube
Thank you. A comment has been added
feeding’ is defined as within 24 hours of to the clinical introduction
admission. We suggest that there
should be a qualifying statement relating
to issues of consent; the Mental
Capacity Act; Advance Directives and
that feeding should be commenced as a
trial to support the patient within the
early days or weeks post stroke and it’s
continuation should be assessed as
38
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
being the patient’s best interest.
SH
British Dietetic Association
20 Full
Algorithm
2 - Acute
stroke and
TIA
algorithm
2 - stroke
pathway
The wording of the box ‘screen patients Thank you the guideline has been
on admission for nutritional supplements amended accordingly.
using a validated tool e.g. MUST’ should
read ‘screen patients on admission for
malnutrition using a validated tool e.g.
MUST’ as that is what MUST actually is
validated as screening for. One of the
outcomes of the screening might be an
evaluation of the use of supplements,
but this is covered in the box about
routine use of supplements.
SH
British Dietetic Association
21 Full
Algorithm
2 - Acute
stroke and
TIA
algorithm
2 - stroke
pathway
SH
British Paramedic Association
– College of Paramedics
We are not sure about the swallow
screening coming after screening for
malnutrition. We appreciate that this
might have been done to make the flow
chart easier to write, but in reality you
would screen for swallow before
screening for malnutrition as the
outcome of the swallow test would effect
the outcome of the nutrition screen.
Especially as the guideline is for early
swallow screening, preferably within 24
hours. The nutrition screening box
should really come on both the yes and
no arms after the swallow screening.
The data figures quoted are from 1999,
nearly ten years ago. Considering the
rapidly ageing and increasing
population, can more contemporary data
be quoted?
The document quotes the NAO 2005
report in relation to incidence of TIA –
20,000 annually. However the total
numbers of actual and suspected stroke
& TIA are unclear. It may be more
appropriate to quote the incidence as
NICE
Incidence
NICE
Incidence
1
SH
British Paramedic Association
– College of Paramedics
2
39
Thank you we have amended the
algorithm for further clarification.
This is the most up to date data available
Accurate data is very difficult to obtain.
The GDG feel that the NAO data is
robust
Typ
e
SH
Stakeholder
N
o
British Paramedic Association
– College of Paramedics
Docum
ent
Section
number
NICE
Patientcentred
care
NICE
Patientcentred
care
NICE
Patientcentred
care
3
SH
British Paramedic Association
– College of Paramedics
4
SH
British Paramedic Association
– College of Paramedics
5
Developer’s Response.
Comments
detailed within the National Stroke
Strategy document – p20 section 6. ‘It is
estimated that 20,000 strokes a year
could be avoided….’ and p22 Key Fact
1: ‘Around 150,000 people per year
have a suspected TIA or minor stroke’
as this may more accurately reflect
demand for stroke and TIA services
Where the document states: “People
with acute stroke or TIA……” – The
document should probably reflect the
need to engage family, friends and
carers too, as often third parties access
stroke services on behalf of friends or
loved ones.
The second paragraph talks about the
patients suffering acute stroke and TIA
to have the ‘opportunity’ to make
informed decisions about their care and
treatment. We feel that it is sometimes
very difficult for patients, or indeed
friends/carers/family to be able to make
informed decisions at a time of crisis;
such as at the time of onset of acute
stroke/TIA. Consequently we feel that
the systems and procedures must
ensure access to highest possible levels
of patient care.
The third paragraph talks about
information that patients are given being
‘culturally appropriate’ and accessible ‘to
people who do not speak or read
english’. This is not clear although
appears to allude to information being
presented in a number of languages.
In the pre hospital and sometimes life
threatening situations communication
can be a major challenge to both the
professionals and to the patients. It
40
We have amended the introduction to
include this
We agree but this is standard NICE
template text
The provision of patient information is
beyond the scope of this guideline. We
agree with the government’s position and
feel that this requirement of all
documentation and not just stroke related
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
should be acknowledged that it is not
always possible to achieve the ideal
outcome when trying to manage cultural
and first language difficulties.
We feel that it may be preferable not to
coalesce the requirement of disabled
patients, carers and relatives with the
disparate issues relating to cultural
diversity and language barriers into one
paragraph or indeed sentence. As in the
case of disability the Disability Equality
Duty mandates the NHS to make all
information accessible, regardless of
disability.
Given the government’s position on
production of documents in each and
every language, it is not clear how to
progress this particular issue, given that
the costs of producing all information in
many different languages may detract
funds from patient care.
SH
British Paramedic Association
– College of Paramedics
NICE
Patientcentred
care
6
SH
SH
SH
British Paramedic Association
NICE
7
– College of Paramedics
British Paramedic Association
NICE
– College of Paramedics
8
1
British Paramedic Association 9 NICE
1.1.2.1
1.1.1.3
The penultimate paragraph talks about
the families’ and carers’ involvement in
decision-making. We feel that it would
be useful to acknowledge that the
consent of the patient may be difficult to
ascertain at the time of an acute episode
and also where the stroke or TIA is
presenting the patient with
communication problems.
There is a missing hyperlink in the
opening paragraph.
We would like to see the piloting and
evaluation of the ROSIER amongst prehospital clinicians, including Paramedics
We would like to see introduction of
41
Thank you the section has been
amended
Thank you yes this will be added at the
same time the full guideline is published.
Thank you this is beyond the remit of a
clinical guideline.
This is beyond the remit of the clinical
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
– College of Paramedics
SH
British Paramedic Association
– College of Paramedics
NICE
1.1.2.2
NICE
1.1.2.3
NICE
1.1.2.3
10
SH
British Paramedic Association
– College of Paramedics
11
SH
British Paramedic Association
– College of Paramedics
12
Comments
Developer’s Response.
ABCD2 risk scoring to Paramedics and
other pre-hospital clinicians, this
supports Quality Marker 5 of the
National Stroke Strategy and is backed
up in p23-24 points 5 and 6 of the
strategy document.
This is essential to ensure rapid referral
to TIA services as with increased
publicity, public and professional
awareness, the patients likely first point
of contact with the NHS is going to be
the 999 system rather than primary care
and GPs.
This group of patients (considered at
high risk of Stroke) should be admitted
earlier to hospital for investigation and
assessment where appropriate via the
ambulance service.
We believe there would be further clarity
achieved if the following was re-worded
accordingly:
‘…an ABCD2 score of less than 4…’replace with‘…an ABCD2 score of 3 and below..’
We believe this is absolutely the correct
approach for this group of patients. We
feel that there should be a nationally
(therefore reproducible and
geographically equitable) agreed
pathway for all patients that extends to
all appropriate disciplines and
professions, particularly including the
Paramedic profession.
Pre-hospital clinician referrals to TIA
clinics and immediate initiation of aspirin
by the same group will achieve the best
possible patient care and avoid delays
that could be experience by referral
through the primary care system.
guideline. We will discuss this with the
NICE implementation team
42
We assume that these patients will be
referred to hospital
Thank you. The text has been amended
Thank you
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
British Paramedic Association
NICE
– College of Paramedics
13
1.2.2.3
SH
British Paramedic Association
– College of Paramedics
NICE
1.2.2
NICE
1.2.2.3
British Paramedic Association
NICE
16
– College of Paramedics
British Paramedic Association
NICE
– College of Paramedics
1.2.3.1
14
SH
British Paramedic Association
– College of Paramedics
15
SH
SH
1.2.3.1
17
SH
British Paramedic Association
– College of Paramedics
NICE
1.3.1.1
British Paramedic Association 19 NICE
1.3.1.1
18
SH
Developer’s Response.
Comments
Systems and care pathways should be
replicable across Paramedics, GPs and
other pre-hospital clinicians. Best
patient care must be afforded from interdisciplinary working.
As comment number 11 above
See response to comment number 11
(‘Thank you. The text has been
amended’)
We like to see the feasibility and
Thank you for your comment, however
appropriateness of Paramedic (possibly this is outside the remit of the guidance
extended practitioner level) referrals for and evidence review.
imaging, where appropriate, prior to
specialist consultation explored. This will
fit nicely with the development for
advanced Paramedics with an extended
scope of practice.
We would suggest that imaging should
Service provision is beyond scope of
be available and tie in with the time for
guideline
referral specialist clinic, thus ensuring
the best available use of time and best
patient care and patient experience.
As comment number 15 above
Please see the response above
We suggest that considering the
stakeholder involved it may be useful to
explicate the pathophysiology and
presentation of such patients such as to
ensure the wide audience for these
guidelines fully appreciate the issues.
It is widely acknowledged that there are
significant advantages associated with
‘streaming’ patients into specialist care
pathways. We are keen to see that all
care pathways are developed conjointly
with acute stroke units and TIA services.
We believe that it is important hyperacute stroke units are able to all three
elements at one location.
We would like to see the development of
43
Thank you for your comment. This is a
clinical guideline and not a text book on
acute stroke and TIA
Thank you for your comment. It is not
possible to include any
comment/discussion on service provision.
Service provision is outside the remit of
this clinical guideline
Thank you for your comment. It is not
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
– College of Paramedics
SH
British Paramedic Association
– College of Paramedics
NICE
20
1.3.2.1
Comments
Developer’s Response.
direct referrals into acute stroke wards
by pre-hospital clinicians, including
Paramedics and GPs.
possible to include any
comment/discussion on service provision.
Service provision is outside the remit of
this clinical guideline
We believe that best patient care and
experience, together with maximum
financial and organisation efficiencies
are achieved by ensuring quickest
access to specialist care and
consequently direct access should be
explored.
We have concerns that the
We agree and have stated “immediately
recommendation to perform brain
(ideally the next slot and definitely within
imaging within 1 hour may create the
1 hour, whichever is sooner)”
impression that waiting for up to an hour
is acceptable practice. The
consequence of setting this standard is
that Imaging service level agreements
and standards could be set at 1 hour
with the decision being based on ‘on the
NICE guidelines’.
We believe that imaging services for
acute, particularly in possibly
thrombolysis eligible patients, needs to
be designed on a very next slot basis
and that it may include moving an
existing patient off of the scanner where
there may be significant delays.
The importance of early intervention is
firmly evidenced by the following paper
that details neurone death at a rate of
1.9million cells per minute in acute
stroke
(http://stroke.ahajournals.org/cgi/content
/full/37/1/263).
Moreover, in the context of the license
44
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
for alteplase (the tPA drug license for
use in acute stroke) which states a
maximum duration for onset to treatment
time of 3 hours –
Building a delay of a maximum of 1 hour
for imaging, combined with an estimated
45 minutes from 999 to getting a stroke
patient to a hyper-acute treatment centre
and that it takes up to an hour to access
the patient for thrombolysis by the acute
stroke team, the total time left for the
patient to call 999 after the onset of new
symptoms, may be as little as 15
minutes.
This possible delay in access to
scanning might present the only
obstacle, preventing access to
thrombolysis. Each of the factors in the
patient journey provide an immediate
next available response so we feel that
the same must be said for imaging to
ensure maximum numbers of patients
treated with the best possible clinical
outcome.
SH
SH
British Paramedic Association
NICE
21
– College of Paramedics
British Paramedic Association 22 NICE
1.4.1.3
1.4.5.2
Therefore the for the first bullet-point
‘indications for thrombolysis….’, we feel
that an immediate imaging response
must be specified. However, for the
other presentations listed, we feel that a
maximum of an hour’s response would
be acceptable, although desirably
immediate where acute intervention may
provide the best opportunity for a good
clinical outcome.
Does the use of clopidogrel extend to
Thank you. There is no evidence of this.
the treatment of TIA?
We would like to see the addition of a
Thank you for your comment. It is not
45
Typ
e
Stakeholder
N
o
Docum
ent
– College of Paramedics
SH
British Paramedic Association
– College of Paramedics
NICE
22
SH
British Paramedic Association
NICE
– College of Paramedics
23
SH
British Paramedic Association
– College of Paramedics
NICE
24
SH
British Paramedic Association
– College of Paramedics
NICE
25
SH
British Paramedic Association
NICE
– College of Paramedics
26
SH
British Paramedic Association
– College of Paramedics
NICE
27
Section
number
Developer’s Response.
Comments
bullet-point requiring centres that offer
thrombolysis for acute ischaemic stroke
to offer a full compliment of acute stroke
unit and TIA clinics, including access to
the full range of imaging modalities (CT,
CTP, DWMRI, etc).
1.4.5.2
We feel that it would be beneficial to
specify that FASTrack pre-hospital
pathways from the ambulance service
and its clinicians need to be in place to
offer a thrombolysis service.
1.4.6.1
Is it possible to further define the
timescales relating to ‘following’ in the
context of ‘following acute stroke’
1.5.1.1
There is not clear evidence to suggest
that there are clear benefits to receiving
or not receiving oxygen therapy.
Consequently we should like to see this
fully investigated in the context of an
RCT.
1.5.3
This section states that blood pressure
manipulation is not recommended based
on the evidence base. This could
however be confusing as the guidelines
for the use of alteplase in thrombolysis
in acute ischaemic stroke indicate the
use of labetolol where the systolic blood
pressure is over 185mmHg-1 or the
diastolic is more than 110mmHg-1 and
the use of sodium nitroprusside in the
case of lower elevate blood pressures.
1.5.3
We feel that it may be useful to refer to
the SCAST study, due to report in 2009,
in the NICE version.
4
We are keen to see the development of
Research research that involves a full range of the
recommen disciplines involved in stroke care
d-dations delivery. We are particularly keen to
examine the benefit of increased input of
46
possible to include any
comment/discussion on service provision.
Service provision is outside the remit of
this clinical guideline
Thank you for your comment. It is not
possible to include any
comment/discussion on service provision.
Service provision is outside the remit of
this clinical guideline
Thank you. There is no evidence
available.
Thank you for your comment this has
been noted by the developers
We have added reducing BP to enable
thrombolysis to this section
The developers do not wish to add this to
the NICE version of the guideline.
Thank you
Typ
e
SH
Stakeholder
N
o
British Paramedic Association
– College of Paramedics
Docum
ent
Section
number
NICE
4.2
Full
3.2
28
SH
British Paramedic Association
– College of Paramedics
29
SH
British Paramedic Association
– College of Paramedics
Full
3.2
Full
3.2
30
SH
British Paramedic Association
– College of Paramedics
31
Developer’s Response.
Comments
pre-hospital clinicians in terms of clinical
outcome.
We concur that this is a valuable
research recommendation although we
feel that this research question should
be extended to include TIA and
investigate the administration of these
drugs by pre-hospital clinicians where
appropriate.
It is felt that the sensitivity and specificity
of the FAST test is not appropriate as
the ‘broad clinical sieve’. Indeed the
guidelines identify that the ROSIER will
identify more patients than FAST. FAST
test very well identifies patients with
motor deficits and dysphasia/dysarthria
however is less able to identify finer
symptoms including sensory loss and
visual field defects. Further
development of an increased sensitivity
pre-hospital stroke identification tool is
required.
Algorithm 1 – all stroke symptoms
presenting to the healthcare system,
until resolved or rapidly resolving will be
treated as stroke (as opposed to TIA) –
it is unlikely therefore that a TIA would
be suspected prior to excluding
hypoglycaemia.
We feel that it would be beneficial to
combine the algorithms for the prehospital elements of stroke and TIA such
as in the attached pathway utilised by
South East Coast and South Western
Ambulance Trusts.
Appendix 1 - SECAmb & SWAST
Integrated Pre-Hospital Stroke & TIA
Pathway
47
We agree that this would be a useful
research study but it is not appropriate to
widen 4.2 ; it would be a separate
research question
Thank you FAST is specified as an
example of a validated tool.
Thank you for your suggestion however
this area of research was not highlighted
as imperative by the guideline
development group
Thank you. We have removed this from
the algorithm
Thank you for this useful information. The
developers have clarified the definition of
Stroke and TIA and amended the
algorithm accordingly. The developers do
not think the pathways should be
combined.
Typ
e
Stakeholder
SH
British Paramedic Association
– College of Paramedics
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
Full
3.1
We believe that a key priority for
implementation is the development of
interdisciplinary working particularly
involving pre-hospital and ambulance
service clinicians in the emergency
response and early treatment and
referral of stroke and TIA patients.
We note that the use of the MEND
examination (www.asls.net) was not
assessed.
We note that the ROSIER scale was
show to have a higher sensitivity than
the FAST test.
This paragraph states that the ROSIER
assessment if validated for use in A&E.
It also states that ROSIER ‘would not be
practicable to do outside hospital.’ We
are concerned that this assertion has
been made without evidence.
Paramedics and other pre-hospital
clinicians routinely measure blood
glucose levels and are licensed to
correct hypoglycaemic episodes with the
use of intravenous glucose or
intramuscular glucagon, furthermore
detailed history is taken and would
include the noting of fits or seizures such
as relevant in Todd’s paresis. We note
that the visual field assessments have
not previously been used pre-hospitally
but these skills could easily be
developed amongst pre-hospital
clinicians.
Thank you, we agree; however is is not
an issue to be specifically addressed
within the guideline. This is an issue that
is important for implementation and we
will discuss this further with the NICE
implementation team.
32
SH
British Paramedic Association
Full
– College of Paramedics
33
5.1.2
SH
British Paramedic Association
Full
– College of Paramedics
34
5.1.4.14
SH
British Paramedic Association
– College of Paramedics
5.1.5.3
Full
35
We believe that evidence for the use of
ROSIER out of hospital should be
gained from appropriate studies,
particularly in light of the higher
sensitivity associated with this tool.
48
Unfortunately we were only able to review
some of the ‘screening’ tools and MEND
has not be validated
Noted. The FAST study is used in clinical
practice
Thank you. We have removed this
sentence.
After prehospital assessment we have
inserted “including blood glucose”.
The FAST tool has been validated tool
used by paramedics in order to screen
whether pts need rapid transfer to a
stroke unit. Please see section 5.1.5.1
Typ
e
Stakeholder
SH
British Paramedic Association
– College of Paramedics
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
Full
6.3.4.1
The term ‘cerebrovascular accident’ is
used in this paragraph. We had
understood that this term was
considered obsolete.
Full
9.1.1.1
The literature quoted in this paragraph,
the JRCALC 2006 pre-hospital
guidelines does not state that oxygen
supplementation by mask is usual
practise. In fact the guidelines state that
oxygen should only be given to maintain
a peripheral saturation of over 95%. We
acknowledge though that for many years
oxygen therapy often through nonrebreather masks has been delivered to
all stroke and TIA patients in the prehospital setting.
This wording is taken directly from the
paper and is quoted directly because the
authors do not define specifically what
patients were included under this
‘diagnosis’
Thank you the text has been altered
36
SH
British Paramedic Association
– College of Paramedics
9.1.5.1
37
SH
British Paramedic Association
– College of Paramedics
Full
9.1.6
38
SH
SH
British Paramedic Association
Full
39
– College of Paramedics
British Paramedic Association
Full
– College of Paramedics
40
10.2.2.8
13.1.6.1
We feel that this statement could only be
made following a survey of current
practise and that guidelines could be
made with the benefit of an RCT, looking
at the administration of Oxygen in acute
stroke, with the randomisation
commencing pre-hospitally. We also
feel that there may be some benefit in
investigating the use of cerebral
pulsoximetry
This recommendation appears not to
benefit from a strong evidence base and
we feel that without further studies in this
area it is difficult to make the
recommendation.
There appears that there may be some
detail missing here ‘[Power]’
We would like to see a consistent
national protocol for referrals to regional
neurosurgical and neuroradiological
interventionalist centres developed to
49
Thank you. The recommendation was
agreed through consensus. This area of
further research was not agreed by the
group.
This paper has not been published and
has been removed from the guideline
This is beyond the remit of the scope.
Typ
e
SH
SH
Stakeholder
British Paramedic Association
– College of Paramedics
N
o
Docum
ent
Section
number
Full
13.2.1.1
Full
14
NICE
General
41
British Paramedic Association
– College of Paramedics
42
SH
British Psychological Society, 1
The
Comments
ensure equity of patient care across the
country.
We would like to see a consistent
national protocol developed to ensure
early appropriate referrals ensuring
equity of patient care across the country.
We feel that the following subject should
be added to the list of research
recommendations:
 Referrals to TIA services by
paramedics and other prehospital clinicians, including the
administration of early aspirin
and other early platelet therapy
 The use of Oxygen therapy in
acute stroke, a national RCT for
the first 72 hours
 The use of early intravenous
fluids to counter chronic
dehydration in acute stroke
There is no consideration of
psychological issues.
There needs to be a section on
Psychological Care /Patient Centred
Care. Please see below:
Developer’s Response.
Please see response above (‘This is
beyond the remit of the scope’)
These areas of research were not agreed
as high priority by the guideline
development group
Thank you.
We agree that the psychological support
of people with chronic conditions is
important however the main scope of the
guideline was to address the ‘initial and
Patient-Centred Care and Psychological early management aimed at reducing the
Support
ischaemic brain damage and in the case
of TIAs preventing subsequent stokes’.
(i) Patient-Centred Care
We were unfortunately unable to cover all
As with all treatment, it is essential that areas for acute stroke and focused upon
the care of the acute stroke patient is
those that people / stakeolders initially
done in a patient-centred way1’2, which
suggested as critical areas to be
means the clinicians involved should
addressed.
ascertain:
1
2
DoH (2000). The NHS Plan. Department of Health: London.
General Medical Council (2007). Consent: patients and doctors making decisions together – a draft for consultation. GMC: London
50
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments


the wishes and preferences of
the patient in order that they can
be key decision makers.
the patient’s strengths in order
that they do not become an
untapped resource3.
A comment about the need for patient
centered care has been added to section
1.1 1 2
We will ensure that your suggestion is
passed on to the NICE Topic Selection
Panel as a suggestion for a topic for a
future guideline ‘the psychological
Every effort should be made to establish support of people with chronic
these, using a variety of communication conditions’.
modalities as needed eg verbal,
gestures, signboards, sign language,
etc. Where, after exhaustive efforts have
been tried and failed directly with the
patient, their carer(s) should be asked
about the patient’s likely preferences
and strengths.
Psychologically-informed approaches,
such as solution-focused, are well
placed to deliver these principles of
patient-centred care.
(ii) Psychological support
Delivering patient-centred care is a form
of psychological support in itself, and all
clinicians should be practising these
principles.
Clinicians with training such as
specialised Physiotherapists and
Occupational Therapists will be in a
position to offer more formal
psychological support to patients/carers
and to advise less specialized MDT
51
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
colleagues; these specialised clinicians
should in turn be supported by mental
health specialists such as Counsellors,
Clinical/Health Psychologists and
Psychiatrists, who will also offer input to
a minority of more psychologically-needy
patients/carers.
This can be readily be represented by a
four-tier model, as in the NICE (2004)
Cancer Guidance4 (below):
Table removed and placed at the end
of the comments table as not
readable here
LEVEL
GROUP
1
All
health
and
social
care
professi
onals
ASSES
SMENT
S
Recogn
ition of
Psychol
ogical
Needs
52
INTER
VENTI
ONS
Effectiv
e
informat
ion
giving,
compas
sionate
commu
nication
and
general
psychol
ogical
support
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
2
Health
and
social
care
professi
onals
with
addition
al
expertis
e
Screeni
ng for
psychol
ogical
distress
3
Trained
and
accredit
ed
professi
onals
Assess
ed for
psychol
ogical
distress
and
diagnos
is of
some
psycho
patholo
gy
Mental
health
specialis
tsclinical
psycholo
gists
and
psychiatr
ists
Diagno
sis of
Psycho
patholo
gy
4
SH
British Psychological Society,
The
2 Full
Developer’s Response.
Comments
Psychol
ogical
interven
tions
(such
as
anxiety
manage
ment
and
problem
solving
Counse
lling,
Cognitiv
e
behavio
ural
therapy
(CBT)
and
solution
focused
therapy,
Speciali
st
psychol
ogical
and
psychia
tric
interven
tions
6.2.6
It would be useful to include a timescale
Recomme of when patient is to be assessed by a
53
See section 5
Typ
e
SH
SH
Stakeholder
N
o
Docum
ent
British Psychological Society, 3 Full
The
British Psychological Society, 4 Full
The
Section
number
Comments
ndation
6.2.6.1 8.1.7.1
specialist.
It would be useful if the bullet points
included timing.
General
Some of these recommendations should
stand out more in the text to highlight
P.71 + 72 their importance.
10.2.6.1 + a clear deadline for assessment by
10.2.6.2
appropriately trained healthcare
professionals should be included to
indicate the maximum length of time that
should lapse from admission.
General
NICE should consider including applied
psychologists as part of teams to
develop future guidelines.
SH
British Psychological Society, 5
The
Full
SH
British Psychological Society, 6
The
Full
SH
British Psychological Society, 7
The
Full
General
Development of the guidelines would
benefit from having more than two
patient representatives.
SH
British Psychological Society, 8
The
Full
General
There were no references to
maintaining
the patients confidence in their
recovery or the role of perceived
control, and care givers. Stroke
survivors do better with a regular care
giver. Given that these two issues are
very relevant to the stroke survivors
54
Developer’s Response.
Thank you. The text has been amended
accordingly
Thank you for your comment which has
been noted by the developers
This evidence was not considered by the
group.
Thank you for your comment. Applied
psychologists are not excluded from
participating in the development of NICE
guidelines. We did not have an applied
psychologist as part of the acute stroke
and TIA guideline development group
because this area was outside the remit
of the scope. For more information of
guideline development recruitment please
see the NICE guidelines development
manual.
Thank you for your suggestion, when
convening the guideline development
group we have tried to ensure that the
GDG is a workable size to ensure
individuals are able to contribute
effectively whilst ensuring the need for a
broad range of experience and
knowledge
Thank you for your comment. We agree
that the maintaining the patients
confidence in their recovery is important
however the main scope of the guideline
was to address the ‘initial and early
management aimed at reducing the
ischaemic brain damage and in the case
of TIAs preventing subsequent stokes’.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
recovery in the long run, these should
be addressed in the acute stages.
SH
British Society of
Interventional Radiology
1
Full
General
We were unfortunately unable to cover all
areas and focused upon those that
people initially suggested as critical areas
for address. Please refer to the Royal
College of Physicians Intercollegiate
Stroke Working Party guideline which
looks at this area in more detail. We did
not review any evidence that stroke
survivors do better with a regular care
giver.
The draft consultation document “Stroke: Thank you
diagnosis and initial management of
acute stroke and transient ischaemic
attack (TIA)” from the National Institute
of Clinical Excellence is a welcome step
towards improving services for patients
suffering “brain attacks” and highlights
the department of Health’s commitment
to the prioritisation of stroke services
within the NHS.
As a National Guideline the document is
intended to provide clinicians, managers
and service users with summaries of
evidence and recommendations for
clinical practice. It has the potential,
therefore, to have a profound influence
on processes of care and clinical
outcomes.
In the “Introduction” section of the
document it is stated that “this guideline
is a stand-alone document, but is
designed to be read alongside the
Intercollegiate Working Party (ICWP)
National Clinical Guideline on stroke”
(National clinical guidelines for stroke.
Second edition. Prepared by the
Intercollegiate Stroke Working Party.
55
Thank you for your comments. The
ICSWP are currently in the process of
updating their 2004 version of the
guideline and we have liaised with this
group to ensure consistency. We have
amended the introduction to clarify the
context of this document with there
update.
Where possible we have tried to ensure
that the guideline is consistent with the
Typ
e
SH
Stakeholder
British Society of
Interventional Radiology
N
o
2
Docum
ent
Full
Section
number
General
Comments
Developer’s Response.
