Skip to main content


26 November 2015 (Last updated: 24 Jan 2020 17:04)

Recent feedback to National Institue for Health and Care Excellence (NICE) on Intra-arterial Thrombectomy guidance update has been made on UKNG’s behalf both via British Society of Neuroradiologists (BSNR)/ Royal College of Radiologists (RCR) and also by UKNG as a separate stakeholder.

The final version of Standards for providing safe acute ischaemic stroke thrombectomy services were published in September 2015.

Latest documents

Standards for providing safe acute ischaemic stroke thrombectomy services
(September 2015)

PM White, A Bhalla *, J Dinsmore †, M James ‡, N McConachie +, C Roffe *, G Young ‡
On Behalf of the British Association of Stroke Physicians *, British Society of Neuroradiologists□,
Intercollegiate Stroke Working Party ‡, Neuroanaesthesia and Neurocritical Care Society of GB &
Ireland †, UK Neurointerventional Group +

These groups are recognised Special Interest Groups of their respective Royal Colleges and the short life
working party was operating under the oversight of the Intercollegiate Stroke Working Party

Read More



Recommendations for Endovascular Therapy for Acute Ischaemic Stroke – organisation &

processes of care

Recommendations for individual departments and trusts

Recommendations for individual endovascular practitioners

Implementation of standards and quality benchmarks

Anaesthetic management of endovascular treatment of acute ischaemic stroke

Models of safe stroke IA endovascular therapy provision



Appendix 1 – DH IAT Consensus document


Stroke is the third leading cause of death and the leading cause of disability in Europe. The
management of acute ischaemic stroke is a major healthcare challenge but improving outcomes for
acute stroke patients offers major benefits to patients, healthcare systems and society as a whole.
The immediate aim of treatment of acute ischaemic stroke is to recanalise an occluded vessel as
quickly, safely and effectively as possible so as to restore reperfusion to the ischaemic brain region.
Currently the standard treatment for acute ischaemic stroke for patients presenting up to 4.5 hours
after onset is intravenous thrombolysis using tissue plasminogen activator (IVT) [1]. This treatment
results in a recanalisation rate of only 50% of patients with distal vessel occlusions. However,
recanalisation rates of large proximal vessel occlusion are disappointing at 24 hours after IVT
treatment, with rates of 14% for internal carotid arteries and 35% for middle cerebral arteries being
reported [2]. Over one third of patients with anterior circulatory ischaemic stroke will have large
vessel occlusion.

The prognosis for patients with clinically severe stroke secondary to proximal occlusion is poor with
the NINDS trial demonstrating that only 10% of patients with an NIHSS score of 20 or more
achieved independence at three months [2]. Previous randomised controlled studies reported in
2013 (IMS III, MR RESCUE, SYNTHESIS Expansion) evaluating endovascular approaches in acute
ischaemic stroke caused by large vessel occlusive stroke showed no additional benefit to
endovascular approaches [3, 4, 5], however despite this in many locations outside the UK these
procedures have been incorporated into usual clinical practice [6, 7].

Reasons for the neutral results include use of older, less effective recanalisation devices, longer
time window from onset to intervention and the inability to effectively identify patients with large
vessel occlusion with appropriate angiographic imaging in a timely fashion. The rationale to deploy
endovascular therapies for ischaemic stroke is to potentially improve outcomes by facilitating early
recanalisation of an occluded large artery as quickly as possible. Data from the IMS III study
confirmed that time window for treatment is a crucial factor and subgroup analysis from this study
suggested that improved outcomes (although not statistically significant) were observed in patients
who received endovascular therapy within 90 minutes of IVT if thrombolysis treatment was
commenced within 2 hours of stroke onset.

Following this, NICE (2013) [8] made the following recommendations stipulating that patients who
are suitable for thrombolysis and satisfy the criteria for a research study (e.g. Pragmatic Ischaemic
Stroke Thrombectomy Evaluation-PISTE) [9] should be considered for enrolment into a randomised
controlled trial and finally patients for whom thrombolysis has failed or is unsuitable due to
contraindications can be considered for thrombectomy if not eligible for a trial. This statement was
supported by the NHS Improvement Stroke Working Party document in 2012 [10].

