SYNOPSIS

Content and changes during the programme

The programme had two work packages (WPs) with matching objectives.

Work package 1

The aim was to determine the content, clinical effectiveness and cost-effectiveness of an enhanced Paramedic Acute Stroke Treatment Assessment (PASTA) trial to facilitate emergency stroke treatment.

The objectives were to:

1. develop an enhanced paramedic role for assessment of patients with acute stroke symptoms by a review of relevant literature and qualitative assessment of factors influencing the role from public and professional perspectives
2. examine the paramedic intervention by a cluster randomised trial of cost-effectiveness and qualitative process evaluation of professional and public experiences
3. report a within-trial economic evaluation of the enhanced role compared with standard care.

Work package 2

The aim was to determine the clinical effectiveness, costs, cost-effectiveness and affordability of delivering intra-arterial therapy (IAT) for acute ischaemic stroke patients in England.

The objectives were to:

1. develop a conceptual model of potential care pathways for IAT patients across NHS services, including pre-hospital, secondary and tertiary care settings
2. convert the conceptual model into a mathematical model, identify the evidence for parameterising key decision points and estimate outcomes
3. understand patient, public and relevant professional groups’ views on possible IAT service designs
4. estimate the effectiveness, incremental cost per quality-adjusted life-year (QALY) and other outcomes from an NHS and societal perspective of a national IAT service for stroke
5. develop an implementation plan for IAT in English stroke services that optimises access.

During the programme, there were a number of changes made in response to emerging evidence for treatment effectiveness, evolution of clinical services and challenges for trial recruitment. These are summarised below.

Changes to work package 1

1. The original proposal included a short phase to test the feasibility of delivering the intervention and data collection in a clinical setting. However, because of the logistical and training challenges created by designing and delivering a separate pilot study within part of a participating ambulance service, the objectives were incorporated into the main clinical trial as an internal pilot phase. This was approved by the Trial Steering Committee (TSC). Recruitment began in December 2015 and the pre-set pilot criteria were confirmed by the TSC as achieved in April 2016.
2. During the main PASTA trial phase examining the cost-effectiveness of an enhanced ambulance stroke pathway, there were delays in training sufficient numbers of intervention paramedics to achieve the planned recruitment target. In April 2017, it was agreed with the funder that the primary outcome should change from a health outcome [i.e. 3-month modified Rankin Scale (mRS) score] to a process outcome (i.e. administration of thrombolysis), as the latter measured the intended impact of the intervention, but required fewer patients to show a clinically important impact. The original sample size of 3640 patients was replaced by a new estimate of 1297. A major amendment was approved by NHS Ethics in October 2017.
3. During the parallel process evaluation, it quickly became evident that patients who had recently experience acute stroke were unable to provide views on the PASTA intervention that might inform its acceptability. After discussion with the Programme Steering Committee, it was agreed that attempts to identify patients for this purpose should cease, and notification was given to NHS Ethics.

Changes to work package 2

There were no significant changes to the planned model purpose or development, but the following aspects were altered in response to events outside the programme:

1. The systematic review of IAT effectiveness also included a trial sequential analysis to understand the impact of the most recent trials.1–3
2. NHS England issued guidance4 for commissioners during the programme (featuring early work undertaken in the programme), which pre-empted part of the intended dissemination activity. The dissemination focus changed to providing commissioners with directly relevant information about choices for their area via an online configurable tool: Interface for Thrombectomy Economic Modelling and outcomeS in stroke (ITEMS).

These changes were agreed by the Programme Steering Committee. There were no implications for research permissions.

Background

Stroke is the single largest cause of adult disability and the third leading cause of death in England, but outcomes are significantly improved when patients are quickly admitted to specialist care for time-critical treatments and multidisciplinary care.5,6 Acute stroke management across the NHS improved substantially following the publication of a National Stroke Strategy in 2007 and National Institute for Health and Care Excellence (NICE) guidance in 2008,7,8 but emergency provision of the only licensed emergency drug treatment has remained variable and below aspirational targets. Known as ‘intravenous thrombolysis’, effective treatment requires administration of intravenous recombinant tissue plasminogen activator to selected ischaemic stroke cases within 4.5 hours of symptom onset, thereby promoting breakdown of any thrombus responsible for a sudden reduction in cerebral blood flow. Earlier treatment is more likely to reduce future dependency, but there is also a 3% risk of deterioration as a result of symptomatic intracranial haemorrhage.9 As well as requiring rapid assessment, individual patients must be carefully selected based on a combination of clinical and brain-imaging information, which provides an indication of their potential to benefit from treatment.

