Page 310

Karger_ESC London_2013

22. European Stroke Conference 60 Stroke prognosis Transcranial Doppler CO2-reactivity does not predict recurrent ischemic stroke in patients with symptomatic carotid artery occlusion W.M.T. Jolink1, R. Heinen2, S. Persoon3, A. van der Zwan4, L.J. Kappelle5, C.J.M. Klijn6 University Medical Center Utrecht, Utrecht, THE NETHERLANDS1, University Medical Cen-ter Utrecht, Utrecht, THE NETHERLANDS2, University Medical Center Utrecht, Utrecht, THE NETHERLANDS3, University Medical Center Utrecht, Utrecht, THE NETHERLANDS4, Univer-sity Medical Center Utrecht, Utrecht, THE NETHERLANDS5, University Medical Center Utrecht, Utrecht, THE NETHERLANDS6 Background: Patients with a transient ischemic attack (TIA) or minor disabling ischemic stroke as-sociated with occlusion of the internal carotid artery (ICA) have a higher risk of recurrent stroke in case of a compromised cerebral blood flow. The aim of our study was to investigate whether tran-scranial Doppler ultrasonography with measurement of CO2-reactivity is an independent predictor of recurrent ischemic stroke in a large cohort of patients with symptomatic ICA occlusion who re-ceive optimal medical treatment. Methods: We included consecutive patients with symptomatic ICA occlusion and transient or minor disabling stroke (modified Rankin scale ≤3) within the last 6 months between July 1995 and November 2007. CO2-reactivity was measured in all patients within three months after presentation. Patients were followed up until October 2010. We determined the predictive value of CO2-reactivity for recurrent ischemic stroke using Cox proportional hazard analysis. Results: We included 193 patients with a median follow-up time of 6.9 years. Mean CO2-reactivity was 15% (SD 20). The annual rate for ipsilateral ischemic stroke was 2.0% (95% confidence inter-val (CI) 1.4- 2.9) and for any recurrent stroke 3.1% (95% CI 2.3-4.2). We did not find a significant relationship between CO2-reactivity and the risk of ipsilateral (HR 1.01, 95% CI 0.99-1.02) or any recurrent ischemic stroke (HR 1.01, 95% CI 1.00-1.03). Conclusion: CO2-reactivity does not predict recurrent ischemic stroke in patients with recent tran-sient or minor disabling ischemic stroke that is associated with an ICA occlusion during a time that rigorous control of vascular risk factors has been widely implemented in clinical practice. 310 © 2013 S. Karger AG, Basel Scientific Programme 62 Stroke prognosis The performance of prognostic models for predicting occlusive vascular events after stroke: a systematic review D.D. Thompson1, G.D. Murray2, W.N. Whiteley3 University of Edinburgh, Edinburgh, UNITED KINGDOM1, University of Edinburgh, Edin-burgh, UNITED KINGDOM2, University of Edinburgh, Edinburgh, UNITED KINGDOM3 BACKGROUND: Prediction models for recurrent ischaemic stroke or myocardial infarction (MI) after ischaemic stroke may be useful in targeting treatment. We aimed to systematically review the available predic-tion models. We studied (i) the methodological quality of the models and (ii) their related measures of predictive power. METHODS: We searched Medline, EMBASE, reference lists and forward citations of relevant articles from 1980 to the 7th of February 2012. We included articles which developed a multivariate statistical model to predict recurrent stroke and MI after ischaemic stoke. We extracted data in duplicate using a val-idated data extraction form. We assessed model quality using pre-defined criteria and aimed to pool performance metrics (calibration and discrimination) using random-effects meta-analysis. RESULTS: We identified ten model development studies and ten evaluation studies. Investigators often did not report effective sample size, beta coefficients, handling of missing data; typically categorised con-tinuous predictors; and used data dependent methods to build models (e.g., univariate screening of predictors). Multiple evaluations were only available for two of the ten models. The pooled area un-der the receiver operating characteristic curve (AUROCC) estimate for the Essen Stroke Risk Score (ESRS) was 0.60 (95% CI 0.59 to 0.62) and for the Stroke Prognosis Instrument II (SPI-II) was 0.62 (95% CI 0.60 to 0.63). CONCLUSIONS: Few prediction models for recurrent stroke and MI after ischaemic stroke have been developed. The available models discriminated only modestly between patients with and without a recurrent stroke or MI. The performance of those models not evaluated remains uncertain; however, given the meth-odological flaws that we identified, it appears that improvements in model development are required before evaluation studies demonstrate better performance.


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