However since this recommendation, there have been five published randomised controlled trials
[11-15] evaluating the effects of interventional thrombectomy in addition to IVT compared with
standard treatment (i.e. IVT alone delivered within 4.5 hours). These all demonstrated significantly
improved outcomes with endovascular therapy. These studies were heterogeneous in terms of
patient selection, imaging triaging criteria used for large vessel occlusion and salvageable brain
tissue (CT Angiography, CT or MR Perfusion), rescue therapy in failed IVT or contraindication to
IVT, onset to treatment of endovascular therapy (within 6,8 or 12 hours). This creates issues around
generalisability. The consistent theme of improved functional independence (Rankin Score 0-2)
from these studies was driven by improved patient selection by neuro-imaging with rapid
demonstration of large vessel occlusion, enrolment of patients with moderate to severe stroke
(NIHSS of 17), change in technology with stent retriever devices as well as an improved heightened
response of door to intervention time (stroke onset to groin puncture < 4.5 hours). The numbers
needed to be treated by thrombectomy to prevent one case of functional dependency or death
ranged from 3 to 7. Only some trials reported a positive effect on mortality. The summary of the
trials are described below:

MR CLEAN (N= 500) [11] confirmed that endovascular therapy as rescue therapy in failed IVT
delivered within 6 hours of ischaemic stroke resulted in a 13.5% absolute increase in functional
independence with no difference in symptomatic haemorrhage or mortality. Patients underwent CT
imaging and CT angiography to identify large vessel occlusion prior to randomisation but there is
uncertainty whether alternative imaging modalities were used to select patients further.

ESCAPE (N=316) [12] confirmed that a policy of using complex imaging assessment (Multiphase
CT angiography) to identify patients with large vessel occlusion suitable for treatment with
endovascular therapy within 12 hours of onset (15% randomised beyond 6 hours) was beneficial
and resulted in 24% absolute increase in functional outcome. Of interest, patients who could not
receive IVT also appeared to benefit, although this needs to be confirmed in future studies.

EXTEND IA (N=70) [13] again used complex neuro-imaging to identify potentially salvageable brain
tissue with treatment completed within 8 hours, however although there were significant differences
favouring endovascular therapy (31% absolute increase in functional independence), the patients
were highly selected and the results not generalisable.

SWIFT PRIME (N=196) [14] confirmed that using a complex imaging strategy of CT/CT
Angiography and CT/MR Perfusion imaging to identify patients treated within 6 hours of onset
resulted in 25% absolute increase in favourable outcome. Of note, the median time of arrival in the
emergency department to groin puncture was 90 minutes.

REVASCAT (N=206) [15] demonstrated that in patients who failed to revascularise after 30 minutes
of IVT and with evidence of large vessel occlusion and large ischaemic core (Perfusion Imaging),
delivery of endovascular therapy within 8 hours of onset resulted in a 15% absolute increase in
functional outcome.

An important pragmatic question to answer is whether there is strong evidence for early adjunctive
endovascular therapy using a strategy of CT imaging and CT angiography alone rather than
deploying complex neuro-imaging to further triage patients to identify those where thrombectomy
may well result in futile recanalisation. The former policy has the potential to be delivered quicker
within the current NHS structure. This strategy was deployed in the THRACE study [16]. THRACE
(N=414) evaluated endovascular therapy in patients where there was no or minor neurological
improvement (< 5 points of NIHSS) after IVT who had evidence of large vessel occlusion evaluated
by CT angiography. Interventional treatment had to commence within 5 hours of stroke onset and
be completed within 6 hours. Although not yet published (Presented European Stroke Organisation,
April 2015), intermediate results at 90 days suggest a 12.1% absolute difference in functional
outcome, consistent with the growing evidence for endovascular therapy.

In the presence of current data, it can be argued that equipoise no longer exists for selected
patients with moderate to severe stroke with large vessel occlusion. However future research is
required to evaluate the cost benefits [25] of delivering this complex intervention across the stroke
pathway as well as ensuring we have a pragmatic paradigm in delivering such a process which is
generalisable within the NHS. Estimating the percentage of patients who may benefit from
endovascular therapy also needs to be recognised in the context of resources required to evolve the
hyper-acute stroke infrastructure. The published trials to date are not generalisable and the
numbers recruited were small in total but there is suggestion that 5 to 10% of the ischaemic stroke
population may now be eligible for such treatment [17]. That % would rise if time windows for IAT
are extended by on going trials and proposed studies into IAT in stroke of unknown time onset using
advanced imaging triage were also positive. NICE will be required to update its review and
technology appraisal on these interventions thus providing guidance, quality standards and cost
effectiveness data for these processes and will provide a platform for specialist commissioning
arrangements in the future.

