Journal
NEUROLOGY
Volume 73, Issue 21, Pages 1759-1766Publisher
LIPPINCOTT WILLIAMS & WILKINS
DOI: 10.1212/WNL.0b013e3181c34a7d
Keywords
-
Categories
Funding
- UK Department of Health's National Institute for Health Research Biomedical Research Centers funding scheme
- Stroke Association
- Higher Education Commission of Pakistan
- Inverness Medical (Biosite, Inc.)
- Medical Research Council, UK
- Reta Lila Weston Trust for Medical Research
- Samantha Dickson Brain Tumour Trust
- Brain Research Trust
- Department of Health/Higher Education Funding Council for England
Ask authors/readers for more resources
Objective: Brain microbleeds on gradient-recalled echo (GRE) T2*-weighted MRI may be a useful biomarker for bleeding-prone small vessel diseases, with potential relevance for diagnosis, prognosis (especially for antithrombotic-related bleeding risk), and understanding mechanisms of symptoms, including cognitive impairment. To address these questions, it is necessary to reliably measure their presence and distribution in the brain. We designed and systematically validated the Microbleed Anatomical Rating Scale (MARS). We measured intrarater and interrater agreement for presence, number, and anatomical distribution of microbleeds using MARS across different MRI sequences and levels of observer experience. Methods: We studied a population of 301 unselected consecutive patients admitted to our stroke unit using 2 GRE T2*-weighted MRI sequences (echo time [TE] 40 and 26 ms). Two independent raters with different MRI rating expertise identified, counted, and anatomically categorized microbleeds. Results: At TE = 40 ms, agreement for microbleed presence in any brain location was good to very good (intrarater kappa = 0.85 [95% confidence interval (CI) 0.77-0.93]; interrater kappa = 0.68 [95% CI 0.58-0.78]). Good to very good agreement was reached for the presence of microbleeds in each anatomical region and in individual cerebral lobes. Intrarater and interrater reliability for the number of microbleeds was excellent (intraclass correlation coefficient [ICC] = 0.98 [95% CI 0.97-0.99] and ICC = 0.93 [0.91-0.94]). Very good interrater reliability was obtained at TE = 26 ms (kappa = 0.87 [95% CI 0.61-1]) for definite microbleeds in any location. Conclusion: The Microbleed Anatomical Rating Scale has good intrarater and interrater reliability for the presence of definite microbleeds in all brain locations when applied to different MRI sequences and levels of observer experience. Neurology (R) 2009; 73: 1759-1766
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available