XVII

XVII. European Stroke Conference
Nice, France

Poster Session: Acute stroke: early management and stroke units

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


Reduction of inpatient mortality: comparison between a mobile stroke service and a comprehensive stroke care unit
L.C.Weir    D.A.De Silva    P.J.Hand     B. Yan     S.M .Davus

Royal Melbourne Hospital, University of Melbourne

AUSTRALIA

Background: Stroke Care Units (SCU) are coordinated by dedicated multidisciplinary teams and geographically located in one area. The Stroke Unit Trialists’ Collaboration found a significant reduction in death and dependency with organized care in a SCU. There have been few direct comparisons of inpatient mortality rates between SCU care and mobile stroke teams. In 1995 a comprehensive, dedicated and localized SCU was established at The Royal Melbourne Hospital (RMH). This replaced a multidisciplinary mobile stroke service, which consulted on all stroke patients, located in various wards. We aimed to compare stroke mortality before and after the implementation of the SCU at RMH. Methods: We compared data from 2 prospective databases of consecutive acute ischemic and hemorrhagic stroke patients at RMH, before (1987-1989) and after development of the RMH SCU (Year 2006 inclusive). The outcome measure was inpatient mortality. Results: In the first time epoch (1987-1989), inpatient mortality with mobile team treatment was 15.3% for the 695 ischemic stroke patients and 34.0% for the 144 hemorrhagic stroke patients. After implementation of the SCU, inpatient mortality in 2006 was 11.4% for the 403 ischemic stroke patients and 27.7% for the 141 hemorrhagic stroke patients. For ischemic stroke, the absolute risk reduction (ARR) was 3.9% and relative risk reduction (RRR) 25% in inpatient mortality after SCU implementation (p=0.095). Excluding those receiving intravenous tPA, a potential outcome confounder, inpatient mortality for ischemic stroke patients in 2006 was 11.4%, with an ARR of 3.9% and RRR of 25% after SCU implementation. For hemorrhagic stroke, the ARR was 6.3% and RRR 19% in inpatient mortality after implementation of the SCU (p=0.300). Discussion: The reduction in mortality rates over a 20 year period supports the replacement of a mobile stroke team by a SCU. The ARR and RRR of mortality in our clinical setting are concordant with Stroke Unit Trialists Collaboration and support geographical SCUs as the standard of care for stroke.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


HEAD POSITION AND CEREBRAL HAEMODYNAMICS IN STROKE PATIENTS WITH SEVERE CAROTID ARTERY DISEASE
L. Benavente    S. Calleja    C.H.Lahoz

Hospital Universitario Central de Asturias

SPAIN

BACKGROUND: Acute stroke patients are routinely positioned with the head of the bed elevated at 30º despite de lack of evidence for increased intracranial pressure. An adequate brain perfusion pressure is needed to supply impaired vasoreactivity tissue, but intracranial hypertension must be avoided. This issue takes special relevance in case of large vessel stroke, because of the haemodynamic mechanism. OBJECTIVES: We aim to determine the effect of head position on blood flow velocity and resistence to residual flow in case of occlusion or high-grade stenosis of internal carotid artery with acute stroke. METHODS: Middle cerebral artery (MCA) mean flow velocity (MFV) and pulsatility index (PI) were monitored with transcranial Doppler by headframe. Cerebral haemodynamic reserve was calculated with Brain holding Index (BHI) and related to other parameters at 45º, 30º and 0º head position. RESULTS: Fifteen patients were evaluated. MCA MFV increased in all patients with lowering head position an average of 22% (range 4 to 60%), 14% from 45 to 30º and 7% from 30 to 0º. PI decreased by 16% (mean rate at 45º 1,0; 30º 0,9; 0º 0,84), indicating no increase in resistance to blood flow. MFV increase was no related to BHI, but this increase was higher as poorer the initial MFV (<=30cm/s) (p=0.05). CONCLUSIONS: Acute stroke patients with high-grade carotid stenosis benefit from lower head position independently of their cerebral hemodynamic reserve, but most benefit occurs when basal MFV is low.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