2004). Reference is also made to the
recently published “National Stroke
Strategy” which is essentially a
framework for the achievement of a high
quality service. Clearly the guidance
issued in this draft consultation
document must be put into appropriate
context with regards these two important
texts and not appear to contradict them.
message from the DOH stroke strategy
however there are some instances where
following the review of the evidence the
GDG have resulted in slightly different
conclusions. Where are differences this
has been highlighted in the appropriate
from evidence to recommendations
sections of the guideline.
General Point 1. The terminology used
Thank you. We did not review any RCT
in the consultation document should be evidence for carotid stenting within the
consistent throughout the document and ‘acute’ two week period of the guideline
consistent with the terminology
employed in the National Stroke
Strategy.
Throughout the National Stroke Strategy
document the term “carotid intervention”
has been employed to describe
intervention by surgical or endovascular
means (carotid endarterectomy or
carotid stenting). The Stroke
consultation document on the other
hand appears to use the terms “carotid
endarterectomy” – CEA or “carotid
intervention” randomly (sometimes in the
same paragraph (6.3.6) which is at the
very least confusing) and makes no
specific reference to carotid stenting as
an alternative treatment strategy. In the
interest of both consistency and clarity
the generic term “carotid intervention”
should be used in place of “carotid
endarterectomy”. On first use within the
text it could be explained as follows;
“carotid intervention to mean carotid
endarterectomy or carotid stenting”.
56
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
Accordingly, “Carotid stenting” should
appear alongside “Carotid artery” and
“Carotid Endarterectomy” in section 4
(the Glossary). Perhaps “Carotid
stenting; a minimally invasive procedure
to “brace back” (secure) atheroma within
the carotid artery” would be appropriate.
This mirrors the definition given in the
NICE guidance on carotid artery stenting
(IPG 191) (3).
Rationale for Point 1:
The NICE guidance (IPG 191)
recognises and accepts that carotid
stenting in the United Kingdom is not
just confined to randomised trial
comparisons against carotid
endarterectomy for example, “carotid
stent procedures performed outside of
the ICSS trial (International Carotid
Stenting Study) should be submitted to
the Endovascular Carotid Registry held
by the British Society of Interventional
Radiology and the Vascular Society of
Great Britain and Ireland”. Furthermore,
the Intercollegiate Working Party (ICWP)
guideline on stroke indicated that
‘carotid artery angioplasty or stenting is
an alternative to surgery but should only
be carried out in specialist centres
where outcomes of these techniques are
routinely audited’. There are a number of
such units in the United Kingdom where
outcomes are regularly audited, and, as
per NICE guidance (IPG 191) the
decision for carotid intervention by
surgical or endovascular means is
routinely made in the setting of a
57
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
multidisciplinary team environment,
protective for both the patient and the
practitioner.
The respective roles of carotid
endarterectomy and carotid stenting
remain to be defined but it is clear from
the available evidence that carotid
stenting is a viable, less invasive and in
selected cases safer intervention than
carotid endarterectomy. It has and will
continue to have an important role to
play and it would be unfortunate,
therefore, if a document from the
National Institute for Clinical Excellence
were to give the impression that one
intervention was to be preferred over
another in all cases. National Clinical
Guidelines that seek to provide
comprehensive patient care should
present all the viable treatment options
to the patient and should thus offer
carotid stenting as an alternative to
carotid endarterectomy.
This statement cannot be argued
strongly enough, particularly given the
statement in the “Aims” section of the
draft consultation; the “aim of the
National Collaborating Centre for
Chronic Conditions (NCC-CC) is to
provide a user-friendly, clinical evidencebased guideline for the National Health
Service In England and Wales” that
amongst other considerations, “Takes
into account patient choice and informed
decision-making”.
SH
British Society of
3
Full
General
General Point 2. The development
58
Thank you for your comment. The
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Interventional Radiology
Comments
Developer’s Response.
process and membership of the
Guideline Development Group (GDG) is
explained but the process of
appointment is not.
For purposes of credibility appointment
to the group should be transparent.
Perhaps the document should detail how
individuals were selected.
developers do not feel that this level of
detail is not required within the clinical
guideline because it is covered in detail in
the NICE guideline development manual.
A GDG is never designed to represent
the interest of any specific interest group,
nor could all be represented. It is
clinicians and professionals with
knowledge of the field who are tasked
with evaluating evidence when such
evidence exists. The GDG members do
not represent a professional body.
For the Stroke document to be
unbiased, the GDG must provide a
balanced opinion and judgement. All
specialist clinicians involved in the
prevention and management of stroke
should be included. It is noted with
interest that both the Vascular Society
and the Society of British Neurological
Surgeons is represented within the GDG
(by at least one surgeon who performs
carotid endarterectomy), whereas there
is apparently no representative with
experience of carotid stenting. The
Royal College of Radiologists is
represented by a Neuroradiologist but
there is no representation from the
British Society of Interventional
Radiology. This imbalance may lead to
bias and should therefore be addressed.
SH
British Society of
Interventional Radiology
4 Full
2.6
SH
British Society of
Interventional Radiology
5 Full
6.3.1.1
In section 2.6, Carotid Artery Stenting
(IPG 191) should be added to the list of
“Related NICE guidance”.
Mention is made of carotid
endarterectomy for those “with a TIA or
minor or recovered stroke involving the
anterior circulation who are fit and willing
for surgery”. This is misleading. It might
imply that carotid stenting should be
reserved for those who are surgically
59
IPG 191 covers interventions which are
outside the time period of this guideline
Thank you. These papers were excluded
in the evidence review because they
compare CEA with CS. Thank you. We
did not review any RCT evidence for
carotid stenting within the ‘acute’ two
week period of the guideline
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
unfit. Whilst there is level-1 evidence
demonstrating better outcomes for
carotid stenting than for endarterectomy
for “high risk” populations (SAPPHIRE
trial) (Yadav J, Wholey MH, Kuntz RE et
al. Protected carotid-artery stenting
versus endarterectomy in high-risk
patients. N Engl j Med 2004;351:14931501.), carotid stenting is not limited to
patients at high risk for surgery (National
clinical guidelines for stroke. Second
edition. Prepared by the Intercollegiate
Stroke Working Party. 2004).
SH
British Society of
Interventional Radiology
6 Full
6.3.5
The health economic evidence
statement for this section is based on
Wardlaw et al (Wardlaw JM, Chappell
FM, Stevenson M et al. Accurate,
60
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.
The Sapphire Trial does NOT represent
grade 1 evidence of equivalence or
superiority of carotid stenting in patients
with Stroke or TIA. Less than 30% of the
patients had suffered a symptomatic
event and the 30 day endpoint included
non fatal MI. At the conventional endpoint of 30 day any stroke or death, the
symptomatic patients (less than 30% of
enrolled patients) were not different (the
difference in the main trial was solely
based on non fatal myocardial events).
Indeed the trial was not powered to
detect such a difference. The statement
in the accompanying commentary in the
NEJM by Cambria (NEJM 351:1566) is
pertinent “However the small sample size
and the study end-points preclude major
conclusions about the relative role of
endarterctomy and carotid artery stenting
in the treatment of carotid artery
stenosis.”
We do not agree that the model was
misrepresented but we do acknowledge
that the results were overly brief and
over-simplified. The GDG did not intend
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
practical and cost-effective assessment
of carotid stenosis in the UK. Health
Technol Assess 2006;10:iii-iv, ix-x, 1182). The statement: “using a threshold
of £30,000 per QALY, the most costeffective strategy was to conduct an
ultrasound scan and then offer
endarterectomy to all patients with a 5099% stenosis. As well as offering cost
savings by avoiding confirmatory tests
after ultrasound scanning, this strategy
minimised the average time to
endarterectomy, thus maximising the
health gain” misrepresents the
conclusions of the paper and, although it
does not overtly recommend a strategy
of carotid endarterectomy based on a
single ultrasound investigation, it implies
that this is the most cost-effective and
therefore preferred strategy.
to recommend a specific diagnostic
strategy and we have now made this
more explicit. We have revised this
section to emphasise the implications of
speedy access to diagnosis.
Wardlaw et al also state that if society is Yes
willing to pay only £20,000 per QALY,
the optimal strategy is ultrasound
followed by CT angiography or MRA
where ultrasound shows 70-99%
stenosis with patients offered
intervention when tests agree. Where
there is disagreement, the decision to
intervene should be based on CEMRA.
The Executive summary of the paper
Yes
actually states the following: “In the costeffectiveness model, on current UK
timings, strategies (which) allowed more
patients to reach endarterectomy very
quickly, and where those with 50-69%
stenosis would be offered surgery in
addition to those with 70-99% stenosis,
61
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
prevented most strokes and produced
greatest net benefit. This included most
strategies with ultrasound as first or
repeat test, and not those with intraarterial angiography. However, the
model was sensitive to less invasive test
accuracy, cost and timing of
endarterectomy. In patients investigated
late after TIA, test accuracy is crucial
and contrast-enhanced magnetic
resonance angiography (CEMRA)
should be used before surgery”. The
authors concluded that “In the UK, less
invasive tests could be used in place of
intra-arterial angiography if radiologists
trained in carotid imaging are available.
Imaging should be carefully audited”.
We believe that these consequences are
Wardlaw et al, point out in a number of
taken in to account in the modelling by
sections of the full monograph that
Wardlaw et al.
CEMRA has the highest diagnostic
accuracy of the less invasive tests, with
the highest sensitivity and specificity and
the least heterogeneity. A strategy
based on a single ultrasound
investigation will therefore inevitably
lead, in some patients, to unnecessary
and potentially hazardous interventions
whilst denying interventions to others
who would benefit. Such unnecessary
interventions are not included in cost
effectiveness modeling nor are those
patients who do not undergo intervention
when they should.
We respectfully suggest either that the
document makes no reference to the
62
We have expanded our description of the
model by Wardlaw to highlight the
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
analysis of cost-effectiveness of
qualifications that you refer to. We now
Wardlaw et al or quotes their
quote their conclusions in full.
conclusions in full, as per the “Executive
Summary”.
SH
British Society of
Interventional Radiology
7 Full
6.4.4.4
Bearing in mind that this guideline is
designed to be read alongside the
Intercollegiate Working Party (ICWP)
National Clinical Guideline on stroke
(National clinical guidelines for stroke.
Second edition. Prepared by the
Intercollegiate Stroke Working Party.
2004) it is noted with interest that the
ICWP recommends that duplex findings
be confirmed by MRA (or a second
duplex) in the assessment of carotid
artery stenosis.
Yes.
It is clear that patients with a history of
recent TIA or minor stroke should be
investigated in a timely fashion to
determine whether or not they would
benefit from intervention. The GDG must
surely agree, however, that investigation
should be as sensitive and specific as
possible so as to ensure that
intervention is offered appropriately. This
need not delay intervention if, for
example, targeted local investment
allows provision of same-day MRI
imaging slots, in addition to carotid
duplex ultrasound to support
neurovascular/TIA clinics.
Diagnostic strategies need to be costeffective. The most accurate test is often
although not always the most costeffective one. However, we acknowledge
that it is difficult to assess costeffectiveness in this area. The best
strategy for imaging of the carotid artery
was not a question considered by the
GDG and therefore not one that they can
make a recommendation on.
It is stated that “the effects of surgery
are modified by time since last event,
gender and age such that the benefit
63
Thank you. We have amended section
6.4.2 to include a paragraph that was
omitted from this section but included in
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
statistically decreases as the time since the evidence tables.
last symptoms increases, is statistically
greater in males than females and in the
elderly”. The section further explores the
outcomes of carotid endarterectomy in
the setting of acute stroke. The
arguments made in favour of early
surgery are based on limited data. The
consultation document references the
ECST and NASCET with a total
population of 5893 patients. The actual
number of patients in the ECST with a
70-99% stenosis that were randomised
to surgery was 750 and only 14.5%
(108) were randomised within two weeks
of index symptom (time from
randomisation to surgery is not given)
(AR Naylor, PM Rothwell, PRF Bell.
Overview of the principal results and
secondary analyses from the European
and North American Randomised Trials
of Endarterectomy for Symptomatic
Carotid Stenosis. Eur J Vasc Endovasc
Surg 2003;26:115-129). In the NASCET,
there were 328 patients operated for a
70-99% stenosis and only 25.9%
(around 85) were randomised within two
weeks of symptom onset (time from
randomisation to surgery is not given).
The consultation document further
presents a prospective case series of
238 patients, of which 55% with greater
than 50% stenosis (i.e. around 130)
were operated on within two weeks of
symptom onset. The systematic review
used to support the case for early
endarterectomy in the setting of stroke
does not include randomised trial
evidence and reports on only 271
64
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
patients with “unstable” symptoms (to
include stroke in evolution, non-specified
“urgent” cases and crescendo TIA). The
total number of neurologically stable
patients presenting with stroke in series
comparing perioperative hazard for
patients undergoing endarterectomy
between less than three weeks and
more than three weeks from stroke is
252 (Fairhead JF, Rothwell PM. The
need for urgency in identification and
treatment of symptomatic carotid
stenosis is already established.
Cerebrovascular diseases 2005;19:355358). In addition, a large series from the
US demonstrated that when
symptomatic patients are treated with
CEA within 4 weeks of the index
symptom the adverse event rate can be
expected to be 4 times higher than when
patients are treated after 4 weeks
(Rockman CB, Maldonado T,
Jacobowitz GR, Cayne NS, Gagne PJ,
Riles T Early endarterectomy in
symptomatic patients is associated with
poorer perioperative outcomes. J Vasc
Surg 2006;44;480-487). Clearly there is
a fine balance between peri-operative
risk and benefit of early intervention, but
when data are conflicting some care
must be taken in the setting of
recommendations based on relatively
limited data. Furthermore, when
considering early intervention following
stroke, perioperative hazard will relate to
the neurological stability of the patient
being treated and this should be
specified in any analysis of early
intervention.
65
This was not included because it is an
editorial
This paper was not included because it is
a retrospective case review and the
evidence review was restricted to RCTs
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
It is stated in section 6.4.5.1 that no
studies were identified on early carotid
stenting i.e. within two weeks of index
symptom and it is implied that carotid
stenting may be less safe in the acute
phase. However, there is at least one
paper that investigated median delay
before intervention in 320 patients
undergoing carotid stenting. Median
delay between symptoms and
intervention was 19 days (interquartile
range 10-36). Time delay was not
significantly associated with periprocedural complications, regardless of
whether this variable was dichotomized
to  14 days or ≥ 14 days, separated
into interquartile ranges or analysed as a
continuous variable (Gröschel K, Knauth
M, Ernemann U, et al. Early treatment
after a symptomatic event is not
associated with an increased risk of
stroke in patients undergoing carotid
stenting. Eur J Neurol. 2008 Jan;15(1):25. Epub 2007 Nov 14). An analysis of
527 carotid stenting procedures
performed in a high throughput carotid
stenting centre (Sheffield) demonstrated
no significant difference between
procedural complications for patients
treated within two weeks compared with
those treated after two weeks of
symptom onset. In due course, post hoc
analyses of the pooled results from the
randomised trials SPACE, SAPPHIRE,
EVA3S, ICSS and CREST, when the
latter two trials are completed, may
provide additional data on outcomes for
carotid stenting in the setting of acute
66
This paper was published outside of the
literature review cut-off date and is a case
series. We have clarified that the papers
included were restricted to RCTs
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
symptoms.
SH
British Society of
Interventional Radiology
8
Full
10.3.1
In summary, although it is self-evident
that any intervention performed with
prophylactic intent should be performed
sooner rather than later (since delay
inevitably increases the risk that the
event to be prevented will occur before
the intervention is performed), there is,
as yet, no level-1 evidence that either
carotid surgery or stenting is preferable
in the setting of acute stroke. Thus we
would respectfully request that the
committee considers carefully the weight
given to the recommendation that all
patients be treated early after any form
of neurological event. Where
intervention is to be recommended, a
choice of Carotid Endarterectomy or
Carotid Artery Stenting would appear to
be appropriate.
This section examines tube feeding via
nasogastric or percutaneous endoscopic
gastrostomy/jejunostomy (PEG/PEJ)
tube. It is stated that PEG/PEJ requires
an endoscopy, which carries a small
risk, especially in patients with chest
problems. Radiologically-inserted
gastrostomy tubes (RIGs) are not
mentioned and although they are used
more often in the setting of head and
neck and upper gastrointestinal cancers,
they can be used in patients who cannot
swallow following stroke and have
distinct advantages in patients with
chest problems. A meta-analysis of the
literature on radiologic, endoscopic and
surgical gastrostomies indicates that
RIGs are associated with a lower
67
We have also amended the FETR to
further clarify the GDG’s rationale. The
group have not reviewed evidence to
support making a recommendation on the
use of carotid stenting. The post hoc
analysis data is still considered as level 1
evidence and whilst the number of
patients randomised was much smaller
than the total population of patients the
total number of patients randomised were
still significant for acute stroke studies.
Thank you. Please see the amended
section of the guideline.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
British Society of
Neuroradiologists
1
NICE
1.3.2.1
SH
British Society of
Neuroradiologists
2
NICE
1.3.2.2
SH
British Society of
Neuroradiologists
3
NICE
1.2.1.1 to
1.2.2.3
Developer’s Response.
Comments
infection rate, require less sedation, are
associated with lower rates of aspiration,
can be performed in patients with
oesophageal strictures and are
associated with a lower incidence of
colonic transfixion and possibly lower
incidence of prion transmission
(Wollman B, D’Agostino HB, WalusWigle JR et al. Radiologic, endoscopic
and surgical gastrostomy: an institutional
evaluation and meta-analysis of the
literature. Radiology 1995;197:699-704).
Overall we agree with the great majority
of the guideline and congratulate the
NICWE working group on it; but we do
have a few specific points to make:
We believe that clarification is needed in
the list of indications for urgent brain
imaging.
A depressed level of consciousness
needs defined e.g. Glasgow Coma
Score <13, would be appropriate.
Likewise fever is not defined and the
recommendation could be clearer e.g.
“sustained Temp >37.5”
There needs to be clear recognition a)
that in a small group of stroke patients
e.g. terminally ill from other condition,
brain imaging may be inappropriate, and
b) that if a patient presents late (days to
weeks) after symptom onset, brain
imaging within 24h is not indicated
should be overtly stated
We feel that the current wording is
entirely appropriate except that “CT
techniques” would be better than just CT
as currently worded. This potentially
encourages wider use of more advanced
techniques such as CT
68
These papers were not included in the
evidence review because they are not on
acute stroke patients
Thank you we have stipulated Glasgow
Coma Score < 13. We have not defined
fever in more detail as we feel this is a
clinical definition
The introduction has been amended to
state that the guidelines may not be
appropriate for all patients eg those who
are terminally ill
Thank you, we did not review evidence
for techniques other than plain CT scan
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
British Society of
Neuroradiologists
4
NICE
1.2.3.1
SH
British Society of
Neuroradiologists
5
NICE
1.2.3
Developer’s Response.
Comments
Angiography/Perfusion in patients for
whom MRI is contraindicated or not
available.
The up to 1 week timeframe for carotid
imaging seems contrary to the evidence
that most benefit accrues in the group of
TIA/acute non-disabling stroke treated
by CEA within 2 weeks of symptom
onset. It is also at odds with 1.2.4 where
NICE is advocating CEA within 2 weeks.
High risk TIA / ANDS patients should
have carotid imaging within 24h if there
is to be an early referral for CEA and a
chance of receiving carotid intervention
(better phrase to use) within 2 weeks.
This would improve clarity as it would
also be in line with National Stroke
Strategy.
As worded in the draft, NICE guidance
will potentially encourage a 1 week
carotid imaging policy- making it very
difficult indeed to achieve carotid
intervention within 2 weeks.
Carotid intervention is a better
terminology than CEA as increasing
numbers of patients are treated by
carotid stenting- even if at present this
should mainly be for cases where CEA
is difficult or within context of a trial such
as ICSS.
Thank you, we agree that early carotid
imaging is necessary to ensure that
carotid endarterectomy is available within
2 weeks.
See response to comment 62 See
response 62 (‘Thank you. We did not
review any RCT evidence for carotid
stenting within the ‘acute’ two week
period of the guideline
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.’)
SH
British Society of
1
Full
General
The BSRM welcomes the acute stroke
69
This is outside of the scope.Many of
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Rehabilitation Medicine
SH
British Society of
Rehabilitation Medicine
2 Full
7.1.1
/General
Comments
Developer’s Response.
guidance. We understand that the scope
of the document is concerned with the
acute medical interventions for stroke,
which have been employed in well
designed trials, and for which a strong
evidence base exists. Our response
would wish to highlight the importance of
other concurrent aspects of care in such
an acute unit, and also immediately
afterwards, which will allow these
improvements in interim outcomes to be
fully realized in terms of meaningful long
term improvements in reduction of
disability and maximum quality of life for
stroke survivors.
these aspects are covered by the ICSWP
guidelines. A comment to underline the
importance of patient centred
rehabilitation has been added to the
introduction
We see that the proposed improvement Thank you, we have added a comment
in the acute stroke care will increase the about early rehabilitation and patient
number of survivors of stroke and
centred care to the introduction
reduce the secondary damage which
may accrue from such an event. We
suggest that this will substantially
increase the number of the population
who will have good potential for recovery
and long term survival. At the end of
such acute treatment a patient’s ability
may be regarded as potential, as more
robust long term outcomes such as
dependency, return to economic activity
and institutionalization depend also on
the realization of this potential. The
WHO in its international classification of
function defines the interaction between
a disease state, its effects on structure
and function, activity limitation and
participation restriction, and how this is
not always a direct path but influenced
by modifying and mediating factors, it is
the role of rehabilitation to make sure
70
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
that by maximising potential and
modifying these factors a truly good
outcome can be obtained.
As survival from acute care increases
this will increase the demand for high
quality rehabilitation services. For those
who have survived the initial stroke,
moving on is essential both for
themselves and in order to maintain the
integrity of service. Unless pathways are
in place to move individuals through the
services, this will restrict access to the
acute beds.
Even with effective acute care, many
stroke survivors will experience
significant impairment following the
acute event and will benefit from timely
rehabilitation in order to maximise their
potential for recovery. Others will survive
their stroke where they might not have
before the advent of such organised
care, and will remain substantially
disabled. This is analogous to the issues
surrounding the survival of special care
babies, and recipients of intensive care
services. People who are left with a
disability will require the attention of
rehabilitation services to provide safe
and timely discharge from hospital and
subsequent long term access to services
which will prevent long term
complications from disability.
SH
British Society of
Rehabilitation Medicine
3
Full
General
Requirement for accurate diagnosis of
neurological disabilities and of
Rehabilitation Potential
71
Thank you.
We agree that the rehabilitation needs
are important however the main scope of
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
British Society of
Rehabilitation Medicine
4
Full
11.1
SH
British Society of
Rehabilitation Medicine
5
Full
/10.2
General
Comments
Developer’s Response.
We recommend that rehabilitation needs
are always held in mind along the whole
length of the stroke pathway, and that
rehabilitation principles are observed.
Although we appreciate that the NICE
guidance is confined to acute care, we
recommend specific recommendation to
be included to mandate the provision of
skilled rehabilitation assessment,
including assessment by doctors with
skills in the area of rehabilitation of
stroke patients as early as survival is
established. We would like the guidance
to specify the entitlement of all stroke
survivors to full multidisciplinary team
treatment within organised services. We
would point out that such an approach is
mandated in the Government’s National
Service Framework for Long Term
Conditions, and also in the
Government’s recent Stroke Strategy.
the guideline was to address the ‘initial
and early management aimed at reducing
the ischaemic brain damage and in the
case of TIAs preventing subsequent
stokes’. We were unfortunately unable to
cover all areas and focused upon those
that people / stakeholders initially
suggested as critical areas for address.
We recommend that the guidelines
reflect the importance of the early
identification of rehabilitation needs and
recognize the potential improvement in
outcomes which can be achieved by
timely input of rehabilitation of sufficient
intensity
We are pleased that early mobilisation
has been included, along with
identification of swallowing problems.
We would also wish for guidance to
reflect the need for rapid attention to
preventable physical, behavioural or
neuropsychological complications of
stroke, as well as the medical ones
which have been well covered. This
includes, but is not exclusively covered
72
Please refer to the Royal College of
Physicians ICSWP guideline which is
currently being updated and will be
published at the same time as this
guideline. The ICSWP guideline covers
the issue of rehabilitation.
Thank you
Thank you for your comment.
The main scope of the guideline was to
address the ‘initial and early management
aimed at reducing the ischaemic brain
damage and in the case of TIAs
preventing subsequent stokes’.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
by issues of pain, mood, contracture,
spasticity, and adequate attention to
posture, seating, inattention
management, rapid attention to
behavioural and cognitive difficulties and
problems of adjustment of both patient
and their close family.
SH
British Society of
Rehabilitation Medicine
6
Full
General
SH
British Society of
Rehabilitation Medicine
7
Full
General
Many of the outcome measures which
are used in stroke trials do not cover the
nuances of high quality of life following
stroke. We also feel that even those
people who have a good physical
recovery would have further needs for
vocational and social support, return to
fitness and economic activity. Therefore
the recommendation for expert
assessment after leaving an acute
stroke service is equally important for all
groups irrespective of level of disability.
The BSRM as a professional
organisation would specifically like to
make the following points which
summarize our position.
The BSRM feels that Rehabilitation
Medicine (RM) has a vital role to play in
the immediate and long term needs of
individuals who have sustained stroke.
We would emphasize the substantial
role that RM specialists contribute to
stroke services, and also the important
role we play to identify, champion and
maintain expertise within community
teams, including early stroke discharge
teams and community teams.
SH
British Society of
Rehabilitation Medicine
8
Full
General
One area where RM physicians have
particular expertise is with Younger
73
We were unfortunately unable to cover all
areas and focused upon those that
people / stakeholders initially suggested
as critical areas for address.
Thank you for your comment. This
recommendation is outside the remit of
our scope. Please refer to the Royal
College of Physicians ICSWP guideline.
The updated 3rd edition is due to be
published at the same time as this
guideline. we will pass these comments
to the ICSWP guideline group
Thank you for your comments. Please
see the response for comment 116
(‘Thank you for your comment. This
recommendation is outside the remit of
our scope. Please refer to the Royal
College of Physicians ICSWP guideline.
The updated 3rd edition is due to be
published at the same time as this
guideline. we will pass these comments
to the ICSWP guideline group.’)
Thank you for your comments. Please
see the response for comment 116
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
stroke survivors who have particular
needs and indeed high expectations of a
good quality of life following survival
from stroke. This includes vocation:
indeed they are likely to be in a position
where other people may be
economically dependent on them, and
they may have other social roles
including parenting.
(‘Thank you for your comment. This
recommendation is outside the remit of
our scope. Please refer to the Royal
College of Physicians ICSWP guideline.
The updated 3rd edition is due to be
published at the same time as this
guideline. we will pass these comments
to the ICSWP guideline group’)
SH
British Society of
Rehabilitation Medicine
9
Full
General
BSRM believes that there is no single
Thank you. This has been noted by the
model which will fit all stroke patients
developers.
along the whole course of their illness
trajectory. We would point out that in the
North West effective referral pathways
have been established so that those
who have been identified as being best
served in Neurological Rehabilitation
Units are readily identified. This may
particularly include those below the age
of 40, those with good physical recovery,
but ongoing cognitive problems or
dysexecutive syndrome., or people with
very severe disability for instance the
“locked in Syndrome”, which will require
very careful and expert planning to
ensure timely and safe discharge with
the access to specialist skills such as
communication aids, environmental
control systems and integrated assistive
technology.
