London, United Kingdom 2013 18 Acute cerebrovascular events (ACE): TIA and minor strokes Magnetic resonance (MR) diffusion-weighted imaging (DWI) in patients with TIA: implica-tions for stroke prevention and reclassification in ICD11: Systematic review and meta-analysis M. Brazzelli1, F.M. Chappell2, H. Miranda3, K. Shuler4, M.S. Dennis5, P.A.G. Sandercock6, K. Muir7, J.M. Wardlaw8 University of Aberdeen, Aberdeen, UNITED KINGDOM1, University of Edinburgh, Edinburgh, UNITED KINGDOM2, Santiago Hospital, Santiago, CHILE3, University of Edinburgh, Edinburgh, UNITED KINGDOM4, University of Edinburgh, Edinburgh, UNITED KINGDOM5, University of Edinburgh, Edinburgh, UNITED KINGDOM6, University of Glasgow, Glasgow, UNITED KING-DOM7, University of Edinburgh, Edinburgh, UNITED KINGDOM8 Background: Patients with TIA are at high risk of stroke. Clinical diagnosis of TIA and distinction from TIA mimic may be difficult. MR with DWI might confirm ischaemic aetiology in suspected TIA if positive in most TIA patients. Re-definition of stroke to include DWI positive TIAs (ie symp-toms lasting <24 hours and an acute lesion on DWI) as stroke, not TIA, is being considered by the WHO. Methods: Using multiple overlapping methods, we searched the literature from 1995 to 2012 for all studies reporting the proportion of patients with TIA and an acute ischaemic lesion on MR DWI. We used PRISMA guidelines and 2 observers extracted data. We meta-analysed the pooled proportion with a DWI lesion (univariate random effects) and explored heterogeneity in sensitivity analyses. Results: 47 studies including 9,078 (range 18-1693) patients were included; 17 studies used retro-spective case ascertainment, diagnosis was by a stroke specialist in 26/47 studies and all studies ex-cluded TIA mimics. Most patients were scanned within 48 hours of TIA. The pooled proportion of TIA patients with a DWI lesion was 34% (95% CI31%-38%), range 9-67%, with substantial hetero-geneity (Isqu=89.3%). Larger studies (n>200) had lower DWI positivity (29%, 95%CI 23.2-34.6%) than studies with n<50 (40.1%, 95% CI 33.5-46.6%; p=0.035); study size was the only factor that accounted for heterogeneity; the doctors’ specialty, retro vs prospective, population vs hospital based, time to scanning and inclusion of TIA+stroke vs TIA only, had no effect on heterogeneity. Conclusion: The commonest DWI finding in patients with specialist-diagnosed TIA is a negative scan (66%): evidence-based stroke prevention treatment should not be denied on the basis of DWI findings. The 7-fold difference in DWI positivity and substantial unexplained heterogeneity means re-classifying DWI positive TIAs as strokes will add unmeasurable but large variance to estimates of stroke diagnosis and outcome in research and healthcare data E-Poster Session Blue Cerebrovasc Dis 2013; 35 (suppl 3)1-854 575
Karger_ESC London_2013
To see the actual publication please follow the link above