Opioid binding to the cerebral blood vessels may affect vascular responsiveness and hence confound interpretation of blood oxygen level-dependent (BOLD) responses, which are usually interpreted as neuronal in origin. Opioid binding varies in different brain regions. It is unclear whether opioids alter neurovascular coupling, or whether their effects are purely neuronal. This study used BOLD functional magnetic resonance imaging (FMRI) to investigate the effect of a mu-opioid agonist remifentanil, on cerebrovascular CO(2) reactivity (being one component of neurovascular coupling). Hypercapnic challenges were delivered to human volunteers, while controlling potential opioid-induced respiratory depression. The BOLD signal increase to hypercapnia was compared before and during remifentanil administration. Remifentanil was shown not to have a generalised effect on CO(2) responsiveness in the cerebral vasculature. However, it caused a significant reduction in the positive BOLD response to hypercapnia in the bilateral primary sensorimotor cortices, bilateral extrastriate visual areas, left insula, left caudate nucleus, and left inferior temporal gyrus. We conclude that remifentanil does not modulate cerebrovascular CO(2) reactivity, as we saw no difference in BOLD response to hypercapnia in areas with high opioid receptor densities. We did however see a focal reduction in areas related to motor control and putative task activation, which we conclude to be related to changes in neuronal activity related to the sedative effects of remifentanil. Our method of controlling CO(2) levels effectively mitigated the potential confound of respiratory depression and allowed comparison over a similar range of CO(2) levels. We suggest that similar methodology should be used when investigating other potentially vasoactive compounds with FMRI.