Clinical specialist statement. Thank you for agreeing to give us a statement on your view of the technology and the way it should be used in the NHS.

Transcription

1 Clinical specialist statement Thank you for agreeing to give us a statement on your view of the technology and the way it should be used in the NHS. Healthcare professionals can provide a unique perspective on the technology within the context of current clinical practice which is not typically available from the published literature. To help you in making your statement, we have provided a template. The questions are there as prompts to guide you. It is not essential that you answer all of them. Please do not exceed the 8-page limit. About you[0] Your name[0]: Kennedy R Lees, Professor of Cerebrovascular Medicine, Consultant Stroke Physician & Clinical Director - Acute Stroke Unit & Cerebrovascular Clinic, Western Infirmary, Glasgow and University of Glasgow. Name of your organisation (if applicable): Safe Implementation of Thrombolysis in Stroke (SITS) group British Association of Stroke Physicians Are you (tick all that apply): - a specialist in the treatment of people with the condition for which NICE is considering this technology? - a specialist in the clinical evidence base that is to support the technology (e.g. involved in clinical trials for the technology)? - an employee of a healthcare professional organisation that represents clinicians treating the condition for which NICE is considering the technology? If so, what is your position in the organisation where appropriate (e.g. policy officer, trustee, member etc.)? - other? (please specify) Executive committee member of European Stroke Initiative (EUSI), the independent educational arm of European Stroke Council (ESC), European Federation of Neurological Societies (EFNS) and European Neurological Society (ENS), which is responsible for writing guidelines on stroke care in Europe and for organising educational activities in stroke. Executive International Steering Committee member and joint UK coordinator for the Safe Implementation of Thrombolysis in Stroke group (an academic collaboration). 1

2 What is the expected place of the technology in current practice? Patients with acute stroke are admitted to hospital as soon as possible, and wherever possible directly to a specialist stroke unit. Evidence for benefits of specialist care is compelling and so this forms part of government and professional guidance. Once admitted, patients undergo brain imaging with either CT or MR scanning to diagnose the stroke type (haemorrhage or ischaemia) and to assess aetiology, prognosis etc. Apart from use of alteplase, acute treatment consists of measures to maintain hydration, oxygenation and other physiological parameters within normal ranges. Hence, there is close, non-invasive monitoring in a high dependency but not usually intensive care setting for a period of hours to days. During this time further investigation will be undertaken to establish the cause of the stroke and to assess disability, to prevent and/or treat complications such as infection, aspiration and peripheral thrombosis, and to educate the patient and family. Patients with ischaemic stroke are commenced on aspirin and dipyridamole as antiplatelet agents, a statin to reduce cholesterol (and possibly for its anti-inflammatory effects) and antihypertensives even if blood pressure appears relatively normal. Specific causes such as tight carotid stenosis or cardiac embolism may be treated surgically. The timing of these various measures depends on extent of infarction, the level of dependency of the patient, the return of swallowing function and on the waiting time for investigations. In practice, a typical patient in UK will wait 1-2 days for brain imaging and will be started on aspirin within approximately 48 hours of admission; most units have an agreed policy for more urgent scanning in selected cases and this will be offered to 10-15% of admissions. Throughout the investigation and early treatment phase, the patient is assessed by rehabilitation staff and commences physiotherapy, occupational therapy and/or speech and language therapy. Therapy may continue for many weeks after stroke, often requiring transfer to a rehabilitation institution. Ultimately, patients either die, are discharged home or transferred to a nursing home for long term care. Variation in practice centres around 3 elements: first, urgency and extent of scanning varies. Some specialist stroke centres will offer immediate scanning to all patients. This is the norm across USA, most of Europe and Australia but is relatively uncommon in UK despite evidence that it is the most cost-effective approach. Most Uk hospitals use only plain brain CT imaging, but MRI is usually more informative and so major stroke centres will opt for MRI in a high proportion of cases either as first line investigation or after initial management directed by CT. CT perfusion is now being used in a small number of UK centres to establish the extent of at-risk tissue and to consider treatment outwith usual guidelines. Second, thrombolysis using alteplase for acute ischaemic stroke within 3 hours is available in approximately 30 UK hospitals, with interhospital transfer extending this to some other regions. UK practice lags behind many other countries in this regard. Third, intra-arterial treatment with thrombolysis is offered at some specialist stroke centres probably only a handful in UK especially for basilar artery occlusion. A few centres also offer hemicraniectomy as decompressive surgery for malignant middle cerebral artery occlusion a procedure recently recognised to be lifesaving and to assist with functional recovery. Thus, alteplase is the only option for the majority of patients with acute ischaemic stroke. No other measure reduces disability. Whilst intra-arterial treatment may be available in a handful of hospitals, the delays in arranging angiography, the procedure-related risks and costs, and the lack of confirmatory trial evidence beyond that from the PROACTII trial 1 render this an unattractive option for routine treatment. In future, it is possible that intravenous use of alteplase could be 2