In light of the prevailing evidence, the ESO Karolinska Consensus Stroke Update Consensus
statement (2015) [18] has made a number of recommendations:

Mechanical thrombectomy, in addition to intravenous thrombolysis when eligible is provided to patients with large artery occlusions in the anterior circulation up to 6 hours after symptom onset (Grade A, Level 1a).

If IVT is contraindicated (warfarin with therapeutic INR), thrombectomy is recommended as first line of treatment for patients with large vessel occlusion (Grade A, Level 1a).

Regarding patient selection, intracranial vessel occlusion must be diagnosed with noninvasive imaging prior to consideration of thrombectomy (Grade A, Level 1a), perfusion maging determining penumbral size may be used for further selection (Grade B, Level 1b) and high age is alone is not a reason to withhold thrombectomy (Grade 1, Level 1a).

The consensus supports the continued collection of patient data systematically into an established
national programme audit (SSNAP or SITS TBY) or the Scottish Stroke Care Audit Programme
embedded within special arrangements for clinical governance, audit and quality improvement. The
recent results from the published randomised controlled trials could potentially impact on
recruitment to ongoing trials (PISTE has been temporarily halted by its steering committee) but it is
important that patients are considered for enrolment into trials where there is on going uncertainty.
In the advent of increasing evidence supporting endovascular therapy albeit for only a modest
percentage of the stroke population, it is important to produce a consensus on the minimum
standards of care, reflecting the methodology used in the trials, required to support such a service in
terms of specialist staffing/skill mix and organisation of services. It is recommended that such
interventions should only be undertaken within specialist stroke centres that fulfil agreed standards
of care for both staff and process and meet any additional requirements of a RCT or Registry. Every
trust where such procedures are undertaken has a duty to ensure that safe arrangements are in
place [19]. This document has been developed to aid this process by describing the key
requirements for an endovascular acute ischaemic stroke therapy service, including the minimum
service support requirements and basic performance standards that should be met.

Recommendations for Endovascular Therapy for Acute Ischaemic Stroke

Organisation of Care

If endovascular therapy is being considered for the treatment of acute ischaemic stroke this should
be confined to neuroscience centres incorporating hyperacute stroke units (HASU) embedded
within a high quality comprehensive stroke service with access to neurosurgical, neurocritical care
and specialist in and out-patient stroke services. The findings from the trials are generalisable to
only those centres that have access to advanced brain imaging facilities and appropriate
[neuro]endovascular expertise with efficient in-hospital hyper-acute pathways.

The key features of what a HASU should incorporate include:

24 hour availability of an experienced consultant stroke physician [20]

Immediate access to brain imaging including CT Angiography/Perfusion & MRI as required [21]

Direct admission to HASU from A&E < one hour

Continuous physiological monitoring (ECG, oximetry, blood pressure)

Specialist stroke physician ward rounds 7 days per week

Acute stroke protocols/guidelines

Nurses trained in swallowing screening, stroke neurological assessment (including the

NIHSS assessment), eligibility assessment for thrombolysis and administering thrombolysis treatment

Endovascular stroke therapy should only be carried out in the context of high quality hyperacute
stroke services with the appropriate experience in delivering such interventions. Evidence suggest
that hospitals with an annual thrombolysis volume ≥ 50 cases per annum achieve significantly better
onset to needle, door to needle and door to imaging time than thrombolysis volumes < 50 cases [22]
and this may enhance endovascular delivery times. Sites participating in randomised controlled
trials for neuro-endovascular therapy were from comprehensive stroke centres performing between
40 and 60 procedures annually.

Skill Mix
The decision to undertake endovascular stroke therapy should be made jointly by a multidisciplinary
team compromising a consultant stroke physician, [neuro]interventionist (with the necessary
experience and skills) and an anaesthetist (preferably experienced in neurological care).
The stroke physician undertaking the decision to consider endovascular therapy must satisfy the
criteria the BASP criteria for a stroke specialist including:

Completion of specialist training or recognised expertise (existing specialists)

Ongoing active involvement in stroke management (at least 5PA of which 3 PA is direct clinical care)

Annual attendance of at least one specific training event

Evidence of continued professional development in the field of stroke medicine

Participation in national stroke related audit

Basic research skills (participation or facilitation of stroke research)

The stroke physician should be trained in delivering thrombolysis and in the monitoring of any
complications associated with thrombolysis and endovascular therapy.

