Page 652

Karger_ESC London_2013

22. European Stroke Conference 581 Acute stroke: emergency management, stroke units and complications Are decisions for IV thrombolysis reliable? G. MUTLU1, E. MORVAN2, Y. SAMSON3 SERVICE DE NEUROLOGIE ET UNITE NEUROVASCULAIRE - CENTRE HOSPI-TALIER VERSAILLES, LE CHESNAY, FRANCE1, URGENCES CEREBROVASCULAIRES - SALPETRIERE, PARIS, FRANCE2, URGENCES CEREBROVASCULAIRES - SALPETRIERE, PARIS, FRANCE3 Background. Intravenous thrombolysis for stroke has become a routine treatment. However decisions for treat-ment (or not) sometimes remain difficult, which could affect homogeneity of patient care. Studies on reliability of decision for thrombolysis appear scarce. We sought to measure reliability of decision taken by stroke-neurologists for a panel of cases extracted from real-life. Methods Real life clinical cases were extracted from a stroke patients database, either treated with thrombol-ysis or not, from 2009 to 2010, in Urgences Cerebrovasculaires, Salpetriere Hospital, Paris, France. Short cases (with clinical features, past medical history, brain imaging and biology results) were presented to 13 stroke-neurologists (8 senior and 5 junior). Raters should answer to the questions « do you treat the patient with IV thrombolysis ? » and « Is it a common situation or a difficult case ? », blinded to the decision really taken. Reliability was measured with Fleiss kappa, and considered « Poor » (kappa = 0), « Slight » (0-0.20), « Fair » (0.21-0.40), « Moderate » (0.41-0.60), « Substantial » (0.61-0.80), or « Almost Perfect » (0.81-1). Reliability of the decisions for « Common cases » and « Difficult cases » were compared using bootstrap resampling method. Results The stroke neurologists individually rated the 37 cases from october 10th to 19th (2012), with a mean total rating time of 42 min. Cases were considered as « Common case » for 24/37 and « Diffi-cult cases » for 13/37. Fleiss Kappa was 0.70 (« substantial ») for the « Common cases » and 0.27 (« Fair ») for « Difficult cases ». The poorer reliability for « Difficult cases » was significant (p < 0.05). Conclusion Decisions for IV-thrombolysis remain faced with variability, even in a single intensive stroke unit, with a poorer reliability observed for « Difficult cases ». This result should be taken in account when considering homogeneity of patient care and when designing procedures to minimize it. 652 © 2013 S. Karger AG, Basel Scientific Programme 582 Acute stroke: emergency management, stroke units and complications Withdrawn!


Karger_ESC London_2013
To see the actual publication please follow the link above