Abstract:
HTML
Purpose: Several studies have compared CBF activation maps to BOLD activation maps using simple visual or motor paradigms (Davis et al. 1998; Kim et al. 1997, 1999; Li et al. 2000; Luh et al. 2000; Malonek et al. 1997). CBF and BOLD maps generally overlap, though BOLD receives contributions from surface veins, while CBF more closely approximates active gray matter (Pfeuffer et al. 2002). Correspondingly, recent studies have localized task-correlated CBV changes in cortical layers II-V which receive the greatest density of dendritic input (Harel et al. 2002). We sought to determine whether these relationships between BOLD and perfusion fMRI maps would hold when the complexity of the motor task is increased, specifically with tasks involving greater input to the motor cortex. Additionally, we addressed the effect of field strength on the hemodynamic response parameters in BOLD and Perfusion fMRI for different motor tasks.
Methods and Materials: Four healthy adult subjects were imaged on a Siemens Allegra 3T MR system or a GE Signa Echo-Speed 1.5T system. At 3T, BOLD images were acquired using a GRE EPI sequence and the Perfusion images with an EPI inversion recovery sequence. At 1.5T, BOLD images were acquired using a spiral GRE sequence, and Perfusion images with a spiral FAIR sequence. Subjects performed a series of motor tasks for Perfusion and BOLD imaging: finger-tapping, exploration of a sphere, and a point-to-target task with the dominant hand. The raw data was reconstructed off-line, motion corrected and statistical maps of activation generated in AFNI. ROIs encompassing the left primary motor cortex (M1) and the supplementary motor area (SMA) were analyzed for spatial correspondence, peak signal intensity and time course of signal change across different lags of the TR.
Results: Activation maps for the BOLD and Perfusion images overlapped in all cortical areas, though the Perfusion maps generally showed a smaller number of activated pixels. There were no significant differences in the peak signal intensity in M1, but differences were observed in the SMA. The time course of the hemodynamic response was consistent between BOLD and CBF in M1, whereas the SMA showed a variable response depending on task.
Conclusion: The results of this study suggest that BOLD and Perfusion based fMRI maps may show regionally specific hemodynamic parameters. The possibility of regionally specific hemodynamic parameters has important implications for the evaluation of stroke and predicting responses to surgical or radiotherapeutic intervention.
Questions about this event email: kmosier@iupui.edu
Mosier DMD, PhD, K,
Variability of Hemodynamic Responses in Different Motor Tasks: Comparison of Perfusion-based fMRI and BOLD fMRI. Radiological Society of North America 2003 Scientific Assembly and Annual Meeting, November 30 - December 5, 2003 ,Chicago IL.
http://archive.rsna.org/2003/3105087.html
