The evolution of stroke units—towards a more intensive approach?
S. Sinha , E.A. Warburton
DOI: http://dx.doi.org/10.1093/qjmed/93.9.633 633-638 First published online: 1 September 2000
Introduction
The last five to ten years have seen an intensive research effort to find novel treatments for acute stroke with many more large trials of thrombolytic1,,2 and neuroprotective agents.3,,4 Unfortunately, none of these treatments have proven effective enough to recommend routine use in acute stroke. Although results from thrombolysis trials are promising,1,,2 uncertainty persists about more widespread usage, and it is likely that only a small proportion of stroke patients will ever be eligible for treatment because of the short time window for triage and treatment.5,,6 Perhaps the most significant advance in stroke management therefore has not been pharmacological, but concerns the process of care for stroke patients, with convincing evidence that changing the approach to the way stroke patients are managed has a beneficial impact on both mortality and morbidity.7 The concept of organized care on geographically‐defined units has given rise to a more disease‐specific approach to the management of stroke. In practical terms, this has cut across the more generic approach to general medical admissions within acute hospitals, particularly in the UK, and creation of stroke units often requires a substantial reorganization of beds within hospitals. This article aims to define which aspects of stroke unit care make the difference and how the process of care for stroke patients may evolve further in the future.
The evidence for the efficacy of stroke units
In 1993, results from all existing randomized controlled trials comparing the outcome of patients managed in defined stroke units with outcomes from conventional settings of care were examined in a meta‐analysis.8 This showed that care on a stroke unit reduced mortality by 28% and also reduced the risk of patients requiring institutionalized care at a median of 12 months after stroke. Subsequently, the Stroke Unit Trialists' Collaboration within the Cochrane Collaboration has been formed, to keep an updated database for all randomized controlled trials of stroke‐unit care.7,,9 Analysis from this larger dataset has confirmed the original findings that care within stroke units reduces mortality (19% over the first 12 months) and furthermore reduces dependency and disability amongst survivors.9 This data immediately raises important questions about this process of service delivery, namely what exactly is a stroke unit, do all strokes benefit, and what precise interventions on a stroke unit make the difference over conventional management?
What does the term ‘stroke unit’ mean?
The term stroke unit means different things to different people. To some it is analogous to a coronary care unit, offering a more ‘intensive care’ setting for acute stroke. To others, a stroke unit is a dedicated rehabilitation unit often providing many weeks of remedial therapy to stroke patients. What are the characteristics of stroke units included in the trialist's collaboration? In the original stroke unit overview, trials were included that admitted stroke patients to stroke units anywhere between 72 h and 2 weeks following acute assessment. From the collaborative data, those with an acute admissions policy (within one week) have been compared to trials where admission was delayed until 1–2 weeks.7,,10 Both show improved outcomes when compared to a standard service, with no significant differences between them.
Trials have also been assessed according to their discharge policies—i.e. one trial with a one week duration, six with discharge limits set between 4–16 weeks and twelve trials with no discharge limit.7,,10The results show that the effectiveness of stroke units is apparent where the duration of admission lasts several weeks, and is sustained after discharge to the community.12 Furthermore, the units included were situated in different hospital departments—general medicine, medicine for the elderly, neurology and rehabilitation medicine—but there is no evidence that this is a factor determining the effectiveness of stroke unit care.7,,10
Which types of patient benefit?
Should all stroke patients be admitted to a stroke unit? The stroke unit trials have been analysed within three subgroups stratified according to age (<75 years >75 years), sex and stroke severity (i.e. mild, moderate and severe), and the efficacy of the stroke unit care was found not to vary significantly.10More specifically, patients with mild strokes gain no benefit in terms of increased survival, but more of the survivors gain physical independence. Patients with strokes of moderate severity benefit in terms of increased survival and there are an increased number of independent survivors. Those with severe strokes show the greatest apparent benefit from the stroke unit care in terms of survival, with a small increase in the numbers of independent survivors. A caveat to these results, of course, is that most trials used some selection criteria and excluded patients with very mild and very severe strokes from their trials. Can the results from such trials be reproduced in routine practice? A recent study suggests that they can—the greatest benefit being in those who had previously lived independently,11 although those who had impaired consciousness at admission benefited least.
