22. European Stroke Conference 9 Acute cerebrovascular events (ACE): TIA and minor strokes 9:50 - 10:00 Predictive factors for transient ischaemia attacks (TIA) and stroke in an open access TIA clinic Y.K. Kee1, S. Mahmood2, E. Lawrence3 Croydon University Hospital, London, UNITED KINGDOM1,Croydon University Hospital, London, UNITED KINGDOM2, Croydon University Hospital, London, UNITED KINGDOM3 Background: To determine the predictive factors for transient ischaemic attack (TIA) or stroke in patients attending an open access TIA clinic. Methods: Data were collected prospectively over a 14-month period from consecutive atten-dances in an open access TIA clinic. Clinical and demographic data were collected, including previous history of TIA/stroke, migraine, hypertension, hypercholesterolemia, diabetes, smok-ing, and obesity. Data regarding clinical features such as headaches, vertigo, dysphasia, dysar-thria, dysphagia, limb weakness, sensory loss, visual disturbance, and diplopia were also col-lected. Results: 851 patients were included. 264 patients (31.0%) had a final diagnosis of TIA or stroke. Hypertension was found to be the most predictive of a diagnosis of CVA or TIA (OR=1.50, 95% CI 1.11 to 2.04, p=0.009), followed by a previous history of TIA/CVA (OR=1.46, 95% CI 1.02 to 2.08, p=0.04). When modelled with other risk factors, the associa-tion remained significant (p=0.04). The clinical features with highest predictive value were dys-arthria (OR=2.50, 95% CI 1.68 to 3.71, p<0.0001), dysphasia (OR=5.24, 95% CI 3.45 to 7.96, p<0.0001), and limb weakness (OR=2.04, 95% CI 1.44 to 2.89, p=0.0001). These factors also remained highly significant when modelled with other clinical features. (p<0.0001) In this co-hort, clinical signs and symptoms alone, were superior to the ABCD4 risk score established at the time of referral (OR=1.15, 95% CI 1.01 to 1.31, p=0.04). Conclusions: The ABCD2 score is a clinically useful score in identifying a raised risk of stroke patients with TIA. However, in an open access TIA clinic, this score was less useful. Individual risk factors such as hypertension, past history of TIA/CVA, or clinical features such as dysar-thria, dysphagia and limb weakness, were more predictive of a diagnosis of TIA or stroke. A simpler scoring system, utilizing the above features, may be more useful screening patients at-tending an open access TIA clinic. 8 Acute cerebrovascular events (ACE): TIA and minor strokes 9:40 - 9:50 Performance of ABCD2 score in secondary stroke prevention; meta-analysis of recurrent stroke, proportions with risk factors and ‘effect per 1000 patients triaged’. M. Brazzelli1, F.M. Chappell2, H. Miranda3, K. Shuler4, M.S. Dennis5, P.A.G. Sandercock6, J.M. Wardlaw7 University of Aberdeen, Aberdeen, UNITED KINGDOM1,University of Edinburgh, Edin-burgh, UNITED KINGDOM2, Santiago Hospital, Santiago, CHILE3, University of Edinburgh, Edinburgh, UNITED KINGDOM4, University of Edinburgh, Edinburgh, UNITED KING-DOM5, University of Edinburgh, Edinburgh, UNITED KINGDOM6, University of Edinburgh, Edinburgh, UNITED KINGDOM7 Background: The ABCD2 score is intended to triage TIA patients at high risk of stroke for rapid investigation and treatment. Many guidelines now recommend triage on ABCD2 score >/=4 vs <4, but the effect of ABCD2 score triage on patient care is unclear. Methods: We used multiple methods to identify all published studies between 2005-2012 on recurrent stroke after TIA, proportions of true TIAs and mimics and with risk factors, dichoto-mised at ABCD2 >/=4. Two reviewers extracted data. We calculated the recurrent stroke rate, proportion with risk factors and of true TIAs or mimics triaged as < v>/=4 in a hypothetical co-hort of 1,000 patients attending TIA clinics. Results: 26 studies (28 reports), included 12,586 TIA patients (range 69-1679). All studies used time-based TIA definition, 42% calculated ABCD2 score retrospectively, none said if carot-id stenosis or AF were treated. The pooled risk of recurrent stroke at 7 and 90 days in all 26 studies had high heterogeneity: in 9 studies reporting at both times it was: ABCD2 >/=4, 4.7% (95%CI 2.4-8.7%) and 8.2 (4.7-14%); ABCD2 <4, 1.6% (1-3.4%) and 2.7% (1.5-4.7%); at 7 days, sensitivity was 85.8% (80.4-90.0%), specificity 36.1% (30.6-42.1%). 20% of patients with ABCD2 <4 had key risk factors, eg tight carotid stenosis or AF. ABCD2 did not distin-guish TIA mimics: 64% of true TIAs and 35-41% of mimics had ABCD2 score >/=4. Amongst 1,000 patients attending a TIA clinic including mimics, 52% of patients would have an ABCD2 score >/=4. Conclusion: ABCD2 does not discriminate between TIAs and mimics, between those at high and low risk, or help identify carotid stenosis and AF. Ist use in triage does not significantly reduce the total number of patients who need specialist assessment since many patients with ABCD2<4 are at significant stroke risk and require urgent treatment, TIA services need the ca-pacity to assess all patients properly and promptly to avoid high risk patients being denied im-mediate evidence-based treatments. 104 © 2013 S. Karger AG, Basel Scientific Programme
Karger_ESC London_2013
To see the actual publication please follow the link above