SH
British Society of
Rehabilitation Medicine
10 Full
General
We would be concerned if
commissioners treated stroke as a
strictly homogeneous condition with
services which were dominated only by
the considerations of acute care, rather
we would prefer a network of specialties
and services which would collaborate to
74
Thank you.
We agree that the rehabilitation needs
are important however the main scope of
the guideline was to address the ‘initial
and early management aimed at reducing
the ischaemic brain damage and in the
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
provide seamless service which were
commissioned together.
case of TIAs preventing subsequent
stokes’. We were unfortunately unable to
cover all areas and focused upon those
that people / stakeholders initially
suggested as critical areas for address.
Please refer to the Royal College of
Physicians intercollegiate stroke working
party guideline which is currently being
updated and will be published at the
same time as this guideline. The
intercollegiate stroke working party
guideline covers the issue of
rehabilitation.
SH
SH
British Society of
Rehabilitation Medicine
College of Emergency
11 Full
1 Full
General
7.2
BSRM would like to draw the
Development Group’s attention to the
following references:
Thank you these references were noted
by the developers.
1. Kwakkel, G et al (1999).
Intensity of leg and arm training
1. Excluded because it is a pilot
after primary middle cerebral
study
artery: a randomised trial. The
Lancet, 354, 191-96.
2. The International Classification
of Functioning, Disability and
2. This paper did not specifically
Health (ICF) Core. Stucki and
address the question for the
Cieza Ann Rheum Dis. 2004;
clinical evidence review
63: ii40-ii45.
3. www.dh.gov.uk/en/Healthcare/N
ationalService
3. NSF (could not be included in the
Framework/Longtermconditions/
evidence review)
index.htm
4. www.dh.gov.uk/en/Publicationsa
ndstatistics/
Publications/PublicationsPolicyA
4. Policy document (as for point 3.)
ndGuidance/
DH_081062
The guidance recognises that ‘access to Thank you. This issue was discussed at
75
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Medicine
in
particular
7.2.6
Comments
Developer’s Response.
brain scanning has been difficult in the
past because of a perceived lack of
urgency for scanning.’
The recommendations here have the
potential to allow that situation to
continue.
length by the GDG and it was felt that the
recommendations set challenging but
realistic targets.
The clinician faced with a patient with an
uncertain diagnosis has to find a way to
persuade the radiologist that the patient
fits the criteria for an urgent scan. The
default position should be to scan unless
it is clear that the clinical decision will be
utterly unaltered by a scan- a terminally
ill patient for example. The National
Stroke Strategy recognised that some
patients will not be eligible to receive
thrombolysis they may still benefit from
rapid access to stroke services. This
needs the basic imaging to have been
performed so that the patient can go to
the right place. The patient
representatives on the Emergency
Response group of the National Stroke
strategy were vocal in their belief that all
patients should have immediate access
to CT scanning unless there were major
reasons why not – that clinicians should
have to justify NOT scanning, not the
other way round. Emergency
Department clinicians also want rapid
access to scanning for all so that
appropriate clinical disposal decisions
can be made at the ‘front door.’ The
document states that NICE clinical
guidelines do not cover issues of service
delivery (2.5.1.1.) However by giving a
list of indications (3.1.1.4) only in the
Key messages section and then a 24 hr
76
Typ
e
SH
Stakeholder
College of Emergency
Medicine
N
o
2
Docum
ent
Full
Section
number
General
and
2.1.1.1
SH
College of Emergency
Medicine
3 Full
9
and 3.1
SH
College of Emergency
Medicine
4 Full
3.1.1.3
&
7.1.7.1
Developer’s Response.
Comments
window for scanning for the other
patients (7.2.6.2) this guidance will
encourage imaging services to remain
difficult to access.
The guidance covers the available
evidence in a comprehensive manner. It
also mentions the major components of
NHS care provision for the acute
management of stroke and TIA.
Emergency Medicine teams would be
keen to see more emphasis on
collaborative working with pre-hospital
and Emergency Department staff to
achieve a streamlined patient pathway
of care as part of basic ‘organised stroke
services’.
The guidance contains evidence for best
practice in the maintenance and
restoration of homeostasis but the Key
priorities only mention swallow
assessment and are very similar to
previous recommendations. This seems
a missed opportunity to encourage best
practice from the start of the clinical
assessment. A Key priority about
aspirin, hyperglycaemia management
and hydration would help to emphasis
the importance of urgent basic clinical
care in all stroke patients not just those
eligible for thrombolysis. There is no
section on management of acute
pyrexia.
The guideline states that “all patients
should be admitted directly to a
Specialist Stroke Unit.”
Whilst reference to the flow algorithm
shows that patients should be formally
assessed and diagnosed prior to
transfer, the brevity of the text may lead
77
Thank you for your comment we agree
and hope that recommendations made
within this guideline will help to ensure
this.
Thank you for your comment. The key
priorities for implementation were voted
on by the GDG.
Thank you we have amended the
recommendation to include “directly from
the community or A7E or emmergency
room”.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
to an incorrect assumption of patients
being admitted directly from ambulances
into a stroke unit.
The College would have concerns about
such practice as stroke units are not
usually staffed or equipped to manage
unfiltered and potentially unstable
patients.
SH
College of Emergency
Medicine
5 Full
8.5
in
particular
8.5.1.2
It would therefore be preferable for the
body of the text to be clarified, reflecting
that patients are referred to a stroke unit
once formally assessed by either a
General Practitioner or within an
Emergency Department.
With reference to the comment “used in Thank you we have amended the text
full accordance with its marketing
authorisation.”
This requires Alteplase only be given by
a “Physician specialised in neurological
care.”
At face value this excludes all
Emergency Care Physicians and has
prevented many hospitals from providing
this treatment.
In the UK only 0.2% of patients with
ischaemic stroke currently receive
thrombolysis, compared to 10% in
Australia. Considerable ground must
therefore be gained if a 50-fold increase
is to be achieved.
Although Alteplase is licensed for use
any time within 3 hours of ischaemic
stroke onset, its therapeutic benefits are
known to be greater the earlier that
treatment can be given: 1.9 million
78
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
neurones are lost every minute from the
onset of ischaemic stroke. This results in
an NNT (number needed to treat to
show benefit) at 2 hours of just 2.5,
compared to an NNT of 11 at 3 hours.
Necessitating a Stroke Physician will
restrict treatment with Alteplase,
especially in centres where few are
employed: patients either being
excluded or receiving it later within the
allowed time window.
Emergency Medicine is the natural
choice for the provision of stroke
thrombolysis, offering:
1) Rapid Access
2) 24/7 staffing: consisting of
greater seniority of medical
cover out of hours than
anywhere else in the hospital.
3) Expertise: Emergency
Physicians see more strokes
within 3 hours of onset than any
other specialty.
4) Experience with thrombolysis of
more than 10 years in the
treatment of myocardial
infarction.
5) Ready access to CT scanning
and skills in CT interpretation.
Professor Roger Boyle, National Director
for Heart Disease & Stroke, has recently
clarified the position. He states that
given appropriate training and under
local agreement with Stroke Physicians,
Emergency Care Physicians could take
the lead in the use of Alteplase in
79
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
patients with acute ischaemic stroke.
The College of Emergency Medicine
acknowledges that the way forward for
this service is by active collaboration
with Stroke Physicians and the
development of regional decision
support networks. However, there will
still be occasion where, having
previously satisfied training
requirements and under local clinical
governance arrangements, the only
person available to provide thrombolytic
therapy is an Emergency Physician
acting independently.
Although this lies within the reinterpretation of the licence, it is contrary
to the letter of NICE guidelines as they
stand currently. This will leave
Emergency Physicians who are trying to
act in their patients’ best interest, feeling
vulnerable and exposed. The practical
result of which may be the withholding of
treatment in such cases.
SH
College of Emergency
Medicine
6 Full
9.3
in
particular
9.3.6.1
SH
College of Occupational
1 Full
5.12
It is therefore imperative for NICE
guidance to reflect Professor Boyle’s
comments and to explicitly include
trained non-stroke physicians, including
Emergency Physicians, if they are to be
fully engaged in this process.
Blood pressure manipulation is also a
possibility in those patients being
considered for thrombolysis and should
be added to the list. (Blood pressures in
the range >180/105, may be
manipulated for the facilitation of
thrombolysis).
FAST is the preferred method of
80
Thank you the text has been amended.
Thank you section 5.1.2 has been
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Therapists
Developer’s Response.
Comments
identifying a stroke. Whilst we
amended.
understand that this is a well developed
tool, have their been any tools identified
or can there be some guidance for those
strokes that do not fit this picture (visual
perceptual or cognitive deficits from the
stroke without the obvious physical
signs), but come into wards etc with
acute confusion and not appropriately
diagnosed as a stroke or are diagnosed
a long period after admission?
SH
College of Occupational
Therapists
2
Full
General
The role of accurate cognitive or
perceptual assessment in diagnosing
stroke. Perhaps this needs to be
identified as an ‘area of uncertainty or
controversy requiring further research’.
This area was not reviewed by the group
SH
College of Occupational
Therapists
3
Full
11
Mobilisation as a title – perhaps this
could be changed to activity
engagement as mobilisation is an
activity but not all activity comes under
mobilisation, hence this limits and does
not illustrate the importance of activity in
general.
We would further suggest that the
search question in Appendix A (page
131) MOBIL1 ‘Does early mobilisation
versus treatment as usual reduce
mortality and morbidity in patients with
acute stroke?’ is expanded to include
engagement in activity.
Thank you for your comment.
Functional assessment to differentiate
stroke – often helps with identifying
functional overlay, picks up other subtle
changes. Perhaps this needs to be
identified as an ‘area of uncertainty or
controversy requiring further research’?
Thank you for your comment. This area is
outside the scope
SH
College of Occupational
Therapists
4
Full
General
81
We have amended the wording of the
recommendation for greater clarification.
However we are unable to make the
change to appendix A as it sets out a
retrospective audit trail for the conducted
literature searches. These searches
were based upon the specified questions
as cited.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
SH
College of Occupational
Therapists
5
Full
General
We note that engagement in activity and
cognitive/perceptual screening have not
been mentioned as tools for assessment
or early intervention.
We recognise there might not be much
evidence for this but was it looked into?
We also noted that overall there was not
a significant amount of evidence for
many items, hence do these two points
at least require a mention, so their
importance is not forgotten?
Thank you for your comment. This area
was not looked at by the group. Please
refer to the Intercollegiate Stroke working
party guidelines.
SH
College of Occupational
Therapists
6
Full
14
In reference to comments 2-5 above we
would suggest that consideration is
given to adding the following research
questions:
‘What is the efficacy of early
cognitive/perceptual/mood screening
after stroke?’ Particularly on disability
and QOL outcomes. And similar
questions for early assessment of
function and communication.
Please see the response to comment 132
(‘Thank you for your comment. This area
was not looked at by the group. Please
refer to the Intercollegiate Stroke working
party guidelines.’)
In our view, this section places
limitations on the types of people to be
considered for thrombolysis. In our
opinion, this is slightly at odds with the
National Stroke Strategy(NSS), which
does not.
You may be aware that the Purchasing
and Supply Agency (PASA)’s centre for
evidence- based purchasing are looking
at Diffusion MRI – they are likely to
recommend that if CT scanning is
inconclusive for stroke, then MRI should
be used. We feel that perhaps it may be
worthwhile for NICE and the team
preparing this for CEP to hold a
We acknowledge the differences between
an evidence based guideline such as this
and a policy document. The developers
of this guideline highlight that the
guideline recommendations are derived
from a sound evidence based.
Thank you for this information. We will
discuss this further with NICE.
&
Appendix
C
SH
Department of Health
1
NICE
1.3.2.1
SH
Department of Health
2
NICE
1.3.2
82
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Department of Health
3
NICE
1.4.5
SH
Diabetes UK
1
Full
General
SH
Diabetes UK
2
Full
General
Developer’s Response.
Comments
discussion.
Regarding the use of alteplase, we
believe that it may be worth stating that
this could take place in the Emergency
Department. In our view, it is not the
location but the training and specialist
input that is important.
Recommendations are needed
throughout the guidance that emphasise
the importance of keeping individuals as
informed as possible throughout the
process of assessment and treatment of
TIA and Stroke and of any subsequent
diagnoses and referrals. Where clinically
possible and appropriate, individuals
should be able to make informed
decisions in partnership with their
healthcare professional regarding
treatment and interventions.
A recommendation is needed regarding
the identification of further
cardiovascular risk factors and
conditions. Following on from such a
recommendation, Diabetes UK
recommends that people presenting with
TIA/Stroke are screened for diabetes
using recognised screening methods
and an urgent referral made once a
positive diagnosis has been confirmed.
Details of recommended screening
methods can be found in the Diabetes
UK Position Statement: Early
identification of Type 2 diabetes. 1
Stroke is a recognised risk factor for
Type 2 diabetes1 and it is known that
people with diabetes have a 3 fold
increased risk of stroke and a 6 fold risk
83
Thank you for your comment. It is not
possible to amend the wording of the
NICE TA in section 1.4.5.1. We have
added this clarification to section 1.4.5.2.
Thank you. This has been added to the
introduction
Thank you for your comment. This area is
outside the scope. Please refer to the
intercollegiate stroke working party
guideline.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
of TIA. 2
1.
http://www.diabetes.org.uk/About_us/Ou
r_Views/Position_statements/Early_iden
tification_of_people_with_Type_2_diabe
tes/
2. Diabetes UK (2007) The Diabetes
Heartache
SH
Diabetes UK
SH
Diabetes UK
3 Full
4
NICE
5.2.6.2
1.1.2.2
People with diabetes have a 3 fold
increased risk of stroke compared to the
general population and are included as
one of the factors in the ABCD2
calculation. People with poor diabetes
control or the presence of diabetic
complications should be considered
within the high risk group in the light of
a) the links between HbA1c and
cardiovascular disease and
b) the additional negative impact of a
stroke on a person with diabetes who
already has complications of their
diabetes.
People with diabetes have a 3 fold
increased risk of stroke compared to the
general population and are included as
one of the factors in the ABCD2
calculation. People with poor diabetes
control or the presence of diabetic
complications should be considered
within the high risk group in the light of
a) the links between HbA1c and
cardiovascular disease and
b) the additional negative impact of a
stroke on a person with diabetes who
already has complications of their
diabetes.
84
Thank you; the studies detailing use of
ABCD2 scoring system did not address
the issue of poor diabetic control
Please see the response above.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
SH
Diabetes UK
5
NICE
1.5.2.1
Thank you for your comment. The
developers were unable to find any
evidence that indicates that for people
with acute stroke blood glucose needs to
be so tightly controlled
SH
Diabetes UK
SH
Diabetes UK
SH
Diabetes UK
8 Full
9.2.6.1
SH
Diabetes UK
9 Full
9.2.2.2
SH
Diabetes UK
10 Full
9.2.5.1
A recommendation is needed
specifically for people with diabetes that
outlines the optimal blood glucose range
of fasting 4-6 mmol/l.
Blood glucose control is particularly
important for people with diabetes and
ranges too high or low could lead to
acute complications such as hypo and
hyper glycaemia which could become
dangerous if left uncorrected.
A recommendation is needed
specifically for people with diabetes that
outlines the optimal blood glucose range
of fasting 4-6 mmol/l. Blood glucose
control is particularly important for
people with diabetes and ranges too
high or low could lead to acute
complications such as hypo and hyper
glycaemia which could become
dangerous if left uncorrected.
It is our understanding that people
without diabetes are unlikely to have a
blood glucose measure above 10 mmol/l
It is our understanding that people
without diabetes are unlikely to have a
blood glucose measure above 10 mmol/l
Is the word “hypoglycaemia” that follows
the phrase “mild to moderate” meant to
read “hyperglycaemia”
Diabetes UK questions why Type 2
diabetes has been singled out
specifically within this statement as
people with Type 1 diabetes are also at
risk of CVD.
SH
Diabetes UK
11 Full
9.2.5.1
6 Full
7
NICE
9.2.6
1.5.2.1
Please see the response above.
Thank you for your comment. We agree.
Please see the response above.
Thank you for your comment. The
guideline has been corrected.
Thank you for your comment. We have
added to the ‘from evidence to
recommendation’ section of the guideline
a sentence that refers the reader to the
type 1 diabetes guideline. This links in
the recommendation taken from the type
1 diabetes guideline in section 9.2.6.2.
Diabetes UK queries the reference to
We did not find any evidence that tighter
HbA1c targets in this instance as HbA1c control was important.
is a measure taken over a longer period
85
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Diabetes UK
12 Full
9.2.5.1
SH
Education for Health
1
General
SH
Essex Cardiac Network
1 Full
5.2.6
SH
Essex Cardiac Network
2 Full
8.6.6.1
SH
Essex Cardiac Network
3 Full
6.4.6
Full
Developer’s Response.
Comments
of time. It would be better to refer to the
optimal target range as follows: fasting
4-6 mmol/l
Clarification is sought regarding the
This section refers to both. We feel that
intended recipients of 9.2.5.1. It is not
this is implicit in the the current wording.
clear whether this is specifically for
people with diabetes, the general
population, or both as content that
appears to be for the general population
is interspersed with information about
people with diabetes.
There is a need to include something
about the education and awareness
raising for primary care staff to use
things like FAST and ABCD2
1. There is enough evidence (express
,faster) that both aspirin and clopidogrel
should be used in tia for 1-3 months.
(5.2.6)
This is beyond the remit of the evidence
review and is an implementation issue
which the developers discuss with the
NICE implementation team.
We disagree. The FASTER and
EXPRESS studies are not RCT’s and did
not specifically assess the use of aspirin
and clopidogrel in comparison with any
other intervention
2. I think the wording should be changed Please see the amended wording of the
in the use if statins in acute stroke; I
recommendation.
would agree that there is no evidence
showing benefit in acute statin use, but
there seems to be no risk and clearly all
patients with iscahemic stroke should go
onto a statin if they have a lifeexpectancy
> 2years; I think therefore we should not
be discouraged from starting
statins early. (8.6.6.1)
If statins not started in hospital a large
number of patients will not eventually be
on them
3) we should be recommending carotid Thank you. The guideline
endarterectomy for all patients with
recommendations are based upon
symptomatic stenosis>70% within 3
published trial evidence.
months, but also symptomatic stenosis
86
Typ
e
Stakeholder
N
o
4
Docum
ent
Full
Section
number
SH
Essex Cardiac Network
10.3.6
SH
Essex Cardiac Network
5 Full
9.2.6
SH
Essex Cardiac Network
6 Full
5.2.62
SH
Essex Cardiac Network
7 Full
5.2.6.2
Developer’s Response.
Comments
50-69% within 4 weeks (%risk reduction
of ischaemic stroke 0-2 weeks 13.8%,
2-4 weeks 3.4%) (6.4.6)
4. I don't necessarily agree that all
patients should receive NG feeding
within 24 hours if swallow unsafe. - no
evidence for this(10.3.6)
5. stroke patients do worse if blood
sugar>7.8mmol/l, therefore could
recommend titrating acute stroke
patients to a level 4.4-7.8 mmol/l.
(9.2.6)
Need to ensure primary care aware and
using ABCD2 tool
There has been much debate between
the clinicians in relation to whether anti
platelets should be given prior to a CT
scan in high risk TIA see below
Consultant 1
The document is very useful and fills a
huge gap for reference. However I have
a problem with the antiplatelet therapy in
the High Risk TIA (evolving stroke)
without prior imaging. This is especially
true if we are going to apply the
recommendations with this group of
patients to have a CT scan within 24
hours. I believe that HIGH RISK TIA
should have a scan followed by aspirin
(the window of benefit in acute ischemic
stroke allows for this). Over the years I
have come across quite few patients
who had symptoms < 24 hours or even
< few hours but had either a tumour or
haemorrhage. I am sure this will be the
case with many colleagues.
87
The group reached a consensus that
early feeding was likely to be beneficial in
the majority of patients
Thank you for your comment. The
developers are unaware of the evidence
which you are citing
This is beyond the remit of the evidence
review and is an implementation issue
which we discuss with the NICE
implementation team.
Where patients symptoms have fully
resolved we are recommending that they
are offered / have been given aspirin.
Where the symptoms are unresolved,
patients should be treated as within the
stroke pathway.
This comparison was not identified within
the clinical question and therefore no
literature search was undertaken.
We have reviewed the evidence and
there is no evidence to suggest that
patients with TIA where symptoms have
fully resolved need scanning before
giving aspirin
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
However the draft ICP for TIA is
recommending aspirin followed by scan.
On the other hand, the case for another
pathology than ischemia when the
symptoms are of brief duration, are slim
Consultant 2
However, in the EXPRESS study,
patients were given aspirin AND
clopidogrel, and a CT was only obtained
prior to starting treatment for patients
with incomplete resolution of symptoms.
This study showed an 80% reduction in
risk of early recurrent stroke. It is
unlikely that any of the treatments used
in the trial were harmful, in view of the
low risk of major bleeding or recurrent
stroke (ischaemiac or haemorrhagic). In
the FASTER study, a CT/MRI was
obtained first, but aspirin and clopidogrel
seem to benefit patients with TIA/minor
stroke NIHSS<3.
I would suggest that all high-risk TIA
patients are started on aspirin AND
clopidogrel AS LONG AS THEIR
SYMPTOMS HAVE FULLY
RESOLVED. They will require brain
imaging within 24 hours anyway, and if
there is an alternative diagnosis, the
medication can be stopped immediately.
If there are persisting symptoms,
Imaging should be done first.
Controversial?
Consultant 3
High risk patient should be seen and
scanned before antiplatelet therapy TIA
is a clinical diagnosis but scanning is
88
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
essential to rule out stroke mimics as
Consultant 4
I am yet to find a clinical guideline or
policy recommending using aspirin &
clopidogrel together on all TIA/Stroke
patients except where there is a recent
coronary event and one can argue in
cases with peripheral vascular disease
and/or diabetes.
Apart from individual controlled studies
which showed such a combination
beneficial in EXPRESS and adverse in
MATCH - I haven't seen it been used in
ALL especially BEFORE brain Imaging
even when the symptoms have
resolved.
If we follow a simplistic atherothrombotic
pathology for a majority of TIA - there is
an argument for high dose antiplatelet
regime especially in the initial stages but before we roll it out as a guideline we would need stronger evidence
It is important to get some evidence for
functioning platelet activity for such a
combination - especially since aspirin
irreversibly blocks the cox1 site pretty
much permanently and any aspirin
resistance can be easily overcome by an
increased dose (75mg is sufficient for
overcoming in a majority of the
population)
I personally am not bold enough to
prescribe blindly unless a strong body of
practioners support InLow risk TIA which
89
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
we see after some time (as in my TIA
clinic) is a different story
This issue needs to be clarified
SH
Essex Cardiac Network
8
Full
SH
Essex Cardiac Network
9
Full
SH
Essex Cardiac Network
10 Full
SH
Essex Cardiac Network
11 Full
SH
Essex Cardiac Network
12 Full
SH
Essex Cardiac Network
13 Full
SH
Essex Cardiac Network
14 Full
Algorithm1 IN the wake of the rapid response
strategy to strokes, TIAs can only be a
retrospective diagnosis (ie only if
symptoms resolve rapidly). Therefore,
one cannot suspect a TIA (start of the
algorithm) and if one does in spite, will
end up not treating a stroke!
Algorithm1 Rosier does not apply to a TIA…it is only
for strokes
Algorithm1 Regarding the comment to ‘consider
blood pressure management’ it may be
dangerous to treat high blood pressure
when a severe carotid stenosis has not
been excluded
Algorithm1 ABCD2 score should come immediately
after FAST
Algorithm1 Positive screen should be worded –
positive history. Anybody with a positive
screen at the time of assessment should
be in the stroke pathway
Algorithm1 In the high ABCD score group,
endarterectomy should be considered
ASAP(in under 2 weeks ideally)
Algorithm1 Aspirin should be the only intervention
prior to the referral of the patient(this
applies exclusively to the group where
the symptoms and signs have resolved)
– or else the GP who is greatly
disadvantaged without any
investigations to back up, would have to
commit too much into the diagnosis of a
TIA when alternatives may remain to be
excluded. There is no evidence of any
other intervention preventing recurrence
90
Thank you. Please see the amended
algorithm for further clarification
Thank you this has been amended
accordingly.
Thank you. Theoretically that is correct
however it has not been shown in
practice
Thank you please see the amended
algorithm.
Thank you the algorithm has been
amended.
Thank you. The developers think the
algorithm is clear and correct i.e that the
surgery has to take place within 2 weeks.
Thank you. It is reasonable if a GP sees
patients initially that they can consider
simple interventions such as statins prior
to referral
Docum
ent
Section
number
Developer’s Response.
Typ
e
Stakeholder
N
o
Comments
SH
Essex Cardiac Network
15 FUll
SH
GE Health Care
1
NICE
1.1
This section talks about the rapid
Thank you for your suggestion. This has
recognition of symptoms and diagnosis been noted by the developers who would
and refers to FAST, ROSIER and the
like the to keep these sections separate
ABCD scoring. However, Imaging is
discussed in another Section namely 1.2
as a separate heading called Imaging in
TIA and Stroke even though imaging
forms part of diagnosis. We therefore
think it should be one of the
methodologies of diagnosis after the
identification of symptoms and should be
integrated into the section on diagnosis.
SH
GE Health Care
2
NICE
1.1
The draft guidelines states that “People We disagree. With a good history it is
with a suspected TIA who require brain often possible to be confident of the
imaging (i.e. those in whom vascular
clinical territory.
territory or pathology is uncertain)
should undergo MR with DWI (magnetic
resonance with diffusion-weighted
imaging) except where contraindicated,
in which case CT (computed
tomography) should be used.” However
since in most TIA patients vascular
territory or pathology is uncertain it
should be recommended that imaging is
of TIA or occurrence of a stroke.
Algorithm1 Re: FAST negative – consider alternate Thank you. Please see the amended
diagnosis
algorithm for further clarification
This can confuse GPs and other
potential users of this algorithm – this
needs to be removed. There could be
patients with amaurosis fugax, posterior
circulatory symptoms or those in whom
symptoms and signs have resolved in
minutes. So at the time of assessment if
FAST is negative but the history is good,
it should still be deemed a TIA until
proven otherwise
91
Typ
e
SH
SH
Stakeholder
GE Health Care
GE Health Care
N
o
3
Docum
ent
NICE
4 Full
Section
number
1.2.2.1
7.1.12
Developer’s Response.
Comments
done as a matter of course for all TIA
patients to ensure proper diagnosis and
treatment.
The guidelines states that people with a
suspected TIA whose symptoms and
signs have completely resolved should
be assessed by a specialist before a
decision on brain imaging is made. This
needs to be incorporated into Sections
1.2.2.2/3 which clarifies it by saying
“whom vascular territory or pathology is
uncertain should undergo urgent brain
imaging”. Also, it does not give the time
frame mentioned in the subsequent
points and sections about the urgency of
imaging required in some cases to
prevent a stroke.
It is noted that you stated that “there is
much less trial evidence available for the
efficacy of acute stroke units than for
rehabilitation units.” Section. 7.1: further
states that 3 studies demonstrated that
patients admitted to a stroke unit
received the therapeutic interventions
and investigations more appropriately
and quickly compared to those in a
general ward.
However, it should be noted that
identification of stroke and TIA
symptoms is currently low and that with
an awareness campaign and quicker
response; people would be going to
stroke units much earlier than is
currently happening. This would then
result in increased benefit to patients
going directly to a stroke unit. This is
because even though there are quite a
92
Thank you. The developers would like to
keep this as a stand alone
recommendation.
We have defined time for interventions,
defining how quickly brain imaging needs
to be done.