The crucial common factor identified in all trials is that care within a stroke unit is administered by a multidisciplinary team of people, all with an interest in stroke disease. This usually comprises a physician, nurses, physiotherapists, occupational therapists and speech and language therapists, with support from other disciplines, and it is perhaps this feature that gives stroke units effectiveness over a period of weeks when compared to the standard service. Generally this was achieved in the trials by reorganization of existing resources rather than increased staffing within the stroke units.
Why do stroke units reduce mortality?
What is it that this multidisciplinary approach to stroke care does to reduce the death toll from stroke? Secondary analyses of data from the stroke unit trials have analysed both the timing and cause of death in those dying within the first few months following stroke. Despite problems with the accuracy of documented cause of death, and lack of data from post mortems, this has revealed some interesting results.14 In both the stroke unit and conventional management group, there is a rapid rise in the number of deaths during the first 6 weeks, but after that it declines. The apparent difference in death rate between the stroke unit and control groups occurs largely between 1 and 4 weeks after the stroke. Deaths after these points are again similar between the two groups. Deaths occurring within a week of stroke tend to be direct neurological sequelae, and those beyond the first 6 weeks due to recurrent vascular events.15,,16 However the period between 1 and 4 weeks is primarily when most of the medical complications of stroke occur, and when most of the deaths are attributable to the medical sequelae (e.g. pneumonia and PE). A major factor in the success of stroke units therefore is likely to be that these medical complications are better managed. Specific measures suggested are: (i) a more intensive approach to the assessment and management of dysphagia that may reduce the risk of aspiration and subsequent pneumonia; (ii) improved management of incontinence reducing the need for urinary catheters and incidence of UTI; (iii) the earlier detection and more intensive treatment of infections in general; and (iv) earlier mobilization that may help reduce the risk of thromboembolic complications. Hence it would seem sensible that any stroke service (defined stroke unit or not) should develop guidelines and protocols for the detection and management of these common complications.
Why do stroke units reduce dependency?
How can stroke unit care reduce eventual disability after stroke? Within the stroke unit trials, almost all used some form of dependency score, and therefore it has been possible to document how many survivors were either dead or dependent. Not all of the trials used assessors who were ‘blind’ in terms of the study, hence introducing a risk of detection bias. However, analysis of the whole group as well as trials where there were blinded assessors showed that the patients in the stroke unit group were less likely to be dependent at the end of follow‐up than were those managed in a conventional setting.10Excluding trials with different periods of follow‐up did not alter these results; nor did different randomization procedures or looking at different models of stroke unit care.7,,10 In a recent study, factors differing significantly in the stroke unit from the general ward were: treatment (increased use of oxygen, heparin, intravenous saline and antipyretics), education, integrated physiotherapy and nursing and earlier mobilization.13 A key feature may be that patients on a stroke unit receive more intensive remedial therapy, particularly physiotherapy and occupational therapy, because the stroke team nurses and in some instances, carers, become more involved with the rehabilitation process.13,,14
In general, there is no consensus as to when physiotherapy should start, what the optimum intensity is and how long it should continue. Certainly stroke units are more conducive to earlier assessment and continuity of therapy input. There is some evidence that more intensive physiotherapy does improve outcome,17–,19 although there is debate as to whether any additional benefit is sustained beyond the first few months.19 Similar issues exist within speech and language therapy, and the move towards the more intensive management of dysphagia is tending to take speech therapists away from aphasia work.20 Perhaps a major benefit arising from the development of stroke units will be that carefully designed controlled trials of different kinds and patterns of therapy will be able to be more rigorously tested.
Why are stroke units difficult to set up?
It is now seven years since the publication of the meta‐analysis showing the efficacy of stroke unit care. However, the latest surveys of stroke services in the UK showed that in general stroke services were still poorly co‐ordinated.21–,23 In another recent survey, only 50% of departments responding to a questionnaire were working in a hospital with a stroke unit.20 In any hospital, perhaps the first requirement is to find a physician to run the stroke service and in particular a stroke unit. In many hospitals, the integration of geriatric medicine with general medicine has left geriatricians in particular with little time to develop stroke services. Within neurology, the demands of general neurology leave neurologists with little scope to take on additional stroke work. At the moment there are few posts in stroke medicine per se, but the development of subspecialty training programmes in stroke medicine and the creation of the Association of Stroke Physicians in the United Kingdom may improve this. Creation of geographically‐defined stroke units that are not age‐related often requires reorganization within medical directorates, and can cause problems where other parts of the general medical service are more age‐related.
