Article Text

Download PDFPDF
E-061 Rapid blood brain barrier permeability imaging in a canine ischemic re-perfusion model
  1. R King,
  2. J Caroff,
  3. M Gounis
  1. Radiology, University of Massachusetts Medical School, Worcester, MA


Introduction Ischemic stroke is often found to cause breakdown in the blood brain barrier1 (BBB) that persists even after reperfusion, which has been linked to an increased risk of hemorrhagic transformation. Currently in clinical practice CT is used calculate BBB permeability; however, a more precise assessment of this risk would be very useful in cases of large infarct territories or when a stent is delivered. Using MRI based T2W-EPI approaches permeability of the BBB can be calculated, but they require an accurate arterial input function (AIF) and field inhomogeneity correction. Here we combine multiple different MR sequences to solve both issues in a canine reperfusion model.

Methods T2W perfusion MRI was performed on four canines after a 45 min temporary MCA occlusion. MR imaging was achieved using a T2W-EPI based sequence, consisting of 60 dynamics, acquisition parameters were: TR 1500, TE 20, FA 40o, pixel size 1.1 mm × 1.1 mm, slice thickness 3 mm. Contrast injection parameters was set to 0.2 mg/kg gadopentate (Magnevist, Bayer Healthcare) at a rate of 2 mL/s, followed by 20 mL of saline, also at 2 mL/s, started immediately after the 2nd dynamic. Once the MR data is acquired, it is exported to MATLAB for calculation of permeability (Ktrans).

In order to calculate Ktrans, first the AIF was extracted from a RoI consisting of the ICA, selected from ToF MRA. Once the AIF was extracted it was modeled as a double gamma variate2 allowing for matching of both the initial contrast injection, and the recirculating bolus. With the modeled AIF pixel by pixel calculation of Ktrans can be performed under the assumption of the Tofts3 two compartment model with both contrast flux into the extra-vascular space, and back flux into the vascular space (Ktrans and Kep). Due to inherent noise, small RoIs are drawn to compare Ktrans. Three general areas were chosen for comparison of Ktrans: areas of ischemic lesion (confirmed by DWI), area within the MCA territory that did not show DWI abnormality, and tissue that remained fully perfused throughout the temporary occlusion.

Results In three of the four canines all three regions of analysis were available, in the last canine there was no area of stroke confirmed on the DWI. When comparing the three regions of interest, territory that was supplied by the occluded MCA was found to have a 91% increase in Ktrans over the contralateral side; while the tissue that was within the DWI lesion was found to have a 366% increase in Ktrans. Due to small sample size (n=4, 4, 3 respectively) no statistical significance was found.

Conclusions Ktrans imaging of the brain was achieved using a modified post processing of a standard T2W perfusion imaging MR sequence. It can be seen that tissue within areas of ischemic lesion have a large increase in permeability of the BBB compared to regions that remained fully perfused, and that tissue that was not fully infarcted, but did have reduced perfusion saw a smaller, but noticeable increase in permeability.


  1. JNIS 2014;6:139–143.

  2. JMRI 2006;24:288–296.

  3. JMRI 1997;7:91–101.

Disclosures R. King: None. J. Caroff: None. M. Gounis: 1; C; Phillips Heathcare.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.