Thank you we agree that acute stroke
units have many benefits and the GDG
made a consensus recommendation
about this. This is an implementation
issue and we will discuss your comment
with the NICE implementation team
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
number of stroke units, TIA is still not
being identified early enough to be dealt
with before a stroke.
SH
GE Health Care
5 Full
7.1.6.2
In developing a randomised trial in this
case to gather the evidence it will be
necessary to go beyond comparing
direct admission to an acute stroke unit
vs. admission to a medical ward,
because stages of symptoms may be
different, brain “damage” may be at
different levels so it would need a
thorough assessment of patients to
ensure they were at comparable stages
of stroke to ascertain the differences.
Thank you for your comments. We agree
these points are very valid. Your
suggestion been noted by the
developers. These suggestions would be
considered at the stage of designing a
clinical trial.
Certainly in terms of staff expertise and
to improve “patient experience”;
admitting patients to an acute stroke unit
should be the preferred
recommendation.
SH
GE Health Care
SH
GE Health Care
6 Full
7
Full
5.2.4
/ 5.2.6
General
Thank you this is certainly the case for
the recommendation made in section
7.1.7
This is not a health economic study
however the reduction in stroke incidence
is noted and has been liaised with the HE
team within the NCC-CC. Please see
section 5.2.5.10 for further clarification.
The study by Rothwell et al. also
showed costs savings to the NHS can
be realised through the more intensive
and appropriate treatment of TIA
through a shift in care towards daily
clinics. These clinics routinely
incorporated an imaging exam and
reduced the incidence of future Strokes
by 80%. Therefore the GDG should note
that the evidence from the Rothwell et al
EXPRESS study showed potential cost
savings.
GE Healthcare would like to applaud the Thank you
work of the Guideline Development
Group for their thoroughness in
93
Typ
e
SH
SH
Stakeholder
Greater Manchester and
Cheshire Cardiac Network
Greater Manchester and
Cheshire Cardiac Network
N
o
Docum
ent
Section
number
Full
General
Full
General
1
2
2.4
2.5
2.6
2.8
SH
SH
SH
Greater Manchester and
Cheshire Cardiac Network
Greater Manchester and
Cheshire Cardiac Network
Greater Manchester and
Cheshire Cardiac Network
3 Full
4
Full
5 Full
3.1
General
5.1.2.1
5.1.6
5.2.5.2
5.2.6
Developer’s Response.
Comments
assessing the evidence and developing
recommendations for the diagnosis and
initial management of Acute Stoke and
Transient Ischaemic Attack.
The Greater Manchester and Cheshire
Stroke and Cardiac Network welcomes
the opportunity to review this extensive
document. The Guidelines appear both
clear, concise and comprehensive. It is
beneficial to have cross reference links
with the National Stroke Strategy and
RCP Guidelines which are included
within the document.
This ensures the aims and scope of the
document are clear.
It is of ongoing importance to involve
both stroke patients and carers to obtain
their views and we commend you on
achieving this within the document.
It is helpful to have the limitations of the
guidelines within the document.
It is helpful to have a list of other
relevant cross referenced NICE
Guidance to refer to.
It is helpful to refer to this within the
document
This compliments details within the
National Stroke Strategy.
This appears comprehensive to both
patient carer and professionals.
It is helpful to discuss the different prehospital assessment tools that are
available in depth.
Reflecting on the review of the evidence,
the recommendations are clear.
It is important to informing patients of
their potential risk following a TIA and
awaiting a further appointment.
The recommendations in the Guidance
94
Thank you
Thank you
Thank you
Thank you
Thank you
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Greater Manchester and
Cheshire Cardiac Network
6
Full
General
Developer’s Response.
Comments
are clear and concise.
The Guidance appears very
comprehensive.
6.3.7.1
SH
Greater Manchester and
Cheshire Cardiac Network
7
Full
General
SH
Greater Manchester and
Cheshire Cardiac Network
8
Full
General
SH
Greater Manchester and
Cheshire Cardiac Network
9
Full
General
SH
Greater Manchester and
Cheshire Cardiac Network
10 Full
General
SH
Greater Manchester and
Cheshire Cardiac Network
11 Full
General
11.1.6.1
SH
Greater Manchester and
Cheshire Cardiac Network
12 Full
12.1.6
Could this recommendation be
expanded to include details on the
identification of candidates for carotid
intervention ?
Comprehensive, robust evidence
provided within the Guidance for brain
imaging.
Comprehensive evidence and
recommendations provided for
pharmacological treatments.
Comprehensive evidence and
recommendations provided for
maintenance of homeostasis.
Comprehensive evidence and
recommendations provided for
hydration, nutrition, assessment of
swallow function etc.
Comprehensive evidence and
recommendations provided for
mobilisation.
Would it be helpful to expand on this to
state that it should be a trained
physiotherapist or is this outside the
remit of NICE Guidance ?
Thank you this is detailed in section
6,1.1.1
It is helpful that although no
recommendations as such could be
made based on the lack of evidence, the
Guidance had identified that further
research recommendations should be
made in this area of stroke.
Thank you for your comment. After
additional discussion following the
stakeholder consultation the group
agreed to insert a recommendation. A
research recommendation has also been
made please see section 14 of the
guideline. According to the NICE
Technical Manual, the GDG were only
95
Thank you
Thank you
Thank you
Thank you
Thank you for your comments you are
correct it is outside the remit of the NICE
clinical guideline to name specific
professions. We have inserted the words
“appropriately trained healthcare
professional” to ensure that it is only a
professional such as a physiotherapist
with the relevant expertise who is
involved in mobilisation.
Typ
e
Stakeholder
N
o
Docum
ent
SH
Greater Manchester and
Cheshire Cardiac Network
13 Full
SH
Greater Manchester and
Cheshire Cardiac Network
14 Full
SH
Greater Manchester and
Cheshire Cardiac Network
15 Full
SH
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
1
SH
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
Full
2 Full
Section
number
Developer’s Response.
Comments
General
Comprehensive evidence and
recommendations provided for surgical
interventions
General
The Guidelines appear concise, clear
and comprehensive, the research
recommendations made are relevant to
the contents discussed throughout the
document.
General
The Greater Manchester and Cheshire
Stroke and Cardiac Network supports
the guidance provided in the NICE
document which compliments the
National Stroke Strategy document.
1.1.1.3
The definition of TIA is incomplete. It
should include monocular visual loss
(otherwise amaurosis fugax is excluded
Definitions by your definition)
(page 5) in
NICE, and
section
1.1.1.3 in
Full
6.1
The evidence in support of using MRIDWI as the primary cerebral imaging
in Full,
technique in the investigation of TIA is
section
very weak (only level 3 evidence). There
1.2.1.1 in is no data comparing MRI with CT
NICE
directly, and no evidence is presented to
show that MRI actually changes patient
management in a way that cannot be
achieved with CT. Early CT will exclude
haemorrhage, and the important TIA
mimic of a mass lesion. The example
given that MRI can help to differentiate
TIA from migraine is not much of a
justification as MR may well be normal in
96
able to prioritise 5 research
recommendation for inclusion in the NICE
guideline
Thank you
Thank you
Thank you
Thank you the text has been amended
CT scan is more practical in acute stroke
and is sensitive to early haemorrhage.
There is clear evidence to show that MRI
with DWI/FLAIR is more sensitive to
small ischaemic events. We agree that
MR is more sensitive in the detection of
posterior circulation events but it is
unlikely that MR will become the
investigation of choice for acute stroke in
the foreseeable future.
Typ
e
Stakeholder
N
o
Docum
ent
SH
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
3 Full
SH
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
4 Full
SH
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
5 Full
SH
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
6 Full
Section
number
Developer’s Response.
Comments
both conditions.
If the consensus view is that MRI is
superior to CT in the investigation of
TIA, why does the group not make a
similar recommendation in the
investigation of acute stroke? In that
group, the argument about identifying
the vascular territory where there is
uncertainty is perhaps stronger, and MRI
is certainly better for identifying precisely
posterior circulation events.
4
RCT = Randomised Controlled Trial, not
Glossary Randomised Clinical Trial.
and
Punctuation is poor – for example in the
Definitions definitions of TIA and tPA.
In the definition of thrombolysis, tPA and
Alteplase are both listed as thrombolysis
(sic) drugs. They are the same.
5.2.5.5
In the full version the point is made that
in addition to patients with ABCD2 score
and
of 4 or above, other high risk patients
5.2.6.2 in who should be seen urgently include
Full,
those with recurrent TIAs or who are
section
anticoagulated. Should this not therefore
1.1.2 in
be added to the recommendations?
NICE
8.1.7
In terms of the effect of aspirin on
in Full,
platelet function, an initial dose of
section
300mg followed by 75mg daily should be
1.4.1.1 in enough to achieve a full anti-platelet
NICE
effect, so is there any justification for
recommending treatment with doses of
150-300mg daily for the first 2 weeks
8.8.6.3
Given the uncertainty surrounding this
(as outlined in section 8.8.6.3) is it
in Full,
sensible to have such a prescriptive
section
recommendation regarding withholding
1.4.8.2 in anticoagulants for 14 days?
NICE
Did the GDG also consider the group of
97
Thank you for your comments. These
sections have been corrected.
Thank you. The text has been amended
- an additional recommendation has been
added to this section to add clarification.
Thank you for your comment. IST/CAST
studies showed benefit for 300mg aspirin
for 14 days
This was agreed by the developers taking
into consideration risk of
thromboembolism for an individual which
the developers felt was clearly important
and needed to be taken into account.
Typ
e
SH
SH
SH
SH
SH
Stakeholder
N
o
Docum
ent
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
7
NICE
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
9
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
10 Full
Hammersmith Hospital NHS
Trust (now Imperial College
11 Full
8 Full
Full
Section
number
General
9.3.6
in Full,
section
1.5.3 in
NICE
13.2
in Full,
section
1.9.2 in
NICE
6.3.7
in Full,
section
1.2.3.1 in
NICE
6.4
in Full,
Comments
Developer’s Response.
patients with stroke post MI and mural
thrombus visualised on ECHO? Can
they wait 14 days for anticoagulation?
There is no recommendation about DVT
prophylaxis in the guidelines. Do the
GDG think this should be included?
There is no explicit recommendation
about whether or not existing antihypertensive medication should be
continued. Do the GDG think this
should be included?
Stroke patients post MI would have been
excluded from the evidence review.
With massive MCA territory infarct,
development of ’malignant’ spaceoccupancy may not occur until 3 – 5
days after the ictus. Indeed, one of the
European trials quoted is randomising
patients up to 4 days (though the pooled
analysis only included patients from that
trial who were treated within 2 days).
Therefore, is it sensible for the
recommendations to limit
decompressive craniectomy to 48 hours
after the onset of symptoms? Should
this window not be extended. Also
patients with only MCA involvement (not
MCA plus ACA) and with right (nondominant) hemisphere stroke may be
better candidates for this procedure
(more likely to recover lower limb
function and are not aphasic)
Should the recommendation about
access to carotid imaging specify
patients with symptoms likely to be in
carotid territory?
The GDG considered the evidence from
the pooled analysis. Further trials may
delineate in more detail those patients
most likely to benefit
Both NASCET and ECST used cerebral
angiography as their tool for the
The evidence discussed used NASCET
and ECST. Angiographically determined
98
Thank you. No evidence was reviewed by
the GDG pertaining to this as it was
outside of the guideline scope
This is the subject of current trials
Thank you the text has been amended.
Typ
e
Stakeholder
N
o
Docum
ent
Healthcare NHS Trust)
SH
SH
Hammersmith Hospital NHS
Trust (now Imperial College
Healthcare NHS Trust)
Heart UK
SH
Heart UK
Section
number
Comments
Developer’s Response.
section
1.2.4 in
NICE
assessment of carotid stenosis. Most
units now use non-invasive techniques
(carotid duplex, MRA, CTA) and
establishing how the measurements of
stenosis made relate to NASCET and
ECST criteria is something of a
minefield. For example some units rely
on velocity criteria for assessment of
stenosis. Is it therefore appropriate to
be using these angiographically
determined criteria in the
recommendations?
Also, shouldn’t the recommendation for
CEA patients include a reminder that
they too should receive best medical
treatment?
The definition of the ABCD2 score here
has unfortunately had diabetes omitted.
criteria was not looked at by the group.
Thank you comment added to 2.6.1.1
The developers agree & this has been
added to the recommendation.
Thank you. We have added diabetes to
the definition
12 Full
Appendix
C
1
General
H·E·A·R·T UK welcomes the NICE
guideline on the acute management of
stroke. However acute therapy often
becomes chronic management and
many patients do not have their therapy
significantly changed after discharge
from secondary care. The necessity for
comprehensive management of
cardiovascular risk factors in the longterm needs to be stressed in the
guideline as it is critical to the reduction
of subsequent strokes and other
cardiovascular events[1 2]. This could
be done by making reference to other
groups working in this field for NICE
(Post-MI; Hyperlipidaemia; Diabetes;
Familial Hypercholesterolaemia).
8.6.1
The evidence for the association of lipids We are cross referencing to the lipids
and stroke is controversial and
modification guideline. Most of the data
Full
2 Full
99
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
convoluted. To reduce it simply to total
cholesterol is a gross over-simplification
as this neglects the contribution made
by triglycerides, low HDL-C and other
factors associated with the metabolic
syndrome (hypertension,
hyperuricaemia, elevated blood glucose)
to the pathogenesis of TIAs and stroke
which increase risk by 1.5-fold[3 4]. For
separate risk factors meta-analysis
supports the role of HDL-C as a
protective factor against stroke[5]. The
association of triglycerides with
cardiovascular events has also been
substantiated[6]. As the other guidelines
relevant to atherosclerosis (postmyocardial infarction; hyperlipidaemia;
diabetes and likely familial
hypercholesterolaemia) all
recommended the measurement of full
lipid profiles it is disappointing that these
are not included in the stroke guideline
especially as many patients are more
likely to have a secondary coronary
rather than cerebrovascular event (e.g.
data from SPARCL[7]).
you are suggesting where it is stroke
specific refers to long term secondary
prevention rather than initial treatment
within the first two weeks. Long term
secondary prevention is covered within
the ICSWP guideline.
It should be noted that as most patients
will be admitted in a non-fasting state,
measurements of both total and HDL-C
cholesterol are reasonably reliable in
this condition while non-fasting
triglycerides can have prognostic
significance[6]. If there is any doubt
about the validity of the measurement
then recourse can be made to
apolipoprotein A1 and B-100 which are
far les affected by prandial state and
may offer superior prediction to lipid
100
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
subfractions of atherosclerotic disease[8
9].
In addition the association between prothrombotic lipid sub-fractions and stroke
is not addressed though the details of
this may be beyond the scope of the
guideline. There is evidence that highly
elevated lipoprotein(a) levels are
associated with increased risk of stroke
and peripheral vascular disease[10 11
12]. It is routine clinical practice to
measure these lipid risk factors in
patients presenting with anticardiolipin
syndromes along with other
prothrombotic risk factors. There are no
specific intervention trials to address the
additional risk due to lipoprotein (a)
though some evidence from the
coronary heart disease suggests the
need to lower LDL-C levels to modern
targets[13].
SH
Heart UK
3 Full
8.6.2
We welcome the recommendation to
continue statin therapy in those patients
admitted acutely with stroke[14] and
agree with the recommendation that all
patients with ischaemic stroke as the
manifesting sign of their atherosclerosis
should be discharged on statin therapy
as recent meta-analysis supports the
benefits for both cardiovascular
disease[2 15] and stroke[16]. The data
for haemorrhagic stroke is more
controversial with an increase seen in
one trial using aggressive statin
therapy[17] though not on metaanalysis[16]. There seems to be little
specific role for other lipid lowering
101
Thank you
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
drugs in stroke[18 19].
SH
Heart UK
SH
Heart UK
4 Full
5
Full
8.6.6.3
The total cholesterol initiation target of
3.5 mmol/L differs from those suggested
in the hyperlipidaemia draft guideline for
initiation. No target for attainment is
given but the hyperlipidaemia guidelines
recommends TC = 4mmol/L; LDL-C =
2mmol/L; as does the diabetes draft
guideline. The lack of a LDL-C target is
disappointing as total cholesterol levels
may underestimate risk in patients with
low HDL-C (e.g. Indian Asians, Africans
and other ethnic groups at elevated
cardiovascular risk due to insulin
resistance/ metabolic syndrome) and
also be confounded by high HDL-C
especially in women. Efforts should be
made by NICE to standardise
recommendations across all
cardiovascular guidelines to aid internal
consistency of recommendations and
simplify clinical practice guidelines.
Thank you. We have taken out this
recommendation as secondary
prevention is not in the remit of our
scope. The developers have referred
readers to the NICE lipids guideline in the
clinical introduction
General
Reference List
Thank you for this information
1 Sever PS, Dahlof B, Poulter NR,
et al. Prevention of coronary and
stroke events with atorvastatin in
hypertensive patients who have
average or lower-than-average
cholesterol concentrations, in the
Anglo-Scandinavian Cardiac
Outcomes Trial--Lipid Lowering
Arm (ASCOT-LLA): a multicentre
randomised controlled trial. Lancet
2003 Apr 5;361(9364):1149-58.
2 Baigent C, Keech A, Kearney PM,
102
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
et al. Efficacy and safety of
cholesterol-lowering treatment:
prospective meta-analysis of data
from 90,056 participants in 14
randomised trials of statins.
Lancet 2005 Oct
8;366(9493):1267-78.
3 Boden-Albala B, Sacco RL, Lee
HS, et al. Metabolic syndrome and
ischemic stroke risk: Northern
Manhattan Study. Stroke 2008
Jan;39(1):30-5.
4 de SG, Devereux RB, Chinali M,
et al. Prognostic impact of
metabolic syndrome by different
definitions in a population with
high prevalence of obesity and
diabetes: the Strong Heart Study.
Diabetes Care 2007
Jul;30(7):1851-6.
5 Amarenco P, Labreuche J,
Touboul PJ. High-density
lipoprotein-cholesterol and risk of
stroke and carotid atherosclerosis:
A systematic review.
Atherosclerosis 2007 Oct 6.
6 Bansal S, Buring JE, Rifai N, et al.
Fasting compared with nonfasting
triglycerides and risk of
cardiovascular events in women.
JAMA 2007 Jul 18;298(3):309-16.
7 Amarenco P, Bogousslavsky J,
Callahan A, III, et al. High-dose
atorvastatin after stroke or
103
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
transient ischemic attack. N Engl J
Med 2006 Aug 10;355(6):549-59.
8 Walldius G, Jungner I, Holme I, et
al. High apolipoprotein B, low
apolipoprotein A-I, and
improvement in the prediction of
fatal myocardial infarction
(AMORIS study): a prospective
study. Lancet 2001 Dec
15;358(9298):2026-33.
9 Sniderman AD, Furberg CD,
Keech A, et al. Apolipoproteins
versus lipids as indices of
coronary risk and as targets for
statin treatment. Lancet 2003 Mar
1;361(9359):777-80.
10 Smolders B, Lemmens R, Thijs V.
Lipoprotein (a) and stroke: a metaanalysis of observational studies.
Stroke 2007 Jun;38(6):1959-66.
11 Chien KL, Hsu HC, Su TC, et al.
Lipoprotein(a) and Cardiovascular
Disease in Ethnic Chinese: The
Chin-Shan Community
Cardiovascular Cohort Study. Clin
Chem 2007 Dec 18.
12 Ariyo AA, Thach C, Tracy R. Lp(a)
lipoprotein, vascular disease, and
mortality in the elderly. N Engl J
Med 2003 Nov 27;349(22):210815.
13 Maher VM, Brown BG, Marcovina
SM, et al. Effects of lowering
104
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
elevated LDL cholesterol on the
cardiovascular risk of
lipoprotein(a). JAMA 1995 Dec
13;274(22):1771-4.
14 Blanco M, Nombela F,
Castellanos M, et al. Statin
treatment withdrawal in ischemic
stroke: a controlled randomized
study. Neurology 2007 Aug
28;69(9):904-10.
15 Kearney PM, Blackwell L, Collins
R, et al. Efficacy of cholesterollowering therapy in 18,686 people
with diabetes in 14 randomised
trials of statins: a meta-analysis.
Lancet 2008 Jan
12;371(9607):117-25.
16 O'Regan C, Wu P, Arora P, et al.
Statin therapy in stroke
prevention: a meta-analysis
involving 121,000 patients. Am J
Med 2008 Jan;121(1):24-33.
17 Goldstein LB, Amarenco P,
Szarek M, et al. Hemorrhagic
stroke in the Stroke Prevention by
Aggressive Reduction in
Cholesterol Levels study.
Neurology 2007 Dec 12.
18 Saha SA, Kizhakepunnur LG,
Bahekar A, et al. The role of
fibrates in the prevention of
cardiovascular disease--a pooled
meta-analysis of long-term
randomized placebo-controlled
105
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
clinical trials. Am Heart J 2007
Nov;154(5):943-53.
19 Birjmohun RS, Hutten BA,
Kastelein JJ, et al. Efficacy and
safety of high-density lipoprotein
cholesterol-increasing
compounds: a meta-analysis of
randomized controlled trials. J Am
Coll Cardiol 2005 Jan
18;45(2):185-97.
SH
Intercollegiate Stroke
Working Party
Full
General
1
Overall, we consider the guidelines
Thank you
provide a thorough review of the
relevant literature, which has been
appropriately interpreted. The
recommendations are comprehensive
and all have our full support. The text
clearly needs editing as several typos
are still present but we have ignored
these.
We have the following comments, where
we consider that additional points,
literature or recommendations should be
considered.
All the following comments refer to the
full version, but we have included the
relevant NICE version paragraph
numbering as well.
SH
Intercollegiate Stroke
Working Party
Full
2
1.1.1.3
This section needs to emphasise that
the WHO definition is outdated, given
that clinicians should not be waiting 24
hours to make a diagnosis of stroke and
diagnosis these days should be based
on imaging – either an abnormal CT or
MRI if CT normal.
106
Thank you. We have kept the WHO
definition for completeness but expanded
this section to state that anyone with
continuing neurological signs at the time
of assessment should be assumed to
have had a stroke, whatever the time
since onset. We do not feel that there is
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
Similarly, there are more modern
definitions of TIA that use imaging to
distinguish between true TIA and
transient symptoms with cerebral
infarction
evidence to support a definition of stroke
or TIA based on imaging.
The guideline could take a bold step and We agree that this might be helpful but
recommend revising the definition of TIA feel that redefining stroke and TIA is
and stroke in the UK – this would have
beyond the scope of the guideline.
the effect of emphasising the change in
culture underlying the National Stroke
Strategy and these guidelines.
It should be noted that brain attack is a
useful term to describe the presentation
as an emergency of stroke and TIA,
especially to lay people by analogy with
heart attack (but is not sufficient as a
diagnosis, since it includes stroke
mimics e.g. epilepsy).
The implication in the document that
Stroke and TIA have different pathways
e.g. the separate algorithms, risks
causing confusion to the public and
particularly the ambulance services
about the emergency management of
TIA. It is not necessary for the
ambulance services to try and
distinguish between TIA and stroke and
certainly not practical for them to
administer the ABCD2 score. We
therefore recommend that consideration
should be given to a section on the
emergency management of “brain
attack” in the community, which would
emphasise the need to treat the sudden
onset of neurological symptoms that
107
We agree that brain attack is a useful
term and have now referred to it in this
section
Thank you for your comment. The
pathways are separate because they look
at whether or not a patient has residual
symptoms. To help clarify this we have
amended the clinical introduction and
algorithms to outline the definition
between acute stroke and TIA to help
understand that these are separate
pathways.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
could turn out to be stroke or TIA as an
emergency by calling 999. This could be
incorporated into the TIA and Stroke
algorithms.
SH
Intercollegiate Stroke
Working Party
Full and 3.1.1.2
NICE
version
“Specialist assessment within 24 hours Thank you the section has been
of symptoms” for high risk TIA is not
amended accordingly
sufficient without access to urgent
investigations. This would be better
worded “Specialist assessment and
relevant investigations within 24 hours of
onset of symptoms”
Full
The recommendation that patients
should be “admitted directly to a
specialist acute stroke unit” would be
clearer if it specified “directly from the
community or Accident & Emergency
Department or Emergency Room”.
3
SH
Intercollegiate Stroke
Working Party
And
NICE
version
and
NICE
version
1.3.1.1.
3.1.1.3
and
7.1.7.1
This recommendation assumes that
patients will receive specialist
assessment from a doctor specialising in
stroke on the Stroke Unit and the time
scale is not specified. Moreover, in
reality many patients with stroke will not
be admitted directly to stroke units (even
if this becomes hospital policy), but
instead will stay on Acute Admissions
Units until a bed becomes available on
the unit. We would therefore recommend
adding an additional recommendation
stating: “All patients with suspected or
confirmed stroke should have specialist
assessment on arrival in hospital.”
4
SH
Intercollegiate Stroke
Working Party
5
Full
3.2
We have major reservations concerning
the TIA algorithms.:
108
Thank you the section has been
amended accordingly
We do not agree that this
recommendation should be included. We
have defined what an a stroke unit
includes and we feel that this
recommendation would allow some
people to opt out of directly admitting to a
stroke unit.
Thank you for your comment
Typ
e
Stakeholder
N
o
Developer’s Response.
Docum
ent
Section
number
Comments
and
NICE
version
1.1.1.1
Algorithm
1
1. A large proportion of TIAs will have
recovered by the time they are seen by
health professionals. The FAST and
ROSIER screens will then be negative
(and in any case they have not been
validated for TIA). If a suspected TIA
patient has a positive FAST or ROSIER
test, then they should be on the Stroke
algorithm, not the TIA pathway. Positive
screens should either lead to the Stroke
Algorithm, or they should be removed
from the TIA algorithm. Instead there
should be a single diamond to cover
diagnosis reading “Is history compatible
with TIA?”
2. All patients with a confirmed diagnosis
of TIA after specialised assessment
should have brain imaging, since the
pathology is always uncertain unless the
patient has previously been investigated
for the same symptoms. Those with high
ABCD2 scores definitely require imaging
to find out if they have actually had
infarction rather than ischaemia. (see
comment number 9 for further
comments re imaging after TIA).
Thank you we have amended the
algorithm for further clarification.
Thank you for your comment. This area
was debated extensively by the group
and the views you put forward were put
forward by some members of the group.
These comments have been reviewed by
the developers who disagree with your
suggestion. This is a consensus based
recommendation based on little evidence.
We have amended the FETR section for
further clarification.
3. The stenosis level should use the
NASCET criteria i.e. should be 50-99%,
since the evidence is based on the
combined analysis of the trials
performed by Peter Rothwell, in which
NASCET measurements were used.
More importantly, the criteria for
assessing stenosis on carotid imaging
(especially ultrasound) are based on
NASCET criteria, not ECST criteria. The
widespread assumption that we should The evidence discussed used NASCET
109
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
use the ECST criteria of 70-99% to
select patients for treatment, means that
a substantial proportion of patients who
have ultrasounds showing 50-69%
stenosis and who would benefit from
treatment, are not being considered for
treatment by physicians and surgeons
unfamiliar with the details of the trials
and ultrasound validation studies.
and ECST. The GDG considered your
comment but feel that their is no evidence
to state that one criteria should be
recommended over the other. We have
added in a recommendation for further
clarification that states that the criteria
used to determine stenosis level should
be reported.
4. The best medical treatment boxes
should be identical whether or not the
patient has carotid imaging.
Thank you this has been amended
5. The persons referred for carotid
endarterectomy should also have a link
to a best medical management box.
Thank you this has been amended
6. Stenting is an emerging alternative to
carotid endarterectomy and it would
therefore be appropriate to replace
“carotid endarterectomy” with “carotid
intervention”.