This must be underpinned by regular quality and audit meetings within a quality improvement group
incorporating stroke physicians, neurointerventionists, neuroradiologists, intensivists ± emergency
medicine physicians [6].

There should be provision of 24/7 consultant cover provided by at least 6 BASP thrombolysis
trained consultants on a rota able to make thrombolysis and hyper acute treatment decisions (1).
BASP have recommended that for a hyper-acute service admitting 600 new admissions annually,
10.5 direct clinical care PA would be required to support such a service [23].

Emergency Medical Services (EMS) play a vital role in hyper-acute stroke care and local protocols
and algorithms should be in place for dispatch, assessment, pre-notification and transport strategies
All potentially eligible patients for intravenous thrombolysis initially should be transferred
immediately to a centre with hyper-acute stroke services with early notification of the specialist
stroke team. There is currently no existing literature on training EMS in identifying potential patients
eligible for endovascular therapy. There is no evidence that EMS should directly transfer patients
outside the intravenous thrombolysis window directly to centres with endovascular capability.

Inter Hospital Transfer:
Transfer from a primary hospital to a centre providing endovascular therapy can potentially result in
considerable delays in administrating treatment which may impact on outcome. This needs to be
considered in the decision whether to transfer or not. However data from SWIFT PRIME suggested
that although patients who were treated with IVT from an outside hospital had less favourable
outcome overall, they still benefited from treatment regardless of which site IVT was provided [14].
Regardless of this, it is important that provision of intravenous thrombolysis is not delayed under
any circumstances.

‘Hub and Spoke’ models have been used regionally to deliver IVT successfully however, regional
strategies with a Network approach for endovascular therapy will need to be considered in light of
level 1 evidence for this procedure.

If algorithms are to be developed for secondary transfer of patients eligible for endovascular
therapy, then local protocols for transferring and accepting patients need to be clearly outlined.

Process of Care

Endovascular Therapy Recommendations and Quality Benchmarks:
Eligibility for mechanical thrombectomy should not delay the initiation of IVT where this is indicated.

Consider endovascular stroke therapy in the following patients:

A)- Those with proximal intracranial large vessel occlusion in a symptomatic territory leading to a disabling neurological deficit (NIHSS ≥6) of known onset and <5 hours ago* (< 4 hours if considering inter-hospital transfer)

B) Those with contraindications to intravenous thrombolysis therapy (e.g. recent surgery, warfarin treatment with therapeutic INR)

C) Those failing to respond to intravenous thrombolysis (However the time window defining failed IV tPA has not been established and further trials may be required to define the criteria for failed IV. THRACE defined this as no improvement or minor neurological improvement of < 5 points on the NIHSS and REVASCAT defined this as failure of revascularisation after 30 minutes of IVT)

D) Those suitable for entry into randomised controlled trial where there is clinical uncertainty

*If the major vessel occlusion is in posterior circulation, time window may be up to 12-24h
depending on clinical status of patient

Thrombolysis should not delay initiation of the thromebectomy pathway. To be able to meet the door
to groin target of less than 90 minutes, thrombectomy needs to be considered in patients who meet
the criterion A and are considered fit for the intervention. If thrombolysis is given and the patient
responds (Criterion C NOT fulfilled), preparations for thrombectomy can be halted.

The metric of ‘arrival to start of treatment’ should be within 90 minutes wherever possible, unless the
indication for intervention only developed after arrival (e.g. deterioration in NIHSS after

Door to Imaging:
All patients with an acute stroke being evaluated for endovascular therapy should be imaged
immediately (ideally the next imaging slot).

Use of CT angiography or MR imaging/angiography should not unduly delay therapy with
intravenous thrombolysis or delay door to arterial puncture. In practice hospitals that undertake CTA
routinely in acute stroke achieve this by review of plain CT occurring whilst CTA is performed and then not waiting for CTA report before instituting IVT (where indicated).

General Anaesthesia:
If general anaesthesia is used due to clear indications and patient safety concerns during the
procedure, then ongoing physiological monitoring including airway management should be
undertaken in the appropriate setting until stable. This should occur either within high dependency
or intensive care settings, where necessary with specialist stroke support.

