A novel noninvasive magnetic resonance imaging (MRI) method was developed to determine in vivo blood oxygen saturation and its changes during motor cortex activation in small cerebral veins. Specifically, based on susceptibility measurements in the resting states, pial veins were found to have a mean oxygen saturation of Yrest=0.544+/-0.029 averaged over 14 vessels in 5 volunteers. During activation, susceptibility measurements revealed an oxygen saturation change of DeltaYsusc=0.14+/-0.02. Independent evaluation from blood flow velocity measurements yielded a value of DeltaYflow=0.14+/-0.04 for this change. These results validate the blood oxygenation level-dependent (BOLD) model in functional MRI (fMRI).

rCBV maps can be obtained in a standard clinical setting and can be used to reveal information on local blood volume.

High-resolution functional imaging experiments at 0.95 Tesla have been performed to determine the changes in oxygen saturation in pial veins during motor activation by measuring both flow and susceptibility changes in the blood. Averaging across subjects, mean values for the change of the oxygenation level, deltaY = 0.16 +/- 0.08 (n = 7) and deltaY = 0.13 +/- 0.09 (n = 4), were obtained from the susceptibility sensitive and the flow sensitive acquisitions, respectively. The results suggest that the increase in blood flow is largely uncoupled from the oxygen consumption. The quoted errors reflect mainly the intersubject variability. In addition, low-resolution echo planar imaging (EPI) measurements were performed on the same volunteers to quantify signal intensity changes. Using the measured change in oxygenation, the observed signal changes in the EPI experiments can be attributed to a 5% venous blood volume.

No abstract

A three-dimensional (3D) T1-weighted sequence was used to acquire high spatial resolution whole brain images in rats before and after the injection of an intravascular contrast agent. These T1-weighted images were used to estimate regional cerebral blood volume (rCBV) as a percentage of blood volume in each voxel. Ventilation was manipulated to investigate the effects of altered arterial carbon dioxide tension (PaCO2) on rCBV. In addition, different doses of a hypertonic mannitol solution were used to investigate the sensitivity of the proposed method in a serial monitoring paradigm. An rCBV of 2.40% +/- 0.34% was obtained before any physiological manipulation, in good agreement with literature values using alternative techniques. Using this method, it was found that there exists a linear relationship between PaCO2 and rCBV (R2 = 0.77) and that rCBV increased in a dose and time dependent fashion in mannitol-treated rats. High signal-to-noise was available due to the substantial increase in blood signal from the intravascular contrast agent.