Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Acute Stroke service change in London-is it applicable elsewhere
- Paul D Syme (11 June 2009)
-
The London Stroke Strategy is comprehensive and will be a quantum leap forward
- Nick A Losseff, Diane Ames, Geoff Cloud, Gill Cluckie, Patrick Gompertz, Binnie Grant, Hugh Markus and Martin Brown (18 June 2009)
-
Fairness and False Accounting in Stroke Service Reorganisation
- David H Barer (26 June 2009)
-
Assisting PCTs in spending wisely and well on stroke services
- Nigel Dudley (30 June 2009)
-
Further doubts on the NAO’s stroke models
- Nigel Dudley (23 October 2009)
|
|
|||
|
Paul D Syme, Consultant Physician NHS Borders & University of Edinburgh
Send response to journal:
|
I support most of the views expressed by Sudlow and Warlow in their recent article "Getting the priorities right for stroke care BMJ 2009; 338: b2083" but I would like to make some comments. Firstly, I want to correct their statement regarding the number of patients admitted to the Borders General Hospital (BGH) during the Scottish Borders Stroke Study. The figure of 91% corresponds to the total First-ever-in-a-lifetime strokes (FES) reviewed in the BGH which included outpatients attending our neurovascular clinic in addition to inpatients. As shown in our paper 368 out of a total of 596 FES were admitted to hospital (62%) which is very similar to the Oxvasc study and reflects an active neurovascular clinic preventing stroke admissions. The outcome for inpatients was much worse with a hazard ratio for death at 28 days of 3.6 (95% CI 2.2 to 5.7). This demonstrates appropriate selection of patients requiring admission based on the severity of their stroke. The authors focus on the new proposals for Acute Stroke care in London and state that an emphasis on thrombolysis at the expense of standard Stroke unit care and aspirin would be counterproductive. Although I fully agree with this, having reviewed the new proposals for acute stroke care in London this would not appear to be the intention of this new service. The proposed changes will make 24/7 availability of thrombolysis a reality which will be a tremendous achievement. High risk Transient ischaemic attacks (TIA) as identified by the ABCD2 scale will also be reviewed within 24 hours and low risk TIAs within a week. This will also be a major step forward. Treating both TIA and Stroke as medical emergencies in this way will not only increase thrombolysis but will also allow secondary prevention to be started at an earlier stage and carotid endarterectomy to performed when it is most beneficial (within the first two weeks). Patients with the Malignant Middle cerebral artery syndrome are also more likely to be detected and treated and increased thrombolysis rates may reduce the incidence of this syndrome. In London, patients will still be treated in standard Stroke units for their rehabilitation. The authors suggest that the changes proposed in London could result in an increased number of admissions to hospital. This may be the case so it is important that TIA and minimally-disabled stroke patients are identified as part of this process and their admission prevented. The TIA clinics proposed in the London plans will be available 7 days a week. This may prevent the unecessary admission of this group of patients. Early supported discharge will also need to take place in London. However, the new London proposals may reduce length of stay which may reduce the number of beds required to safely treat and rehabilitate stroke patients. This could provide a more cost-effective service. The main issue I have with these proposals which was not discussed by the authors is why London have chosen a 72 hour stay in a Hyper-acute stroke unit. The evidence supporting this length of time in a stroke unit is weak and not all stroke complications appear in this early period. We have recently completed SIGN 108 "Management of patients with Stroke or TIA: Assessment, investigation, immediate management and secondary prevention" which provides evidence supporting some of the changes proposed in London. However, the challenges that we need to face in order to implement SIGN 108 in Scotland result mainly from the geography of Scotland which is predominantly rural. These challenges also apply to rural areas of England and Wales. In Scotland, although we have the same aims of the London stroke service, the methods required to achieve these aims clearly need to be different. Hyperacute stroke centres could be developed in cities in Scotland but the geography would prevent patients in rural areas outside of these towns reaching these units in sufficient time to have thrombolytic therapy. This will be the same in rural areas of England and Wales. District General Hospitals, like the Borders General Hospital, could provide this service but they are unlikely to have the trained staff to provide a 24/7 thrombolytic service service and setting up a limited service (9 am to 5 pm, Monday to Friday) in these hospitals although feasible, is unlikely to treat many patients. With limitations set by the infrastructure of District general hospitals, it is highly likely therefore that the proportion of patients receiving thrombolysis in rural areas of the UK will be much less than that