3 followed by rescue therapy for those who do not respond, using intra-arterial thrombolysis or physical thrombectomy (eg with the MERCI retreival device). There is some debate amongst professionals on the place of alteplase in stroke. All professionals agree that it is effective and safe in qualified hands for the licensed indication. The debate has centred on 3 points: first, licensed use is restricted to patients up to 80 years old who may be treated within 3 hours of onset. Some groups feel this is too restrictive and wish to extend the time and age limits widely. The EMEA-mandated ECASS-III trial is testing for extension of the time window to 4.5 hours. Clinicians vary in their attitude to the age restriction: most apply a biological rather than arithmetic limit. Second, there was concern that safety of use outwith experienced centres may be lower. The recently published results of the SITS-MOST trial have now confirmed that routine use, even in UK, is safe (see below). 2 Third, many clinicians believe that treatment may be appropriate in additional groups of patients but disagree over how best to identify these. The third international stroke trial, IST-III, is based on the premise that since treatment is effective and safe in selected patients within 3 hours, then it should be reasonable to extend alteplase use to later treatment, to older patients and to inexperienced centres. It is recruiting slowly. Other observers favour an imaging-based approach to select patients in whom infarction is incomplete but in whom blood flow is demonstrably limited patients with evidence of persisting salvageable tissue rather than working on a strictly time-based approach. Treatment appears to carry similar benefits across the wide range of age and severity of stroke patients, but early treatment will have greatest benefit. Prognosis after stroke is worse in the elderly, the more severe strokes and in diabetics, but treatment per se does not interact with most factors. There is evidence of treatment interaction only for patients with prior stroke and/or diabetes; however these are cautions in the product licence. Prior use of aspirin has been questioned as a risk factor, but in a meta-analysis of all trials submitted to EMEA, aspirin use did not independently influence good outcomes or interact with treatment. It had a small effect on the risk of symptomatic intracerebral haemorrhage but this appeared only when treatment was delayed beyond 3 hours. Around 30% of patients are already taking aspirin when they suffer stroke, and almost all patients are given aspirin after ischaemic stroke. In practice, clinicians at present delay giving further aspirin until hours after alteplase use, when a further scan has excluded significant bleeding. Trial data from IST and CAST suggest that a delay of 24 hours in aspirin prescription would have no or insignificant (0.07%) effect on treatment benefit and would still be given earlier than in most non-thrombolysed patients in UK. Patients with especially large infarcts or in whom infarction is already well established at time of treatment are at increased risk of haemorrhagic transformation. These patients can be readily identified by clinical examination (eg an NIHSS score of 25 or more) and by brain imaging (hypodensity affecting more than one third of the MCA territory, or an ASPECTS score <7). These points are noted in the product licence. This is not a treatment for use in primary care or non-specialist secondary care: access to imaging, specialist understanding of stroke assessment and management, and access to appropriate specialist monitoring are needed; also, it is desirable to maintain an audit of local results, just as one would for surgery. The need to maintain expertise implies that only specialist centres will have sufficient throughput to staff rotas with qualified personnel. However, it is also clear that middle-grade medical staff and possibly even specialist nurses can be trained to 3

