Dose-dependent regional cerebral blood flow changes during remifentanil infusion in humans: a positron emission...

The current study investigated dose-dependent effects of the mu-selective agonist remifentanil on regional cerebral blood flow (rCBF) in volunteers using positron emission tomography (PET).

Ten right-handed male volunteers were included in a 15O-water PET study. Seven underwent three conditions: control (saline), low remifentanil (0.05 microg x kg(-1) x min(-1)), and moderate remifentanil (0.15 microg x kg(-1) x min(-1)). The remaining three participated in the low and moderate conditions. A semirandomized study protocol was used with control and remifentanil conditions 3 or more months apart. The order of low and moderate conditions was randomized. Cardiovascular and respiratory parameters were monitored. Categoric comparisons between the control, low, and moderate conditions and a pixelwise correlation analysis across the three conditions were performed (P < 0.05, corrected for multiple comparisons) using statistical parametric mapping.

Cardiorespiratory parameters were maintained constant over time. At the low remifentanil dose, significant increases in relative rCBF were noted in the lateral prefrontal cortices, inferior parietal cortices, and supplementary motor area. Relative rCBF decreases were observed in the basal mediofrontal cortex, cerebellum, superior temporal lobe, and midbrain gray matter. Moderate doses further increased rCBF in mediofrontal and anterior cingulate cortices, occipital lobe transition, and caudal periventricular grey. Significant decreases were detected in the inferior parietal lobes. These dose-dependent effects of remifentanil on rCBF were confirmed by a correlation analysis.

Remifentanil induced dose-dependent changes in relative rCBF in areas involved in pain processing. At moderate doses, rCBF responses were additionally detected in structures known to participate in modulation of vigilance and alertness. Insight into the mechanisms of opioid analgesia within the pain-processing neural network may lead to a better understanding of antinociception and opioid treatment.