Acceleration of door-to-imaging-time in acute stroke patients by implementation of an all-points alarm
C.H.Nolte    C.J.Ploner    G.J.Jungehulsing     M. Möckel     Y. Kühnle    S. Roll     F. Knollmann     J. Muller-Nordhorn

Charité University Medical Center in Berlin

GERMANY

Background: Thrombolysis of acute ischemic stroke is licensed for use within 3h only. The less the time delay to treatment, the greater the benefit. The crucial diagnostic step to initiate treatment is cerebral imaging. Methods: Having established a clinical pathway, the interdisciplinary quality assessment group for stroke management planned and assessed an additional intervention to further improve the intrahospital time delay (period between arriving at the emergency department (ED) and cerebral imaging). The intervention consisted of an all-points alarm informing all involved staff simultaneously (the neurologist, ED-physician, radiologist, radiographer and intrahospital transport) about patients potentially eligible for thrombolysis. Time delay as well as sociodemographic and clinical data were assessed four months before and eight months after implementation. Differences in time delay were investigated by analysis of covariance (ANCOVA). Results: During the assessment, 689 patients with symptoms of stroke arrived at our hospital (69±13 years; 49% female). Among those, 111 (16%) were potentially eligible for thrombolysis (71±13 years; 56% female). Patients differed slightly before (N=34) and after (N=77) the implementation with respect to age, gender, insurance, NIHSS, vascular risk factors and prehospital delay. ANCOVA revealed significant reduction due to the intervention (p=0.0001) on differences in time delay adjusted for gender, age, insurance and NIHSS. Time delay [min] was 58.6±3.9 (adjusted mean±SE) before and 37.1±2.8 after the implementation adjusted for gender, age, type of insurance and NIHSS. Conclusion: The implementation of an all-points alarm can result in an additional acceleration of intrahospital delays for acute stroke patients.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


The Significance of Blood Pressure Variation for the Development of Hemorrhagic Transformation in Acute Ischemic Stroke
Y. Ko    J.H.Park    M.H.Yang     S.B.Ko     M.K.Han    C.W.Oh     S.H.Park     J. Lee    H.J.Bae

Seoul National University Bundang Hospital

SOUTH KOREA

Background: It is known that high blood pressure (BP) provokes hemorrhagic transformation (HT) of cerebral infarction and may contribute to its poor prognosis. However, BP is a very dynamic and complex phenomenon, especially in acute stage of stroke, and high BP itself may not be enough to explain the contribution of BP to HT. This study was aimed to elucidate the various aspects of BP, which were measured and defined in several ways, with respect to the risk of HT. Method: Based on the prospective stroke registry, the consecutive series of stroke patients who were hospitalized from 2004/1 to 2007/8 within 24 hour from onset and had relevant ischemic lesions on MRI were selected. Among them, those with no HT on initial MRI were recruited in this study and their BP measurements during the first 72 hours were gathered. Mean, standard deviation (SD), maximum, minimum, difference between maximum and minimum (DF1), maximum difference between successive BP measurements (DF2), and its variation (SV) were defined and calculated. With respect to HT on the follow-up images during 2 weeks after onset, the odd ratio (OR) of those parameters were calculated adjusting initial systolic BP (SBP) and other potential confounders. Results: Among 863 patients who met the eligibility criteria, 70 (8.1%) showed HT on follow-up images. The mean (adjusted OR, 1.25; 95% confidence interval, 1.02 to 1.53 per 10 mmHg), maximum (1.32; 1.13 to 1.54), DF1 (1.15; 1.04 to 1.28), and DF2 (1.10; 1.03 to 1.18 per 5 mmHg) of SBP were independent predictors of HT respectively. The statistical significance of the maximum, DF1, and DF2 did not change despite additional adjustment for the mean SBP. The analyses on diastolic BP also showed similar results. Conclusions: This study suggests that most parameters reflecting BP variation can contribute to HT of ischemic stroke independent of the initial and mean BP during the first 3 days. To prevent HT, we may pay attention not only to the absolute value of BP but also to its variation.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