The following outcomes should be measured from the SITS TBY data collection set:

Revascularisation achieved- with a TICI grade 2b or 3 in ≥60% of patients

Rankin 0-2 at 3 months achieved in ≥33% of cases

Symptomatic Intracranial Haemorrhage using SITS definition (parenchymatous haemorrhage type 2 combined with deterioration of NIHSS score≥ 4 points at 22-36 hours (in no more than 12% SICH should be expected)

Early deaths (within 72 hours)

Recommendations for individual trusts

Neuroradiology departments

1. Recognition that existing interventional neuroradiology services should be supported to
develop/expand ability to provide stroke endovascular therapy within the NHS subject to NICE
guidance update.

Whether this is currently (June 2015) confined to normal working hours or extended hours will depend upon local circumstances & funding

[Urgent] regional planning for future delivery of extended hours services should be undertaken

2. There should be clarity within the trust and among referring clinicians and service commissioners
about what endovascular stroke services are available, when they are available and what happens
when the service is not available

Local patient pathways should exist, be clear and widely available

Local protocols and SOPs for each step of the pathway should be accessible

The trust’s clinical governance committee and relevant referring clinicians need to be aware of the situation

Delivery of intra-arterial interventions in acute stroke should only be in the context of established, high quality stroke services, where all patients have timely access to specialist stroke units

Consideration for intra-arterial interventions in acute stroke should not compromise the timely delivery of intravenous thrombolysis in eligible patients

3. Service provision must be subject to a formal rota

It is not safe or sustainable to rely on ad hoc arrangements as such arrangements do not ensure that all eligible patients are treated [24]

Nor is it acceptable to assume another centre will be willing or able to provide the service without official and agreed service level agreements (A mechanism should be in place for informing clinical teams about when service is available

4. There should be recognition of the resource implication on Interventional Neuroradiologists (INR)
consultants of supporting even a limited hours stroke endovascular service and also of the knock on
impact on diagnostic neuroimaging services

Appropriate diagnostic neuroimaging support and protocols should be in place

A whole support team of staff is required to deliver endovascular stroke therapy and this needs to be IMMEDIATELY available

Anaesthetic staff with appropriate training/experience in neurocritical care plus ODP support

Angio suite staffing – nursing and radiographic

After care is a critical requirement – neuro-critical care facility should be immediately available post procedure

5. Services should have regular clinical-radiological MDT meetings where endovascular stroke
patients can be reviewed/discussed. As well as enrolment into RCT/Registry, appropriate local audit
processes should be in place

Quality improvement processes for IAT should be in place

Quality improvement group could include a combination of interventionists, neurologists, stroke physicians, intensivists & diagnostic neuroradiologists. Additional members might include representative(s) from quality assurance/improvement or risk management teams

6. IA intervention should be delivered in specialist stroke centres that fulfil the following attributes

adherence to defined standards (as section above)

consultant-led & delivered services with adequate volumes of activity

formal commissioning arrangements in place

demonstrable research capability and activity where clinical activity extends beyond core evidence from the randomised trials underpinning NICE guidance

Recommendations for individual endovascular operators

1. All doctors are bound to adhere to GMC guidance and comply with principles and values set out
in Good Medical Practice

2. Operators should not normally carry out procedures with which they are unfamiliar (in the UK at
present this, to all intents and purposes, essentially limits provision of endovascular intracranial
stroke therapy to consultant interventional neuroradiologists)

If a procedure is required on a reasonably regular basis then individual operators must maintain the necessary skills

There will inevitably be a risk benefit assessment to be made in any individual case and patients and presentations do vary considerably. The risk of any patient transfer, presence or absence of alternative therapies and INR experience will all need to be taken into account.

This risk benefit analysis should be reflected in the consent/assent obtained and documented.

3. Operators should recognise that ad-hoc rotas are not in the best interest of patients [RCR ref 24]

This form of service provision may conceal lack of safe, reliable service provision

There must be a safe environment for performing the procedure and close liaison with the appropriate clinical team(s)

4. Operators should participate in multidisciplinary case review

5. It is the duty of the operator to report any risk management concerns to the trust’s clinical
governance committee

Implementation of standards and quality benchmarks

Departmental leads should ensure the following:

Local agreement is reached amongst INRs about what service is provided

Process for maintenance of skills is in place (this may include updating practical skills by

spending time in other departments)

Mechanism for information to be available to clinicians on regular basis about when service

is available

Formal contracts exist with referring trusts

Agreed local protocols should be evidenced based (within limits of evidence)