achievable in London. Without an alternative, this would result in an inequity of provision of Stroke care between large towns and rural areas. Fortunately this scenario can be avoided. With the development of Telemedicine links it should be possible to share expertise across the country to allow all stroke patients access to expert-directed stroke care including thrombolysis. London should be congratulated on the changes proposed to their acute stroke service which will hopefully result in better access to specialist stroke teams and speedier treatment of stroke. However, these changes are unlikely to be applicable to the majority of the UK including most District General Hospitals in rural areas. In order to achieve the same provision of care (rapid access 24/7) in these settings, Telemedicine is likely to be pivotal . Dr Paul Syme, Bsc Hons MBChB MD FRCPEdin Consultant Physician NHS Borders, Part-time Senior lecturer University of Edinburgh, Principal investigator of the Scottish Borders Stroke Study, Chair of SIGN Guideline 108 Competing interests: None declared |
|||
|
|
|||
|
Nick A Losseff, Interin London Clinical Director for Stroke Portland House, London SW1, Diane Ames, Geoff Cloud, Gill Cluckie, Patrick Gompertz, Binnie Grant, Hugh Markus and Martin Brown
Send response to journal:
|
Cathie Sudlow and Charles Warlow (1) are quite correct that we need to balance provision within a comprehensive stroke service, and their concerns about the London Stroke Strategy will be listened to. However they assume that the benefits of the proposed hyperacute stroke units (HASUs) are limited to the delivery of thrombolysis. The hyperacute arm of the London Stroke Strategy is not just about thrombolysis but aims to provide emergency assessment of all suspected acute stroke patients (whether eligible for thrombolysis or not) by specialists in acute stroke, deliver emergency brain imaging reported by neuroradiologists (including MRI if needed) and to place patients immediately into an environment of careful clinical and physiological monitoring and support provided by stroke trained nurses and therapists, which will benefit all patients, not only those receiving thrombolysis. It is only through such systems that early intensive care and investigation can be implemented cost effectively. For example, immediate brain imaging has been shown to be the most cost effective approach for all stroke patients(2. It would simply not be possible to provide this standard of care 24 hours a day, 7 days a week in every existing stroke unit without a massive expansion in stroke trained ward staff and consultants. The strategy is therefore underpinned by issues specific to London, including its dense geography. Sudlow and Warlow’s concentration on the cost effectiveness of thrombolysis, ignores the much greater savings to be made from centralising hyperacute stroke services e.g. in the cost of consultant and junior staff cover and emergency imaging. The HASU concept has not been subject to a randomised trial, but the benefits shown in the trials of stroke unit care (which are predicated on rehabilitation as the intervention) and the prescription of aspirin or thrombolysis, do not on their own explain the alarming differences observed in case mix adjusted mortality after stroke found in the studies comparing outcome in the UK with that in other European countries, such as the BIOMED project. Mortality after stroke was significantly higher in stroke units in the UK (42% in one UK unit) than stroke units in other European countries (19% in France)(3). The difference was not explained by thrombolysis, but rather appeared to be associated with more active management of a wide variety of clinical and physiological parameters in the continental European centres (4). Within centre studies have also shown improved outcomes associated with intensive physiological monitoring compared to care on a conventional stroke unit (5). Although most data showing effectiveness of stroke units is from rehabilitation units, a large study from Italy demonstrated markedly improved outcome with acute stroke unit care, with most of the reduction in mortality being seen very soon after stroke (6). Currently less than 50% of stroke units described in the 2008 National Sentinel Audit provide the 5 characteristics deemed to constitute quality acute care, and the number of patients that receive the type of physiologic supportive interventions described in the American Heart Association Guidelines for Stroke, is simply unknown. Moreover, the UK contributed very few patients to the trials of stroke thrombolysis conducted in USA and Europe that provide the evidence base for this treatment. The introduction of HASUs in London and elsewhere will provide a unique opportunity for the UK to develop, and play a leading role, in hyperacute stroke treatment and research. This is only possible in units which have a large critical mass able to provide specialised imaging and expert clinical assessment on a 24 hour basis. Developing this research will present challenges, particularly in following patients after repatriation, but with appropriate investment is achievable. Sudlow and Warlow are concerned that expertise in good centres will be lost if they no longer provide hyperacute