A Case-Controlled Comparison of the Management of Acute Stroke Patients between Acute Stroke Unit and Acute Medical Unit of an Urban Teaching Hospital
G. Ray    T.J.Quinn    K.R.Lees

Western Infirmary

UNITED KINGDOM

Background: – The superiority of Acute Stroke Unit (ASU) to general medical care is established. Our ASU appears to be of insufficient size at times, necessitating admission of acute stroke patients to the acute medicine unit (AMU) which follows the same stroke protocol. We hypothesized that outcomes may differ in patients not admitted to ASU, influenced by inequality of access to investigation and / or specialist supportive care. Methods: For each acute stroke patient admitted to AMU between January and May 2007, the most recently admitted patient from ASU was taken as control. We described outcomes in terms of: in-hospital mortality; duration of acute admission; need for intravenous antibiotics; time to access medical and non-medical interventions that are likely to affect outcome. Proportions were compared using chi square statistics; time differences with paired t testing. Results: Forty stroke patients were admitted in AMU (21 males ) and 40 control patients (22 males) were selected from ASU. There was no significant difference between groups for for key prognostic variables for stroke - Age , Glasgow Coma Scale and National Institutes of Health Stroke Scale . Mortality ( 20% versus 5% P=0.043) and antibiotic use (27% vs 10% P =0.01) were higher in AMU patients. Mean duration of hospital stay was similar on both groups ( 7.7 vs 7.3 days P=0.82). Time to Brain imaging (1.1 vs 1.4 days P= 0.04) , Anti-platelets ( 1 vs 1.6 days P=0.01), Statin ( 1 vs 1.4 days P=0.01),TED stockings (1.1 vs 2.3 days P=0.001),Physiotherapy (1.8 vs 2.9 days P=0.001), Occupational therapy ( 2 vs 3.2 days P=0.001) and Speech Therapy ( 1.4 vs 1.5 days P=0.01) all show that ASU patients received medical and non medical interventions sooner. Discussion: Despite a common protocol for care, patients in our specialist acute stroke unit received approved management more rapidly and experienced significantly better outcomes than those denied access. Limiting specialist resources may be a false economy.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


The Liverpool presentation study: factors responsible for prehospital delays after acute stroke.
M. Koufali    E. Bacabac    C. Woodhall     R. Kumar     R. Durairaj    A.K.Sharma

Aintree Stroke Team for Audit & Research, Aintree University Hospitals, Liverpool

UNITED KINGDOM

Background: Early admission and treatment are determinants of the level of acute stroke intervention and outcome. Our study aimed to examine the factors associated with delays from onset to admission following an acute stroke. Methods: Data were prospectively and consecutively collected from patients who presented to the Acute Stroke Unit (ASU) during 2003-2006. Univariate and multivariate regression analyses were conducted to analyse factors influencing Delay from Onset to Arrival (DOA). Results: A total of 1,821 patients were admitted to the ASU during that period, but only 1,407 patients with a complete dataset and a confirmed diagnosis of ischaemic stroke were included in the study. Median DOA was 5.0 (interquartile range 1.23 to 15.20) hours. Delays in arrival time were, as reported by previous studies, not significantly associated with older age, female sex or previous TIA/stroke risk factors. According to our multivariate regression model a prehospital delay of less than 2.3 hours is mostly associated with immediate use of the emergency services thus bypassing the GP service (OR 4.45 95% CI =1.6 to 5.9, p=0.03), loss of consciousness (OR 2.6 95% CI =1.1 to 6.1, p=0.0270) and the right side being affected by the stroke (OR 2.9 95% CI =1.2 to 6.9, p=0.016). Conclusions: Future educational campaigns should focus on the use of emergency services, as this proves to be the single modifiable and most important factor determining early admission to the stroke unit and access to appropriate care including thrombolysis services.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