The GDG have found no evidence to
support the use of stenting in the acute
setting.The developers therefore do not
agree with your suggestion. For further
information please see section 6.4 of the
FULL guideline.
7. See also comment number 2 above re
incorporating brain attack into the
The term brain attack is refered to in the
algorithm.
introduction of both the FULL and the
NICE guideline. We do not wish to
incorporate this term into the algorithm as
the terms for stroke and TIA would also
need to be incorporated withi this to avoid
confusion.
110
Typ
e
Stakeholder
SH
Intercollegiate Stroke
Working Party
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
Full
3.2
In the Stroke Algorithm, it would be
better to replace the text “Surgical
Intervention” with “Consider Surgical
Intervention” since not all of the patients
are candidates for surgery.
Thank you, amended
Algorithm
2
6
Patients with cerebellar haematoma
See amended version
should be included on the list considered
for surgical intervention.
There is a typo in “stensosis” at the
bottom right hand corner.
SH
Intercollegiate Stroke
Working Party
7
SH
Intercollegiate Stroke
Working Party
SH
Intercollegiate Stroke
Working Party
Full
and
NICE
version
1.1.2.2
5.2.6.2
Full
5.2.6.3
and
8 NICE
version1
.1.2.3
Full
6.2.1.1
and
NICE
version
1,2
9
Specialist assessment is not sufficient
Thank you for your comment we have
without access to urgent investigations. amended the guideline accordingly.
One of the main barriers to timely
treatment of patients with TIA has been
delay in access to CT and carotid
imaging. This recommendation would be
better worded “Specialist assessment
and relevant investigations within 24
hours of symptoms”
Ditto, within 1 week
Please see the response above (‘Thank
you for your comment we have amended
the guideline accordingly.’)
We strongly disagree with the emphasis
in the statement in 6.2.1.1, that “not all
patients with TIA need brain scanning”.
A proportion of TIAs have subdural
haematomas, brain tumours, AVMs,
small haemorrhages and infarcts without
any symptoms other than an apparently
typical TIA. There is as much
justification for scanning all TIAs as
there is for scanning all strokes. Some
experts in our group felt strongly that the
Thank you for your comment. This area
was debated extensively by the group
and the views you put forward were put
forward by some members of the group.
These comments have been reviewed by
the developers who disagree with your
suggestion. This is a consensus based
recommendation based on little evidence.
111
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
6.2.6.2
and
6.2.6.3
Developer’s Response.
Comments
guidelines should not discriminate
against TIAs in this way and should
match the guideline that all strokes
should have brain imaging. We therefore
recommended having a guideline that
reads: “All TIA patients should have
brain imaging as part of their
assessment”. Not all experts in the
group supported this blanket
recommendation, pointing out that
patients with isolated retinal TIA and
some low risk TIAs may not require
imaging. A compromise
recommendation was suggested to read:
“All TIA patients should have brain
imaging considered as part of their
assessment”. All experts agreed that all
high risk TIAs require brain imaging to
exclude infarction. Hence, we suggest
The evidence and clinical consensus did
that as a minimum there should be a
not support this recommendation.
recommendation reading: “All high risk
TIA patients should have brain imaging
as part of their assessment”.
Similarly, the recommendations that only
patients with TIA “in whom vascular
territory or pathology is uncertain”
should have brain imaging implies that
the pathology is often certain. This is
misleading. Fortunately, only a small
proportion of patients have pathology
other than ischaemia on CT or MRI, and
hence those who do not scan TIAs
because they think they know the
pathology will not be wrong very often.
However, that does not mitigate the risk
of missing serious pathology. We would
therefore strongly advise rewording the
recommendations by deleting the phrase
112
The clinical consensus of the group was
that not all patients groups specified
within the two recommendations required
brain imaging. For clarification we
included a box to explain the population
of people who imaging may be helpful.
We did not feel that imaging should
replace clinical judgement and removing
the wording you suggest would be delay
access to imaging in those patients in
whom imaging may be required.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
“in whom vascular territory or pathology
is uncertain” from recommendations
6.2.6.2 and 6.2.6.3.
SH
Intercollegiate Stroke
Working Party
Full
And
NICE
version
1.2.4.1
10
6.4.4.1
This paragraph overstates the evidence
and misquotes the findings of reference
42 (see Table 2 in ref 42). The
comparison of patients undergoing CEA
less than one week since symptoms
only included 6 events from patients
operated early, and thus the confidence
intervals were very wide. Three out of 4
studies which compared treatment less
than 3 weeks after symptoms with those
treated more than 3 weeks after
treatment showed slightly worse
outcomes in those treated early. In those
treated either side of 4 weeks there was
no difference and it was only the 2
studies comparing less than 6 weeks
with more than 6 weeks that showed a
trend to better outcome with earlier
treatment. It should be noted that these
patients included TIAs and all had to be
neurologically stable. The paper (table
1) shows that patients with unstable
neurological symptoms did worse when
operated early.
We would therefore recommend that
paragraph 6.4.4.1 should be reworded to
read: “The systematic review reported
that there was no statistical difference
for the outcome of perioperative stroke
and death when comparing patients who
were neurological stable undergoing
CEA early (1-6 weeks) after symptoms
than those undergoing the procedure at
a later time. Patients operated early with
113
Thank you.We have amended the text
quoting the number of events for one
week vs. greater than one week and
included the ORs and CIs. We have
added a sentence to indicate which
patients did worse if operated on early.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
unstable neurological symptoms (stroke
in evolution, non-specified ‘urgent’
cases, and crescendo TIA) did worse if
they were operated in the acute phase
compared to later operation.
SH
Intercollegiate Stroke
Working Party
Full
6.4.4.2
It would be helpful to have the wording
Thank you. The wording has been made
in this paragraph clarified. The analysis clearer
of 5-year ARR refers to the delay from
symptoms to randomisation, but in
places seems to have been interpreted
as the delay between symptoms and
performance of CEA, which is not the
same and can not be calculated from the
trial data.
Full
6.4.5.1
We note the statement that “No
evidence for early carotid stenting
(within the two week period of the
guideline) was identified”. However, we
are aware of 2 publications addressing
this question. The first (Topakian et al,
Eur J Neurol 2007;14:672-678)
describes a case series of 77 patients
with symptomatic carotid stenosis
treated by stenting. Those treated less
than 2 weeks after symptoms had a
significantly higher 30 day rate of stroke
or death. The second (Groschel et al.,
Eur J Neurol 2008;15:2-5) reports a
much larger series in which treatment
less than 2 weeks from symptoms was
not associated with an increased rate of
complications compared to those treated
later.
11
SH
Intercollegiate Stroke
Working Party
12
Carotid stenting is increasingly being
used as an alternative to carotid
endarterectomy in patients with contra114
This paper was published outside of the
literature review cut-off date and is a case
series. We have clarified that the papers
included were restricted to RCTs within
the two week period of the guideline.
This is a case series; although the
numbers of patients are greater in this
paper than in the previous paper quoted,
the authors of the paper concede that
there are potential biases in the selection
of patients for inclusion; for example in
the earlier part of the study only patients
thought not to be suitable for cea were
recommended for stenting. Subsequently,
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
indications to surgery and in patients not
willing to undergo surgery. It also
continues to be tested in a randomised
comparison with carotid endarterectomy
in symptomatic patients in the
International Carotid Stenting Study
(ICSS). The data monitoring and
steering committees of the trial have
recently encouraged investigators in the
trial to randomise and treat stable
patients as soon as possible after
symptoms.
patients were offered a choice of CEA or
stenting which may have introduced
some bias. Whilst we recognise that on
occasion case series are helpful, the
GDG did not feel that the evidence fwas
reliable & hence on this occasion was
excluded from the guideline
NICE has previously issued a guideline
on carotid stenting (Interventional
procedure guidance 191, 2006). We
recommend that the NICE Acute Stroke
and TIA guideline should refer to this
earlier guideline and should include a
recommendation concerning carotid
stenting, matching the earlier NICE
Interventional procedure guidance on
the following lines: “Clinicians offering
carotid stenting as an alternative to
carotid endarterectomy should ensure
that patients understand the uncertainty
about safety and the long-term efficacy
of the procedure, and should preferably
include the patients in a randomised
clinical trial.”
SH
Intercollegiate Stroke
Working Party
Full
13
6.4.5.1
The IP guidance was not applicable to
the management of acute stroke and is
outside the scope of this guideline. The
GDG did not find any evidence to support
a recommendation on the use of stenting
outside the remit of this guideline. There
is evidence supporting the use of stents
in long term treatment of stroke. This
timeframe falls within the intercollegiate
stroke working party guideline
The sentence in this paragraph reading Thank you we have amended the section
“There is less benefit from early surgery accordingly.
in patients who are medically unfit” is not
115
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
derived from the evidence. It should be
reworded to read: “There is evidence
showing that patients with unstable
neurological symptoms (stroke in
evolution, non-specified ‘urgent’ cases,
and crescendo TIA) may be harmed by
early surgery.”
SH
Intercollegiate Stroke
Working Party
Full
6.4.6
In view of the above, both the first lines
of recommendations 6.4.6.1 and 6.4.6.2
should read: “People with stable
neurological symptoms from acute nondisabling stroke or TIA…”
Thank you we have amended the
wording of the recommendation
accordingly for greater clarification
Full
6.4.6
Considerable confusion emanates from
the fact that there are 2 methods of
measuring stenosis (see comment re
algorithm 1, above) and radiologists and
ultrasound technicians rarely state what
method has been used in their reports.
It would help to avoid some of this
confusion if NICE included a
recommendation that read: “Reports on
carotid imaging should state the method
used (ECST or NASCET) to calculated
the stenosis measurements.”
The evidence based looked at 2 methods
of measuring stenosis. The GDG did not
feel that there was evidence that one
method should be recommended over the
other.
The recommendation has been inserted
for greater clarification and to avoid
confusion.
6.4.6
The final bullet point of 6.4.6.2
concerning best medical management,
should be moved to a separate
recommendation on its own, since it
applies equally to the patients requiring
carotid endarterectomy.
The mention of aspirin and dipyridamole
in square brackets should be removed,
or clopidogrel added. There is some
evidence that the combination of aspirin
and clopidogrel is beneficial in short
term use prior to carotid endarterectomy
Thank you we have added the last bullet
of the recommendation as a third bullet of
the first recommendation. Please see
changes accordingly.
14
SH
Intercollegiate Stroke
Working Party
And
NICE
version
1.2.4.1
15 And
NICE
version
1.2.4.2
SH
Intercollegiate Stroke
Working Party
Full
16
116
We have removed reference to specific
anti-platelet agents
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
in patients with recent symptoms from a Thank you. This paper was not included
randomised trial known as CARESS and in the evidence review because it
NICE may wish to review and
compares aspirin and clopidogrel
incorporate this evidence into the
guideline (Circulation. 2005 May
3;111(17):2233-40)
SH
Intercollegiate Stroke
Working Party
Full
8.1.1.1
The last sentence of this paragraph is
muddled. Embolism is usually the result
of thrombosis. Thrombosis and
embolism from atherosclerosis is
associated with platelet rich thrombus,
and cardiac thrombo-embolism with
thrombin rich thrombus. We suggest
deleting the last sentence.
Thank you
Full
and
NICE
version
1.4.1.1
8.1.7.1
Although we understand that the GDG is
basing its recommendations on the IST
and CAST data, it seems to us that there
is no particular logic in delaying starting
dipyridamole until 2 weeks after onset,
or in continuing 150-300mg aspirin after
a loading dose, when we know from
other studies that 75mg is sufficient. The
one randomised trial of early
dipyridamole vs control showed no
additional harm from the combined
preparation in acute stroke and other
trials have shown that dipyridamole does
not cause cerebral haemorrhage. The
danger of delaying “definitive long-term
antithrombotic treatment until 2 weeks”
is that it will be forgotten. It is much
more effective to start long term
treatments on admission, than to plan to
start them later, given that in practice
such plans are often neglected. We
would suggest that a consensus
recommendation might be better
Thank you. We have added that patients
being discharged earlier can be started
on definitive secondary prevention (eg
aspirin and dipyridamole earlier
17
SH
Intercollegiate Stroke
Working Party
18
117
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
worded: “Patients should be given a
loading dose of aspirin 300mg;
thereafter they should receive a
combination of aspirin (50-75mg daily)
and dipyridamole MR 200mg b.d.”
SH
Intercollegiate Stroke
Working Party
Full
8.2.1.1
Full
and
NICE
version
1.4.2.1
8.2.6.1
Full
and
NICE
version
1.4.4.1
8.4.1
19
SH
Intercollegiate Stroke
Working Party
20
SH
Intercollegiate Stroke
Working Party
21
8.4.2
MR venography is not particularly
reliable as hinted at in this paragraph.
CT venography is superior and should
be mentioned here.
.
Both the randomised trials examining
the treatment of cerebral venous
thrombosis studied the use of heparin
(one with standard intravenous heparin
alone, and one a low molecular weight
heparin for 3 weeks followed by
warfarin.) The recommendation should
therefore read: “People ……should be
fully anticoagulated with heparin acutely,
followed by full-dose oral anticoagulation
(INR 2-3)….”
Thank you the guideline has been
amended accordingly
Thank you the guideline has been
amended accordingly
A distinction could be made between the Thank you for your comment. We did not
full blown antiphospholipid syndrome
review any evidence to enable us to
associated with lupus like disorders and make this distinction.
simply the association between
antiphospholipid antibodies and stroke.
Most experts would anticoagulate and
immunosuppress the former, but this
management is outside the scope of this
guideline. It would therefore be better in
paragraph 8.4.1 to say “The clinical
question remains as to whether patients
with acute stroke found to have
antiphospholipid antibodies without other
major features of the syndrome should
be anticoagulated…”
118
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
8.4.5.1
Developer’s Response.
Comments
The NICE researchers have missed the
pivotal study that addressed this
question, which was a nested cohort
study within a randomised comparison
Thank you. This paper does not address
of aspirin versus warfarin in stroke
the acute stroke population (within the
prevention conducted by the APASS
first 2 weeks) and hence was excluded
Investigators (JAMA 2004;291:576-584).
This showed that the risk of recurrent
stroke in those with antiphospholipid
antibodies was identical in patients
treated with aspirin or warfarin. The
authors concluded: “The presence of
aPL among patients with ischemic
stroke does not predict either increased
risk for subsequent vascular occlusive
events over 2 years or a differential
response to aspirin or warfarin therapy.
Routine screening for aPL in patients
with ischemic stroke does not appear
warranted.”
The above reference supports the
recommendation as being evidence
based and the foot note should therefore
be deleted.
SH
Intercollegiate Stroke
Working Party
Full
and
NICE
version
1.4.7.1
8.7.6.1.
The recommendation is that clotting
levels should be restored to normal as
soon as possible and the evidence
quoted shows that PCC acts much
quicker than other treatments. We
therefore consider that the
recommendation should add “Using
PCC”. The algorithm should also make
this recommendation and suggest a
treatment regime.
Thank you. We have amended the
guideline accordingly
Full
and
9.3.6.1
Concerning the exceptions to blood
pressure manipulation, we would add
Thank you. There was no evidence
reviewed for this suggestion. People with
22
SH
Intercollegiate Stroke
Working Party
23
119
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
NICE
version
1.5.3.1
Comments
Developer’s Response.
“hypotension e.g. systolic BP less than
90mm”.
hypotension would be managed
according to best medical practice.
We would add a consensus
recommendation on the following lines:
“When hypertension requires
manipulation in the context of stroke,
close monitoring of blood pressure and
the effect of any agent is essential (e.g.
on an ITU) to avoid over-rapid reduction
in blood pressure or hypotension.”
SH
Intercollegiate Stroke
Working Party
Intercollegiate Stroke
Working Party
25
Thank you the guideline has been
amended accordingly
Thank you. This has been amended
Full
13.1.4.3
There is an important error concerning
the benefit of early surgery in STITCH
with haematomas less than 1cm from
the surface, in line 2: “unfavourable”
should read “favourable”.
Full
and
Nice
version
1.9.1.3
and
1.9.1.4
13.1.6.3
The recommendation should include
Thank you the recommendation has been
cerebellar haemorrhage as well as lobar amended for further clarification
haemorrhage, so that it reads: “…have a
lobar haemorrhage or cerebellar
haematoma with hydrocephalus or are
deteriorating neurologically…”
24
SH
We did not find any evidence to support
this statement specifically in acute stroke
13.1.6.4
Patients who present in coma from
cerebellar haematoma often make an
excellent recovery after surgical
evacuation of the haematoma.
Recommendation 13.1.6.4 suggesting
that patients with a GCS of less than 8
rarely require surgical intervention
should therefore not apply to cerebellar
120
Thank you the guideline has been
amended accordingly
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
haematoma. We therefore recommend
changing the last bullet point to read: “a
GCS of less than 8, unless due to a
cerebellar haematoma”
SH
Johnson & Johnson Medical
1
NICE
1.2.2.2
SH
Johnson & Johnson Medical
2
NICE
1.2.2.4
Imaging of these patients for brain
imaging within 24 is clearly necessary,
but could put pressure on imaging
services. Would it be possible to
estimate what the extra demand per
100,000 of population might be, so that
under resourced areas can quickly
identify what they need to do to meet
this target?
Specification of ‘referred for carotid
endarterectomy’ appears to rule out
carotid stenting and should be revised to
‘be assessed and referred for carotid
endarterectomy or stenting within 1
week of onset of symptoms’.
Thank you. This is beyond the remit of
the clinical guideline.
See response to comment 61(‘Thank
you. We did not review any RCT
evidence for carotid stenting within the
‘acute’ two week period of the guideline
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
reason to modify the advice on the basis
of the available evidence.’)
SH
Johnson & Johnson Medical
3
NICE
1.3.2
Same question as per comment 1
above.(‘ Imaging of these patients for
brain imaging within 24 is clearly
necessary, but could put pressure on
imaging services. Would it be possible
to estimate what the extra demand per
100,000 of population might be, so that
under resourced areas can quickly
identify what they need to do to meet
this target?’)
121
See response to comment 61 (‘Thank
you. We did not review any RCT
evidence for carotid stenting within the
‘acute’ two week period of the guideline
Until the ICSS trial reports which is likely
to complete recruitment later in 2008 (and
CREST somewhat later), there is no
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
reason to modify the advice on the basis
of the available evidence.’)
SH
Johnson & Johnson Medical
4
NICE
General
SH
Joint Royal Colleges
Ambulance Liaison
Committee
1
Full
General
SH
Joint Royal Colleges
Ambulance Liaison
Committee
2
Full
1.1.2.1
SH
Joint Royal Colleges
Ambulance Liaison
Committee
3
Full
1.5.1.1
SH
Joint Royal Colleges
Ambulance Liaison
Committee
4 Full
3.1
SH
Joint Royal Colleges
Ambulance Liaison
Committee
5 Full
3.1
5.2.6
8.1.6.2
8.1.7.1
Other than these points, the guideline
looks solid and should be a step forward
for stroke care in England and Wales.
Overall the Joint Royal Colleges
Ambulance Liaison Committee
welcomes the proposed guidance on
Stroke: Diagnosis And Initial
Management of Acute Stroke and
Transient Ischaemic Attack (Tia).
While FAST has been studied and
validated for ambulance use, ABCD2
has not. We therefore question whether
there is sufficient evidence of safety of
this assessment by ambulance
personnel, particularly if used to
determine treatment with aspirin or
leaving a lower risk patient at home.
Agree that oxygen should be restricted
to use in those with hypoxia (and refer to
new the new British Thoracic Society
Guideline for emergency oxygen use in
adult patients).
JRCALC would like to emphasize the
need for urgent transfer particularly
when stroke is being upgraded in
ambulance despatch priorities.
The recommendations for the
administration of aspirin are unclear, see
below,
3.1 recommends that patients at high
risk ” should receive immediate initiation
of aspirin”, and 5.2.6 recommends
similarly “immediate initiation of aspirin”
for those both with high and low risk of
stroke. Then in 8.1.6.2”...aspirin should
be recommended” but “should be given
122
Thank you
Thank you
We presume you mean 5.2.6.1. We
assume that any patient with TIA referred
to ambulance service would be referred
for immediate specialist assessment. We
agree there is no evidence for the use of
this scoring system by ambulance
personnel
Thank you for your comment.
Thank you for your comment
Thank you we have amended the text to
clarify a dose of 300mg aspirin. Where
there are no continuing symptoms within
24 hours of onset a diagnosis of TIA
there is no need to delay aspirin for
imaging. Where there are continuing
symptoms, however soon after onset,
exclusion of haemorrhage by scanning is
required before aspirin administration.
This has been clarified in the text.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Joint Royal Colleges
Ambulance Liaison
Committee
6
Full
General
SH
Joint Royal Colleges
Ambulance Liaison
Committee
Lundbeck Ltd
7
Full
General
1
NICE
1.1
and 1.2
SH
Developer’s Response.
Comments
as soon as possible after haemorrhage
had been excluded…” Finally 8.1.7.1
says “All people presenting with acute
stroke who have had a diagnosis of
primary intracerebral haemorrhage
excluded by brain imaging should be
given aspirin (orally or enterally). TIA
and stroke are differentiated only by the
latter having symptoms that do not
resolve within 24 hours.
This may lead to confusion, as to
whether paramedics are advised to give
it here or not; for a good example, refer
to the NICE Acute Coronary Syndrome
guideline where the administration of
aspirin is precisely specified.
On specialist hyperacute stroke centres,
any recommendation would need to take
into account the likely impact on
ambulance costs (longer journeys and
repatriation) since to our knowledge
there has been no health economic
evaluation that had addressed this.
Ultimately the success of enacting the
intentions of this document will depend
critically on clear and explicit education.
We notice that compared to the classical
definition of TIA listed in guideline,
newer approaches to definitions on TIA
have been devised [e.g. "A brief episode
of neurological dysfunction caused by
focal brain or retinal ischemia with
clinical symptoms typically lasting less
than one hour and without evidence of
acute brain infarction"], i.e. specifically
paying less attention to time (durationresolving) of symptoms and more to a
faster imaging verification although to
our knowledge such potential new
123
This is beyond the remit of a clinical
guideline
Thank you
Thank you the definitions have been
expanded
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
definition has not been adopted by the
WHO. The concern is that with the
existing definition there is a risk that one
could wait up to 24 hours and if the
symptoms continue then this could be a
serious condition or a stroke. From our
point of view, whenever stroke
symptoms are recognised the patient
should be immediately transferred to
hospital and imaging performed first to
exclude bleeding but also to identify
ischemia and/or may even consider
imaging/angiography to detect
occlusion. Subsequent management can
then be guided by imaging results.
Clinical trials are ongoing to expand time
window of thrombolytics as guided by
brain imaging.
SH
Lundbeck Ltd
2
NICE
1.5.3.1
Thank you. The text has been amended.
1.5.3.1. Blood pressure manipulation in
people with acute stroke is not
recommended except where there is a
hypertensive emergency or any of the
following serious concomitant medical
issues:
 hypertensive encephalopathy
 hypertensive nephropathy
 hypertensive cardiac
failure/myocardial infarction
 aortic dissection
 pre-eclampsia/eclampsia
 intracerebral haemorrhage with
systolic blood pressure >200
mmHg.
We recommend that the guideline takes
into account management of blood
pressure within the context of
124
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
administering thrombolytic agents
(NINDS-study group) which requires that
blood pressure is lowered prior to
treatment to avoid augmenting possible
bleeding complications. One option
would be to allow (e.g. add as
exception) blood pressure manipulation
if the stroke patient is candidate for
thrombolytic therapy.
SH
Merck Sharp & Dohme Ltd
1
Full
8.6
As a general comment, we are
disappointed and concerned that section
Statin
8.6 focuses exclusively on statins as the
Treatment only recommended option for lowering
in People lipid levels in stroke patients.
with Acute
Stroke
Other agents apart from statins are
licensed and used in the UK for lowering
lipid levels, most notably ezetimibe
(EZETROL®, INEGY®).
In November 2007, NICE recommended
that ezetimibe be used for the treatment
of hypercholesterolaemia in the following
circumstances:

Ezetimibe monotherapy is
recommended as an option for the
treatment of adults with primary
(heterozygous-familial or nonfamilial) hypercholesterolaemia who
would otherwise be initiated on
statin therapy (as per NICE
guidance TA 94 in adults with nonfamilial hypercholesterolaemia) but
who are unable to do so because of
contraindications to initial statin
therapy.
125
This is beyond the scope of the guideline.
We have amended the clinical
introduction to cross refer to the NICE
lipid guideline. This guideline considered
the current NICErecommendation on
ezetimibe and made recommendations
accordingly.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments


Ezetimibe monotherapy is
recommended as an option for the
treatment of adults with primary
(heterozygous-familial or nonfamilial) hypercholesterolaemia who
are intolerant to statin therapy (as
defined in section 1.6).
Ezetimibe, coadministered with
initial statin therapy, is
recommended as an option for the
treatment of adults with primary
(heterozygous-familial or nonfamilial) hypercholesterolaemia who
have been initiated on statin therapy
(as per NICE guidance TA 94 in
adults with non-familial
hypercholesterolaemia) when:
o
o
serum total or low-density
lipoprotein (LDL) cholesterol
concentration is not
appropriately controlled (as
defined in section 1.5) either
after appropriate dose
titration of initial statin
therapy or because dose
titration
is limited by intolerance to
the initial statin therapy (as
defined in section 1.6)
and
o consideration is being given
to changing from initial statin
therapy to an alternative
statin.

When the decision has been made
to treat with ezetimibe
126
The developers agree and will cross
reference to the NICE lipid guideline. This
guideline cross referred to the ezetimibe
TA.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
coadministered with a statin,
ezetimibe should be prescribed on
the basis of lowest acquisition cost.

For the purposes of this guidance,
appropriate control of cholesterol
concentrations should be based on
individualised risk assessment in
accordance with national guidance
on the management of
cardiovascular disease for the
relevant populations.

For the purposes of this guidance,
intolerance to initial statin therapy
should be defined as the presence
of clinically significant adverse
effects from statin therapy that are
considered to represent an
unacceptable risk to the patient or
that may result in compliance with
therapy being compromised.
Adverse effects include evidence of
new-onset muscle pain (often
associated with levels of muscle
enzymes in the blood indicative of
muscle damage), significant
gastrointestinal disturbance or
alterations of liver function tests.
(NICE TA 132).
We would strongly recommend that this
Clinical Guideline specifically crossreference to the NICE TA 132, and
recommend the use of ezetimibe as an
alternative to statins in circumstances
where cholesterol concentrations are not
being appropriately controlled by statins
alone or where statins are
127
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
contraindicated, not tolerated, or are
otherwise inappropriate.
In addition, we believe that it would be
appropriate to delete the word “statins”
and substitute the term “lipid-lowering
agent” throughout the whole of section
8.6, except where the reference is
clearly intended to refer to the statin
class or a particular statin specifically for
example, when discussing a clinical trial
of a statin).
SH
Merck Sharp & Dohme Ltd
2
Full
8.6.6.2
See comment 1 above. We believe that
this recommendation should be
amended to read “People with acute
stroke who are already receiving one or
more lipid lowering agents should
continue with their treatment”.
We would also urge the Guideline
Development Group to consider
recommending a target total cholesterol
level of < 3.5 mmol/l in patients with
acute stroke who are already receiving
lipid-lowering therapy.
SH
Merck Sharp & Dohme Ltd
3
Full
8.6.6.3
The GDG reviewed statin evidence only
Thank you we have cross referred to the
lipid modification guideline within the
clinical introduction. The last statin
recommendation containing the total
cholesterol target has been removed as
this is outside the remit of our guideline.
