XVII. European Stroke Conference
Nice, France

Oral Session:

TIA 
Date: Wednesday 14 May 2008  
Time: 8:30 - 8:40 -  Room: Hermes
Chair: A.Carolei, Italy and P. Koudstaal, The Netherlands

01
Use of the ABCD2 score in screening for minor stroke or transient ischaemic attack (TIA) in referrals to a fast track clinic
G. Ray    F. Wright    D.J.Stott    P. Langhorne                                          
 

Glasgow Royal Infirmary

UNITED KINGDOM

Background : The ABCD2 score predicts the risk of early stroke recurrence after minor stroke or TIA Although it has been developed as a prognostic tool it has been suggested part of its value is diagnostic ( that is identifying true TIA or minor stroke). Our aim was to see whether a high referral ABCD2 score predicted the diagnosis of stroke or TIA in patients referred to our Fast Track TIA Clinic. Methods – We did a prospective audit of all new patients attending the weekly Fast Track TIA Clinic in Glasgow Royal Infirmary from July – September 2007. All patients had their ABCD2 score checked on first clinic visit. A final diagnosis of stroke or TIA was made on clinical grounds ( supported by neuroimaging) by independent physicians. Results - A total of 75 new patients (30 males) were seen with a mean age of 62.1 years (Standard Deviation 12.3). Referrals were from General Practitioners (n=56), Accident and Emergency (n=11), and other sources (n=8); 43/75 (57.3%) were confirmed to have a diagnosis of stroke or TIA. Median ABCD2 score for diagnosis of stroke and TIA was 4 and for other diagnosis was 2 . The area under the Receiver Operating Characteristic curve for the ABCD2 score for diagnosis of a cerebrovascular accident was 0.80 { 95% Confidence Interval (CI) = 0.69-0.91}. The sensitivity of an ABCD2 score of greater than or equal to 3 for stroke or TIA was 88 % with an Odds Ratio of 16.7 (95% CI =5.1-55.2). Discussion - The ABCD2 score may be an useful tool not only to predict recurrence of minor stroke or TIA but also in screening for diagnosis. Patients with ABCD2 score of 3 and above might be targeted for rapid assessment and treatment if there are limited clinic resources. These results require validation in larger patient groups.

TIA 
Date: Wednesday 14 May 2008  
Time: 8:40 - 8:50 -  Room: Hermes
Chair: A.Carolei, Italy and P. Koudstaal, The Netherlands

02
Value of DWI-MR and TIA etiology to improve prediction of early risk of stroke after TIA patients by ABCD2 score. Value of DWI-MR and TIA etiology
D. Calvet    E. Touzé    G. Turc    C. Oppenheim    C. Lamy    J.-L.Mas                            
 

Hôpital Sainte-Anne

FRANCE

Background: The ABCD2 score has been shown to predict the early risk of stroke after TIA. The additional predictive value of diffusion-weighted imaging (DWI) and TIA etiology is not well known. Methods: From January 2003 to June 2007, 343 consecutive patients with recent (<48h) TIA were admitted to our stroke unit. On admission, all patients had DWI and a standard etiological work-up. They were followed up to 90 days. The predictive value of the ABCD2 score, DWI lesions, large artery disease (LAD) and atrial fibrillation (AF) with respect to occurrence of ischemic stroke at 1 week and 3 months was assessed with the use of log rank test and Cox proportional hazards model. Results: The median time (IQR) from TIA to DWI was 19.5h (8.0-28.3). DWI was positive in 136 patients (40%). The risk of stroke was 1.5% (95% CI; 0.3-2.7) at 1 week, and 2.9% (1.1-4.7) at 90 days. In univariate analysis, the ABCD2 score and the presence of DWI lesions were significantly associated with an increased risk of stroke at 1 week and 3 months. LAD was associated with an increased risk of stroke at 3 months (Table). In multivariate analysis, ABCD2 score > 5 (HR=10.1; 1.1-93.4) and DWI lesions (HR=8.7; 1.1-71.0) were independent predictors of stroke at 90 days. LAD (HR=3.4; 1.0 -11.8) also predicted stroke at 90 days, whereas AF did not. Discussion: Taking into account DWI and TIA etiology in addition to the ABCD2 score improve the prediction of the early risk of stroke after a TIA at 90 days.

 
http://www.eurostroke.org/ni_graphics/t_aid3019.htm

TIA 
Date: Wednesday 14 May 2008  
Time: 8:50 - 9:00 -  Room: Hermes
Chair: A.Carolei, Italy and P. Koudstaal, The Netherlands