National guidelines are adhered to

Appropriate case review and quality improvement processes are in place

Quality Benchmarks for Endovascular stroke therapy

1. Local protocol specifies reasonable indications for the procedure to be undertaken:

Proximal large artery occlusion in symptomatic territory plus symptom onset known and <5 hrs ago

Time window may be extended within local protocol if: Fails to respond to IVT/IVT CAI and imaging favourable (cf ESCAPE trial), time window greater if vertebrobasilar occlusion (specify in local protocol)

Account is taken in indication protocol of NIHSS and patient factors – co-morbidity At least 95% of patients treated with IA therapy should meet the institutional selection criteria 100% of patients have the required process and outcomes data entered into a trial or registry


Lysis to groin puncture time of <90 minutes

Puncture time to start of revascularisation: <45 minutes in at least 65%

Puncture time to end of revascularisation: median ≤60 minutes


Revascularisation achieved: at least 60% should have TICI grade 2b or better

Clinical outcome: at least 33% of patients with strokes treated by endovascular methods should have a mRS score of 0–2 at 90 days. Large artery occlusive stroke treated with IVT alone results in good outcome in trials in range 19-42%

SICH: rate should be <12% and all cases with symptomatic ICH should be reviewed

All deaths within 72h of procedure are reviewed

Anaesthetic management of endovascular treatment of acute ischaemic stroke

1. Recommendations for provision of care

The anaesthetic care of these patients should be supervised by Neuroanaesthetists with skilled assistance. It should be consultant led.

If a patient is sedated the responsible anaesthetist must be present in the procedure suite.

2. Recommendations for pre assessment

These patients present as time critical emergencies, akin to evacuation of an extradural haematomaor category 1 Caesarian section. Delays have detrimental effects on patient outcome. Time to groin puncture after IVT thrombolysis should be <90 mins. Pre assessment of the patient must be done as quickly as possible to avoid delay.

3. Recommendations for anaesthetic management

The choice of anaesthetic should be tailored to the individual patient based on neurological status, airway control and treatment plan in close communication with the INR.

Local anaesthesia should be aimed for, if feasible, in patients who are cooperative and can protect their airway.

General anaesthesia is recommended in patients with a reduced level of consciousness, uncooperative or agitated patients, those who cannot protect their airway or those already intubated.

Patients receiving local anaesthesia with sedation should be monitored and provision made to enable rapid conversion to a general anaesthetic if necessary.

4. Recommendations for airway management

Tracheal intubation is recommended for those patients with reduced level of consciousness, signs of brain stem dysfunction, those unable to protect their airway, with active nausea and vomiting before intervention and patients who become hypoxic or develop airway obstruction under sedation.

Supplemental oxygen administration is recommended during sedation.

All patients should be monitored with pulse oximetry and capnography.

FiO2 should be titrated to maintain SpO2 > 94%. Ventilation should be adjusted to maintain normocapnia under anaesthesia. Hypercapnia should be avoided in patients undergoing sedation.

5. Recommendations for haemodynamic management

Haemodynamic monitoring should include ECG and continuous blood pressure or, if noninvasively, measured at least once every 3 minutes.

Continuous invasive arterial monitoring is recommended for all interventional procedures as long as arterial cannulation will not delay intervention. The femoral artery cannulated by the neurointerventionist may sometimes be used to provide continuous arterial monitoring if necessary.

In patients having general anaesthesia systolic blood pressure should be maintained within 10% baseline with fluids and / or vasopressors. Suggested targets are >140mmHg but < 220mmHg (< 180mmHg in patients who have received IV tPA).

These blood pressure targets may need adjustment in communication with the stroke physician/INR.

6. Recommendations for Postoperative Care

Critical care should be contacted to establish the availability of level 2/3 facilities for postoperative care. However this should not delay the start of intervention. If necessary, alternative facilities can be sought whilst the procedure is performed.

Patients who have received GA should be managed on NICU or High dependency care / stroke unit postoperatively to continue physiological monitoring and neurological monitoring.

7. Protocols

Irrespective of anaesthetic technique the two most important anaesthetic goals are minimising any time delay and haemodynamic control. Institutional protocols can assist in the safe and timely delivery of care. These protocols should address:

Choice of anaesthetic agents

Timeliness of induction

Blood pressure parameters

Postoperative care

8. Audit

Institutions should audit their practice including type of anaesthetic, monitoring, timing, anaesthetic agents used and complications.

