Analysis of fluctuation in cerebral venous oxygenation using MR imaging: Quantitative evaluation of vasomotor function of arterioles

Abstract

Purpose: Cerebral arteriolar vasomotor function plays an important role in brain health. Since respiratory changes in the partial arterial pressure of CO2 (PaCO2) cause arterioles to vasodilate or vasoconstrict, resting-state arteriolar vasomotion results in the fluctuation of venous blood oxygenation, which can be monitored by observing magnetic resonance (MR) signals. Focusing on the superior sagittal sinus as the largest cerebral vein, we developed a method to elucidate the respiratory fluctuation of cerebral venous oxygenation that may reflect the vasomotor function. Methods: Single slices of varying thickness (7–15 mm) taken perpendicular to the superior sagittal sinus of five volunteers were imaged by spin-echo echo-planar imaging using a 1.5-T MR system. The time series of the signal intensity at the superior sagittal sinus was Fourier-transformed, and the spectral fluctuation intensity (SFI) at respiratory frequency was obtained. The amplitude of the respiratory fluctuation in the cerebral venous oxygenation was calculated from the gradient of the relation between the SFI and the average signal intensity, which increased proportionally with an increase in slice thickness. The amplitude of the fluctuation in cerebral venous oxygenation at low (<0.1 Hz) and cardiac pulsation frequencies was also calculated for comparison with the respiratory fluctuation. Results: The amplitude of respiratory fluctuation in the cerebral venous oxygenation was quantified as 1.2%, demonstrating the validity of our method via the highest significant correlation (r = 0.82) in the plot of SFI and average signal intensities; the correlations at low and cardiac pulsation frequencies were 0.60 and 0.63, respectively. Conclusion: We have successfully demonstrated cerebral venous oxygenation fluctuation at respiratory frequencies in the resting state. This fluctuation was non-invasively evaluated as 1.2%, representing the control value for the arteriolar vasomotor function of a healthy human.