Immuno-inflammatory and thrombotic/phybrinolytic markers as a diagnostic panel of acute ischemic stroke
A. Tuttolomondo    A. Pinto    S Corrao     D Di Raimondo     R. Di Sciacca    G. Licata

Biomedical Department of Internal and Specialist Medicine, University of Palermo

ITALY

Introduction: Accumulating evidences suggest that inflammation plays an important role in the development of cardiovascular and cerebrovascular disease. The aim of this study is to evaluate the predictive value of a bound of candidate serum immunoinflammatory and thrombotic/phybrinolitic moleculae towards acute ischemic stroke diagnosis. Materials and Methods : were enrolled 120 consecutive patients with a diagnosis of acute ischemic stroke and 123 consecutive hospitalized control patients without a diagnosis of acute ischemic stroke. Were evaluated plasma levels of IL-1alfa, TNF-alfa, IL-6 and IL-10, E-selectin, P-selectin, sICAM-1 and sVCAM-1 as markers of immunoinflammatory activation, vWF plasma levels as a marker of endothelial dysfunction, TPA antigen and PAI-1 plasma levels as a marker of a prothrombotic state. Results:, TNF-alfa, PAI-1 and TPA on bivariate logistic regression were highly correlated to stroke diagnosis. Among the other variables maintained in the final model IL1beta, Selectin E, were significantly associated with acute ischemic stroke diagnosis , whereas IL-6 , VICAM-1, ICAM-1 and neutrophil percentage showed only a slight or no association with stroke diagnosis . Furthermore the continuous values of TNF-alfa, PAI-1 and TPA showed a significative predictive value, likelihood ratio, with an area under the ROC curve respectively of 98,6%, 97,1% and 99,9%. Discussion: Our findings could suggest the possible high diagnostic power of these immunoinflammatory and thrombotic/phybrinolitic variables in patients with acute ischemic stroke Although our results are encouraging, additional studies are needed to establish the validity of this approach and samples are expected to enhance our control population for marker panel analysis.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


Reductions in length of Hospital stay among non-stroke patients in a stroke unit after the introduction of thrombolytic treatment.
C. Sølling    S.P.Johnsen    L.H.Ehlers     L. Østergaard     G. Andersen

Aarhus University Hospital

DENMARK

Background and Purpose: Implementing thrombolytic therapy in a stroke unit (SU) profoundly affects the infrastructure available to patients admitted for suspected acute stroke. Highly specialized units with rapid access to neuroimaging and staff dedicated to stroke care could also benefit non-stroke patients who experience faster and more efficient diagnosis and management. We examined the benefits of an acute stroke service to non-stroke patient in terms of length of hospitalization, and estimated the economic impact in terms of derived costs. Methods: We performed a historical follow-up study of 792 non-stroke patients admitted to our SU for suspicion of stroke before, during, and after implementing thrombolysis as part of our service. Data on demographic and clinical characteristics, including imaging data and final diagnoses, and length of stay (LOS) were collected prospectively. Multivariate logistic regression analysis was performed to identify variables associated with LOS. Results: Median LOS for non-stroke patients in the SU decreased from 43.8 (interquartile range, 19-96) to 23.5 (16-44) hours after implementing thrombolytic therapy. Total hospital LOS for non-stroke patients decreased from 52.7 (22-147) to 28.7 (21-124) hours during the same period. Initial MRI was associated with shorter LOS in the SU. Derived cost reductions from LOS reduced the costs of implementing recombinant tissue plasminogen activator treatment. Conclusion: Stroke care reorganization following the introduction of thrombolytic treatment was associated with a 50% reduction in LOS for non-stroke patients admitted to the SU. Reduced LOS in the SU for non-stroke patients could further add to the cost-effectiveness of thrombolytic treatment.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


