Figure 1: A graph showing perfusion change in the motor and visual cortex relative to the pre-treatment baseline.

Functional magnetic resonance imaging (fMRI) contrast is derived from changes in oxygenation of the blood. These changes are a result of increases in blood flow and blood volume in the neighborhood of the brain required to accomplish the task. Caffeine is a cerebral vasoconstrictor. The introduction of 200 mg of caffeine (equivalent to two to three cups of coffee) causes a global reduction in cerebral perfusion by about 20 to 30 percent. The impact on the functional imaging contrast is actually positive. The baseline blood oxygen level dependent (BOLD) signal amplitude is reduced due to vasoconstriction, increased concentration of deoxyhemoglobin. The result is a larger percent signal change on the order of 20 to 40 percent (relative to baseline). The benefit of this signal increase can be used to improve the acquisition and detection and/or applied to paradigms that will have reduced contrast.

Figure 1 shows how the introduction of caffeine causes reduced perfusion in the motor and visual cortex. This was measured with an arterial spin labeling MRI method. It is a resting state measurement that occurs over a three-minute period.

Figure 2: BOLD changes in the motor and visual cortex due to caffeine relative to the pre-treatment measurement.

In Figure 2, the BOLD contrast is increased in both the motor and visual cortex. These changes were measured during a visually cued motor task conducted in an event related design with a long trial interval. The peak BOLD signal response was measured. An example time course before and after caffeine are shown in Figure 3 (below). Note that the peak response is larger with caffeine but the post-stimulus undershoot is the same.

Functional maps were made before (Figure 4, top row) and after (Figure 4, bottom row) the introduction of caffeine during a visually cued motor task. Both maps have the same correlation threshold. Note the larger extent associated with the caffeine study.

Figure 3: BOLD time course changes due to caffeine.

Figure 4: An fMRI map without caffeine (top row) and an fMRI map with caffeine (bottom). The subject performed the same simple task and the same statistical threshold was used to make the maps.