Separating neural and vascular effects of caffeine using simultaneous EEG–FMRI: Differential effects of caffeine on cognitive and sensorimotor brain responses

The effects of caffeine are mediated through its non-selective antagonistic effects on adenosine A(1) and A(2A) adenosine receptors resulting in increased neuronal activity but also vasoconstriction in the brain. Caffeine, therefore, can modify BOLD FMRI signal responses through both its neural and its vascular effects depending on receptor distributions in different brain regions. In this study we aim to distinguish neural and vascular influences of a single dose of caffeine in measurements of task-related brain activity using simultaneous EEG–FMRI. We chose to compare low-level visual and motor (paced finger tapping) tasks with a cognitive (auditory oddball) task, with the expectation that caffeine would differentially affect brain responses in relation to these tasks. To avoid the influence of chronic caffeine intake, we examined the effect of 250 mg of oral caffeine on 14 non and infrequent caffeine consumers in a double-blind placebo-controlled cross-over study. Our results show that the task-related BOLD signal change in visual and primary motor cortex was significantly reduced by caffeine, while the amplitude and latency of visual evoked potentials over occipital cortex remained unaltered. However, during the auditory oddball task (target versus non-target stimuli) caffeine significantly increased the BOLD signal in frontal cortex. Correspondingly, there was also a significant effect of caffeine in reducing the target evoked response potential (P300) latency in the oddball task and this was associated with a positive potential over frontal cortex. Behavioural data showed that caffeine also improved performance in the oddball task with a significantly reduced number of missed responses. Our results are consistent with earlier studies demonstrating altered flow-metabolism coupling after caffeine administration in the context of our observation of a generalised caffeine-induced reduction in cerebral blood flow demonstrated by arterial spin labelling (19% reduction over grey matter). We were able to identify vascular effects and hence altered neurovascular coupling through the alteration of low-level task FMRI responses in the face of a preserved visual evoked potential. However, our data also suggest a cognitive effect of caffeine through its positive effect on the frontal BOLD signal consistent with the shortening of oddball EEG response latency. The combined use of EEG–FMRI is a promising methodology for investigating alterations in brain function in drug and disease studies where neurovascular coupling may be altered on a regional basis.