Should stroke units be more ‘intensive stroke care units’?
Can the process of stroke care be extended to include a more intensive approach within a stroke unit during the acute period following the stroke onset? Can more intensive monitoring and aggressive management of the ‘extracranial’ parameters provoked by stroke produce further improvements in mortality and morbidity? In some centres, particularly in the US and Europe, patients are admitted to stroke units with facilities for intensive monitoring and interventions are made to correct abnormalities.24Modalities measured include blood pressure, temperature, oxygen saturation, cardiac rate and rhythm, and in some units cerebral oedema is actively managed with mannitol or glycerol.24,,25 To date there have not been any randomized controlled trials of such units, which by definition, require increased staffing levels and are more expensive to run than conventional stroke units. The rationale for intensive monitoring only exists if any resulting interventions are going to influence the eventual outcome. Evidence for the beneficial effect of individual interventions in stroke patients is lacking, although data from animal models of stroke suggest that optimizing physiological parameters may be inherently neuroprotective. For example, hypoxaemia has been shown in animal stroke models to increase the amount of cerebral damage.26 Hypoxaemia is common following stroke, due to central effects on the control of respiration and from the systemic complications of aspiration and pneumonia.27 Continuous monitoring of oxygen saturation by pulse oximetry is simple and non‐invasive, and it would seem reasonable to monitor and maintain the oxygen saturation above 95% by nursing patients sitting up and by using supplemental oxygen if necessary.30,,31 However in animal stroke models, excessive oxygen administration tends to worsen ischaemic damage, possibly due to the generation of free radicals and enhanced lipid peroxidation.28 As yet there is no clear evidence that routine supplemental oxygen improves eventual stroke outcome in patients.29
Similarly, it is known that pyrexia following stroke is an independent predictor of a poor outcome.32 This is not solely the result of infective complications, but is also a direct central effect of the cerebral ischaemia. In animal models, cerebral damage is worsened by hyperthermia, which increases the number of ischaemic depolarizations, causes exaggerated oxygen radical production, more extensive blood‐brain barrier breakdown and worsening of cytoskeletal proteolysis.35,,36 Conversely, hypothermia in animal stroke models protects the ischaemic cortex from disruption of ionic homeostasis provoked by ischaemia.33,,34 This animal work suggests that post‐stroke pyrexia should be treated with antipyretics;37however, there is no convincing evidence that hypothermia should be induced in acute stroke.
Hyperglycaemia is also an independent poor prognostic factor following stroke.38 Randomized controlled trials are in progress to determine whether regular monitoring and treatment of hyperglycaemia with insulin will have a favourable impact on eventual outcome. A pilot study has already demonstrated the feasibility of treatment of mild‐to‐moderate hyperglycaemia.39
What to do about blood pressure after stroke is not clear.40 Blood pressure is usually elevated in the acute phase, and resolves spontaneously within 7 days.41,,42 There are changes in diurnal blood pressure, particularly in cortical strokes, where the central modulation of blood pressure control is altered due to diminished cortical baroreceptor sensitivity.43 Theoretically, untreated hypertension may worsen cerebral damage by increasing the production of vasogenic oedema or provoking haemorrhage. Conversely, even modest reductions in blood pressure may critically reduce cerebral perfusion pressure and worsen cerebral ischaemia where there has already been loss of autoregulation of cerebral blood vessels.45,,46 At the moment there is no clear rationale for intervention with antihypertensive agents in the first 24–72 h following stroke unless the blood pressure exceeds 220 mmHg systolic and/or 120–130 mmHg diastolic.44 Hypotension is less common after acute stroke but can occur, particularly as a result of dehydration,47 and is defined as cerebral perfusion pressure <80 mmHg in head injury patients.48
Continuous cardiac monitoring has shown the existence of transient arrhythmias, repolarization anomalies and conduction delays associated with cerebral damage.49,,50 These are generally transient and are rarely markers for sudden death.49 Whether it is worth monitoring all patients in the acute phase is not known.