stroke care. However, this level of care is currently only available in very few centres. The current plans will dramatically improve stroke unit care in many district hospitals leading to significant improvements in care during both the subacute and rehabilitation phases. Establishing close links with HASUs and surrounding district hospitals, with consultants working across Trusts as is already starting to happen, should lead to improved links and truly integrated networks. One of the advantages of London is that NHS Trusts are often geographically close to each other, providing the opportunity to rationalise services so long as the expert staff are willing to work at another site (e.g. by consultants joining on the HASU attending rota) for the benefit of patients. We await the decision of the JCPCT on the suggested London designations, but it is noteworthy that many organisations have already come together to embrace the proposals with some forming larger comprehensive stroke services and moving expertise from two previous operating hyperacute sites to one HASU, with shared protocols for transfer to the local stroke unit for further care if needed. These initiatives show a dedication to patient care and organisational altruism. In addition stroke networks are working closely to learn from each other and set up functioning communities of providers to meet the rigid performance standards required in the new system. The stroke reforms are part of Lord Darzi’s larger vision and constitute a quantum leap if achieved. Although some details remain to be finalised, we are signed up and committed to this change that will improve patient care greatly. Rehabilitation is a recognised and significant part of this strategy. There is one opportunity to get this right and transform stroke care in London to provide the best comprehensive stroke services in the world. It would be a disaster for patients if the proposals were derailed by the resistance to change that has previously characterised UK stroke services. Dr Nick Losseff – Consultant Neurologist and Interim London Clinical Director for Stroke. Dr Diane Ames – Consultant Stroke Physician and Joint North West Thames Clinical Lead for Stroke Dr Geoff Cloud – Consultant Stroke Physician and South West Thames Clinical Lead for Stroke Gill Cluckie – Stroke Clinical Nurse Specialist and South East Thames Clinical Lead for Stroke Dr Patrick Gompertz - Consultant Stroke Physician and North East Thames Clinical Lead for Stroke Binnie Grant – Stroke Co-ordinator and Joint North West Thames Clinical Lead for Stroke Hugh Markus, Professor of Neurology and Clinical Lead, South East England Stroke Research Network Martin M Brown, Professor of Stroke Medicine and Clinical Lead, Thames Stroke Research Network References 1. Sudlow S, Warlow C. Getting the priorities right for stroke care. BMJ 2009;338:1419-1422 2. Wardlaw JM, Seymour J, Cairns J, Keir SL, Dennis MS, Sandercock PAG. Immediate Computed Tomography Scanning of Acute Stroke Is Cost-Effective and Improves Quality of Life. Stroke 2004; 35: 2477-2483 3. Wolfe CDA, Tilling K, Beech R, et al. Variations in death and disability from stroke in western and central Europe. Stroke 1999;30:350- 356 4. Bhalla A, Tilling K, Kolominsky-Rabas P,et al. Variation in the management of acute physiological parameters after ischaemic stroke: a European perspective. Eur J Neurol. 2003;10:25-33 5. Cavallini A, Micieli G, Marcheselli S, Quaglini S. Role of monitoring in management of acute ischemic stroke. Stroke 2003;34:2599- 2603 6. Candelise L, Gattinoni M, Bersano A, Micieli G, Sterzi R, Morabito A; PROSIT Study Group. Stroke-unit care for acute stroke patients: an observational follow-up study. Lancet. 2007 Jan 27;369(9558):299-305. Competing interests: None declared |
|||
|
|
|||
|
David H Barer, Cons / Prof in Stroke Medicine QUeen Elizabeth Hospital, Gateshead NE9 6SX
Send response to journal:
|
Cathie Sudlow and Charles Warlow [1] have efficiently demolished the spurious health economics evidence behind the NAO report [2], used to justify the heavy emphasis on ‘hyperacute stroke care’ in the National Stroke Strategy (NSS), and particularly the plans for reorganising stroke services in London [3]. No-one would deny the benefits of the NSS in raising the profile of the UK’s Cinderella stroke services and providing much needed resources, but it is essential to apply not only sound evidence but principles of fairness and clinical common sense, and to think through the likely effects on real patients of separating hyperacute care from other parts of the service. It is natural to want to emulate the glamour of modern high-tech treatment for patients with acute coronary disease by setting up hyperacute stroke centres, but the analogies are superficial. The facilities required by stroke patients are completely different and there has been no convincing reassurance that the focus on hyperacute care for a small minority will not harm care for the majority. The fact is that, however worthy the motives, inadequate research evidence has been puffed up to serve a political agenda: • We have no idea why studies have consistently shown higher fatality rates for stroke in UK hospitals than in other European acute care centres. This is certainly not due to differences