4 deliver thrombolysis safely and efficiently, and there is no reason why A&E medical staff should not work alongside specialist stroke teams to deliver treatment. There is wide variation in use of thrombolysis with alteplase for stroke in the NHS. For example, my colleagues and I have treated over 250 patients in the last 3 years in Glasgow, and isolated hospitals in Newcastle, London and Cambridge have each treated over 50, whereas many large cities have offered treatment to under 10 patients. Two factors account for this. First, whilst over 300 stroke physicians have been trained in use, many hospitals have been waiting until a full 24/7 service could be offered before agreeing to allow any treatment. It takes time to negotiate the combination of rapid assessment, immediate CT scanning and specialist stroke rotas. Second, results from SITS-MOST the safety monitoring study were published only in February The SITS-MOST study has shown that treatment is largely restricted to the licensed indications. 2 Age is the only criterion which is regularly disregarded in some centres: just under 10% of UK patients were aged years old. There is no evidence that age per se should be a barrier to alteplase use, however, despite the restricted licence. Time to treatment 70 exceeds 3 hours by minutes in a 60 small number of cases, though metaanalysis reassures us that alteplase treatment may be safe and beneficial up to at least 4.5 hours. These delays in AGE administration can arise from a number of practical issues over interpretation of CT scan, discussions with patients and relatives, venous access, portering etc. They also occur in clinical trials. Frequency Guidelines cover both use of imaging and use of thrombolysis: The Royal College of Physicians Guideline states: Brain imaging should be undertaken as soon as possible in all patients, within 24 hours at most of onset unless there are good clinical reasons for not doing so Brain imaging should be undertaken as a matter of urgency if the patient has inter alia indications for thrombolysis or early anticoagulation The SIGN guideline on Management of Patients with Stroke (1997) recommends that all patients with acute stroke should undergo CT brain scanning as soon as possiblepreferably within 48 hours-and no later than seven days. A local protocol for more urgent scans (e.g. patients receiving anticoagulant or recent thrombolytic therapy) should be available. The EUSI (European Stroke Initiative) guideline states: 4

5 Intravenous rtpa (0.9 mg/kg; maximum of 90 mg), with 10% of the dose given as a bolus, following by an infusion lasting 60 minutes is recommended within 3h of onset of ischaemic stroke (LoE I) The benefit from the use of i.v. rtpa for acute ischaemic stroke beyond 3 h after onset of the symptoms is smaller, but present up to 4.5 hours (LoE I) Intravenous rtpa is not recommended when the time of stroke onset cannot be ascertained reliably; this includes persons whose strokes are recognised upon awakening (LoE IV) Data on the efficacy and safety of any other intravenously administered thrombolytic drugs are not available to provide a recommendation The American Stroke Association guideline was reviewed in 2005 as follows: The recommendation for the intravenous administration of rtpa within 3 hours of onset of stroke in carefully selected patients should not be changed (grade A, no change from 2003). The evidence is strong that all delays in treating patients should be avoided (grade A, new recommendation). Although intraarterial thrombolysis alone or in combination with intravenous thrombolysis holds great promise, the use of these approaches is preferable in the setting of randomized clinical trials. A correction is needed in Table 7 of the 2003 Guidelines. Patients with an INR level of 1.7 or below can be treated with rtpa. The Royal College of Physicians Guideline states: Thrombolysis has the potential to improve outcome of patients with cerebral ischaemia,however it is a high-risk treatment and should only be administered by personnel trained in its use, in a centre equipped to investigate and monitor patients appropriately. Evidence from Phase IV studies on intravenous thrombolysis in North America has shown that unless the protocols for treatment are strictly adhered to outcomes are worse. The evidence for the benefits of intra-arterial thrombolysis remains limited. It recommends: Thrombolytic treatment with alteplase should only be given provided that: 1. it is administered within three hours of onset of stroke symptoms (unless as part of a clinical trial) 2. haemorrhage has been definitively excluded 3. the NINDS criteria have been met 4. the patient is in a centre registered with Safe Implementation of Thrombolysis in Stroke Monitoring Study (SITS-MOST) All of these guidelines have followed modern practice in their development. Evidence was graded according to its level. The contributors declared any interest. Essentially, the evidence derives from the main trials NINDS Parts A & B, ATLANTIS, ECASS-I and ECASS-II and from individual patient meta-analysis of those data. The data and the guidelines have been published in peer-reviewed journals including New England Jounal of Medicine, Lancet, Stroke and Cerebrovascular Diseases. The guidelines mirror the product licence for alteplase in stroke, except that these guidelines do not refer to an age restriction. In my view, we should offer intravenous thrombolysis to all patients who present to a hospital that has a specialist stroke service within 3 hours of stroke onset, provided that there was no pre-existing disability or contraindication and that immediate imaging has excluded both haemorrhage and extensive acute infarction. All terms of the product licence should be followed except that judgement should be applied 5