Models of safe endovascular stroke provision

1. Individual neuroscience centres provide safe endovascular stroke service for patients
admitted via their local ED/hyperacute acute stroke unit (HASU)

ADVANTAGES: more readily deliverable and fairly limited additional resource requirement

DISADVANTAGES: relatively few NHS stroke patients potentially eligible are able to receive endovascular therapy

2. Hub and spoke arrangement with acute stroke patients accepted by neuroscience linked
HASUs for thrombectomy from outside hospitals and where practicable development of the
“London model” of ambulance bypass to take suspected stroke patients to a comprehensive
HASU rather than simply the nearest ED.


i. far more NHS stroke patients able to receive endovascular therapy

ii. deliverable by existing services with additional resource requirement & pathway reconfiguration

iii. need for robust transfer arrangements and close monitoring of times taken will be a driver for service improvement to all stroke patients – as per IVT model

iv. synergy/quality improvement gains of building up existing neuroscience & stroke services rather than starting from scratch to develop a service

v. Enables INRs to work with other practitioners to expand service provision & training whilst mentoring & supporting safe thrombectomy practice


i. more formalised and complex service and governance arrangements needed (but that in fact may well be a good outcome)

ii. additional resource requirement as existing INR & diagnostic neuroradiology services require expansion

3. Local provision of IA thrombectomy by any “qualified” provider


most eligible NHS stroke patients able to receive endovascular therapy


i. Lack support services on site for safe IAT – Neurocritical care, INR & neurosurgery

ii. No robust trial evidence that non INR services can/will attain excellent technical, clinical and safety outcomes required for IAT to be SAFE & cost effective

iii. Multiple small volume centres with relatively inexperienced operators are extremely unlikely to achieve the same quality of outcomes as fewer high volume centres with experienced operators

iv. Current services are all INR based so nobody else presently “qualified”

v. Long lead time for non INR staff to attain skills i.e. become “qualified”

vi. Inadequate IAT numbers in most cases to maintain the skills of a whole team whose only Neurointerventional work is stroke thrombectomy (~40 cases per operator per annum & minimum team for 24/7 cover of 5 operators)

vii. Staff, equipment, consumable & imaging infrastructure costs for multiple small centres operating rotas are all greater than for centralised high volume services

Routes to service delivery


In view of Grade Level 1a evidence, networking of centres to deliver a regional or supraregional
24/7 service may be one approach, however in the short term, extended hours 7 days per week
may be the initial stepwise approach to accommodate such service reconfiguration.

There is currently insufficient stroke workload at present to attain/maintain skills unless operators
are active INR practitioners utilising the same skillset in the treatment of other intracranial
neurovascular conditions on a very regular (weekly) basis. One consideration may be to align both
neuro-interventional services for thrombectomy and for coiling (SAH) in a regional collaboration but
this will depend on local circumstances.

The urgency of the need for stroke intervention militates somewhat against supra regional INR
centre networking – though this might be feasible in London and potentially most of the other 6
metropolitan areas in England (e.g. where units are ≤1h ambulance transfer apart)

~90% of all strokes occur 0430-2300 & so could be managed by a thrombectomy service
operating 0700-2400 (see figure below)

However, of those strokes occurring 2300-0430, many/most will be “wake up” stroke & so a far higher proportion will be stroke of unknown time onset (SUTO) than is the case in strokes occurring 0700-2300. Currently no intervention is proven to be effective in SUTO.

So actually >95% of strokes eligible for thrombectomy under current scientific evidence would be covered by a 0700-2300 hours service

There is thus a clear sensible rationale for gradual extension of IAT cover towards 24/7 rather than attempting to go from “no service” to “24/7” immediately


