Klinische Neurophysiologie 2010; 41 - ID96
DOI: 10.1055/s-0030-1250925

Non-invasive assessment of cerebral blood flow using selective MR-Arterial Spin Labeling (ASL): First Experience

R Wiest 1, K Jann 2, J Gralla 1, M Hauf 1, J Wang 3, H Mattle 4, T Dierks 2, A Federspiel 2
  • 1Inselspital Bern, Neuroradiologie, Bern, Schweiz
  • 2Psychiatrische Universitätsklinik, Psychiatrische Neurophysiologie, Bern, Schweiz
  • 3University of Pennsylvania, Dept. of Radiology and Neurology, Philadelphia, Vereinigte Staaten von Amerika
  • 4Inselspital Bern, Universitätsklinik für Neurologie, Bern, Schweiz

Introduction: The assessment of cerebral blood flow (CBF) using non-invasive MR methods may increase the insight of the pathophysiological mechanism that leads to disease as well as to the understanding of the development of diseases. Arterial Spin Labeling (ASL) is a fast developing technique with a wide range of applications in the field of neuroradiology and neurology. ASL is a fast technique, precise in spatial resolution and it provides basic quantities in units of ml/100g/min. Recent advances in ASL have permitted noninvasive evaluation of vascular territories for the evaluation of collateral blood flow in cerebrovascular diseases.

Methods: We have tested selective vessel-encoded ASL in two healthy controls and five patients with different extra- and intracranial stenoses to assess cross flow of the internal carotid and basilar arteries through anastomoses. The method was paralleled with images of MR- and Digital Subtraction Angiography. Imaging was performed on a Siemens Magnetm Trio 3 T scanner (Siemens, Erlangen, Germany) using a commercial eight-channel head RF and body coil for RF transmission. The FOV was 230mm × 230mm, matrix size 128×128, 5 slices of 8mm thickness with a 2-mm gap between slices. Tagging duration was 1375ms for the tagging pulse train. The postlabeling delay was 1000ms and TR/TE was 3000ms/52ms. Analysis and computation of selective ASL maps were performed using self written (A.F.) MATLAB programs (MATLAB version 7, release 14; The MathWorks, Inc., Natick, USA).

Results: In two healthy volunteers, vessel-encoded ASL detected the physiological vascular territories. Total scanning time resulted in 6min using a three-vessel encoding ASL. In five patients with hemodynamically relevant extra- and intracranial stenoses, cross flow and vascular supply from other territories matched the vascular territories as detected with DSA.

Conclusion: Vessel-encoded ASL labelling is a novel tool to map collateral blood flow in patients with cerebrovascular disorders. The principal advantage of this promising technique is to map vascular territories and their overlap throughout the whole brain non-invasively.