[Skip to Navigation]
Views 229
Citations 0
Comment & Response
June 2017

Managing Patients With Large Ischemic Core—What Is in a Match?

Author Affiliations
  • 1Department of Medicine, Monash University, Clayton, Victoria, Australia
  • 2Monash University, Clayton, Victoria, Australia
  • 3Monash Medical Centre, Clayton, Victoria, Australia
JAMA Neurol. 2017;74(6):746. doi:10.1001/jamaneurol.2017.0168

To the Editor We read with interest the article by Rebello and colleagues1 on endovascular treatment. There is still uncertainty on how best to manage patients who have a large ischemic core and a large artery clot. Using matched controls is an excellent approach when the study is not a randomized clinical trial. The authors modified the approach of Mandava and colleagues,2 who had proposed matching controls based on the National Institutes of Health Stroke Scales, age, and pretreatment glucose level. The authors had substituted an “ischemic core on computed tomography perfusion” for the National Institutes of Health Stroke Scales.1 However, we are uncertain that the matching approach2 used by the authors provided balance between the 2 arms (treated and untreated) since it had not addressed infarct location.3 We cannot find the term registration in the article and presumed, perhaps incorrectly, that the authors had used ischemic core volume and not location. Location analysis requires the coregistration of brain images. Further, the authors had not specified the laterality of lesions, and the matching of volume between the left and right hemispheres may not be appropriate because the associated stroke deficits are not the same.4 For consistency with their hypothesis of using “computed tomography perfusion imaging” data, the authors can perform matching by the coregistration of the segmented lesions to a standard brain image and identify cases as suitable for matching in which there are high numbers of voxels which intersect. Finally, we are unclear how the authors identified “involvement of eloquent areas (sensorimotor, language, spatial attention, or visual functions).” Was this performed qualitatively? There were no template images listed on which to base this decision.