Based on the above, many stroke units have devised pragmatic protocols in order to introduce some multimodal monitoring in the acute aftermath of stroke. A sample protocol might include guidelines about treating pyrexia, maintaining oxygen saturations above 95%, treating random glucose levels above 7 mM and keeping the blood pressure above 90 mmHg systolic and below 220/120 mmHg. It will be interesting to see whether patients randomized to such ‘acute protocols’ will do better than those managed more conventionally. There is some evidence that the important treatment factors in stroke unit care contributing to overall benefit are more intensive treatment of pyrexia and hyperglycaemia, and less variation in systolic and diastolic blood pressure.13 Specific measures to reduce intracranial pressure and cerebral oedema following stroke are unproven, and cannot be recommended routinely.51,,52 Whether such measures have a role in very large middle‐cerebral‐artery territory strokes is not clear, and again has not been the subject of randomized trials.
Implications of thrombolytic therapy for acute stroke
More widespread adoption of acute assessment protocols and triage for thrombolysis will certainly lead to a more intensive approach to acute stroke, beginning with the response to the onset of stroke symptoms in the community. In hospitals where such protocols for thrombolysis are in place, only 6% of acute stroke patients are actually being thrombolysed.54 This is primarily due to delays in presentation, even with patient education programmes and co‐operation from ambulance services. However, any unit offering thrombolysis will require trained staff and facilities for at least some monitoring of vital signs and cardiac status, to administer the infusion and to detect deterioration due to haemorrhage.55
Conclusions
There is convincing evidence that mortality, morbidity and levels of dependency from stroke can be reduced by managing such patients in geographically‐defined stroke units using a multidisciplinary approach. Stroke patients of any age, sex and stroke severity benefit from such care, the benefits being most apparent when compared to conventional care in the first 1–4 weeks following stroke onset. Likely reasons for benefit include a more intensive approach to the management of medical complications and earlier and more focused rehabilitation. It is likely that management of acute stroke will become more intensive, particularly if thrombolysis becomes more widely adopted, and interventions based on multimodal monitoring prove to be effective. All of these changes provide opportunities to increase the research effort in both acute and recovering stroke patients.
Footnotes
- ↵Address correspondence to Dr E. A. Warburton, Box 135, Addenbrooke's Hospital, Cambridge CB2 2QQ. e‐mail: liz.warburtonmsexc.addenbrookes.anglox.nhs.uk
- © Association of Physicians
References
- ↵ Wardlaw JM, Yamagucchi T, del Zoppo G. Thrombolytic therapy versus control in acute ischaemic stroke.Stroke Module of the Cochrane Database of Systematic Reviews. Cochrane Library. Oxford, Update Software, 1997.
- ↵ Wardlaw JM, Warlow CP, Counsell C. Systematic review of evidence on thrombolytic therapy for acute ischaemic stroke. Lancet1997; 350:607–14.
- ↵ Lees KR. Does neuroprotection improve stroke outcome? Lancet1998; 351:1447–8.
- ↵ Devuyst G, Bogousslavsky J. Clinical Trials Update: Neuroprotection against acute ischaemic stroke. Curr Opin Neurol1999; 12:73–9.
- ↵ Alberts MJ. tPA in acute ischaemic stroke: United States experience and issues for the future. Neurology1998;51 (Suppl. 3):S53–5.
- ↵ Villar‐Cordova C, Morgenstern LB, Barnholz JS, Frankowski RF, Grotta JC. Neurologists' attitudes regarding rt‐PA for acute ischaemic stroke. Neurology1998; 50:1491–4.
- ↵ Stroke Unit Trialists' Collaboration. A systematic review of specialist multi‐disciplinary (stroke unit) care for stroke inpatients. In: Warlow C, van Gijn J, Sandercock P, eds. Stroke Module of Cochrane Database of Systematic Reviews. London, BMJ Publishing Group, 1995.
- ↵ Langhorne P, Williams BO, Gilchrist W, Howie K. Do stroke units save lives? Lancet1993; 342:395–8.
- ↵ Stroke Unit Trialists’ Collaboration. Collaborative systematic review of the randomised trials of organised in‐patient (stroke unit) care after stroke. Br Med J1997; 314:1151–9.
- ↵ Dennis MS, Langhorne P. Stroke Units: An evidence based approach. London, BMJ Publishing, 1998: Chapter 4, 33–40.
- ↵ Stegmayr B, Asplund K, Hulter‐Asberg K, et al. Stroke Units in their natural habitat. Can results of randomized trials be reproduced in routine clinical practice? Stroke1999; 30:709–14.
