Prediction of final infarct volume on subacute MRI by quantifying cerebral edema in ischemic stroke.

Journal: Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism
Published Date:

Abstract

Final infarct volume in stroke trials is assessed on images obtained between 30 and 90 days after stroke onset. Imaging at such delayed timepoints is problematic because patients may be lost to follow-up or die before the scan. Obtaining an early assessment of infarct volume on subacute scans avoids these limitations; however, it overestimates true infarct volume because of edema. The aim of this study was to develop a novel approach to quantify edema so that final infarct volumes can be approximated on subacute scans. We analyzed data from 20 stroke patients (median age, 75 years) who had baseline, subacute (fu5d) and late (fu90d) MRI scans. Edema displaces CSF from sulci and ventricles; therefore, edema volume was estimated as change in CSF volume between baseline and spatially coregistered fu5d ADC maps. The median (interquartile range, IQR) estimated edema volume was 13.3 (7.5-37.7) mL. The fu5d lesion volumes correlated well with fu90d infarct volumes with slope: 1.24. With edema correction, fu5d infarct volumes are in close agreement, slope: 0.97 and strongly correlated with actual fu90d volumes. The median (IQR) difference between actual and predicted infarct volumes was 0.1 (-3.0-5.7) mL. In summary, this novel technique for estimation of edema allows final infarct volume to be predicted from subacute MRI.

Authors

  • Aaryani Tipirneni-Sajja
    1 Stanford Stroke Center, Stanford University Medical Center, Stanford, CA, USA.
  • Soren Christensen
    1 Stanford Stroke Center, Stanford University Medical Center, Stanford, CA, USA.
  • Matus Straka
    1 Stanford Stroke Center, Stanford University Medical Center, Stanford, CA, USA.
  • Manabu Inoue
    4 Department of Neurology, Showa University Fujigaoka Hospital, Yokohama, Japan.
  • Maarten G Lansberg
    1 Stanford Stroke Center, Stanford University Medical Center, Stanford, CA, USA.
  • Michael Mlynash
    1 Stanford Stroke Center, Stanford University Medical Center, Stanford, CA, USA.
  • Roland Bammer
    1 Stanford Stroke Center, Stanford University Medical Center, Stanford, CA, USA.
  • Mark W Parsons
    5 Department of Neurology, John Hunter Hospital, University of Newcastle, Newcastle, NSW, Australia.
  • Geoffrey A Donnan
    6 Florey Neuroscience Institutes, Austin Health, University of Melbourne, Melbourne, VIC, Australia.
  • Stephen M Davis
    7 Royal Melbourne Hospital, University of Melbourne, Melbourne, VIC, Australia.
  • Gregory W Albers
    1 Stanford Stroke Center, Stanford University Medical Center, Stanford, CA, USA.