Page 571

Karger_ESC London_2013

London, United Kingdom 2013 28 Acute stroke: emergency management, stroke units and complications Body temperature variability at the acute phase of stroke as a marker of poor outcome after IV-tPA G. Turc1, M. Tisserand2, D. Calvet3, M. Apoil4, E. Touzé5, C. Oppenheim6, J.L. Mas7 Department of Neurology, Université Paris Descartes Sorbonne Paris Cité, INSERM UMR S894, Hopital Sainte Anne, Paris, FRANCE1, Department of Radiology, Université Paris Des-cartes Sorbonne Paris Cité, INSERM UMR S894, Hopital Sainte Anne, Paris, FRANCE2, Depart-ment of Neurology, Université Paris Descartes Sorbonne Paris Cité, INSERM UMR S894, Hopital Sainte Anne, Paris, FRANCE3, Department of Neurology, Université Paris Descartes Sorbonne Paris Cité, INSERM UMR S894, Hopital Sainte Anne, Paris, FRANCE4, Department of Neurology, Université Paris Descartes Sorbonne Paris Cité, INSERM UMR S894, Hopital Sainte Anne, Paris, FRANCE5, Department of Radiology, Université Paris Descartes Sorbonne Paris Cité, INSERM UMR S894, Hopital Sainte Anne, Paris, FRANCE6, Department of Neurology, Université Paris Des-cartes Sorbonne Paris Cité, INSERM UMR S894, Hopital Sainte Anne, Paris, FRANCE7 Background: Elevated body temperature (BT) within the first hours of stroke has been shown to be independently associated with poor outcome. However, the prognostic value of BT variability at the acute phase remains to be determined. Objective: To assess the relationships between BT variability within 24 hours after IV-tPA and clinical outcome at 24 hours and 3 months. Methods: We reviewed prospectively collected clinical and radiological data of all patients treated ≤4.5h by IV-tPA between 2003 and 2012 in our center, where MRI is implemented as first-line diagnostic imaging. Mean, standard deviation (SD), coefficient of variation, and index of variation of BT were determined. 24- hour and 3-month poor outcomes were defined as ≤40% decrease in baseline NIHSS and modified Rankin Scale >2, respectively. Results: 298 patients were treated by IV-tPA during the study period. 240 patients had >= 3 BT measurements within 24 hours after treatment and were included in the analysis. Mean age was 68.4 ±14.4, median NIHSS at admission was 13 (IQR 8-19). 24-hour and 3-month poor outcomes were observed in 140 (58%) and 109 (45%) patients, respectively. Mean ±SD BT was 37.0 ±0.3°C over 24 hours. In univariate analysis, SD BT was associated with 24-hour (OR 10.5; 95%CI 2.5-44.2, per 1 SD increase) and 3-month (OR 10.6; 2.7-41.8) poor outcomes, while mean BT was not (3-month poor outcome : OR 1.4; 0.7-3.0 per 1°C increase). In multivar-iate analyses, SD BT remained associated with 24-hour (OR 7.6; 1.4-40.2) and 3-month (OR 6.4; 1.1-38.3) poor outcomes, after adjustment for mean BT, age, baseline NIHSS, DWI-ASPECTS<=7, baseline glucose level, onset to treatment time, and 24-hour recanalization. Sensitivity analyses using index or coefficient of variation of BT showed similar results. Conclusion: BT variability is more strongly associated than mean BT with 24-hour and 3-month poor outcomes in acute stroke patients treated by IV-tPA. E-Poster Session Blue Cerebrovasc Dis 2013; 35 (suppl 3)1-854 571 27 Acute stroke: emergency management, stroke units and complications High-resolution mapping of the collateral circulation deficit in patients with acute MCA-M1 occlusion by Bayesian processing of MR perfusion-weighted images F. Nicoli1, F. Pautot2, F. Scalzo, PhD3, J.L. Saver, MD4, A. Mitulescu, PhD5, Y. Chaibi, PhD6, N. Gi-rard, MD PhD7, N. Salamon, MD8, D.S. Liebeskind, MD, for the UCLA Stroke Investigators9 Olea Medical, La Ciotat, FRANCE1, Olea Medical, La Ciotat, FRANCE2, UCLA Stroke Center, Los Angeles, CA, USA3, UCLA Stroke Center, Los Angeles, CA, USA4, Olea Medical, La Ciotat, FRANCE5, Olea Medical, La Ciotat, FRANCE6, Department of Neuroradiology, La Timone Univer-sity Hospital, Marseille, FRANCE7, UCLA Stroke Center, Los Angeles, CA, USA8, UCLA Stroke Center, Los Angeles, CA, USA9 BACKGROUND: The normalized Collateral Circulation Deficit (nCCD) has recently been intro-duced as an index of collateral flow deficit (CFD). Inconsistency has limited the establishment of a reliable noninvasive collateral flow imaging method. The purpose of this study is to test if a Bayes-ian processing of DSC MR perfusion-weighted images could provide quantification and mapping of CFD in patients with MCA-M1 occlusion. METHODS The Bayesian processing allowed quanti-fication of collateral perfusion speed from standard perfusion MRI (TR=2sec.) for systemic(n=26) or endovascular(n=57) revascularization via the relative difference of the hemispheric maximal en-hancement rate (rdMER) calculated from Bayesian TTP maps (temporal resolution=0.5sec.). Bayes-ian arterial-tissue delay (ATD) maps (bATD) and volume of tissue with severely increased ATD(Vol- ATD6) were also calculated. Correlations between nCCD, rdMER, relative CBF(rCBF) inside the area with increased ATD and VolATD6 were tested in the IV thrombolysis cohort, and the correla-tion between VolATD6 and angiographic collateral flow grade (ACG) was tested in the IA cohort. RESULTS: rdMER is highly positively correlated with nCCD and VolATD6(logarithmic regression; r2=0.45, p=0.0001, and r2=0.56, p<0.0001 respectively). VolATD6 is significantly correlated to nCCD and rCBF (linear regression; r2=0.72, p<0.0001 and r2=0.63, p<0.0001 respectively). The higher the VolATD6, the higher the nCCD and the lower the rCBF. VolATD6 is negatively correlated with the ACG (Kendall tau=-0.35, p=0.0006). Patients with ACG >2 or>1 have a significantly lower ATD6 value than patients with ACG ≤2 or≤1(Wilcoxon test, p<0.0001 and p=0.044 respectively). The reproducibility of the VolATD6 measurement is almost perfect (ICC (95%CI)=0.994(0.988- 0.997)). CONCLUSION: High-resolution TTP and ATD maps provided by a Bayesian processing of MR PWI data allow a noninvasive and reliable quantification and mapping of CFD in patients with MCA-M1 occlusion.


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