in provision of ‘hyperacute treatment’ as currently understood, as the studies were done in the pre-thrombolysis era. Do rigorous epidemiologists not object to the use of case series from self-selected hospital units to draw conclusions about the comparative cost-effectiveness of stroke care in different countries and thus guide national policy? • Acute stroke unit care should undoubtedly include close monitoring for and correction of physiological insults which could worsen brain damage, but it is facile to attribute huge variation in outcome to differences in the intensity of such monitoring [4, 5]. There is certainly no evidence that effective care in the first 72 hours requires 3.5 wte nurses/bed, as recommended in the London stroke standards [3] – nearly 10 times the national average for acute stroke units [National Sentinel Audit 2009]. Such staffing levels could never be achieved nationwide without cutting resources for large numbers of patients treated on non-hyperacute units. • Not only is the ‘hyperacute hype’ based on misinformation – there is no evidence that intravenous thrombolysis saves lives – and the NAO health economic calculations wrong [1], but current knowledge does not allow true estimates of the overall cost-effectiveness of either thrombolysis or ‘stroke unit care’. We cannot do the sums while trial outcomes are reported in terms of the proportion of patients dying or recovering to full independence - thresholds that the majority of stroke patients are never likely to cross, with or without treatment [6]. Thus large numbers of trial patients are unable to contribute their data to estimates of the overall benefits or harms of these treatments. • More perniciously, the inadequacy of present knowledge excludes many from treatment itself, such as the 40% of stroke patients aged over 80, for whom alteplase is not yet licensed. How will hub-and-spoke services cater for such people and deal with their frustration when they are denied treatment at the centres of excellence? Both clinical trials and common sense indicate that something can be done for almost anyone with a stroke, by adopting a problem solving approach based on realistic goal setting with individual patients – the foundation of multidisciplinary stroke unit team care. Expert teams take years to build up but can be destroyed overnight by thoughtless management decisions. London clearly has special problems and a radical solution may be appropriate, but it would be disastrous to impose such a politically driven, top-down approach in places which have struggled - with far more meagre resources - to develop comprehensive, integrated stroke services, from which no-one need be excluded. 1. Sudlow C, Warlow C. Getting the priorities right for stroke care. BMJ 2009;338:1419-1422 2. Department of Health. Reducing brain damage: faster access to better stroke care. London: National Audit Office, 2005. www.nao.org.uk/publications/0506/reducing_brain_damage.aspx 3. Healthcare for London. Stroke strategy for London. 2008. www.healthcareforlondon.nhs.uk/assets/Publications/Stroke/London20stroke20strategywebversionFINAL.pdf 4. Bhalla A, Tilling K, Kolominsky-Rabas P,et al. Variation in the management of acute physiological parameters after ischaemic stroke: a European perspective. Eur J Neurol. 2003;10:25-33 5. Davis M, Hollyman C, McGiven M, Chambers I, Egbuji J, Barer D. Physiological monitoring in acute stroke. Age Ageing 1999 (Suppl): 33 6. Berge E, Barer D. Could stroke trials be missing important treatment effects? Cerebrovasc Dis 2002;13: 73-5 Competing interests: None declared |
|||
|
|
|||
|
Nigel Dudley, Consultant in Elderly / Stroke Medicine St James's University Hospital LEEDS LS9 7TF
Send response to journal:
|
The fact that Cathie Sudlow and Charles Warlow show that there are problems with the NAO’s arithmetic - that interestingly have not been challenged after two weeks - should be of concern.[1] The NAO’s 2005 stroke report outcomes and savings figures were republished in the Committee of Public Accounts own July 2006 report; that report was made “to the House” so if concerns were suspected about the figures the Committee should have been warned so as to avoid the possibility of making misleading claims in the report. Those reports have been driving stroke service funding decisions and the focus of the National Stroke Strategy on acute care. It is hard to believe that with known NNTs for the effects of thrombolysis and stroke units of the variety pointed out in the tables in Sudlow and Warlow’s paper that it cannot have struck anyone in either the Department of Health or the NAO before July 2006 when the Committee reported to the House that those NAO report outcome and savings figures were inaccurate as demonstrated by Sudlow and Warlow. The two tables in Sudlow and Warlow’s paper should now aid all PCTs in England in making far better choices around how to spend the taxpayers’ money wisely and well rather than relying on either the NAO or Committee stroke reports’ figures and recommendations. The focus of time, energy and resources over recent years on access to acute imaging and thrombolysis has been detrimental to rehabilitation developments as shown by the 2008 Royal College of Physicians audit findings in relation to Early