At a service level, thrombolysis delivery is challenging because both brain imaging and specialist assessment must be rapidly available to confirm treatment eligibility, achieve optimal treatment outcomes and avoid harm. Despite wide dissemination of the National Strategy, NICE guidelines and corresponding national clinical guidelines,10 and significant reorganisation of services in some regions,11,12 the national audit has continued to show large variations in the rate and speed of thrombolysis delivery between services and diurnal variations within services.6,13 At the start of the programme in 2014, only 11% of total stroke admissions in the NHS were being treated against an aspirational target of 20%, with a median door-to-treatment time of 54 minutes, despite a target of < 40 minutes.6 This largely remained unchanged by the end of the programme in 2019, implying that further improvements are unlikely to be achieved by focusing solely on the process of care delivery within hospitals. Even if stroke patients are unsuitable for thrombolysis after rapid processing, they might still benefit from other aspects of early specialist management to avoid complications, such as intravenous blood pressure lowering or reversal of anticoagulation medication, to reduce the risk of intracerebral haematoma expansion.10

Ambulance stroke assessment

The NHS ambulance assessment of suspected stroke patients consists of initial symptom recognition using the Face, Arm, Speech Test (FAST),14 exclusion of hypoglycaemia and urgent transfer with pre-notification to the nearest hyperacute stroke unit (HASU) if onset is believed to be within 4 hours.10 Local pathway variations exist according to the location of the specialist HASU, but the role of the paramedic has fundamentally remained the same for 15 years. Despite proximity to the patient and audit data showing scope for improvement in overall service delivery, the pre-hospital content of the emergency stroke pathway and related training has not been further optimised for thrombolysis decision-making. There have been reports that additional pre-hospital-phase interventions can facilitate thrombolysis treatment, including multiprofessional workforce training,15 raising the service priority level for suspected stroke16 and personalised feedback to paramedics about care quality.17 However, studies were setting specific and/or observational, and generally described short-term improvements in thrombolysis-naive services.

In other specialties, evidence is increasing that imposing a structure on interactions within multidisciplinary teams at specific points along a clinical pathway has a major bearing on the efficiency of care delivery. Simple tools can standardise communication of key information and confirm whether or not essential tasks have been undertaken, including structured formats for paramedic handover to emergency department (ED) staff18,19 and multidisciplinary care process checklists for pre- or post-care delivery.20,21 Enhanced handover and team checklists might, therefore, be valuable during the specific scenario of assessment for thrombolysis eligibility, as well as improving access to other stroke treatments and organised stroke care.

In view of the potential for ambulance personnel to play a more significant role during the initial assessment of suspected stroke patients who may be suitable for thrombolysis, WP1 developed and evaluated an enhanced PASTA pathway.

Intra-arterial thrombectomy

Although clinical services were seeking effective implementation of thrombolysis provision, an evidence base was rapidly developing for a powerful additional treatment suitable for selected patients with moderate to severe ischaemic stroke as a result of large artery occlusion (LAO), known as IAT. Although thrombolysis reduces long-term disability, restoration of cerebral blood flow occurs in only 50% of patients and in only 10% with LAO,22 thereby limiting its effectiveness. During IAT, an arterial catheter is guided into the cerebral circulation by a trained interventionist to extract the thrombus directly, thereby achieving greater success in restoring the blood supply. To reduce disability, this must also be performed as soon as possible, usually within 6 hours of symptom onset and following initial treatment with thrombolysis.23,24

As the IAT procedure requires interventionists and facilities currently only available at regional neuroscience centres, the clinical pathway is more complicated than thrombolysis and the majority of IAT-eligible patients require rapid secondary transfer to the centre following initial assessment at a local HASU.25,26 Advanced symptom checklists for ambulance personnel have been developed in an attempt to identify patients who are more likely to have LAO for selective redirection, but, so far, these have not shown acceptable levels of accuracy for widespread clinical deployment.27,28

At the start of this programme, in 2014, there was no commissioned provision of IAT across > 120 HASUs. Critical issues were still unclear, including how many patients were suitable for IAT based on emerging trial data, the optimal configuration of HASU and IAT centres, and stakeholder preferences regarding the inevitable trade-off between possible health gains at a central site relative to additional travel distance and displacement. This information was essential for services and commissioners to prepare for IAT implementation, but required presentation in a format facilitating comparison of options in a local service context. Hence, WP2 sought to determine the clinical effectiveness, costs, cost-effectiveness and affordability of delivering IAT for acute ischaemic stroke patients.