Megan Singal BS, Presenter: Nothing to Disclose

Anup Alexander, Abstract Co-Author: Nothing to Disclose

Ian Atkinson PhD, Abstract Co-Author: Nothing to Disclose

Sepideh Amin-Hanjani MD, Abstract Co-Author: Nothing to Disclose

Keith Raymond Thulborn MD, PhD, Abstract Co-Author: Owner, Thulborn Associates, Inc

To determine if hemodynamic reserve (HR) is measured reliably by changes in dynamic susceptibility contrast (DSC) perfusion parameters before and after vasodilation by acetozolamide. Blood oxygenation level dependent (BOLD) fMRI, which is dependent on intact HR at the tissue level for activation, is used as a gold standard.

A retrospective study of 62 subjects (age 46.4 ± 16.8, 42F) with known cerebrovascular occlusive disease (CVOD) had technically adequate functional and perfusion MRI performed at 3T. DSC perfusion, performed before and after vasodilatation, used gradient-echo echo-planar imaging to record signal changes during the first pass of the contrast bolus through the brain. The time series were voxel-wise fitted with a multi-variate non-linear least squares routine to obtain mean transit time (MTT) and cerebral blood volume (CBV) parameters. Intact HR was determined in each vascular territory by the significance (unpaired t-test, p<0.05) of the differences in perfusion parameters before and after vasodilation. HR was also evaluated with BOLD fMRI using an audiovisual bilateral hand motor paradigm (4.5 cycles, two alternating 30s conditions of audiovisual stimulated hand movement and central fixation). Activation maps were generated using voxel-wise unpaired t-tests without spatial or temporal filtering. Each vascular territory contains BOLD activation in individuals with normal HR. The percentage of congruence for the presence or absence of HR by perfusion and BOLD activation was calculated across all compromised and all uncompromised vascular territories.

The congruence between techniques across all types of CVOD including single and multi-vessel disease was 61%. Figure 1. Case showing Incongruent DSC and BOLD results in which (a) BOLD activation remains intact in the motor cortex bilaterally but the DSC shows worsening asymmetry in TTT from (b) before to (c) after vasodilation although little change in CBV from (d) before to (e) after vasodilation.

DSC perfusion MRI, as compared to BOLD fMRI, is not reliable for risk stratification of severity of CVOD based on HR. This is most likely due to the violation of the central volume theorem due to an ill-defined arterial input function in the setting of collateral circulation that is the basis of CVOD physiology.

DSC perfusion MRI is not reliable for determining integrity of HR in patients with CVOD.

Singal, M, Alexander, A, Atkinson, I, Amin-Hanjani, S, Thulborn, K, Hemodynamic Reserve in the Setting of Chronic Cerebrovascular Occlusive Disease Is Not Reliably Assessed by Changes in Perfusion Parameters before and after Pharmacological Challenge with Acetozolamide.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12024774.html