03
Population-based study of risk of stroke in the first few hours after a TIA: urgency or emergency?
A. Chandratheva    Z. Mehta    O.C.Geraghty    L. Marquardt    P.M.Rothwell                                   
 

Stroke Prevention Research Unit, Dept. of Clinical Neurology, University of Oxford

UNITED KINGDOM

BACKGROUND Understanding the risk of stroke in the first few hours after TIA or minor stroke is essential for planning stroke services and public education. We therefore conducted the first population-based study of prognosis in the hyper-acute phase. METHODS We did a prospective 5-year, population-based study in Oxfordshire, UK, from 1st April 2002, using multiple methods of case ascertainment and face-to-face follow-up (Oxford Vascular Study). A recurrent stroke was defined as new and persistent neurological symptoms of sudden onset in a patient in whom the initial symptoms had already partially or fully recovered. RESULTS Follow up to 90 days was complete for all 1232 initial events (485 TIAs / 747 strokes). 35 (25% of all strokes within 90-days) occurred within 24 hours. The recurrent event appeared to be in the same arterial territory as the initial event in all 35 cases. In 22 (63%) cases, the patient had sought medical attention prior to their recurrent stroke. The initial event had fully recovered prior to the stroke (i.e. was a “TIA”) in 25 cases. If the 10 acute recurrences that occurred prior to full resolution of the initial event are excluded, the risk of stroke within 24 hours of a TIA was 5.2% (95%CI 3.2-7.2). Among the 485 TIAs thus defined, the ABCD2 score was highly predictive (p=0.00025) of stroke within 24 hours, with a risk of 2.2% at a score <5, 6.6% at a score of 5, 13.8% at 6 and 33% at 7. CONCLUSION Risk of recurrent stroke within 24 hours of a TIA is as high as 5%, even when cases in which stroke occurred prior to full resolution of the initial event are excluded. Most patients sought medical attention prior to the stroke, and the ABCD2 score was predictive of stroke in this hyper-acute phase.

TIA 
Date: Wednesday 14 May 2008  
Time: 9:00 - 9:10 -  Room: Hermes
Chair: A.Carolei, Italy and P. Koudstaal, The Netherlands

04
Risk of bleeding with combination of aspirin and clopidogrel in the acute phase after TIA or minor ischaemic stroke
O.C.Geraghty    A. Chandratheva    L. Marquardt    Z. Mehta    P.M.Rothwell                                   
 

Stroke Prevention Research Unit, Dept. of Clinical Neurology,University of Oxford

UNITED KINGDOM

BACKGROUND: In the EXPRESS Study, early preventive treatment after TIA and minor stroke reduced early recurrent stroke by 80% without increasing bleeding complications. However, safety data were not reported separately for patients given aspirin plus clopidogrel (A+C). In the absence of other data on the safety of acute treatment with A+C in elderly non-trial populations, we determined the 30-day risk of bleeding during and after EXPRESS. METHODS: We studied patients treated in the EXPRESS clinic up to 1/12/07. A+C was given for 30 days in patients presenting acutely. Brain imaging was done prior to treatment of minor strokes. Bleeding events, categorised by the CURE Trial criteria, were identified by regular face-to-face follow-up, diagnostic coding in primary and secondary care, and blood transfusion data. RESULTS: Of 627 patients with TIA or minor stroke (40% aged ≥80) given aspirin alone (AO) or A+C, 6 had minor (n=3) or major (n=3) bleeding (all extracranial) in the first 30 days: 1/418 on AO (0.2%, 95%CI 0-0.7) versus 5/209 (2.4%, 95%CI 0.3-4.4) on A+C (p=0.017). In the A+C group, 2 GI bleeds required hospitalisation and transfusion, but no bleeds were disabling or fatal. The age, sex and other baseline characteristics of the AO and A+C groups were similar, as were the subsequent monthly bleeding risks from day 30-180: 0.21% (0.03-0.39) in the AO group and 0.19% (0.05-0.32) in the A+C group (after stopping clopidogrel). CONCLUSION: A 30-day course of A+C acutely after TIA or minor stroke was associated with a greater risk of non-disabling extracranial bleeding than aspirin alone, but there does not appear to be a high risk of intracerebral haemorrhage. The balance of risk and benefit of acute treatment with A and C after TIA and minor stroke is uncertain.