Reductions in length of Hospital stay among non-stroke patients in a stroke unit after the introduction of thrombolytic treatment
C. Sølling    S.P.Johnsen    L.H.Ehlers     L. Østergaard     G. Andersen

Aarhus University Hospital

DENMARK

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


Very early rehabilitation or intensive telemetry after stroke (VERITAS): a pilot randomised trial
P. Langhorne    A. Knight    D.J.Stott     J. Bernhardt     C.L.Watkins    D. Barer

Academic Section of Geriatric Medicine, University of Glasgow

UNITED KINGDOM

Background: Stroke patients have a better chance of making a good recovery if they receive care in a well-organised stroke unit. However there are uncertainties about how best to provide such care. We studied two key aspects of early stroke unit care; early active mobilisation (EM) and automated monitoring (intensive telemetry; IT) for physiological complications such as hypoxia. Methods: This was an observer blinded, factorial (2 x 2) pilot randomised controlled trial recruiting stroke patients within 36 hours of symptom onset. Patients were randomised to one of four nurse-led treatment protocols; a) standard stroke unit care, b) early mobilisation (EM), c) automated intensive telemetry (IT) for 72 hours, or d) combined EM and IT. The primary outcome was Rankin score at 3 months. We report the initial data on feasibility. Results: We have randomised 31 patients (mean age 65 years; mean baseline NIH score 6; mean time to randomisation 25hrs). EM patients were mobilised sooner after randomisation (median 0 hours versus 5) and were more likely (OR 4.6; 95% CI 1.0-21; P=0.05) to have very rapid mobilisation (within 1 hour of randomisation). The EM group also more likely to spend more than 5% of their day standing and walking (OR 14.0; 1.5-133; P=0.02). The IT group were more likely (OR 12.0; 2.1-65; P=0.004) to have several (>10) pre-defined physiological complication events detected. The EM group had fewer chest infections (OR 0.2; 0.03-1.1; P=0.06). All these associations remained, but were less statistically significant, after correcting for age, baseline NIH score and co-interventions. Discussion: We have demonstrated the feasibility of implementing the EM and IT interventions in a randomised trial setting. The main functional outcomes are currently being collected.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


Predictors of outcome in stroke patients treated with recombinant tissue plasminogen activator: the NICE experience
L. Suissa    E. Bozzolo    S. Lachaud     M.H.Mahagne

Hôpital Saint Roch, CHU Nice

FRANCE

Background and purpose: Some predictors of outcome after intravenous rt-PA in stroke patients have been identified in clinical trials, but the reproducibility of these results in the set-up of clinical routine is still a matter of debate. The aim of this study was to assess the clinical, biological and radiological variables associated with poor outcome and death in patients treated with systemic thrombolysis in the set-up of clinical routine, in our stroke unit. Methods: 150 stroke patients treated with intravenous thrombolysis were included. All patients had a standardized and follow-up procedure. We identified two groups, with good and poor outcome at 3 months, (respectively, modified Rankin scale between 0 and 2 and between 3 and 6). 50% achieved a good outcome. Clinical, radiological and biological markers were analyzed with univariate then multivariate analysis. Results: Independent predictors of poor outcome were: 1) Stroke severity with initial NIHSS > 15 (OR: 5.3, CI 95%: 2.0-13.9) and 2) Absence of early clinical improvement (> 4 points on NIHSS at 24 hours; OR 9.5, CI 95%: 3.5-25.9). Other statistically significant predictors of functional outcome included: internal carotid artery occlusion, atherothrombotic etiology, pre-treatment diastolic hypertension, initial glycaemia >1.1 g/l. Age, sex, side of infarction, radiological data, time from stroke onset to needle were not significantly associated with functional outcome. Conclusion: In this series, only initial severity and absence of rapid improvement can predict bad outcome. These results are comparable to those obtained in the setting of clinical trials and show the reproducibility in clinical routine of results obtained in clinical trials.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