1. National Clinical Guidelines for Stroke 4th Edition 2012. Royal College of Physicians London or available as an application for iphone or android
2. Mortimer AM, Marcus DB, Renowden SA. Managing hyperacute ischaemic stroke with
interventional radiology. BMJ 2012; 344: e857. PubMed 22331279
3. Broderick JP, Palesch YY, Demchuk AM et al. Endovascular therapy after intravenous tPA
versus tPA alone for stroke. N Engl J Med 2013; DOI: 10.1056/NEJMMoa1214300
4. Ciccone A, Valvassori L, Nichelatti et al. Endovascular Treatment for acute ischaemic stroke.
N Engl J 2013; 368: 904-13
5. Kidwell CS, Jahan R, Gornbein J et al. A trial of imaging selection and endovascular
treatment for ischaemic stroke. N Engl J Med 2013; 368: 914-23
6. Multisociety Consensus Quality Improvement Guidelines for Intra-arterial Catheter-directed
Treatment of Acute Ischemic Stroke, from the American Society of Neuroradiology,
Canadian Interventional Radiology Association, Cardiovascular and Interventional
Radiological Society of Europe, Society for Cardiovascular Angiography and Interventions,
Society of Interventional Radiology, Society of Neurointerventional Surgery, European
Society of Minimally Invasive Neurological Therapy, and Society of Vascular and
Interventional Neurology. J Vasc Interv Radiol 2013; 24:151–163
7. Sanyal R, Barrick J, Bhalla A et al. British Association of Stroke Physicians. Survey of
interventional treatments for stroke in United Kingdom. In J Stroke 2013; 8 Suppl A100:62-8.
doi: 10.1111/j.1747-4949.2012.00931.x. Epub 2013 Jan 7
8. National Institute for Clinical Effectiveness (NICE) 2013. Mechanical clot retrieval for the
treatment of acute ischaemic stroke. NICE international procedure guidance 458.
9. Pragmatic Ischaemic Stroke Thrombectomy Evaluation-(PISTE). Clinical identifier:
10. Intra-arterial Interventions in Acute Stroke Consensus Statements. NHS Improvement Stroke
11. Berhemer OA, Fransen PSS, Beumer D et al. A randomised trial of intra-arterial treatment
for acute ischaemic stroke. (MR CLEAN). NEJM 2015; 372: 11-20
12. Goyal M, Demchuk AM, Menon BK et al. Randomised assessment of rapid endovascular
treatment of ischaemic stroke. (ESCAPE). NEJM 2015; 372: 1019-30
13. Campbell BCV, Mitchell PJ, Kleinig TJ et al. Endovascular therapy for ischaemic stroke with
perfusion imaging selection (EXTEND IA). NEJM 2915; 372: 1009-18
14. Saver JL, Goyal M, Bonafe A et al. Stent retriever thrombectomy after intravenous t-PA vs. t-
PA alone in stroke. (SWIFT PRIME). DOI:10.1056/NEJMoa1415061
15. Jovin TG, Chamorro A, Cobo E et al. Thrombectomy within 8 hours after symptom onset in
ischaemic stroke (REVASCAT) DOI: 1056/NEJM0a1503780
16. THRACE. European Stroke Organisation (ESO) Conference 2015. Abstract 180. April 19th
17. Furlan AJ. Endovascular therapy for stroke. It’s about time. NEJM 2015 April 17.
19. Ahmad N, Nayek N, Jadun C, Natarajan I, Jain P, Roffe C. Mechanical Thrombectomy for
ischaemic stroke: First U K Series. PlosOne 2013; 8(12): 10.1371/journal.pone.0082218
20. of a stroke
21. Implementing the National Stroke Strategy- an imaging guide…/dh_085145.pdf
22. Bray BD, Campbell J, Cloud GC et al. Bigger, faster? Associations between hospital
thrombolysis volume and speed of thrombolysis administration in acute ischemic stroke.
Stroke 2013; ;44(11):3129-35
23. Meeting the future challenge of stroke. Stroke Medicine Consultant Workforce Requirements
24. Standards for providing a 24-h interventional radiology service.
25. Ganesalingam J, Pizzo E, Morris S, Sunderland T, Ames D, Lobotesis K. Cost utility
analysis of mechanical thrombectomy using stent retrievers in acute ischaemic stroke
Stroke; 2015: 46: 2591-2598

CS = Conscious sedation
ED = Emergency Department
GA = General anaesthesia
GMC = General Medical Council
IA = Intra-arterial
IAT = Intra-arterial thrombectomy
IMS 3 = Third interventional management of stroke trial
INR = Interventional Neuroradiologist
IV = Intravenous
IVT = Intravenous thrombolysis
mRS = modified Rankin score
NICE = National Institute for Health & Clinical Excellence
NIHSS = National Institutes for Health stroke scale
RCT= randomised controlled trial
SICH = symptomatic intracerebral haemorrhage
SITS = Safe implementation of treatments in stroke
tPA = tissue plasminogen activator