TIA 
Date: Wednesday 14 May 2008  
Time: 9:10 - 9:20 -  Room: Hermes
Chair: A.Carolei, Italy and P. Koudstaal, The Netherlands

05
MRI Should be the Primary Imaging Modality for Minor Stroke and TIA .
S.B.Coutts    M. Eliasziw    M.D.Hill    K.L.Fischer    S. Subramaniam    J.N.Scott    A.M.Demchuk                     
for the VISION study group.

University of Calgary

CANADA

MRI can identify early ischemia more accurately than CT, particularly with small volume ischemia, and can predict clinical outcome in stroke patients. We hypothesized that the usefulness of multimodal MRI in predicting functional outcome was greater among patients with milder stroke. METHODS: Ischemic stroke or TIA patients were prospectively enrolled if: examined within 12hours by a stroke neurologist; pre-morbid mRS 0 or 1; and MRI completed within 24hours. Primary outcome was mRS>1 at 90days. The effect of baseline clinical and imaging parameters on the primary outcome was assessed among those with minor stroke/TIA (baseline NIHSS 0-5) and those with moderate/severe strokes (NIHSS>5). The following variables: clinical (age>60 years, sex, hypertension, diabetes mellitus(DM) and blood glucose>7mM) and imaging (DWI lesion, intracranial vessel occlusion, perfusion/diffusion mismatch(mismatch) and microbleeds) were analyzed for prediction of functional impairment. RESULTS: 334 patients were prospectively enrolled. 229 patients were in the mild group and 105 in the moderate/severe group. Of the clinical parameters we found that DM p=0.01, blood glucose>7.0 p=0.005, and male sex p=0.048 were predictive of functional impairment in the mild group, but not in the more moderate/severe group. Of the MRI parameters we found that DWI lesion presence p<0.001, intracranial vessel occlusion p=0.007, and mismatch p=0.002 were predictive of outcome in the mild group and not in the moderate/severe group. Effect modification was present for DWI lesions p=0.004, vessel occlusion p=0.007 and mismatch p=0.016. We found that only baseline NIHSS predicted outcome in the moderate/severe group. DISCUSSION: In a prospective study we have found evidence of effect modification on the predictive value of MRI according to the NIHSS score. MRI is most useful as a predictor of functional impairment in minor stroke and TIA patients. This suggests that MRI should be considered the primary imaging modality of choice for patients with TIA/minor stroke.

TIA 
Date: Wednesday 14 May 2008  
Time: 9:20 - 9:30 -  Room: Hermes
Chair: A.Carolei, Italy and P. Koudstaal, The Netherlands

06
Impact of Systematic Transcranial Doppler Ultrasonography in a Transient Ischaemic Attack Clinic
E. Meseguer    P.C.Lavallee    M. Mazighi     J. Labreuche    L. Cabrejo    T. Slaoui    C. Guidoux    H. Abboud    P.J.Touboul    P. Amarenco
 

Bichat Claude Bernard Hospital

FRANCE

BACKGROUND: TIA patients with intracranial stenosis (IS) have a high risk of recurrence. Transcranial Doppler (TCD) can detect IS. However the yield of systematic TCD in the diagnosis of TIA has not been evaluated. METHODS: 1085 consecutive patients with TIA suspicion were admitted to the SOS-TIA clinic with round-the-clock access, immediate evaluation lasting <4 hours, and immediate multifactorial treatment interventions. TCD was systematically performed upon admission. IS was defined by high velocities (MCA>150 cm/s; siphon>90 cm/s; V4 and BA>120m/s). All patients had brain imaging and were evaluated for cardiac source of embolism (CSE). Patients were classified into definite with (n=108) and (n=535) without brain tissue damage or possible (n=144) TIA, or minor stroke (n=58) or TIA mimick (n=240). We stratified the 90-day stroke risk according to presence or absence of IS. RESULTS: IS was found in 11% and hemodynamic compromise in 3%. IS was more frequent in patients with definite TIA or minor stroke than in patients with possible TIA or TIA mimick (14% vs. 5%, p<0.001); patients with definite TIA and brain damage have the highest prevalence (20%). Ipsilateral severe carotid stenosis or occlusion, and definite CSE were found in 13% and 4% of IS, respectively. Admission in stroke unit and any modification in treatment of main risk factors were more frequent in patients with IS. 3 patients had urgent intracranial revascularisation because of early recurrence. The 90-stroke rate was 2.7% in IS patients, although expected rate from ABCD² score was 7.5%. CONCLUSION Same day systematic TCD evaluation is feasible and helps detect IS in 1 patients in 9 with a suspicion of TIA. Immediate aggressive treatment and early detection may explain a 60% lower stroke rate than expected.