There is interest in functional magnetic resonance imaging (fMRI) to measure brain activation after exercise. The blood oxygen level dependent (BOLD) fMRI signal varies with cerebral blood flow (CBF), however, it is not known if CBF changes after acute exercise. Previous methods used during exercise (Xe clearance, transcranial doppler) do not provide good spatial localization of CBF. We report the first use of arterial spin labeling (ASL) to detect changes in CBF after exercise. PURPOSE: The aims of this study were to examine: 1) the optimal inversion time to detect changes in CBF in the motor cortex from before to after acute exercise; and 2) if acute exercise alters CBF in the motor cortex at rest or during a finger tapping task. METHODS: Subjects (n = 6) performed 30 minutes of moderate intensity exercise on an electronically braked cycle ergometer (RPE 'somewhat hard'). Before and 10 minutes after exercise relative CBF was measured over four superior contiguous, 6-mm slices (1mm gap) on a 3T GE Signa scanner (8-channel coil) using multiple inversion time (TI) pulsed arterial spin labeling (PASL): FOV=24 cm2, 3.75 mm2 in-plane resolution, FA=90deg, TE=3.2 ms, TR=2000 ms, tag width=600 ms, tag thickness=10 cm, with a 1 cm gap between the tagging band and proximal slice. Two multiple TI runs were obtained at rest and during 4Hz finger-tapping. Four inversion times (675, 975, 1275, 1575 ms) were acquired per run, with 20 interleaved pairs of tag and control images per inversion time (run time 320 s). Heart rate, blood pressure and O2 content, pCO2, and respiration were recorded throughout. RESULTS: Resting CBF increased up to 20% following exercise, with significant differences obtained at an inversion time of 1575 ms (p <.05). Finger tapping-induced CBF in the motor cortex significantly increased from before to after exercise at TI = 1575 ms (p <.01). CONCLUSION: These findings suggest changes in human cerebral blood flow can be detected up to 30 minutes after moderate intensity cycling exercise using pulsed arterial spin labeling at 3T with an inversion time of 1575 ms. This study has important implications for using fMRI after exercise. Prior exercise may introduce variance into the fMRI BOLD signal that reflects increased CBF, not neural activation. Future fMRI studies should take into consideration exercise-induced changes in CBF.
Read full abstract