Supported Discharge and Community rehabilitation teams. NHS London’s decision to spend £21 million of taxpayer’s money “for the acute hospital costs” and just “£1 million for rehabilitation and community care costs as a result of changes in the acute system” [1] looks rather odd when the Department of Health’s own excellent 2007 Impact Assessment work shows that net benefits of the National Stroke Strategy are delivered by Early Supported Discharge and community rehabilitation. At present NHS London would not appear to be commissioning a comprehensive stroke service for the capital. It is possible that there is yet more funding to come for stroke services in London that will soon be announced in July when decisions are made about the location of the eight hyperacute units. The longer term rehabilitation requirements of the local 7.2 million NHS London population do need to be addressed as clearly shown by Sudlow and Warlow’s paper. [1] Sudlow C, Warlow C. Getting the priorities right for stroke care. BMJ 2009;338:b2083 Competing interests: None declared |
|||
|
|
|||
|
Nigel Dudley, Consultant in Elderly / Stroke Medicine St James's University Hospital, LEEDS. LS9 7TF
Send response to journal:
|
Earlier this summer, Oxford’s Department of Public Health published stroke statistics on behalf of the British Heart Foundation and Stroke Association showing that the total economic costs of stroke to the UK in 2006/07 were £4.5 billion.[1] The NAO had claimed that stroke costs to the economy in 2005 in England were £7 billion a year, with direct care costs to the NHS of £2.8 billion.[2] The research team that was commissioned by the NAO for its 2005 economic modelling work published a claim earlier in 2009 that economic costs to the UK in 2005 were £8.9 billion a year.[3] These differences between the Oxford and NAO estimates on stroke costs are important given that after four months the NAO has failed to make any response to the BMJ article by Sudlow and Warlow that raised concerns about its arithmetic, modelling and the transparency of the work. A possible reason for the silence could be that the NAO did not retain full details of the model as explained to the Information Commissioner and outlined in paragraph 35 of the Information Commissioner’s Decision Notice FS50180545.[4] Similarly, the research team that was commissioned by the NAO to carry out the modelling work failed to retain the spreadsheet of the model containing the calculations that gave the 550 deaths avoided figure in the 2005 report (paragraph 47 of FS50180545). Given that the NAO is intending to publish a progress report in Winter 2009 [2], it should be of concern to the public and to MPs - particularly the Members of the Committee of Public Accounts who scrutinised the November 2005 NAO stroke report at its February 2006 evidence session - that the NAO has remained silent for the four months since Sudlow and Warlow first raised their concerns. Do stroke units really prevent more deaths (550) than enable patients to recover independence (205) if an additional 25% of 82,138 stroke patients [5] pass through such units? For every 33 patients treated in a stroke unit an extra one survives; for every 20 treated, an extra one patient is discharged independent. Those numbers needed to treat (NNT) with stroke unit care are the ones stated in paragraph 76 of the Department of Health’s Impact Assessment [6] that accompanied the December 2007 National Stroke Strategy. These NNTs of 33 and 20 suggest that 622 deaths would be prevented and 1026 would recover to independence with stroke unit care. Does thrombolysis really prevent 55 deaths if the rate of thrombolysis increased from 1% to 20%? [7] Surely the research studies carried out on Alteplase show that thrombolysis is an autonomy saving treatment, not a life-saving, death-preventing treatment as claimed by the modelling work. Was the total economic cost of stroke in England in 2005 really £7 billion with £2.8 in direct care costs or have Oxford researchers on behalf of the BHF and Stroke Association grossly underestimated the cost of stroke to the UK in 2006/7? Decisions made by those commissioning health services and responsible for getting the priorities right in stroke service developments depend on reliable and credible information. The BMJ published concerns do need now to be answered by the NAO and its supporting research team. [1] Stroke Statistics 2009. www.heartstats.org/datapage.asp?id=8615 [2] Work in Progress: Stroke Care: a progress report. www.nao.org.uk/publications/work_in_progress/stroke_care_a_progress_report.aspx [3] Saka O, McGuire A, Wolfe C. Cost of stroke in the United Kingdom. Age and Ageing 2009;38:27-32 [4] Decision Notice FS50180545. www.ico.gov.uk/upload/documents/decisionnotices/2008/fs_50180545.pdf [5] King’s College London. Economic burden of stroke in England. National Audit Office, 2005 - table 7, paragraph 52, page 20. [6] Department of Health. Impact Assessment. A new ambition for stroke. Department of Health, 2007 [7] King’s College London. Economic burden of stroke in England. National Audit Office, 2005 - paragraph 56, page 21 Competing interests: I have raised concerns about the NAO figures and effects on stroke service development and investment choices since 2005. |
|||