[Show abstract][Hide abstract] ABSTRACT: Little is known regarding the changes in blood oxygen tension (P(O2)) with changes in brain function. This work aimed to measure the blood P(O2) in surface arteries and veins as well as tissue with evoked somato-sensory stimulation in the anesthetized rat. Electrical stimulation of the forepaw induced average increases in blood flow of 44% as well as increases in the tissue P(O2) of 28%. More importantly, increases in P(O2) throughout pial arteries (resting diameters=59 to 129 microm) and pial veins (resting diameters=62 to 361 microm) were observed. The largest increases in vascular P(O2) were observed in the small veins (from 33 to 40 mm Hg) and small arteries (from 78 to 88 mm Hg). The changes in oxygen saturation (S(O2)) were calculated and the largest increases were observed in small veins (Delta=+11%) while its increase in small arteries was small (Delta=+4%). The average diameter of arterial vessels was observed to increase by 4 to 6% while that of veins was not observed to change with evoked stimulation. These findings show that the increases in arterial P(O2) contribute to the hyper-oxygenation of tissue and, mostly likely, also to the signal changes in hemoglobin-based functional imaging methods (e.g. BOLD fMRI).
Journal of cerebral blood flow and metabolism: official journal of the International Society of Cerebral Blood Flow and Metabolism 10/2009; 30(2):428-39. DOI:10.1038/jcbfm.2009.213 · 5.41 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Stimulation-induced changes in transverse relaxation rates can provide important insight into underlying physiological changes in blood oxygenation level-dependent (BOLD) contrast. It is often assumed that BOLD fractional signal change (DeltaS/S) is linearly dependent on echo time (TE). This relationship was evaluated at 9.4 T during visual stimulation in cats with gradient-echo (GE) and spin-echo (SE) echo-planar imaging (EPI). The TE dependence of GE DeltaS/S is close to linear in both the parenchyma and large vessel area at the cortical surface for TEs of 6-20 ms. However, this dependence is nonlinear for SE studies in the TE range of 16-70 ms unless a diffusion-weighting of b = 200 s/mm(2) is applied. This behavior is not caused by inflow effects, T(2)* decay during data acquisition in SE-EPI, or extravascular spin density changes. Our results are explained by a two-compartment model in which the extravascular contribution to DeltaS/S vs. TE is linear, while the intravascular contribution can be nonlinear depending on the magnetic field strength and TE. At 9.4 T, the large-vessel IV signal can be minimized by using long TE and/or moderate diffusion weighting. Thus, stimulation-induced relaxation rate changes should be carefully determined, and their physiological meanings should be interpreted with caution.
Magnetic Resonance in Medicine 06/2006; 55(6):1281-90. DOI:10.1002/mrm.20918 · 3.57 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Introduction In fMRI, the blood oxygenation level dependent (BOLD) contrast reflects activation-induced changes in the oxygen metabolism (CMRO 2) and the hemodynamic response which consists of the blood flow (CBF) and the cerebral blood volume (CBV). Currently, the fundamental relationship between oxygen metabolism and the hemodynamic responses has not been fully characterized. For example, there are controversies over whether the focal functional CBF increase is caused by the energy metabolism or the neurotransmitter release, and on the physiological source of the post-stimulus undershoot in the BOLD signal. Specifically for the BOLD undershoot, there are conflicting opinions on whether the venous CBV or CMRO 2 remains elevated for a prolonged period of time after the stimulus was turned off, assuming the CBF change quickly returns to the pre-stimulus basal level [1-4]. The discrepancies in the literature may partly be due to the difference in imaging techniques, stimulation type and duration, and spatial localization. Thus, multimodal studies with high spatial resolution would be helpful to clarify the debate. In this study, a cat visual stimulation model was used to investigate the spatial and temporal dynamics of BOLD, CBV, and CBF. Materials and methods fMRI experiments were performed on a 9.4T/31cm MRI (Varian) system. Female adolescent cats (n=11) were anesthetized under ~1.1% isoflurane and kept under normal physiological condition. For all the experiments, coronal images were acquired with 2×2cm 2 FOV and 2mm slice thickness. A T 1 weighted image with 128×128 matrix size was obtained for anatomical reference. The binocular visual stimuli consisted of drifting square wave grating. CBV-weighted studies were performed following BOLD experiments on the same animal, but the CBF changes were measured separately on different animals because of time restriction. BOLD and CBV studies (n=6): BOLD and CBV-weighted images were obtained using a 1.6cm diameter surface coil before and after the injection of 10 mg/kg MION, respectively. Imaging parameters were: 2-segmented GE-EPI, TR=0.5s/segment, 96×96 matrix size zero filled to 128×128. Two TEs of 10ms and 20ms were arrayed for BOLD; TE = 6ms, 10ms for the CBV-weighted studies. The temporal resolution was 2s, and the stimulation paradigm was 60s control, 60s stimulation, and 160s control. About 20 data sets were averaged. CBF studies (n=5): FAIR images were obtained using an actively detuned two coil system and single shot GE-EPI technique with matrix size=64×64, TE=19ms, TR=1.5s, and TI=1.5s. The time resolution was 6s. The stimulation paradigm was 60s control, 60s stimulation, and 162s control. 40-50 data sets were averaged for each study. Data analysis: The CBV-weighted signal change contained the contribution from the blood deoxyhemoglobin content change, which is more significant at the large vessel area and at higher fields. Therefore a BOLD correction was performed to calculate the CBV change . Fractional signal change maps were calculated with a minimal cross correlation coefficient of 0.3 and minimal cluster size of 3 pixels. ROI-based data analysis was performed. For each experiment, two ROIs were drawn from the anatomic image: one at the surface of cortex and the other at the middle of cortex. The surface ROI contains large blood vessels such as arteries or veins, while the middle ROI contains mostly microvessels such as arterioles, capillaries, and venules. Results Fig. 1 showed the normalized temporal dynamics of BOLD (at TE=20ms), CBV, and CBF for the two ROIs. Surprisingly, post-stimulus undershoots were found consistently for both CBV and CBF. Moreover, the spatial and temporal patterns of the normalized CBF and CBV response were very similar. At the surface of the cortex, there is no significant difference in the temporal dynamics of CBV, CBF, and BOLD. At the middle cortical ROI, the BOLD signal decays much faster than that of the CBV and CBF during the 60s stimulation period, and the peaks of the poststimulus undershoot are much more pronounced than those of the CBV and CBF. Compared to the middle ROI, the CBV (CBF) response at the surface ROI was faster: faster increasing to the positive peak, faster decaying during the 60s stimulation, faster decreasing to the peak of the post-stimulus undershoot, and faster recovering from the undershoot to the pre-stimulus baseline.