6 when considering if age should be a restriction. We can achieve thrombolysis rates of 5-10% in UK hospitals without substantial changes in practice. The advantages and disadvantages of the technology There is no alternative to thrombolysis for acute ischaemic stroke. In future, it may be possible to offer intra-arterial thrombolsysis, at considerable extra cost, inconvenience and possibly risk; and to consider physical removal of clot eg with a MERCI retrieval device. However, it is more likely that these measures would be reserved as rescue therapy for patients who fail to respond quickly to iv alteplase. The main requirement is for immediate imaging by CT scan. Scanning is already mandatory in all stroke patients, and guidelines exist for immediate scans in several indications; thrombolysis adds only one additional criterion and so should carry minimal cost implication. An HTA-funded report suggests that immediate scanning is more cost-effective for all patients, however. A second scan at hours is recommended to assess extent of infarction and/or haemorrhagic tranformation, to guide aspirin use and judge prognosis. This carries a small extra cost (perhaps 50) and is easily tolerated by most patients. Concomitant treatment with aspirin would be deferred for 24 hours: this carries no measurable adverse consequences compared to current practice (see above). The guidelines indicated in the product licence are sufficient. Extremely severe stroke and patients with extensive infarction on the initial scan should not be treated; conversely, mild stroke and/or improving symptoms should not be regarded as a barrier to treatment, nor should age per se prevent treatment in an otherwise healthy person. Treatment should be started as soon as possible after stroke onset and imaging, but should not be started much after 3 hours; treatment should be given at a dose of 90 mg/kg of alteplase up to a maximum of 90 mg, with 10% administered as an iv bolus and the remaining 90% over the following hour by infusion. The patient should be monitored in an acute stroke unit or similar high dependency setting for 24 hours, and should be rescanned hours after treatment. Physicians offering treatment should be stroke specialists or have been trained in use of thrombolysis for stroke, should be competent to read a CT scan and assess stroke severity, and should understand the natural course of the condition. Active measures including all usual specialist stroke unit care should be offered. All treated cases should be recorded prospectively for local and national audit purposes. There is a good case for considering additional imaging with either CT perfusion or MR diffusion/perfusion scans to identify patients outwith the 3 hour window or in whom onset time is uncertain, in whom salvageable penumbral tissue is still present and in whom the infarcted area is not excessive. Such patients may benefit from treatment. There is no justification based on current evidence for selecting patients more strictly otherwise who meet the licensed indications. The trials that established the benefits of alteplase for acute ischaemic stroke were conducted in stroke centres around the world that applied care practices at that time which are similar to conditions that pertain now in UK hospitals: discrepancies in care between UK and elsewhere were less apparent at the times of these trials. Quality of imaging and experience in reading CT scans is probably better now in UK than was available during the trials. In addition, the SITS-MOST trial has conducted an extensive audit of all cases treated in routine care across Europe over the last 3 years: the types of patients treated, their demography and the outcomes were closely comparable to the trials. UK patients tend to be more severely affected than those elsewhere in Europe and to be treated later probably 6

7 because we are more conservative in offering treatment and are less well organised at present a situation that the English Stroke strategy may be starting to change. The trials measured functional outcome on a disability scale the modified Rankin Scale. This is the appropriate measure for stroke outcome, since it encompasses all aspects of recovery including any deleterious effect caused by haemorrhagic transformation and/or death. Secondary outcomes in the trials included haemorrhagic transformations and deaths. Regarding the haemorrhages, only symptomatic heaemorrhages are of any consequence; small amounts of haemorrhagic transformation are common after ischaemic stroke and it is only those in which treatment converts the petechial haemorrhage into massive bleeding that cause concern. In practice even many of the so-called symptomatic haemorrhages are of limited consequence, because they occur in patients with massive ischaemic stroke who would suffer fatal brain oedema if they were not thrombolysed; and haemorrhagic transformation probably only occurs in presence of reperfusion, i.e. if treatment fails to open the vessel then harm is unlikely, if it opens the vessel too late then the harm suffered by the patient simply changes from oedema to haemorrhage. The modified Rankin Scale measures net benefit or harm. In practice, we have come to recognise that haemorrhages only matter if they are large occupying most of the territory of the infarct with mass effect and if they are accompanied by a sudden neurological deterioration usually measured as a 4-point worsening on the NIHSS. By this criterion, experience in several thousand patients has now taught us that such events are rare around 1.7%. Remote haemorrhage is uncommon. Other side effects have not been a problem. To a large extent, this good experience reflects the fact that alteplase has been a licensed drug for several decades and so its safety profile was already well known. Any additional sources of evidence The main evidence that will not have been available through systematic review will be that from the SITS register. This was published in the Lancet in Febrary I strongly recommend that this paper is consulted, as it provides the largest and most reliable source of information on safety and effectiveness of alteplase use in stroke. Background The aim of the Safe Implementation of Thrombolysis in Stroke- Monitoring Study (SITS-MOST) was to assess the safety and efficacy of intravenous alteplase as thrombolytic therapy within the first 3 h of onset of acute ischaemic stroke. Under European Union regulations, SITS-MOST was required to assess the safety profile of alteplase in clinical practice by comparison with results in randomised controlled trials. Methods 6483 patients were recruited from 285 centres (50% with little previous experience in stroke thrombolysis) in 14 countries between 2002 and 2006 for this prospective, open, monitored, observational study. Primary outcomes were symptomatic (a deterioration in National Institutes of Health stroke scale score of 4) intracerebral haemorrhage type 2 within 24 h and mortality at 3 months. We compared mortality, the proportion of patients with symptomatic intracerebral haemorrhage as per the Cochrane definition, and functional outcome at 3 months with relevant pooled results from randomised controlled trials. Findings Baseline characteristics of patients in SITS-MOST were much the same as those in the pooled randomised controlled trials. At 24 h, the proportion of patients with symptomatic intracerebral haemorrhage (per the SITS-MOST protocol) was 1 7% (107/6444; 95% CI ); at 7 days, the proportion with the same condition as per the Cochrane definition was 7 3% (468/6438; ) compared 7