See comment 1 above. We believe that Thank you we have cross referred to the
this recommendation should be
lipid modification guideline.
amended to read “People with acute
ischaemic stroke and a total cholesterol
of 3.5 mmol/l or greater should be
initiated on lipid-lowering therapy prior to
discharge.
See above; cross referenced to lipid
In particular, we would ask the Guideline guideline
Development group to note that it will be
difficult for many patients to reach a total
cholesterol target of 3.5 mmol/l or less
128
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Merck Sharp & Dohme Ltd
4
Full
8.6.6.3
SH
NCCHTA - 1
1
Full
General
SH
NCCHTA - 1
2
Full
General
SH
NCCHTA - 1
SH
NCCHTA - 1
3 Full
4
Full
4
General
Developer’s Response.
Comments
with statin therapy alone. It is therefore
imperative that clinicians are permitted
to use other lipid –lowering therapies
that have been proven to reduce total
and LDL-cholesterol, such as ezetimibe.
The draft guideline recommends
initiating statin therapy in stroke patients
with a total cholesterol level of 3.5mmol/l
or greater. This seems to be inconsistent
with current professional treatment
guidelines operating in the UK, such as
the Joint British Societies 2005
guidelines, and could lead to confusion
amongst the clinical community.
The guideline follows the intentions
specified in the scope document
In general the methods used are
consistent with the guidelines. In some
instances the effect size and
corresponding confidence interval are
not presented in the evidence.
The definition of statistical significance
given in the Glossary is incorrect and
should be replaced by the standard
definitions for p-value and significance
level
Throughout, the evidence based is
discussed in terms of presence or
absence of statistical significance. In
many areas the supporting evidence
comes from small studies, or studies
which address the occurrence of rare
events. In both of these situations the
statistical power will be limited. Hence,
it would be advisable also to interpret
the evidence in terms of the magnitude
of the effect (and its uncertainty,
represented by the 95% confidence
129
This recommendation has been removed
as this is outside the scope of our
guideline. We have referred to the NICE
lipids guideline within our clinical
introduction
Thank you
Thank you
Thank you we have amended the
guideline accordingly
Thank-you. .
We have amended the guideline so that
point estimates and confidence intervals
are cited in summary tables.
In the absence of a summary table PE
and CI are provided in the narrative text
when the outcome adds something
to the text and to make a particular point.
evidence tables are made publicly
available and contain all statistical
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
NCCHTA - 1
5 Full
5.2.4.20
SH
NCCHTA - 1
6 Full
5.2.4.23
SH
NCCHTA - 1
7 Full
5.2..4.24
SH
NCCHTA - 1
8
Full
General
SH
NCCHTA - 1
9
Full
10.2.4.9
SH
NCCHTA - 1
10 Full
11.1.4.1
SH
NCCHTA - 1
11 Full
13.2.4.4
SH
NCCHTA - 1
12 Full
Appendix
Comments
Developer’s Response.
interval).
outcomes'
The analysis presented here appears to
be data-driven and may not be
generalisable.
The count 11165 is incorrect. The
tabulation of the range of percentage
risk across validation groups is
uninformative because it will be
dominated by the smaller studies. A
more informative table would pool the
data from the validation groups.
Addition of confidence intervals to this
table would be informative.
Differences between groups are often
referred to as “statistical” differences,
rather than “significant” or “statistically
significant” differences.
Kappa measures should be compared to
the standard reference ranges
corresponding to poor; good; very good;
etc. rather than being described in terms
of statistical significance.
Although this evidence is based on a
study with key limitations, the significant
difference at 30 days followed by a nonsignificant difference at 6 months may
reflect a greater speed of recovery under
early mobilization, which would
presumably be considered beneficial.
Alternatively, the lack of significance at 6
months may be due to the substantial
loss to follow-up.
The non-significant subgroup analyses
described here will have had low
statistical power.
This assumes that the risk of stroke is
constant over the first week.
Noted thank you
130
The count has been amended. We have
added text indicating this to the reader
Unfortunately, none are reported in the
paper
The wording/phrasing has been amended
throughout the guideline
The wording has been changed
The text has been amended to reflect this
The text has been amended to reflect
Our stroke risk data is presented in Table
6. They show a higher risk for days 1
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
C
and 2 than for days 3 to 7.
, above
Table 9
SH
NCCHTA - 1
13 Full
Appendix
C
Table 10
SH
NCCHTA - 1
14 Full
Appendix
C
Sensitivity
analysis C
SH
NCCHTA - 1
15 Full
General
SH
NCCHTA - 1
16 Full
5.2.6
There are a small number of strokes and
so there will be substantial uncertainty
regarding the distribution of outcomes.
Additional evidence could be sought or
the values in table 10 could be varied in
a sensitivity analysis.
The health effects of the drugs
prescribed are not modelled. The
rationale for this should be explained.
Thank you. We are not aware of other
data on stroke outcome after an initial
TIA. We have now added a sensitivity
analysis.
We have added the following rationale to
the text:
In the base case analysis patients
undergoing specialist assessment are
assumed to be prescribed a number of
drugs and yet only the health effects of
aspirin and dypiridamole are modelled.
The model had a time horizon of 90 days
for key events (strokes); unlike aspirin
and dypiridamole, the other drugs are
unlikely to influence stroke rates in the
short-term. Given our time constraints
we were unable to model the longer term
health effects. However, for the key
comparison of immediate versus weekly
clinics, the health impact is not important
since both sets of patients will receive the
long-term benefit.
In general, the recommendations follow
logically from the findings described, and
are justified and complete.
The recommendation here is not
consistent with the findings from the
health economics analysis. The
rationale for this is described in 5.2.5.3,
and is related to limitations in service
131
Thank you
In general the recommendations should
follow the evidence. However, the GDG
felt obliged to consider logistics. While
the base case analysis does indicate that
immediate assessment is cost-effective
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
NCCHTA - 1
17 Full
7.2.6
SH
NCCHTA - 1
18 Full
8.2.5
SH
NCCHTA - 1
19 Full
9.2.6.1
Comments
Developer’s Response.
capacity. In general, should
recommendations follow the evidence
rather than what is considered
achievable in practice.
compared to weekly clinics across all
groups, it also shows that the higher the
ABCD2 score the more cost-effective it is.
The GDG were concerned that by
recommending immediate assessment
for all, the recommendation would not be
adequately implemented, and the
adverse consequences would be felt
mostly by those in the higher ABCD2
groups. Hence a prioritisation was felt
necessary at this time. By the time this
guideline comes to be updated we
envisage that immediate assessment will
be widely implemented for higher groups
and can then be extended to the lower
groups. The recommendation as it
stands is consistent with the National
Stroke Strategy.
These recommendations also do not go
as far as the evidence suggests
regarding immediate scanning for all
patients.
There is little evidence available for
intracranial and extracranial
haemorrhage, as no events were
observed for symptomatic ICH (8.2.4.4)
and the confidence interval for major
extracranial haemorrhage is very wide
(8.2.4.5).
The evidence presented does not
support the recommendation for
maintaining blood glucose between 4
and 11 mmol/L.
This area was debated by the group and
a consensus recommendation was
reached. Please see section 7.2.5
132
Thank you. The phrase “without a
significant increase in haemorrhage…”
has been deleted
Although the trial used a range of 4-7
mmol/L however this was one RCT which
primarily looked at the manipulation of
blood glucose rather than the range of
levels at which glucose should be
maintained. The GDG therefore reached
a consensus based on current clinical
practice.
Section
number
Comments
Developer’s Response.
20 Full
General
Thank you
21 Full
General
In general, limitations of the evidence
are clearly addressed.
Although this is a draft guideline, there
are a considerable number of
typographical errors, spelling mistakes
and incomplete sentences.
Occasionally the opposite meaning to
that intended is inserted, for example in
9.2.2.2 hypoglycaemia instead of
hyperglycaemia and in 13.2.2.2 Rankin
Scale of <=2 instead of >=2.
In general there is clear explanation of
how the recommendations are derived
from the evidence presented.
Further detail could be added to the
research recommendations given.
The data in the line of the table “Definite
TIA + new brain lesion” appear to be
incorrect.
The text in this paragraph is unclear –
perhaps including a table for the data
referred to would help.
It would be informative to describe how
the threshold of 4 on the ABCD score
was chosen.
Is the implication here that the ABCD
scoring would be performed by a nonspecialist?
The table referred to is missing from this
section.
The data given in these two sections are
inconsistent: in 2006, 91% of Trusts in
the UK had a stroke unit (7.1.1.1); 52%
of UK Trusts now have an acute stroke
unit (7.1.1.2). If the difference is due to
the defining characteristics listed in
7.1.1.2, it would be helpful if the same
definition were used throughout.
The comparison for death or
133
Typ
e
Stakeholder
N
o
SH
NCCHTA - 1
SH
NCCHTA - 1
Docum
ent
SH
NCCHTA - 1
22 Full
General
SH
NCCHTA - 1
23 Full
14
SH
NCCHTA - 1
24 Full
5.2.4.7
SH
NCCHTA - 1
25 Full
5.2.4.12
SH
NCCHTA - 1
26 Full
5.2.4.19
SH
NCCHTA - 1
27 Full
5.2.6.1
SH
NCCHTA – 1
28 Full
6.4.4.4
SH
NCCHTA – 1
29 Full
7.1.1.1
and
7.2.2.2
SH
NCCHTA – 1
30 Full
7.1.4.1
The guideline has been checked and
amended where necessary.
These have been amended
Thank you
Thank you we have amended the section
accordingly.
This wording is taken directly from the
paper and refers to patients with TIA who
have a new lesion on their MRI scan
The text has been amended
This is explain within the from evidence to
recommendation section
It could be performed by any
appropriately trained HCP
The table has now been inserted
91% of Trusts had a stroke unit of any
sort (would include rehabilitation units)
only 52% have an acute stroke unit as
defined in 7.1.1.2
Additional text has been inserted to
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
dependency is statistically significant for
acute (semi-intensive) units in the
current Cochrane review, although the
overall comparison is non-significant as
stated in this table.
For the types of stroke unit considered
here in section 7.1.4.1, the Cochrane
review found a shorter length of stay
(p=0.04, standardized mean difference 0.88 [-1.70, -0.06]).
“…The alternative cost-effectiveness
ratio of £496 per additional 1% of death /
institutionalisation avoided….”
A 40 year time horizon seems
inappropriate in the context of acute
stroke.
include the heterogeneity that was
reported
SH
NCCHTA – 1
31 Full
7.1.4.9
SH
NCCHTA – 1
32 Full
7.1.6.4
SH
NCCHTA – 1
33 Full
8.1.5.3
SH
NCCHTA – 1
34 Full
8.8.4.14
SH
NCCHTA – 1
35 Full
8.8.5.1
SH
NCCHTA – 1
36 Full
9.3.4.22
SH
NCCHTA – 1
37 Full
13.2.4.3
SH
NCCHTA - 2
13 Full
Appendix
C
Some aspects of the presentation could
be improved: it is stated in the costing
section that a probabilistic sensitivity
analysis was undertaken, in fact the
study rests solely on a deterministic
sensitivity analysis
Thank you. We have tried to improve the
presentation. We have taken out the
reference to probabilistic sensitivity
analysis.
SH
NCCHTA - 2
14 Full
Appendix
C
Table 21 crosses 2 pages
Thank you for your comment this will be
corrected.
This section is unclear. Is it the
difference in incidence of symptomatic
PE between anticoagulants and
antiplatelets that was non-significant?
“…per quality adjusted life year gained
of $8000.”
It would be helpful to define
“neurological changes” here.
The meaning of the qualifier for aphasia
(above or below 24 hours) is unclear.
134
The text has been amended
Thank you. We have changed this to
read £49,600 per death /
institutionalisation avoided.
Intervention after acute stroke can add
years of life to some patients. We believe
the time horizon may be appropriate,
although we acknowledge that it is more
speculative. Therefore we have
presented both the 40 year and the two
year time horizon’s for comparison.
This has been clarified in the text
Thank you. We have corrected this
sentence
This has been clarified in the text
This has been clarified in the text
Section
number
Comments
Developer’s Response.
15 Full
General
It is strange to see results just stated as
‘significant’ or non significant. It would
be much more insightful if the reader
where given the actual results e.g. odds
ratio (95% CI)
NCCHTA - 2
16 Full
6.2.6.2
There is little direct evidence to justify
the statement that patients with abcd
score of 4 or over should have brain
imaging
SH
NCCHTA - 2
17 Full
Appendix
C
I thought the team made a good job of
the cost-effectiveness model
Point estimates (PE) and confidence
intervals (CI) are provided for all
outcomes in the evidence tables. These
will be made available upon publication of
the guideline In addition within the
guideline PE and CI are cited in summary
tables. In the absence of a summary
table PE and CI are povided in the
narrative text when the outcome adds
something to the text and to make a
particular point. These may be primary or
secondary outcomes that were of
particular importance to the GDG when
discussing the recommendations. The
rationale for not citing all statistical
outcomes in the text is to try to provide a
'user friendly' readable guideline
balanced with statistical evidence where
this is thought to be of interest to the
reader.
This was a consensus recommendation
that patients in whom the pathology or
the vascular territory is uncertain should
undergo early DWI/FLAIR MRI
Thank you
SH
NCCHTA - 2
18 Full
General
Better use could be made of the careful
work done on the above model, by
summarizing its main findings in the
main document. The modeling approach
was only used to address one question it
could have been usefully applied
elsewhere. No justification was given as
to why a cost-effectiveness model was
not performed in issuing other
recommendations
SH
NCCHTA – 2
1
Appendix
C
In general I think the economic model
The GDG feel that the omission of stroke
presented here is an excellent attempt at mimics from the base case analysis was
Typ
e
Stakeholder
N
o
SH
NCCHTA - 2
SH
Docum
ent
Full
135
Thank you. We have added a description
of the model and its findings to the main
document. Time available to conduct
modelling was limited and the GDG
prioritised this question for detailed
economic analysis.
Typ
e
Stakeholder
N
o
Docum
ent
SH
NCCHTA – 2
2 Full
SH
NCCHTA – 2
3 Full
Section
number
Developer’s Response.
Comments
using decision-analysis to address a
complex problem. The main concern I
have is the way that the model does not
explicitly address the issue that ‘ a
proportion of cases’ considered to have
a TIA may in fact have other conditions
(TIA mimics). The impression given is
that this insight came rather last to the
economic modeller and so it is dealt with
in a rather crude way in sensitivity
analysis A by simply doubling the cost of
initial assessment in each strategy. I
would suggest that the base case
analysis should actually include a
realistic proportion of non TIA cases,
and the ensuing costs and
consequences of calling these cases in
for specialist assessment.
6.4
While sensitivity analysis A suggests
/ appendix that the ICER is not sensitive to this
C
assumption it is unclear whether 50%
TIA is realistic, if in fact its 25% or even
10% TIA then the cost-effectiveness of
immediate assessment may decline
considerably. This would seem an
important gap in the evidence that
should be better highlighted by the
economic model, to support the
statements made in the main text about
how clinics may be overwhelmed by
patients without TIAs turning up for
specialist assessment.
6.4
I am curious why for the GP strategy it
/ appendix was not considered feasible for GPs to
c
refer a proportion of cases for
immediate/weekly assessment? I.e. is it
not worth modeling a strategy
somewhere between GPs doing
everything vs specialists going
136
necessary since neither the costs nor the
health consequences are known for this
very heterogeneous group of patients.
We acknowledge that this is a limitation
of the model.
Rates will vary according to the referral
criteria of different centres. The TIA
mimic rate has been estimated to be 50%
using OXVASC data. Although a fairly
recent BMJ editorial suggested a TIA
mimic rate of 30%. We have added these
results to the text. Hence these data
support our main conclusions.
The main aim of the clinical question was
to assess which patients need immediate
assessment rather than who should get
referred. Even so, there seems little to
be gained from modelling such a
strategy, since the cost-effectiveness
would lie somewhere between the other
Typ
e
Stakeholder
N
o
Docum
ent
SH
NCCHTA – 2
4 Full
SH
NCCHTA – 2
5 Full
Section
number
Comments
Developer’s Response.
everything?
two strategies – exactly where would
depend on what proportion of patients
was being referred.
The GDG felt that there was no evidence
to indicate that the benefits would be less
in other centres.
6.4
I am a little concerned that the cost/ appendix effectiveness model may have given a
C
somewhat optimistic statement about
the value of specialist assessment. The
key parameters in the model, such as
reduction in risk from carotid
endartarectomy were based on those
observed in highly specialised units. If
this guideline is rolled out nationally, is it
plausible that such good risk reductions
will be observed in routine clinical
practice when surgery is performed in
less specialist centres etc? Might it not
be prudent to include a sensitivity
analysis to allow the relative risk
reductions from carotid endartarectomy
to fall below those reported in the most
specialist centres?
6.4
The lifetime costs of stroke used appear
/appendix very high. In particular the assumption of
C
£11,292 rehabilitation costs per year
does not appear plausible, and no basis
is provided for this.
SH
NCCHTA – 2
6 Full
6.4
/appendix
C
SH
NCCHTA – 2
7 Full
6.4
The costs are taken from an NHS HTA
report (Vol 10 #30). We believe that they
are reasonable.
The £11,292 was mis-labelled
rehabilitation costs but was in fact all
stroke care costs. This has been
corrected.
The first year cost of stroke (dependant The LOS was taken from an NHS HTA
health state) is based on a mean LOS of report (Vol 8 #1). It was based on 1854
51 days, no basis is provided for this,
first ever stroke patients from the ISD,
and again this appears high certainly as Scotland. We have added some
it should represent a national average.
additional description.
There is no mention of costs for patients
who die (within one year) following the
stroke.
The analysis is stated to take a Health
The costs are taken from an NHS HTA
and personal social services
report (Vol 10 #30). The report states
137
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
/appendix
C
SH
NCCHTA – 2
8 Full
SH
NCCHTA – 2
9 Full
perspective, which should exclude costs
to the patient, but it looks to me as if the
nursing home costs are included
whether they are borne by health, social
services or the patient. A significant
proportion of patients bear their own
nursing home costs and this should be
excluded under this perspective.
There is a slight discrepancy between
6.4
the economic model which suggests that
/Appendix
immediate specialist assessment in a
C
stroke unit is cost-effective for patients
scoring ABCD 2 and above, and the
recommendation in section 6.4 which
suggested that practicalities should also
be considered, including the proportion
of non TIA cases (see above). Hence
the overall guideline had a somewhat
more cautious recommendation that only
those with ABCD score of 3 or above
should have immediate specialist
assessment
8.8.5
The limited health economics evidence
favours warfarin over aspirin for
secondary prevention of stroke, this did
not feed through into the guideline
138
(p82) that the costs were estimated from
the health service perspective.
The GDG felt obliged to consider
logistics. While the base case analysis
does indicate that immediate assessment
is cost-effective compared to weekly
clinics across all groups, it also shows
that the higher the ABCD2 score the more
cost-effective it is. The GDG were
concerned that by recommending
immediate assessment for all, the
recommendation would not be
adequately implemented, and the
adverse consequences would be felt
mostly by those in the higher ABCD2
groups. Hence a prioritisation was felt
necessary at this time. By the time this
guideline comes to be updated we
envisage that immediate assessment will
be widely implemented for higher groups
and can then be extended to the lower
groups. The recommendation as it
stands is consistent with the National
Stroke Strategy.
The economic analysis did not take
account of the increase in haemorrhagic
stroke highlighted in the clinical evidence
statement. We believe that had this
consequence been incorporated in to the
analysis. Warfarin would no longer be
cost-effective compared with aspirin. We
have now noted this incongruity in the
text.
Docum
ent
Stakeholder
N
o
SH
NCCHTA – 2
10 Full
SH
NCCHTA – 2
11 Full
SH
NCCHTA – 2
12 Full
6.4
/appendix
C
The guideline/ cost-effectiveness model
assumes that the ABCD score can be
perfectly assessed/classified, this is
unlikely.
SH
NHS Direct
1
1.1
Disappointed at how small this section
on the rapid recognition of symptoms is
– think this needs more detail around
time frames for intervention etc and
most effective contact
i.e. states specialist care within 24 hours
– this is open to misinterpretation and
needs to also have time frames for initial
intervention i.e. within 3 hours? No
mention of initial interventions or
treatment required.
Who is the most appropriate first
contact? – 999? A/E? GP? Think it
needs to be clearer around this point as
from an NHS Direct perspective it does
NICE
Section
number
Developer’s Response.
Typ
e
Comments
7.1.7
There is little clinical or economic
evidence in favour of acute stroke units
rather than rehabilitation stroke units, it
is odd that such a strong
recommendation was given in light of
the scant evidence
The analysis does not consider whether
6.4
it is appropriate to take US evidence on
/ appendix
the rate of stroke after TIA and apply it
C
to the UK
139
The GDG consensus is that there are
many advantages to acute stroke units
including specialist care and more rapid
access to investigations and treatment.
The data pools together stroke data from
the USA and the UK. The overall stroke
rates were very similar for the UK and
USA. The pooling was necessary to give
greater precision to the estimates of
stroke risk for the individual ABCD2
groups. We have added this explanation
to the guideline text.
The ABCD2 score is relatively
straightforward to calculate – we do not
believe that there will be a great deal of
misclassification. The model does not
necessarily assume perfect accuracy – it
assumes that the classification in practice
will be as accurate as the classification
that took place in the studies of stroke
risk. We believe that this is reasonable,
given the simplicity of the tool.
See section 1.3.2.1 for time to CT scan;
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
not give clarity
SH
SH
NHS Direct
NHS Direct
2 Full
3 Full
NICE
version
5.1.1.1
3.1.1.2
5.2.2
5.2.4
5.2.5
5.2.6
6.2.5
6.2.6
Appendix
C
1.1.1
1.1.2
1.2.2
SH
NHS Direct
4
Full
Stroke
algorithm
SH
NHS Direct
5
Full
TIA
algorithm
SH
NHS Pathways
1
Full
general
States that an urgent clinical
assessment is needed – this needs to
be stated within the NICE version as
often all that is read – but also needs to
go further in relation to ‘what is urgent
care?’ Same day? Next working day?
Thank you we have changed urgent to
immediate in 5.1.1.1
Would have liked to see more on pre
hospital care
ABCD scores are mentioned at several
points within the guidance – there is no
clarity within the NICE version as to
what this scoring is
This is outside the scope.
Thank you. We have amended the NICE
guidance.
Within the full version although there is
some discussion around these scores it
is not readily clear especially to the
general health professional who does
not routinely use these score
mechanisms
Timeframes are mentioned within the
algorithm which should be incorporated
within the guideline
Use of the FAST test may get a negative
response due to resolution of symptoms
– should this not be changed to cover
this
It looks like a very well presented
documented that you couldn’t argue
with; together with clear guidance on
management of the different aspects. It
sets clear targets that will be difficult to
achieve for a lot of acute Trusts 365
days of the year.
140
Thank you. Please see the amended
algorithm for further clarification.
Thank you please see the revised version
of the algorithm
Thank you
Developer’s Response.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
SH
NHS Pathways
2
NICE
1.4.2
Our only comment is around the
Thank you we have expanded the
guidance making a clearer distinction
introduction to include the definition of
between TIA and stroke as the treatment Stroke and TIA.
is different, most notably in the
administration of aspirin. Presumably by
“people with a suspected TIA …” they
mean those who at the time of
presentation have no neurological
symptoms at any time up to 24 hours
after the event. Anyone with any
persisting symptoms within 24 hours
can’t be determined to have had a TIA
until that time interval has passed.
Anyone who presents with symptoms
within 24 hours from the onset of those
symptoms has to be viewed as having
had a stroke as it is impossible to tell
whether those symptoms will persist
beyond the 24 hours.
As far as the paramedic F2F is
concerned it may be a tricky judgement
for them to make re aspirin
administration if someone has called an
ambulance but made a rapid recovery
from a TIA [i.e. within minutes]. P7 of
the shorter guideline advocates the
FAST test to recognise a stroke, which
the crews are trained in. However it
then suggests the ABCD test to assess
the risk of stroke following a TIA before
commencing aspirin, which I don’t think
they will be trained in. In some ways
this probably makes our life easier in
that they will simply need transport them
all, including those who have made an
apparently rapid recovery from a
probable TIA.
141
We agree. We assume that any patient
with a positive FAST test assessed by a
paramedic will be taken to hospital for
further assessment. ABCD2 at the scene
could delay transfer to hospital
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
SH
North East London cardiac
and stroke network
1
Full
general
Thank you
SH
North East London cardiac
and stroke network
2
Full
general
SH
North East London cardiac
and stroke network
3
Full
general
The guideline is welcomed by the clinical
community and a publication date very
close to that of the National Stroke
Strategy may given it a higher profile
and brought into wider consultation than
may have been if it had been published
at a different time.
Comments form the clinicians within the
Network was that the guidance was
useful and easy to navigate
We note that there is no inclusion,
reference, recommendations or
comment on stroke rehabilitation in the
acute phase
SH
North East London cardiac
and stroke network
4
Full
general
SH
North East London cardiac
and stroke network
5
Full
Algorithm
Thank you
Thank you for your comment.
The main scope of the guideline was to
address the ‘initial and early management
aimed at reducing the ischaemic brain
damage and in the case of TIAs
preventing subsequent stokes’. We were
unfortunately unable to cover all areas
and focused upon those that people /
stakeholders initially suggested as critical
areas for address. Please refer to the
Royal College of Physicians
intercollegiate stroke working party
guideline which is currently being
updated and will be published at the
same time as this guideline.
We note there is no inclusion, reference, Please see the response above.
recommendations or comment on levels
of nursing and equipment required in the
acute phase of the stroke care pathway
It was felt by the therapists that the
following should be included in this
–
section of the algorithm – if the answer
Suspected is YES to ‘Is the patient able to take
Stroke
adequate nutrition and fluids orally’
Half way
another question should be:
142
We have amended the algorithm for
greater clarification
Typ
e
Stakeholder
N
o
Docum
ent
SH
North East London cardiac
and stroke network
6 Full
SH
North East London cardiac
and stroke network
7 Full
SH
North East London cardiac
and stroke network
8 Full
SH
Nutricia Ltd (UK)
1
NICE
Developer’s Response.
Section
number
Comments
down the
algorithm
on left
hand side
‘Is the
patient
able to
take
adequate
nutrition
and fluids
orally’
‘Is the patient able to feed themselves’ ?
If NO Refer to OT
If YES continue independently
7.1.7
It was felt that the guidelines need to
expand and emphasis the requirements
in an acute unit for example continuous
patient monitoring etc
8.1.7.1
Comment that further discussion should
be undertaken on the rationale for
continuing high does aspirin for 2 weeks
and that patients discharged before 2
weeks should be reduced to 75mgs on
discharge and not left until the out
patient clinic appt
9.3.6.1
This states that ‘Blood pressure
manipulation in people with acute stroke
is not recommended except where there
is a ‘hypertensive emergency’. It was felt
that defining a treatment threshold for
the BP would be more useful in clinical
practice particularly for junior staff who
may have difficulty in diagnosing for e.g
encephalopathy and for the nurses
monitoring a pts BP and escalating a
change
Key
Under ‘Hydration and nutrition’
priorities
Would suggest one additional bullet
for
point to highlight the importance of
implement identifying nutritional risk:
143
Thank you. we have added a box to
describe the components of an acute
stroke unit
We agree. The text has been amended
We agree. The text has been amended
This is part of good general care and not
specific to stroke
Typ
e
Stakeholder
N
o
Docum
ent
Developer’s Response.
Section
number
Comments
ation
-On admission, and at regular intervals,
screening for malnutrition should be
undertaken and appropriate nutrition
support given to those identified as at
risk
This section lacks a guideline on giving
nutrition support to those who are
identified as at risk of malnutrition
following screening. A NICE CG32
(Nutrition Support) recommendation that
could be included is:
‘Health care professionals should
consider oral nutrition support to
improve the nutritional intake of people
who can swallow safely and are
malnourished or at risk of malnutrition’
This would complement 1.6.1.4.
This statement should indicate that oral
supplementation is not recommended in
those who are adequately nourished and
have a safe swallow.
(This then distinguishes it from the many
patients with a CVA, which although
‘adequately nourished’ on admission,
may require nutrition support if they
have dysphagia/require tube feeding).
We agree with this statement but
suggest the addition of the following:
‘For those at risk of malnutrition, nutrition
support should be initiated, which may
include oral nutritional supplements,
referral to a dietitian for dietary advice
and/or tube feeding’
Thank you. This has been added
The GDG may like to consider an earlier
study by Gariballa et al 1998 undertaken
in malnourished patients with acute CVA
for which there has also been a health
Thank you
These papers were excluded in the
evidence review because they are not on
acute stroke patients
SH
Nutricia Ltd (UK)
2
NICE
1.6.1
SH
Nutricia Ltd (UK)
3
NICE
1.6.1.4
SH
Nutricia Ltd (UK)
4
Full
10.1.1.1
SH
Nutricia Ltd (UK)
5
Full
10.1.2
144
This refers to nutritional supplements not
nutritional support
Agreed, guideline amended and inserted
into the ‘from evidence to
recommendation’ section.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
economic evaluation in the BAPEN
Health Economic Report on Malnutrition
in the UK (p 119-121) by Elia et al 2005.
SH
Nutricia Ltd (UK)
6
Full
10.1.2.2
SH
Nutricia Ltd (UK)
7
Full
10.1.4.2
SH
Nutricia Ltd (UK)
8
Full
10.1.5.1
SH
Nutricia Ltd (UK)
9
Full
10.1.6.4
now
10.3.6.
The guideline needs to reiterate that the
large study of hospital diet and
supplementation by Dennis et al 2006
was in ‘well nourished’ patients.
The reference quoted here is incorrect
(No. 154; Norton et al 1996) as this is
not a study of oral food supplementation
but of tube feeding in those with no oral
intake.
Following on from point 6, suggest that
you check this is the correct study here.
There is evidence from a number of
systematic reviews about the benefits of
nutritional supplementation in acutely ill
elderly patients and references here
could include:
Milne et al 2006 Ann Intern Med 144,
37-48
NICE 2006, Clinical Guideline 32
Stratton & Elia 2007 Clin Nutr Vol 2
Supp 1. p 5-23
(nb. The reference provided, Potter et al
is incorrect)
This guideline provides no
recommendation about what to do when
someone is at risk of malnutrition after
screening. The existing wording also
needs modifying as highlighted in
comment 3 above.
We suggest the following alternative
wording:
‘Nutrition support, including oral
nutritional supplementation, should be
considered for those identified as at risk
of malnutrition with a safe swallow.
145
This has been clarified
Thank-you. We wrongly quoted the
Norton study. The correct study is one by
Garbrialla but which was excluded due to
methodological limitations. The text has
been amended
Thank you
See above
These papers were not included in the
evidence review because they are not on
acute stroke patients
Thank you we have amended the
guideline
Thank you see recommendation10.3.6.6
for further clarification.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Nutricia Ltd (UK)
10 Full
10.2.6.3
SH
Nutricia Ltd (UK)
11 NICE
Algorithm
for CVA
Developer’s Response.
Comments
Routine oral supplementation is not
recommended for people with acute
stroke who are adequately nourished
and can swallow safely on admission’
We suggest adding ‘and referral to a
dietitian’ at the end of this
recommendation (as specialis dietetic
input is required for those requiring
dietary modification or tube feeding.
There is no ‘yes/no’ option following
screening for nutritional
supplementation to indicate that those at
risk of malnutrition should receive
nutrition support.
It is outside the remit of the clinical
guidelines to specify professions.
Thank you please see the amended
algorithm for further clarification.
If there is a no option, would suggest
changing the wording in the current box
to Routine oral supplementation is not
recommended for people with acute
stroke who are adequately nourished
and can swallow safely on admission’
SH
Primary Care Neurology
Society
1 Full
3.2.1.1
Referral to a dietitian should be added to
the ‘initiate NG tube feeding’ box
(following if patient is unable to take
adequate nutrition and fluids orally)
The algorithm assumes that a negative
screen using FAST excludes a diagnosis
of Stroke or TIA. FAST is described as
‘a screening diagnostic tool’ and is not
sensitive to identify all presentations of
stroke, e.g. dysphasia, posterior
circulation symptoms. The algorithm
attributes the function of a diagnostic
tool to FAST. The recommendation is
that the box from the ‘negative screen’
arrow is amended to ‘requires further
neurological assessment’. A further
arrow from that box to the box
146
It is outside the remit to specify named
professions
Thank you please see the amended
algorithm
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Primary Care Neurology
Society
2 Full
3.2.1.1
SH
Primary Care Neurology
Society
3 Full
5.2.6.2
Developer’s Response.
Comments
containing ‘start aspirin etc…’ to direct
those who are diagnosed with a stroke
or TIA (or brain attack) following ‘further
neurological assessment’.
The algorithm for suspected TIA refers
Thank you we have amended the
to lifestyle management in the box which algorithm for further clarification
advises ‘start aspirin 150-300mg’.
The box at the final level of the algorithm
describing ‘best medical treatment’
contains the advice ‘cholesterol lowering
through diet and drugs and smoking
cessation’. Effective secondary
prevention is a combination of medical
treatment and lifestyle measures.
Lifestyle measures include: smoking
cessation, exercise, alcohol
consumption within recommended limits
and dietary measures. The aim of
dietary changes are to lower cholesterol,
weight control or weight reduction
depending on BMI and reduction in
further vascular events The main
recommendations are: salt restriction
especially if hypertensive, five a day fruit
and vegetables, reducing meat and
increasing oily fish intake, reducing or
replacing saturated fats with unsaturated
fats. We would recommend that ‘best
medical treatment’ is replaced by
‘secondary prevention measures’ and
that secondary prevention is a
combination of both lifestyle and medical
measures, a more comprehensive list of
secondary prevention lifestyle measures
is included and that the components of
secondary prevention are listed
consistently in the algorithm and
These recommendations do not include Thank you see amended section 5.2.5.5
the time scale for specialist assessment
147
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
Primary Care Neurology
Society
4 Full
5.2.6.3
SH
Primary Care Neurology
Society
5 Full
5.2.6.2
SH
Primary Care Neurology
Society
6 Full
6.2.6.2
SH
Primary Care Neurology
Society
7 Full
6.2.6.3
SH
Primary Care Neurology
Society
8 Full
6.3.7.1
Developer’s Response.
Comments
for those who have had a suspected TIA
who are at high risk of stroke (ABCD2
≥4 who present more than 24 hours after
the onset of symptoms. The
recommended timescale may vary
depending on the time interval between
onset of symptoms and presentation, for
example 7, 30 or 90 days after the onset
of symptoms.
These recommendations do not include
the time scale for specialist assessment
for those who have had a suspected TIA
who are at low risk of stroke (ABCD2
<4) who present more than 24 hours
after the onset of symptoms. The
recommended timescale may vary
depending on the time interval between
onset of symptoms and presentation, for
example 7, 30 or 90 days after the onset
of symptoms.
There is no dose of aspirin given in this
recommendation. The algorithm states
starting aspirin 150-300mg aspirin. We
would request that this does of aspirin is
included in this recommendation.
These recommendations refer to
investigations within 24 hours of onset of
symptoms. Further recommendations
on investigations and timescales for
those presenting at a later time are
required.
Further recommendations are required
for those who present more than one
week after the onset of symptoms.
See response above (Response to
comment 304 ‘Thank you see amended
section 5.2.5.5’)
Thank you. Please see the amended
section.
Where patients present after more than
one week after their last symptom they
should be treated using the low risk
pathway
Please see the response above
(Response to comment 307.’ Where
patients present after more than one
week after their last symptom they should
be treated using the low risk pathway’
Further recommendation for timing of
Please see the response in comment 307
carotid imaging is required for those who (‘Where patients present after more than
148
Typ
e
SH
Stakeholder
N
o
Princess Alexandra Hospital
Docum
ent
Full
1
SH
Princess Alexandra Hospital
Full
1
SH
Princess Alexandra Hospital
Full
2
SH
Princess Alexandra Hospital
Full
3
SH
Princess Alexandra Hospital
4
Full
Section
number
Comments
Developer’s Response.
present more than one week after the
onset of symptoms.
one week after their last symptom they
should be treated using the low risk
pathway)
TIA
algorithm
1
IN the wake of the rapid response
strategy to strokes, TIAs can only be a
retrospective diagnosis (ie only if
symptoms resolve rapidly). Therefore,
one cannot suspect a TIA(start of the
algorithm) and if one does in spite, will
end up not treating a stroke! The usage
of term is suspected TIA is confusing.
Excuse us if we sound pedantic but the
guideline should make it clear that
suspected TIA means resolved focal
neurological symptoms due to
suspected TIA.
TIA
IN the wake of the rapid response
algorithm1 strategy to strokes, TIAs can only be a
retrospective diagnosis (ie only if
symptoms resolve rapidly). Therefore,
one cannot suspect a TIA(start of the
algorithm) and if one does in spite, will
end up not treating a stroke! The usage
of term is suspected TIA is confusing.
Excuse us if we sound pedantic but the
guideline should make it clear that
suspected TIA means resolved focal
neurological symptoms due to
suspected TIA.
TIa
Sorry for being pedantic. I am not sure
pathway
how one can suspect a TIA and then
algorithm perform a FAST test and ROSIER
TIA
Positive screen should be worded –
algorithm positive history. Anybody with a positive
screen at the time of assessment should
be in the stroke pathway
TIA
ABCD2 score should come immediately
algorithm after FAST
149
Please see comment 424 (‘Thank you we
have clarified the algorithm and
definition.’)
Thank you we have clarified the algorithm
and definition.
Thank you we agree that these tools are
unvalidated in TIA and have removed
them
Thank you we agree please see the
amended algorithm
Thank you please see the amended
algorithm.
Typ
e
Stakeholder
SH
Princess Alexandra Hospital
N
o
Docum
ent
Section
number
Full
,,
,,TIA
Algorithm
Regarding the comment to ‘consider
blood pressure management’ it may be
dangerous to treat high blood pressure
when a severe carotid stenosis has not
been excluded
TIA
Carotid endarterectomy should be done
Algorithm,, ASAP before 2 weeks in high risk TIA
,,
group if they have an appropriate degree
of stenosis
TIA
Aspirin should be the only intervention
Algorithm,, prior to the referral of the patient – or
,,
else the GP who is greatly
disadvantaged without any
investigations to back up, would have to
commit too much into the diagnosis of a
TIA when alternatives may remain to be
excluded. There is no evidence of any
other intervention preventing recurrence
of TIA or occurrence of a stroke
5.2.6.1
There seems to be selective bias
towards presenting data which had
showed its utility. There is some
evidence that it is not useful in certain
populations.
Thank you. Theoretically that is correct
however it has not been shown in
practice
Full
6.2.6.2
and
6.2.6.3
The GDG felt that haemorrhage is an
uncommon cause of TIA likewise there
no evidence to show antiplatelet
treatment is harmful.
Full
6.3.7.1
Full
General
on TIA
5
SH
Princess Alexandra Hospital
Full
6
SH
Princess Alexandra Hospital
Full
7
SH
Princess Alexandra Hospital
Full
8
SH
Princess Alexandra Hospital
9
SH
Princess Alexandra Hospital
10
SH
Princess Alexandra Hospital
11
Developer’s Response.
Comments
There are no recommendations from the
guideline to ascertain whether a
suspected TIA is due to bleeding or
infarction on clinical grounds. If there is
no evidence to do so confidently, every
patient should have imaging
We feel that the best strategy would be
to use MRI and MRA in high risk
patients(ABCD2 >4) within 24 hours and
do the same within a week for low risk
patients. Negative predictive value of
MRA is excellent.
TIA with Atrial fibrillation receives no
priority as do TIAs in people who are
150
Thank you. The developers think the
recommendation is clear and that the
surgery should take place within 2 weeks
Thank you. It is reasonable that if a GP
sees patients initially that they can
consider simple interventions such as
statins prior to referral
We are aware of only one publication
showing that it is not useful in pats
assessed at a median of 12 days post
event however these were not patients
who presented acutely
Thank you. There is no evidence for this.
These have been addressed in the
guideline
Typ
e
SH
Stakeholder
N
o
Princess Alexandra Hospital
Docum
ent
Section
number
Full
6.2.6
Full
7.1.7
Full
7.2.6.2
Full
10.3.6
12
SH
Princess Alexandra Hospital
13
SH
Princess Alexandra Hospital
14
SH
Princess Alexandra Hospital
15
Developer’s Response.
Comments
already taking warfarin but INR
subtherapeutic like individuals with
prosthetic metallic valves, those with
recent myocardial infarction(mural
thrombus) and also those with
suspected arterial dissection.
It is very unlikely that doing dwMRI brain
within 24 hours is going to alter the
course of a patient with what clinically
appears to be a carotid territory TIA. The
feasibility of this in a DGH is to be borne
in mind. One could possibly get dwMRs
on a weekly basis in DGHs. MRI Brain
and MRA would be a good alternative
We would sincerely welcome this and it
would be the best thing to happen for
stroke victims
There is no need to give this 24 hour
allowance. CTs should happen as soon
as people present with a stroke similar
to the head injury guidelines
Please remember that there is a
radiologically introduced feeding tube
which is very useful in frail old patients
who put up a great fight but are not well
enough to undergo a procedure as
invasive as a PEG. The specific
subgroup of patients who do benefit
from RIGs(Radiologically inserted
gastrostomy-where they use more or
less the same balloon tipped tube as a
PEG) are
the frail elderly with multiple active
ongoing comorbidities like chest
infection, cardiac failure, recent
myocardial infarction, acute strokes who
repeatedly pull NG tubes(when it is well
known that inserting PEGs too early
following a stroke is associated with
151
If there is a clear history from which the
arterial territory can be determined there
is no need for MR
Thank you
This has been discussed in detail by the
GDG. Please see section 7.2.5
Thank you we have amended the term to
gastrostomy and this includes RIGS. A
literature search was undertaken and no
evidence was identified.
Thank you for this comment. We
considered these papers but they are not
specifically applicable to stroke. The
developers do not feel that this
information should be extraplolated and
think that the type of gastrostomy used
will depend on the condition of the patient
and local expertise.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
increased mortality and a worse
outcome(FOOD trial, NCEPOD ‘Scoping
our practice – chapter 9 and there are
other randomised clinical trials which
suggest that PEGs should be delayed by
4-6 weeks and if done earlier may cause
more harm) Clinical Nutrition, Volume
23, Issue 3, Pages 341-346
Kindly review the comment that states
‘PEGs should be the procedure of
choice’.It is probably better replace
‘PEGs’ by the term’gastrostomy’ as in
the adult nutrition NICE guidelines. This
would allow stroke physicians who have
access to radiological gastrostomies to
treat a wider range of patients in a timely
and efficeient way and this could also
lead to a better functional outcome for
many(The long PEG waits don’t help
either)
Sorry to be stating the obvious that
radiological gastrostomies can be done
under local anaesthesia with no need for
intubation or sedation. In our hospital at
Harlow, we have successfully treated a
considerable number of frail elderly
patients with radiological gastrostomies
and had very good outcomes where
PEGs were thought to be too risky
Wollman et al 1995Radiology, Vol 197,
699-704, Copyright © 1995 by
Radiological Society of North America
have done a metaanalysis where they
showed radiological gastrostomies to be
more successful and had less
complications than PEGs(I am aware
that it was a few years ago but there are
documents in the literature which
suggest that the 30 day mortality from
152
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
PEGs is increasing Journal of
Postgraduate Medicine, 2005 (Vol.
51) (No. 1) 23-28
Leontiadis GI, Moschos J, Cowper T,
Kadis S. Mortality of percutaneous
endoscopic gastrostomy in the UK. J
Postgrad Med 2005;51:152
SH
Royal College of Nursing
SH
Royal College of Nursing
SH
Royal College of Physicians
1
2
1
Full
General
NICE
1.1.2
Full
General
With a membership of over 400,000
Thank you
registered nurses, midwives, health
visitors, nursing students, health care
assistants and nurse cadets, the Royal
College of Nursing (RCN) is the voice of
nursing across the UK and the largest
professional union of nursing staff in the
world. RCN members work in a variety
of hospital and community settings in the
NHS and the independent sector. The
RCN promotes patient and nursing
interests on a wide range of issues by
working closely with the Government,
the UK parliaments and other national
and European political institutions, trade
unions, professional bodies and
voluntary organisations.
The RCN welcomes the opportunity to
review this draft guideline. The
document is comprehensive and we
support it.
The developers need to clarify their
Agree. We assume that any patient with a
recommendations on the use of ABCD2 positive FAST picked up by paramedic
score in risk-assessing patients with TIA. will be transferred to hospital, we agree
ABCD2 is not validated for use outside
This tool has not been validated for
hospital
ambulance use, so needs more
research.
We consider this to be a very detailed
Thank you
careful, analytical and high quality
153
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
document. The team involved are to be
congratulated.
We have some concerns that aiming at
excellence here could prevent high
quality by asking for specialist
assessment in places where it does not
exist. This could have an unintended
knock on for some staff who may default
to "not trained to do this" with the
consequence that the patient receives a
very raw deal. An example from one
Trust, is that the MAU nurses are not
allowed to do swallowing assessments,
SALT teams will not visit the rapid
throughput medical ward, patients with
stroke may be on nil by mouth for no
good reason all weekend or bank
holiday if they cannot get onto the acute
stroke unit.
SH
Royal College of Physicians
Full
1.1.1.1
Full
Glossary
Full
P14
General
5 Full
5.2.5.3.
2
SH
Royal College of Physicians
3
SH
Royal College of Physicians
4
SH
Royal College of Physicians
We wish to endorse the comments of
the British Geriatrics Society, the
Association of British Neurologists and
the British Society of Rehabilitation
Medicine (submitted separately)
Introduction title should say that stroke is
preventable and treatable and not just
treatable. Arguably, prevention is more
important and has a better evidence
base than treatment.
The glossary should include the ABCD
and ABCD2 scores.
Scoring in the ABCD section has an
error. It should be systolic blood
pressure under 140 and diastolic under
90 it says diastolic over 90.
This recommends immediate specialist
154
We agree and we hope that trusts will
use these guidelines to ensure that the
appropriate training is in place. It is
perfectly possible for MAU nurses to be
trained to do swallow assessments
Thank you the text has been altered
accordingly.
Thank you the text has been amended
accordingly.
Thank you. This has been amended
Service provision is beyond the scope of
Typ
e
SH
Stakeholder
N
o
Royal College of Physicians
Docum
ent
Section
number
Full
General
Full
10.1.6
Full
General
6
SH
Royal College of Physicians
7
SH
Royal College of Physicians
8
Comments
Developer’s Response.
assessment. However, the reality is that
weekends and bank holidays have no
specialist staff present, no carotid
imaging available, no speech and
language therapists and usually no
consultant in stroke care in most
hospitals in the country. Furthermore,
there is no specialist radiologist for MRI
in many places even if they were done
out of hours which they are not because
there is a lack of radiographers to
operate this equipment out of hours.
The guideline group must address the
strategy of stroke assessment out of
hours which is the majority of time for
most Trusts and General Practice. How
do we ensure that we can deliver best
care in bad circumstances? How do we
train non stroke staff in high quality
stroke intervention? This is a very
difficult issue but it could do with some
thought.
The nutrition guidance needs to be more
specific to contribute to routine care.
Those with an abnormal ‘MUST’ score
should be referred to a Nutrition Support
Team.
As NICE guidelines also have a
patient/lay version, we would suggest
that a paragraph should be included
listing other aspects of a patient's
condition in the first 48 hours which
might be noticed by a carer and give rise
to anxiety e.g. personality change. The
paragraph could then indicate that these
are looked out for and will be addressed
as appropriate, but the immediate
concern is to stabilise the patient's
medical condition.
the guideline. The national stroke
strategy details how clinical networks can
resolve these issues.
155
Service provision is beyond the scope of
the guideline. The national stroke
strategy details how clinical networks can
resolve these issues
Thank you we have added a note on
nutritional support
Thank you for your comment.
We will liaise with the patient and public
involvement unit (PPIU) to see if this can
be inserted into the understanding NICE
guidance version.
Docum
ent
Section
number
Comments
Developer’s Response.
Royal College of Radiologists 1
Full
General
Thank you
Royal College of Speech and 1
Language Therapists
Full
General
The RCR wrote in to endorse the
comments of the British Society of
Interventional Radiology and the British
Society of Neuroradiologists.
The RCSLT was very disappointed that
there is a lack of mention of
communication disability within the
document.
Typ
e
Stakeholder
SH
SH
N
o
This is outside the scope of the guideline.
This will be addressed in ICSWP
guideline.
Over 30% of people who have a stroke
have communication difficulties.
Every stroke survivor must receive a
“communication lifeline” to help them
communicate their needs in the hours
following their stroke.
SH
SH
Royal College of Speech and 2
Language Therapists
Royal College of Speech and 3
Language Therapists
Full
General
Full
General
Communication problems must be
acknowledged within the guidance.
There is no mention of multi disciplinary
working
It is disappointing that in a document
which is designed to set high standards
of care for our clients, that only a
medical and physical approach is being
mentioned.
See section 1.1.1.2
Thank you for your comment.
The main scope of the guideline was to
address the ‘initial and early management
aimed at reducing the ischaemic brain
damage and in the case of TIAs
The emphasis is too strongly focussed
preventing subsequent stokes’. We were
on:
unfortunately unable to cover all areas
- Medical issues
and focused upon those that people /
- The medical model of care
stakeholders initially suggested as critical
- Pharmacologically
areas for address. Please refer to the
Royal College of Physicians
This is really surprising in an area where intercollegiate stroke working party
AHP intervention is so high.
guideline which is currently being
updated and will be published at the
The document is biased towards
same time as this guideline.
assessment with no emphasis on
techniques aimed at encouraging
156
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
improvement, which in the area of stroke
(rehab) seems an odd focus.
SH
Royal College of Speech and 4
Language Therapists
Full
General
SH
Royal College of Speech and 5
Language Therapists
Full
General
SH
Royal College of Speech and 6
Language Therapists
Full
General
In practice stroke care is shifting to a
client-centred approach.
The audience of the document is people
who have had a stroke. Therefore an
easy access version of the guideline
needs to be produced for patients and
carers.
Thank you for your comment. An easy
access version of the guideline is
produced for patients and carers on
behalf of NICE. It is not available for
consultation but will be available on
publication. Section 2.8.1.18 refers to this
NB Section 2.8.1.18 does not include an section
easy access version
Involvement of people with stroke and
Thank you for your comment. The
TIA: The draft states that only 2 people developers disagree. Our patient and
were involved. It is questionable how
carer representative were chosen
this can represent wider user views.
because of their direct experience of
acute stroke and TIA and their knowledge
of patient/carer issues.
The importance of intervention from all
Thank you.
the therapies appears to be very minimal
and there is no mention of the impact of We agree that the communication /
communication or psychological
psychological difficulties and support of
difficulties after stroke.
people with chronic conditions is
important however the main scope of the
The role of the Speech and Language
guideline was to address the ‘initial and
Therapist with communication or of the early management aimed at reducing the
Psychologist for support for client and
ischaemic brain damage and in the case
family are totally missing.
of TIAs preventing subsequent stokes’.
The impact of no input for these two
areas has been researched and there is
evidence to say that these two areas will
hinder progress with physical and
medical improvement.
From experience with the other nice
guidelines, the advice and ‘evidence’ or
lack of in the document, will be used by
157
We were unfortunately unable to cover all
areas and focused upon those that
people / stakeholders initially suggested
as critical areas for address.
We will ensure that your suggestion is
passed on to the NICE Topic Selection
Panel as a suggestion for a topic for a
future guideline ‘the psychological
Typ
e
Stakeholder
N
o
SH
Royal College of Speech and 7
Language Therapists
SH
Royal College of Speech and
Language Therapists
Docum
ent
Full
8 Full
Section
number
General
3.1.1.5
Comments
Developer’s Response.
NHS trusts when deciding whether to
commission services. This will directly
impact on the quality of care our clients
receive.
The guidance must emphasize the
sudden and serious impact that stroke
has on the individual and on their family
and friends and the need for support to
cope and begin to come to terms with
this.
Swallow screens should be carried out
by either a speech and language
therapist or nursing staff who have been
trained to perform a swallow screen by a
speech and language therapist.
support of and communication with
people with chronic conditions’.'
Thank you. Please see the amended
section 1.1.1.3
Thank you for your comment. It is outside
the remit of NICE clinical guidelines to
name specific professions.
Further a record of this must be made in
the patient notes.
Same in section 10.2.6.1 and 10.2.6.2
SH
Royal College of Speech and
Language Therapists
9 Full
4
The attempt at glossary and definitions
is still full of complex technical medical
jargon. This assumes that the reader
has knowledge of research methodology
and statistics.
1. The technical terms must be modified
into plain English e.g. change “tunica
intima lining of the arterial wall”,
“nasendoscope”. This particularly
applies to the definition of stroke –
“ischaemic” and “haemorrhagic”.
2. Omissions – the glossary does not
include many acronyms and terms e.g.
“CEA”, “cerebrovascular accident”,
“dysphasia”, “peripheral vascular
158
Recording assessments in patient notes
is part of good clinical practice and not
specific to stroke
Thank you. We have amended the
guideline accordingly
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
disease”, “myocardial infarction”,
“hyperlipidaemia”, “duplex imaging”.
3. Many only have the technical term
e.g. “Deep vein thrombosis”, “Fresh
frozen plasma”, “Middle cerebral artery”,
“diffusion weighted imaging”, “relative
risk”, but no explanation. “Relative risk”
is ambiguous and could therefore be
read as a risk to relatives!
4. There is no punctuation between the
term and the definition e.g. “Transient
ischaemic attack a stroke” – suggest ‘- a
stroke’ or ‘: a stroke’.
5. Change to ‘Dysphagia – difficulty in
swallowing’ rather than “Dysphagic – a
difficulty in swallowing”
SH
SH
SH
SH
Royal College of Speech and
Language Therapists
Royal College of Speech and
Language Therapists
Royal College of Speech and
Language Therapists
Royal College of Speech and
Language Therapists
10 Full
5.1.1.1
11 Full
5.1.2.1
12 Full
5.2.1.4
13 Full
6.2.5.1
SH
Royal College of Speech and 14 Full
Language Therapists
6.2.6
SH
Royal College of Speech and 15 Full
Language Therapists
6.3.2.2
Add ‘Few people are aware of the
symptoms of stroke’
“Emergency room” = American. Perhaps
it should be explained at the beginning.
“Basket” of interventions – suggest
‘selection’
“stroke mimics” is ambiguous –
presumably it refers to conditions that
present in a similar way, but could be
misinterpreted. Suggest a change in
wording.
Same in 7.2.1.1
Text box – “CEA” – this is not in the
definitions/glossary – just “carotid
endarterectomy”. The same applies to
6.3.1.1 “MR angiography” = magnetic
resonance.
“dysphasia” – the term ‘aphasia’ is more
commonly used. Further a definition of
159
This is contained in the text
This has been clarified in the text
We think this is clearer
Stroke mimic is standard terminology
The glossary has been amended
The GDG prefer the term dysphasia. A
definition has been added to the glossary
Typ
e
Stakeholder
SH
Royal College of Speech and 16 Full
Language Therapists
7.1.1.1
SH
Royal College of Speech and 17 Full
Language Therapists
Royal College of Speech and 18 Full
Language Therapists
7.1.2.5
SH
N
o
Docum
ent
Section
number
7.1.5.2
Developer’s Response.
Comments
dysphasia is not in the glossary.
Repace “organised stroke care” with
‘specialist’ which is much more
commonly used.
Same in section 7.1.2.2
Suggest ‘involved patients with
intracranial haemorrhage’
“Mobile stroke team” should be
explained as studies have looked at
general teams, rather than specialist
stroke multi-disciplinary teams.
Same in section 7.1.6.4. It should
explain which disciplines were included
in mobile stroke teams in the studies.
SH
Royal College of Speech and 19 Full
Language Therapists
10
SH
Royal College of Speech and 20 Full
Language Therapists
10.2.1
SH
Royal College of Speech and 21 Full
Language Therapists
10.2.2.1.
SH
Royal College of Speech and 22 Full
Language Therapists
10.2.2
Section ten must also include non-oral
feeding and the issue of re-feeding.
Cross-refer to section 10.3
Suggest a change in sentence from “a
judgement is made about whether the
patient coughs ….” to ‘has altered voice
quality or respiratory pattern, pooling of
water in the mouth, or drooling’.
Add ‘Patients who fail a swallow screen
should be referred to Speech and
Language Therapy’.
Screening should only be carried out by
a speech and language therapist or a
trained member of the nursing staff.
There is a spelling error when
discussing the swallow screen. The
screen is called the “Gugging Swallow
160
This was the phrase used in the NSF
This has been clarified in the text
In this case the stroke team comprised of
a specialist registrar, nurse,
physiotherapist and an occupational
therapist. The team assessed every
patient at admission and recommended a
diagnostic & treatment plan based on
stroke unit guidelines for implementation
by the ward team. We have noted this in
the text
Please see the amended text 10.1.1; refeeding syndrome is beyond the scope of
the guideline
This is in the text
We cannot specify professions in the
guideline
Thank you for your comment. This
section summarises the evidence review.
It is inappropriate in the context of this
section. It is outside the remit of the NICE
clinical guideline to name specific
professions
Thank you. The text has been amended
accordingly.
Typ
e
SH
SH
SH
Stakeholder
N
o
Docum
ent
Section
number
Royal College of Speech and 23 Full
Language Therapists
Royal College of Speech and 24 Full
Language Therapists
10.2.4.11
Royal College of Speech and 25 Full
Language Therapists
General
10.2.5.4
SH
Royal College of Speech and 26 Full
Language Therapists
10.2.6.3
SH
Royal College of Speech and 27 Full
Language Therapists
10.3.1.1
SH
Royal College of Speech and 28 Full
Language Therapists
12.1.5.1
SH
Sanofi-Aventis
1
Full
General
SH
South Asian Health
Foundation
1
Full
General
Developer’s Response.
Comments
Screen” rather than the “Guggling” which
is written in the text.
The sentence beginning “Three studies
reported…” – should “that” be removed?
I agree that those with persisting
dysphagia should have access to
instrumental assessment.
The document does give detailed info
about the importance of assessing for
dysphagia. However it does not mention
therapy such as compensatory
techniques and exercises aimed at
improving the swallowing.
Further the assessment it has
mentioned ‘GUSS’ which is not widely
used.
How will this be compatible with
initiatives such as early supported
discharge and effect on length of stay?
Ethical issues should be discussed in
terms of when and why to introduce nonoral feeding, i.e. consultation with the
person or family, and of the Adults with
Incapacity Act (English & Welsh
legislation)
We support the recommendation for
research regarding outcomes of
modified textures verse access to thin
fluids
This organisation wrote in to say that
they had no comment to make.
There appears to be very little on the
problems and management towards
ethnic minorities with stroke e.g. we
know that only 33% of Bengalis have
their cholesterol measured following a
stroke compared to their Caucasian
161
Thank you. The text has been amended
accordingly.
Thank you
This evidence was not reviewed
We would assume patients will not be
discharged into ESD schemes until their
feeding requirements have been
established
Agree see altered text
Thank you
Thank you
Thank you for your comment. Where the
evidence review has shown differences in
the management of ethnic minorities this
has been included within the guideline.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
South Central Ambulance
Service NHS Trust
1
Full
General
Comments
Developer’s Response.
counterparts who have 77% measured.
The following area has been overlooked. We agree that this is an important area.
We did a search for prehospital tools.
The use of telephone-based software to Thank you for informing us of your
identify patients with stroke and target
abstract. This is uncurrently unpublished
appropriate resources:
and therefore cannot be included in the
Current AMPDS-based software misses guideline. We look forward to reviewing
>50% with acute stroke. There is an
this paper when this guideline is updated.
urgent need to examine telephonebased diagnostic algorithms to identify
key indicators of stroke and develop
pathways with a far greater diagnostic
accuracy than currently exists. (Please
find below an abstract of our stroke
study)
ABSTRACT
Background and purpose: As many as
half the patients presenting with acute
stroke access medical care through the
ambulance service. In order to
effectively identify and triage these
patients as life threatening emergencies,
telephone-based ambulance software
must have high sensitivity and specificity
when using verbal descriptions to
identify these patients. We compared
the software-based diagnosis with the
patient’s final clinical diagnosis of all
patients admitted by ambulance to North
Hampshire Hospital Emergency
Department over a six month period to
establish the ability of telephone-based
diagnosis to accurately diagnose stroke.
Methods: All emergency calls to South
Central Ambulance Service over a six
month period resulting in a patient being
taken to North Hampshire Hospital
(NHH) Emergency Department were
162
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
reviewed. The classification allocated to
the patient by ambulance Advanced
Medical Priority Dispatch software
(AMPDS v11.1) was compared with the
final clinical diagnosis made by a doctor
in the Emergency Department.
Results: A total of 4810 patients were
admitted to NHH during the study
period. Of these, 126 patients were
subsequently diagnosed as having had
a stroke. Sensitivity of AMPDS software
for detecting stroke in this sample was
47.62% and specificity 98.68%. Positive
predictive value was 0.49 and negative
predictive value of 0.986.
Conclusions: Fewer than half of all
patients with acute stroke were identified
using telephone triage on the initial
emergency call to the ambulance
service. This first link in the chain of
survival needs strengthening in order to
provide prompt and timely emergency
care for these patients.
SH
South Central Ambulance
Service NHS Trust
2
Full
General
The following area has been overlooked. This is outside the remit of a NICE clinical
guideline. We believe the DH are going to
Public awareness campaign
initiate a public awareness campaign.
Although the Stroke Association is
promoting public awareness of FAST,
should this be something that the DH
support – such as their campaign
encouraging patients with chest pain to
call for help.
SH
South Central Ambulance
Service NHS Trust
3
Full
General
The following area has been overlooked. Thank you for your comment. The group
did not review any evidence on which to
Treatment on scene
base a recommendation on this
Should there be recommendations that
crews should spend no more than 10
163
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
mins on scene with these patients – as
SCAS does for patients with acute
coronary syndrome? Central
recommendations would make it easier
to implement these sort of
recommendations.
SH
South Central Ambulance
Service NHS Trust
4
Full
General
The following area has been overlooked. It is outside the remit of NICE clinical
guidelines to comment on or discuss
Destination hospitals
service provision.
Not all hospital with Emergency
Departments will be able to deliver
appropriate acute interventional care for
stroke patients (especially in the short
term). NICE should make
recommendations about bypassing
hospitals without appropriate treatment
facilities and clarifying under what
conditions this would be appropriate.
SH
South West Essex PCT
1
Full
1.1.1.6
This seems to negate that very good
service that is offered in some pockets
of the NHS and in some parts of the
service, - managers and teams within
these would consider stroke – a high
priority. This could be reworded more
positively eg new research has
prompted the need to focus on stroke
and management of stroke.
Thank you. Please see section 1.1.1.1.
which has been amended
SH
South West Essex PCT
2
NICE
1.2.3.1
Note that there are discrepancies
between the National Stroke Strategy in
regard to carotid imaging in suspected
We acknowledge the differences between
an evidence based guideline such as this
and a policy document. The developers
164
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
South West Essex PCT
3
NICE
1.2.4.1
SH
Stroke Research Network
1
Full
General
SH
Stroke Research Network
2
Full
14
Comments
Developer’s Response.
stroke/TIA. NICE says within 1 week,
National Strategy says within 24 hrs for
high risk TIA. Please explain this
difference in final version
Note that there are discrepancies
between the National Stroke Strategy in
regard to receiving carotid
endarterectomy. National strategy says
within 48hrs, NICE says within 2 weeks.
Please explain this difference in final
version
We congratulate the GDG on developing
an excellent set of guidelines which we
feel will contribute towards further
improvement in stroke care.
There are 5 research recommendations
but it is not clear how these were
decided.
of this guideline highlight that the
guideline recommendations are derived
from a sound evidence based.
The Stroke Research Network has
identified the following research
priorities:
Acute CSG
 Process: Delivery of immediate
investigation using established
methods (MRI, CT perfusion) and of
optimal (hyper)acute care with
proven treatments (thrombolysis,
intervention for carotid disease);
research into practical aspects of
the healthcare process such as
emergency service protocols.
 Methodological: Design and conduct
of acute trials: selection criteria and
outcome assessment.
 Mechanistic: Better characterisation
and understanding of natural history
of subtypes of ischaemic and
haemorrhagic stroke, to inform a)
165
The evidence is for 2 weeks
Thank you
Thank you for your comment. These will
be considered by the developers.
Areas for further research were identified
by the GDG throughout development
following the appraisal of the evidence.
The GDG are then required by NICE to
vote on their top 5 research
recommendations to be highlighted in the
guideline.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
selection for future trials; b)
predicted outcome for trial planning;
and c) mechanisms and choice of
interventional approach. This may
include exploratory work with
existing datasets. It implies
considerable investment in research
imaging support, especially because
adequate imaging is not available
through standard services, and into
development of new imaging
techniques. Reverse translational
work is required, aimed at
understanding and learning from
failures of clinical trial programmes.
Patient, Carer and Public Involvement
CSG
 Evaluation of the patient, carer and
public involvement in stroke
research: measuring costs and
benefits.
 Evaluation of different methods of
involving patients, carers and the
public in stroke research.
 Seeking the experience of
participants in SRN adopted trials in
order to identify notable practice
Prevention CSG
 Value of very early secondary
prevention after stroke/TIA
 Assessment of secondary
prevention in specific subgroups,
especially lacunar stroke (to prevent
stroke and cognitive decline)
 Use of basic and translational
science to develop novel prevention
strategies, e.g. based on haplotype
166
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments


(e.g. as deCODE have done with
their antiplatelet)
Surgical/neuroradiological
interventions for causes of
haemorrhage, e.g. AVMs,
cavernomas
Why do patients take or reject long
term prophylaxis?
Primary Care CSG
 Research demonstrating
effectiveness of secondary
prevention strategies in
representative populations
 Research on identification (including
screening) and treatment of mood
disorders in people with a past
history of stroke
 Research on strategies for providing
late support/rehabilitation for stroke
patients and carers (overlaps with
Rehabilitation CSG)
 The role of primary care in ensuring
rapid treatment of transient
ischaemic attack
Rehabilitation CSG
 Research focused on improving
participation and long term care
after stroke (survivors and carers)
 Investigation of cognitive and
psychological factors that influence
outcome after stroke
 Applying advances in basic science
and technology to develop new and
improve existing rehabilitation
interventions for people with stroke
e.g. robotics, pharmacology.
167
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
Service Development and Training CSG
 Can groups of smaller district
general hospitals provide care for
patients with acute stroke that is at
least as efficient, cost-effective and
acceptable to patients as subregional stroke specialist centres?
 Can generic therapist and nurses
provide the same care outcomes for
people with stroke, as specialist
stroke therapists and nurses, at
equal cost?
 What is the most effective way of
teaching primary care clinicians
about stroke?
These priorities were identified by
discussion within SRN Clinical Studies
Groups which consist of academics,
clinicians, patients and carers.
We would be grateful if the GDC would
consider including the SRN research
priorities regarding the initial
management of acute ischaemic stroke
and TIA within the guidelines.
SH
Takeda UK
1 Full
9.3.5
Thank you we have amended the text to
The guideline makes the following two
include mention of other similar studies
statements on the SCAST study: ‘it was for completeness
agreed that no specific
recommendations could be made until
the publication of the SCAST trial (which
is due to report in 2009) and ‘The GDG
acknowledged that the SCAST trial
results will not be available until 2009’.
168
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
As there are currently no UK
investigators involved in the SCAST
study, awareness of the study amongst
healthcare professionals is likely to be
relatively low. We would therefore like to
suggest that some brief details on the
SCAST study are added to explain the
purpose of the study and to make the
reader aware this is a larger scale study
designed to provide more robust
outcome data than the candesartan
phase II RCT (Schrader et al, 2003)
referenced in 9.3.4 of the guideline.
The Scandinavian Candesartan Acute
Stroke Trial (SCAST) is a phase III study
investigating whether blood pressure
lowering treatment with candesartan is
effective in reducing the risk of death or
major disability in acute stroke and in
reducing the combined risk of vascular
death, myocardial infarction or stroke
compared with placebo. Patients with
acute stroke (<30 hours) and SBP ≥140
mm Hg will receive candesartan 4 – 16
mg or placebo for 7 days followed by
candesartan treatment for 6 months for
patients who are hypertensive. The
primary outcome is death or major
disability at 6 months and combined
endpoint of vascular death, myocardial
infarction or stroke during the first 6
months. The study aims to recruit 2500
patients.1,2
The results of the trial will provide more
reliable data on the effects of an
angiotensin receptor blocker in patients
with acute stroke.
169
Typ
e
Stakeholder
N
o
Docum
ent
SH
Takeda UK
2
Full
Evidenc
e tables
Developer’s Response.
Section
number
Comments
9.3
The evidence table includes the details
of Schrader et al (2003).
Thank you. The text has been amended
There are a couple of minor technical &
typographical inaccuracies that we
would like to highlight:
 The number of patients is stated as
500. Please note this trial was
stopped prematurely on the
recommendation of the safety
committee when 342 patients had
been randomised (339 were valid).
 Currently the table states Placebo
N=160.There were 173 patients in
the candesartan group and 166
patients in the placebo group.
 Please also note the details on
patient population states ‘age 68.3
vs 67.8 trs; male sex 50% vse
52%’. It is suggested this should
read ‘yrs’ and 50% ‘vs’ 52%.
 In the ‘candesartan cilexetil’ column
it states ‘On day increased to 8 or
16 mg if blood pressure...’. This was
on day ‘2’.
 Also in this column it states ‘In the
first three days, occasional
measurements of blood pressure
were taken throughout the day ay 1
to 2 hr’ (should read ‘at’) ‘and during
the night ‘at’ 2-3 hr intervals.
’
170
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Comments
Developer’s Response.
SH
The Chartered Society of
Physiotherapy
1
Full
General
The following areas including research
questions are relevant to clinical
practice for physiotherapists.
Thank you
SH
The Chartered Society of
Physiotherapy
2 Full
SH
The Chartered Society of
Physiotherapy
3
Full
SH
The Chartered Society of
Physiotherapy
4
Full
SH
The Chartered Society of
Physiotherapy
5
NICE
9.1
-9.16
Suppleme
ntary
Oxygen
Therapy
Recomme
ndation:
11.1
-11.6
Early
Mobilisatio
n
Thank you. This is stated in the text
People who have stroke should receive
supplemental oxygen only if sats. drop
below 95%
Insufficient data to comment on safety of Thank you. This is in the text
early mobilisation. Consensus data
stated early mobilisation had potential
advantages including reducing the risk
of chest infection, preventing D.V.T.,
early access to water fluids and nutrition
plus positive psychological benefit.
People with acute stroke should be
Recomme mobilised asap following an assessment
ndation:
of sitting balance and falls risk by an
appropriately trained healthcare
professional and access to appropriate
equipment.
Research Whether early mobilisation delivered by
Question appropriately trained professionals is
safe and improves outcomes compared
to standard care?
12.1
Remains a contentious issue
Thank you we agree. Following the
-12.6
stakeholder consultation the group have
Aspiration No recommendation as insufficient
included a recommendation.
Pneumoni evidence
a
Does free access to water versus
Recomme withdrawal or oral modification of liquid
ndation
prevent aspiration pneumonia following
Research an acute stroke.
Question
General
Could the research recommendation be Thank you. This has been amended
linked/ referenced in the guidance
accordingly
171
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
The Society and College of
Radiographers
1
Full
5.2.5.3
SH
The Society and College of
Radiographers
2 Full
6.1.5.1
Developer’s Response.
Comments
sections?
The NICE guideline for stroke is a
thorough and comprehensive document
which will contribute to the treatment of
patients with stroke.
However referral of TIA’s within 24 hrs
for specialist assessment. An indication
of the specialist assessment pathway at
this point would be helpful.
Imaging in TIA
Thank you
Thank you please see the amended
section.
Thank you. see expanded 6.1.1.1
It is well established that MR is sensitive
than CT in the detection of vascular
lesions, especially if performed early.
Note. This is not necessarily
true in haemorrhagic vascular lesions.
Additionally DWI MRI would need to be
performed to evaluate ischaemic
vascular lesions. CT perfusion can also
be used to evaluate ischaemic lesions.
SH
The Society and College of
Radiographers
3 Full
6.2.1.1
It is important that brain scanning (MRI) Thank you
does not delay the institution of optimum
secondary prevention, or detection and
treatment of significant carotid stenosis.
Some patients, within certain need DWI
MR within 24 hours.
Note. While we fully support
this, it will be necessary to ensure that
resources are available to enable
departments to meet this requirement.
While there is unlikely to be a large
volume of patients requiring this
imaging, it will need to be available
seven days a week, including weekends
and bank holidays. This will have an
172
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
impact on staff resources in MR
scanners for both imaging and
interpretation. There will need to be a
clear patient pathway to ensure that
confusion is avoided
SH
The Society and College of
Radiographers
4 Full
6.3.1
Carotid imaging CTA, MRA and Carotid
Doppler. Patients who are shown to
have a stenosis of ≥ 50% require
surgery within one week.
Note. This may also have an
additional impact on imaging
departments in relation to staff
resources.
SH
The Society and College of
Radiographers
5 Full
7.1.12
We acknowledge that a significant
proportion of stroke patients have
access to CT within 3 hours when
admitted to hospitals who have either
specialist units or have implemented
acute stroke protocols and guidelines.
SH
The Society and College of
Radiographers
6 Full
7.2.1
Changes in clinical practice will be
required to implement CT
recommendations – 100% of patients
will need to be scanned within 24 hours
(increased availability, changes in scan
request and reporting procedures) will
be required to implement the new
recommendations.
Note. There will need to be
robust patient pathways and protocols to
ensure smooth access and appropriate
referral to imaging. There may be an
173
Thank you
The percentage stenosis depends on
how it was measured. The
recommendation on surgery is within 2
weeks; imaging within one week is
recommended to ensure patients can be
operated on within the 2 week limit.
We accept this may impact on resources
and this will be within the scope of the
implementation group.
Thank you
The GDG were very aware of these
issues - these are issues that will need to
be liaised with the NICE implementation
and costing teams and as such will be
passed on to NICE for consideration by
these teams.
Typ
e
SH
Stakeholder
The Society and College of
Radiographers
N
o
Docum
ent
7 Full
Section
number
8.5.1
Developer’s Response.
Comments
increase in out of hours work to
accommodate this change in practice
and radiographers can contribute
uniquely to this, both ensuring
appropriate pathways and with
appropriate investment in training they
can undertake CT head reporting
Recommendation for emergency CT
would include indications for
thrombolysis.
Thank you. Please see the amended text
Note. For ease of use a
breakdown of indications may be useful
here, such as onset of symptoms have
taken place within 3 hours.
SH
The Society and College of
Radiographers
8 Full
8.5.2.2
Immediate access to imaging and
reimaging will be required when
alteplase is administered, along with
appropriately trained staff to interpret
images.
Thank you this has been noted
Note. There may be an increase
in out of hours work to accommodate
this change in practice and
radiographers can contribute uniquely to
this, both ensuring appropriate pathways
and with appropriate investment in
training they can undertake CT head
reporting.
SH
The Stroke Association
1
NICE
1.3.1.1
SH
The Stroke Association
2
NICE
1.3.2.1
This guideline should include a definition
of a specialist acute stroke unit
We believe that brain imaging should be
performed immediately (ideally the next
slot and definitely within 1 hour,
whichever is sooner) for all people with
174
Thank you. This has been added as a
footnote
This has been discussed at length by
GDG; we feel we have produced a
challenging yet achievable goal.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
Developer’s Response.
Comments
suspected acute stroke not just for those
identified in this paragraph. The full
guidance (para 7.2.5.1) identifies these
patients as ones in whom urgent
scanning will result in immediate
changes in clinical management. It does
not however argue or demonstrate that
others with a suspected stroke could not
also benefit from urgent scanning. It
says scanning for these should be as
soon as possible, so why not within an
hour like the others?
The economic evidence (para 7.2.4.5),
Wardlaw et al (2004) in the full
guidance) not only found that scanning
all patients was the dominant strategy
but that it was less costly and more
effective. The only argument for not
scanning all suspected acute strokes
immediately would appear to be “
problems with access to scanning, or a
lack of radiology or radiography support”
(para 7.2.1.1 full guidance). It is not
acceptable that these organisational
barriers should stand in the way of best
practice or patient centred care. The
NICE guidelines should be aspirational
like the National Stroke Strategy for
England; anything less is in danger of
undermining the philosophy of patient
centred care. As a way of improving
access to scanning, consideration
should be given to training
radiographers to operate CT scanners. It
is playing with people’s lives if a
suspected acute stroke is not treated as
a stroke and scanned immediately. We
would therefore like to see the
recommendation in para1.3.2.1 (and
175
Typ
e
Stakeholder
N
o
Docum
ent
3 Full
Section
number
SH
The Stroke Association
5.2.5.3
SH
The Vascular Society
1
NICE
1.2.11
SH
The Vascular Society
2
NICE
1.2.4
SH
The Vascular Society
3
NICE
1.5.3.1
SH
The Vascular Society
4
NICE
1.2.3
Developer’s Response.
Comments
para 7.2.6.1 in the full guidance)
redrafted to recommend that all people
with a suspected acute stroke are
scanned immediately.
As this document says, the health
economic modelling evidence suggests
that the most cost effective service
design overall is immediate assessment
of patients with TIA. The Stroke
Association would like all those with a
suspected TIA to receive immediate
specialist assessment. While this may
not be practical at present we believe it
should be clearly stated that this is the
long term aim, and as such should be
made clear in the Guidelines
At present in many hospitals there is a
long wait for brain imaging, and it will be
important that early investigation is
enforced somehow
We believe that there is no role for
carotid stenting in the early period
(within 2 weeks) after stroke or TIA.
There is evidence to suggest outcomes
are worse in these patients who often
have soft, unstable plaques that are
more likely to embolise (Circulation
2004; 110: 756). CEA should be the
preferred intervention (outwith trials)
pending further investigation.
We think a further category should be
added – stroke following hyperperfusion
(preceded by seizure)
It would be helpful to define non
disabling stroke – e.g. rapid neurological
recovery, no carotid artery occlusion, no
intracranial haemorrhage, Rankin score
0-2, CT evidence of infarction affecting
<33% of the middle cerebral artery
176
We agree and feel we have produced a
challenging yet achievable goal
Thank you
We agree. We have replaced the term
carotid intervention with carotid
endarterectomy to clarify
This paper was not relevant to the
evidence review undertaken
We are are sorry but we are unclear
about what is meant by this stakeholder
comment?
We feel disabling is clear.
Typ
e
Stakeholder
N
o
Docum
ent
Section
number
SH
The Vascular Society
5 Full
3.1
SH
The Vascular Society
6 Full
6.3.7.1
SH
University Hospital
Birmingham NHS Trust
1
NICE
1.9.2
SH
University Hospital
Birmingham NHS Trust
2
NICE
1.1.2
SH
University Hospital
Birmingham NHS Trust
3
Full
General
Developer’s Response.
Comments
territory
We believe that the key priorities should
include urgent referral for carotid
endarterectomy in appropriate cases.
This could appear between 3.1.1.4 and
3.1.1.5 as: referral of people with carotid
stenosis for carotid endarterectomy next
day. These include patients with TIA and
those with minor stroke.
There remains confusion about patients
with TIA , their urgency of imaging and
referral. It would be worth highlighting
that patients at high risk after TIA
(ABCD2>4) should be scanned and
referred for CEA ideally within 48h
Whilst we agree with the
recommendations, some of our
neurosurgical colleagues were
concerned about the lack of expression
of a dominant hemisphere to determine
the appropriateness of surgery.
Essentially some felt that they would be
more reluctant to operate on dominant
hemisphere strokes but no comment is
made of this
Whilst we agree in principle with the
speed of response needed in TIA
assessment. There is no account taken
of the urgency needed where patients
present to primary care late. For
example should a patient that presents
to their GP 3 weeks after the TIA still be
seen in 24 hours if ABCD2 score is 4?
This needs to be addressed because
this is a frequent occurrence.
The role of the ED in identifying strokes
and therefore leading to direct
admissions is critical. Some of our
colleagues felt that having
177
Thank you for your comment. The key
priorities were voted on by the members
of the guideline development group. The
top 5 key areas for implementation are
included within this section.
Reviewed evidence and covered in
6.4.6.1
This is beyond the scope of the guideline
Please see section 1.1.2.1 which has
been amended
Thank you for your comment. Yes we
agree this would have been helpful. This
point has been noted and will be taken
into consideration for any update of the
Typ
e
Stakeholder
SH
University Hospital
Birmingham NHS Trust
SH
University Hospital
Birmingham NHS Trust
N
o
Docum
ent
4 Full
5
NICE
Section
number
8.7.6.1
1.4.8
Comments
Developer’s Response.
representation from the ED colleges
would have been helpful.
IT would be helpful to have greater
clarity as to whether PCC should be
used routinely. It seems that the
evidence suggest it should correct INR
quickly and your guidelines says do this
as quick as possible but then PCC is not
recommended explicitly. What is the
real recommendation?
Patients in AF who will be
anticoagulated in 2 weeks are to be
given aspirin meanwhile. Are we stating
that they therefore do not need
dipyridamole at all for those 2 weeks.
This would seem right but it is not clear.
guideline.
Table extracted from comment - British Psychological Society - comment 1
LEVEL
GROUP
ASSESSMENTS
1
All health and social care
professionals
Recognition of
Psychological
Needs
2
Health and social care
professionals with
additional expertise
Screening for
psychological
distress
3
Trained and accredited
professionals
Assessed for
psychological
distress and
diagnosis of some
psychopathology
Thank you the text has been amended
Up to 2 weeks patients are given aspirin
300mg; there is no evidence to support
dipyridamole in the first 2 weeks
INTERVENTIONS
Effective information
giving, compassionate
communication and
general psychological
support
Psychological
interventions (such as
anxiety management and
problem solving
Counselling,
Cognitive behavioural therapy (CBT)
and solution focused
therapy,
178
4
Mental health specialistsclinical psychologists
and psychiatrists
Diagnosis of
Psychopathology
Specialist psychological
and psychiatric
interventions
179