- ↵ Inderavik B, Bakke F, Slordahl SA, Rokseth R, Haheim LL. Stroke unit treatment improves long‐term quality of life. A randomised controlled trial. Stroke1998; 29:895–9.
- ↵ Indredavik B, Bakke F, Slordahl SA, Rokseth R, Haheim LL. Treatment in a combined acute and rehabilitation stroke unit: which aspects are the most important? Stroke1999: 30:917–23.
- ↵ Stroke Unit Trialists' Collaboration. How do stroke units improve patient outcomes? A collaborative systematic review of the randomised trials. Stroke1997; 28:2139–44.
- ↵ Bamford JM, Sandercock PAG, Dennis MS, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet1991; 337:1521–6.
- ↵ Millikan CH. Stroke intensive care units: objectives and results. Stroke1979; 10:235–7.
- ↵ Sivenius J, Pyorala K, Heinonen OP, Salonen JT, Riekkinen P. The significance of intensity of rehabilitation after stroke: a controlled trial. Stroke1985; 16:928–31.
- Smith DS, Goldenberg E, Ashburn A, Kinsella G, Sheikh K, Brennan PJ, Meade TW, Zutshi DW, Perry JD, Reebak JS. Remedial therapy after stroke: a randomised controlled trial. Br Med J1981; 282:517–20.
- ↵ Kwakkei G, Wagenaar RC, Koelman TW, Lankhorst GJ, Koetsier JC. Effects of intensity of rehabilitation after stroke. A research synthesis. Stroke1997; 28:1550–6.
- ↵ Warburton EA, Morris K, Rossiter D. Speech therapists or swallowing assessors—a survey of speech and language therapy for stroke in the UK. Age Ageing2000 (submitted).
- ↵ Ebrahim S, Redfern J. Stroke Care—A matter of chance: A National Survey of Stroke Services. London, The Stroke Association, 1999.
- Clinical Standards Advisory Group. Report on clinical effectiveness using stroke care as an example. London, HMSO, 1998.
- ↵ Rudd AG, Irwin P, Rutledge Z, Lowe D, Wade DT, Morris R, Pearson MG. The national sentinel audit for stroke: a tool for raising standards of care. J Roy Coll Phys1999; 33:460–4.
- ↵ Millikan CH. Stroke Intensive care units: objectives and results. Stroke1979; 10:235–7.
- ↵ Hacke W, Schwab S, De Georgia M. Intensive care of acute ischaemic stroke. Cerebrovasc Disord1994;4:385–92.
- ↵ Back T, Khono K, Hossman K‐A. Cortical negative DC deflections following middle cerebral artery occlusion and KCl‐induced spreading depression: Effect on blood flow, tissue oxygenation and electroencephalogram. J Cereb Blood Flow Metabol1994; 14:12–19.
- ↵ Nachtman A, Siebler M, Rose G, et al. Cheyne–Stokes respiration in ischaemic stroke. Neurology1995;45:820–1.
- ↵ Mikel HS, Yashes NV, Kempinski O, et al. Breathing 100% oxygen after global brain ischaemia in mongolian gerbils results in increased lipid peroxidation and increased mortality. Stroke1987; 18:426–30.
- ↵ Ronning OM, Guldvog B. Should stroke victims routinely receive supplemental oxygen? A quasi‐randomised controlled trial. Stroke1999; 30:2033–7.
- ↵ Elizabeth J, Singarayer J, Ellul J, Barer D, Lye M. Arterial saturation and posture in acute stroke. Age Ageing1993; 22:269–72.
- ↵ Paczynski RP, Diringer MN, Hsu CY. Experimental therapies to improve delivery of oxygen and substrate in acute stroke. Curr Opin Neurol1995; 8:6–14.
- ↵ Reith J, Jorgensen HS, Pedersen PM, et al. Body temperature in acute stroke: Relation to stroke severity, infarct size, mortality and outcome. Lancet1996; 347:422–5.
- ↵ Baker C, Onesti S, Barth K, et al. Hypothermic protection following middle cerebral artery occlusion in the rat.Surg Neurol1991; 36:175–80.
- ↵ Sick TJ, Guangping Xu, Perez‐Pinzon M. Mild hypothermia improves recovery of cortical extracellular potassium ion activity and excitability after MCA occlusion in the rat. Stroke1999; 30:2416–22.
- ↵ Ginsberg MD, Busto R. Combating hyperthermia in acute stroke: a significant clinical concern.Stroke1998;29:529–34.