8 with 8 6% (40/465; ) in the pooled randomised controlled trials. The mortality rate at 3 months in SITS-MOST was 11 3% (701/6218; ) compared with 17 3% (83/479; ) in the pooled randomised controlled trials. InterpretationThese data confirm that intravenous alteplase is safe and effective in routine clinical use when used within 3 h of stroke onset, even by centres with little previous experience of thrombolytic therapy for acute stroke. The findings should encourage wider use of thrombolytic therapy for suitable patients treated in stroke centres. Further to the above, I can offer some preliminary information specifically on over 600 UK patients treated with thrombolysis. This also comes from the UK-SITS register and thus has been independently monitored and was conducted without sponsor involvement. The average age was 67.4 years (SD12.9) and NIHSS score was 14.5 (SD6.1). About 95% of patients were treated within hours, and rates of treatment at weekends were about 60% of those on weekdays: in terms of service provision, even investment within social working hours would deliver considerable benefit. Outcomes reflect the higher NIHSS scores of UK patients and poorer outcomes usually seen in this country but are otherwise consistent with the excellent safety profile elsewhere in Europe: 43.5% good outcome, 20.6% TIME BY HOURS mortality and a symptomatic Each symbol represents up to 2 observations. haemorrhage rate of 2.3%. Implementation issues Alteplase has been marketed in UK for several decades and is widely available in hospitals. Stroke units have been established in almost every UK hospital. All stroke units have access to brain imaging, and early scanning has been enouraged by the Stroke Strategy and several government reports. Stroke physicians are in post in most hospitals, and the sub-speciality of stroke has been established with its own training programme overseen by the JCHMT and BASP. Specific training in thrombolysis use is desirable, and Prof GA Ford and I have together run 7 thrombolysis training days for UK stroke physicians under BASP/SITS auspices. We have plans to run a further larger training day in summer 2007 after NICE guidance is issued and then to roll out to self-sustaining regional programmes. Additional resources may be required within individual hospitals to facilitate 24/7 imaging, and /or hospitals may need to review their medical rotas to ensure that on-call staff have necessary skills, but these skills can be learned quickly at the existing thrombolysis training days, and the imaging facilities should be offered under existing stroke guidelines. Thus, resource implications are limited in light of the substantial health benefits that alteplase could offer. This has been recognised by the National Audit Office report on Stroke Care in England, which recommended that NICE provide guidance on how to deliver thrombolysis in practice. 1 Furlan A et al. The PROACT II study. JAMA Dec 1;282(21): Wahlgren, Ahmed, Dávalos, Ford, Grond, Hacke, Hennerici, Kaste, Kuelkens, Larrue, Lees, Roine, Soinne, Toni and Vanhooren for the SITS-MOST investigators. Thrombolysis with alteplase for acute ischaemic stroke in the Safe Implementation of Thrombolysis in Stroke- Monitoring Study (SITS-MOST): an observational study. The Lancet 2007; 369: