London, United Kingdom 2013 Poster Session Red Cerebrovasc Dis 2013; 35 (suppl 3)1-854 307 55 Stroke prognosis Five-simple-variables risk score predicts good and devastating outcome after stroke J.M. Reid1, D. Dai2, K. Thompson3, C. Christian4, Y. Reidy5, C. Counsell6, G.J. Gubitz7, S.J. Phil-lips8 Department of Neurology, Aberdeen Royal Infirmary, Aberdeen, UNITED KINGDOM1, Cen-ter for Paediatric Clinical Effectiveness (CPCE), The Children’s Hospital of Philadelphia,, Philadel-phia, USA2, Dalhousie University, Halifax, CANADA3, Department of Neurology, Queen Elizabeth II Health Sciences Centre, Halifax Infirmary, Halifax, CANADA4, Department of Neurology, Queen Elizabeth II Health Sciences Centre, Halifax Infirmary, Halifax, CANADA5, Division of Applied Health Sciences, University of Aberdeen,, Aberdeen, UNITED KINGDOM6, Department of Neurol-ogy, Queen Elizabeth II Health Sciences Centre, Halifax Infirmary, Halifax, CANADA7, Department of Neurology, Queen Elizabeth II Health Sciences Centre, Halifax Infirmary, Halifax, CANADA8 Introduction Several models and risk scores to predict outcome after stroke have been developed. We previously published outcome prediction models for excellent, good and devastating outcomes 6 months after stroke and here report two predictive risk scores for these outcomes. Methods The Stroke Outcome Study (2001-2) enrolled 538 ischemic and hemorrhagic stroke patients admitted to the Halifax Infirmary, Nova Scotia, Canada. Baseline clinical and demographic variables were col-lected. Outcome was assessed by modified Rankin score (mRS) at 6 months. Previously published models for excellent (mRS<2), good (mRS<3) and devastating outcomes (mRS>4) were used (Reid et al, Age Ageing. 2010;39:360-6; and 2012;41:560-4). Points were assigned to each variable from the models by dividing each regression β-coefficient by the smallest β-coefficient and rounding to the nearest integer. A predictive risk score was assigned to each patient by summing the points from each variable. Results Five baseline clinical variables were used from prior models; age<80 years, pre-stroke functional status, normal verbal GCS, ability to walk unaided, and lift both arms. For the devastating outcome score, the variable total anterior circulation stroke was included and ability to walk excluded. An identical 7 point predictive score was developed for excellent and good out-comes, and a 12-point score for a devastating outcome. Area under curve (AUC) values of 0.859- 0.886 were observed for all three outcomes. The scores were externally tested in the OCSP dataset with AUC values of 0.847-0.884. Optimal cut points dichotomizing each score showed sensitivities of 64-72% and specificities of 85-91% for prediction of each outcome. Conclusion Scores based on five simple clinical variables can be used to predict outcomes 6 months post-stroke. It remains to be seen whether these scores offer additional clinical utility over and above informal physician predic-tion, or are comparable to more complex predictive scores. 56 Stroke prognosis PROGNOSTIC VALUE OF INFARCT VOLUME IN ACUTE ISCHEMIC STROKE. ANAL-YSES OF 220 PATIENTS FROM THE SYNTHESIS EXPANSION TRIAL C. Motto1, S. Lanfranconi2, P. Doneda3, P. Pini4, P. Basilico5, M.L. DeLodovici6, N. Checcarelli7, A. Zini8, L. Malfatto9, P. Nencini10, G. Pero11, L. Quilici12, E. Agostoni13, A. Ciccone14 Synthesis Expansion Investigators Stroke Unit - AO Niguarda Ca’ Granda, Milano, ITALY1, Ospedale Policlinico, Milano, ITALY2, Neuroradiology AO Niguarda Ca’ Granda, Milano, ITALY3, Ospedale Policlinico, Milano, ITALY4, Ospedale Policlinico, Milano, ITALY5, Ospedale di Circolo, Varese, ITALY6, Ospedale Valduce, Como, ITALY7, Ospedale di Baggiovara, Modena, 8, Ospedale S. Martino, Genova, ITALY9,Osped-ale Careggi, Firenze, ITALY10, Neuroradiology AO Niguarda Ca’ Granda, Milano, ITALY11, Neu-roradiology AO Niguarda Ca’ Granda, Milano, ITALY12, Neurology-Stroke Unit AO Niguarda Ca’ Granda, Milano, ITALY13, Ospedale Poma, Mantova, ITALY14 Background: The rationale of reperfusion therapy is to save penumbral ischemic tissue and reduce infarct volume improving clinical outcome. The aim of this study was to evaluate the prognostic value of infarct volume in acute ischemic stroke patients treated with intravenous (IV) t-PA or en-dovascular treatment. Methods: We analyzed 220 patients prospectively included in the Synthesis Expansion study, a randomised multicenter controlled trial on fast track endovascular treatment compared to IV t-PA for acute ischemic stroke. Basal clinical and radiological features, clinical out-come (7-day NIHSS, 90-day mRS), and infarct volume were evaluated. All basal and control CT scans were centrally reviewed blinding to treatment allocation and patient outcome. Lesion volumes were visually estimated and manually traced using freeware software for images analysis (MRIcro 1.40 software). Results: Mean infarct volume on control CT scan (2.4+/-1.2 days) was 76.9+/-118.0 cc. Lesion volume was significantly associated with 7-day NIHSS and 90-day mRS (Spearman co-efficient 0.55 and 0.45, respectively, p<0.0001). A linear correlation between increase of volume and 90-day mRS worsening was observed. Mean infarct volume in patients with 90-day mRS</=1 was 12.6+/-22.1 cc and 110.3+/-132.9 cc in patients with 90-day mRS >1 (p<0.0001, z=7.965). In ROC Analysis, volume predicts 90-day outcome with high accuracy (AUC 0.827), and a volume of 50 cc forecasts unfavorable 90-day outcome with 95% specificity and 58% sensibility. Conclusion: Infarct volume predicts short and long term outcome in patients with acute ischemic stroke treated with IV or endovascular thrombolysis. Although early measurement of volume could overestimate the final infarct lesion due to edema, it could be considered as a surrogate biomarker of prognosis useful in research setting and in clinical practice.
Karger_ESC London_2013
To see the actual publication please follow the link above