- ↵ Busto R, Dietrich W, Globus M‐T, et al. Small differences in intrahemispheric brain temperature critically determine the extent of ischaemic neuronal injury. J Cereb Blood Flow Metabol1987; 7:729–38.
- ↵ De Keyser J. Antipyretics in acute ischaemic stroke. Lancet1998; 352:6–7.
- ↵ Weir CJ, Murray GD, Dyker AG, Lees KR. Is hyperglycemia an independent predictor of poor outcome after acute stroke? Results of a long term follow up study. Br Med J1997; 314:1303–6.
- ↵ Scott JF, Robinson GM, French JM, O'Connell JE, Alberti KGMM, Gray CS. Glucose potassium insulin infusions in the treatment of acute stroke patients with mild to moderate hyperglycemia. The Glucose Insulin in Stroke Trial (GIST). Stroke1999; 30:793–9.
- ↵ Blood pressure in acute stroke collaboration (BASC). Stroke module of the Cochrane Database of Systematic Reviews. The Cochrane Library [database on disc and CDROM]. The Cochrane Collaboration. 3rd edn. Oxford, Update Software, 1997.
- ↵ Powers WJ. Acute hypertension after stroke: the scientific basis for treatment decisions. Neurology1993;43:461–7.
- ↵ Mortis J, Schwartz RS, Poulos R, Howes LG. Natural history of blood pressure change after stroke.Stroke1997: 28:1401–5.
- ↵ Dawson SL, Evans SN, Manktelow BN, Fotherby MD, Robinson TG, Potter JF. Diurnal blood pressure change varies with stroke subtypes in the acute phase. Stroke1998: 29:1519–24.
- ↵ O’Connell JE, Gray CS. Treatment of post‐stroke hypertension. A practical guide. Drugs Ageing1996; 8:408–15.
- ↵ Date H, Hossman KA, Shima T. Effect of middle cerebral artery compression on pial artery pressure, blood flow, and electrophysiological function of cerebral cortex of cat. J Cereb Blood Flow Metabol1984; 4:593–8.
- ↵ Lisk DR, Grotta JC, Lamki LIF, et al. Should hypertension be treated after acute stroke? A randomised controlled trial using single photon emission computed tomography. Arch Neurol1993; 50:855–62.
- ↵ Grotta JC, Pettigrew LE, Allen S, Tonneson A, Yatsu FM, Gray J, Spydell J. Baseline haemodynamic state and response to haemodilution in patients with acute cerebral ischaemia. Stroke1985; 16:790–5.
- ↵ Reilly PL, Lewis SB. Progress in head injury management. J Clin Neurosci1997; 4:9–15.
- ↵ Rem JA, Hachinski VC, Bourghner DR, et al. Value of cardiac monitoring and echocardiography in TIA and stroke patients. Stroke1985; 16:950–6.
- ↵ Kocan MJ. Cardiovascular effects of acute stroke. Progr Cardiovasc Nursing1999; 14:61–7.
- ↵ Rogvi‐Hansen B, Boysen G. IV glycerol treatment in acute ischaemic stroke. In: Warlow CP, van Gijn J, Sandercock P, eds. Stroke Module of the Cochrane Database of Systematic Reviews. The Cochrane Collaboration. Oxford, Update Software, 1995.
- ↵ Asplund K. Haemodilution in acute stroke. In: Warlow C, van Gijn J, Sandercock P, eds. Stroke Module of Cochrane Collaboration Database of Systematic Reviews. London, BMJ publishing, 1995.
- Frank JI. Large hemisphere infarction, deterioration and intracranial pressure. Neurology1995; 45:1286–90.
- ↵ Hacke W, Brott L, Caplan L, Meier D, Fieschi C, von Krummer R, Donnan G, Heiss WD, Wahlgren NG, Spranger M, Boysen G, Marler JR. Thrombolysis in acute ischaemic stroke: controlled trials and clinical experience.Neurology1999; 53 (Suppl. 4):S3–14.
- ↵ Braimah J, Kongable G, Rapp K, Daley S, Bratina P, Sailor S, Barch C, Donnamma R, Spilker J. Nursing care of acute stroke patients after receiving tPA therapy. The NINDS rt‐PA study group. J Neurosci Nursing1997; 29:373–83.
Nenhum comentário:
Postar um comentário