Background and objectives: Controversy exists about the optimal imaging technique in acute stroke. It was hypothesised that CT is comparable with DWI, when both are read systematically using quantitative scoring. Methods: Ischaemic stroke patients who had CT within six hours and DWI within seven hours of onset were included. Five readers used a quantitative scoring system (ASPECTS) to read the baseline (b) and follow up CT and DWI. Use of MRI in acute stroke was also assessed in patients treated with tissue plasminogen activator (tPA) by prospectively recording reasons for exclusion. Patients were followed clinically at three months.Results: bDWI and bCT were available for 100 consecutive patients (admission median NIHSS = 9). The mean bDWI and bCT ASPECTS were positively related (p,0.001). The level of interrater agreement ranged from good to excellent across all modalities and time periods. Bland-Altman plots showed more variability between bCT and bDWI than at 24 hours. The difference between bCT and bDWI was (2 ASPECTS points. Of bCT scans with ASPECTS 8-10, 81% had DWI ASPECTS 8-10. Patients with bCT ASPECTS of 8-10 were 1.9 times more likely to have a favourable outcome at 90 days than those with a score of 0-7 (95% CI 1.1 to 3.1, p = 0.002). The relative likelihood of favourable outcome with a bDWI ASPECTS 8-10 was 1.4 (95% CI 1.0 to 1.9, p = 0.10). Of patients receiving tPA 45% had contraindications to urgent MRI. Conclusion: The differences between CT and DWI in visualising early infarction are small when using ASPECTS. CT is faster and more accessible than MRI, and therefore is the better neuroimaging modality for the treatment of acute stroke.

Results from the present study suggest that quantifying mismatch by the human eye is reproducible but not reliable among observers. This raises doubts about using mismatch for clinical decision making and clinical trial enrollment.

THE CANADIAN JOURNAL OF NEUROLOGICAL SCIENCESMagnetic resonance angiography (MRA) allows visualisation of both extra-and intra-cranial vessels in a fast and non-invasive manner and can play an important role in clinical decisionmaking. Assessment of the intra-cranial vessels 1-5 has led to multiple improvements of the technique with a significant increase in the accuracy of intra-cranial MRA. Prior research has shown benefit of careful review of the 3D TOF source images ABSTRACT: Background and purpose: Accuracy of intracranial magnetic resonance angiography (MRA) and reliability of interpretation are not well established compared to conventional selective catheter angiography. The purpose of this study was to determine the accuracy of MRA in evaluation of intracranial vessels in acute stroke and transient ischemic attack (TIA) patients. Methods: Twenty-nine patients (seven females, 22 males; median age 53) with acute stroke or TIA were enrolled into the study. All patients underwent both MRA using a 3 T clinical magnet and conventional angiography within 48 hours. Median time between MRA and angiography was 263 minutes. Conventional angiography preceded MRA in 15 cases. Fourteen patients received thrombolysis during MRA or angiography. National Institutes of Health Stroke Scale scores were obtained prior to the MR exam. One neuroradiologist rated all conventional angiograms, which were used as gold standard. Five observers, blinded to conventional angiography results and all clinical information except symptom side, rated the MR angiograms. Kappa statistics were used to assess reliability; contingency tables were used to assess accuracy of non-enhanced and enhanced MRA. Results: Two hundred and fifty two intracranial vessels were assessed. Agreement between raters was good for both non-enhanced (κ=0.50) and gadolinium-enhanced (κ=0.46) images. There were a total of 26 vessels occluded by DSA. Overall, the non-enhanced MRA showed sensitivity of 84.2% (95% CI 60.4-96.6) and specificity of 84.6% (95% CI 78.6-89.4). The enhanced MRA showed sensitivity of 69.2 (95% CI 38.6-90.9) and specificity of 73.6 (95% CI 65.5-80.7). Conclusions: Magnetic resonance angiography is a good non-invasive screening tool for assessing intracranial vessel status in acute ischemic stroke. Angiography remains the gold standard for definitive assessment of the intracranial circulation. Le score au National Institutes of Health Stroke Scale a été déterminé avant de faire l'ARM. Un neuroradiologiste a lu tous les angiogrammes conventionnels qui ont été utilisés comme examen de référence. Cinq observateurs qui ignoraient les résultats de l'angiographie conventionnelle ainsi que l'information clinique sauf la latéralité des symptômes, ont lu les angiogrammes par résonance magnétique. La fiabilité a été évaluée au moyen de l'analyse statistique kappa; la précision de l'ARM non rehaussée et rehaussée a été évaluée au moyen de tables de contingence. Résultats: Deux cent cinquante-deux vaisseaux intracrâniens ont été évalués. La concordance entre...