The evaluation of time barriers in intravenous thrombolysis for acute stroke patients in Lithuania
A. Vilionskis    D. Jatuzis    I. Urbelis     M. Sumkauskaite

Vilnius University Emergency Hospital; Vilnius University Hospital Santariskiu clinics

LITHUANIA

Background: Intravenous thrombolysis (IVT) with rt-PA was implemented as the treatment of first choice for patients with acute ischemic stroke patients in Lithuania since 2002. However, only small part of stroke patients undergoes IVT as yet, and dominant reason is overlong time delay from onset to treatment. Aim: To estimate the changes of onset to treatment time (OTT), onset-arrival time (OAT) and door-needle time (DNT) and affecting factors during 2002-2007 years in order to identify strategies to increase IVT access for acute stroke patients in Lithuania. Methods: All consecutive patients with acute ischemic stroke treated with IVT in Vilnius University Emergency Hospital and Vilnius University Hospital Santariskiu clinics since 2002 were included into local Lithuanian thrombolysis register and time intervals reflecting the delay of IVT (OTT, OAT, DNT) were recorded. Time intervals were calculated separately in two groups according to date of IVT (2002-2005 and 2006-2007). Changes of time intervals during recent years were assessed, and findings were compared with summarized SITS-MOST data. Results: 36 and 40 acute stroke patients were treated with IVT during 2002-2005 and 2006-2007 years, respectively. The mean OAT even increased from 66+/-50 min during 2002-2005 to 80+/-39 min during 2006-2007 (p>0.1). The average DNT decreased from 86+/-37 (2002-2005) to 72+/-29 min (2006-2007; p=0.06). The OTT remained unchanged: 152+/-48 vs. 152+/-34 min, respectively. This data are comparable with summarized data from all SITS-MOST centers. Conclusions: Only the DNT has tendency to decline during recent years. Targeted strategies (educational activities in community, optimization of ambulance service, etc.) are required to reduce prehospital delaying factors and to increase IVT access for acute stroke patients.

Session: Poster Session II
Date: Thursday 15 May 2008
Time: 12:30 - 14:00
Room: Agora 2


Quality Assessment in Acute Stroke Care (QuASt) - ClinicalTrials.gov Identifier: NCT00574262
C. Kraemer    T. Warnecke    T. Küttner     N. Roeder     E.B.Ringelstein    W.R.Schäbitz

University Hospital Münster

GERMANY

Background – Almost half of all stroke patients in Germany are admitted to an accredited monitoring stroke unit. Following acute treatment, diagnostic work-up and automated monitoring of vital functions are performed as well as adapted secondary prevention. International studies have established the superiority of stroke unit treatment over conventional care. Methods – On the basis of a new developed clinical pathway, we evaluated the quality of care on a 16 bed stroke unit by measuring several predefined quality indicators based on diagnostic, treatment, secondary prevention and outcome. All patients admitted to the stroke unit of the university hospital Münster, Germany from October 1st, 2007 to March 31st, 2008 with the diagnosis of an acute ischemic stroke or a transient ischemic attack are enrolled in this prospective observational study. By using the clinical pathway the following indicators are measured: Primary outcome measure is the frequency of patients monitored according to the specific german procedure standard (OPS-code 8-981). Secondary outcome measures are the frequency of initiated or improved antithrombotic secondary prevention, statin therapy and antihypertension therapy, and reduction of systolic and diastolic blood pressure by more than 10 mmHg, the frequency of totally completed clinical diagnostics regarding the etiology of the stroke while patient is on the stroke unit, the frequency of the stroke been classified according to the TOAST criteria and the frequency of a reduction on the NIH-Stroke Scale by at least two points. Results – The new developed clinical pathway and the results of the study will be presented. Discussion – Measuring the quality of acute stroke care is important to improve and standardize the treatment on a stroke unit.