Near-infrared spectroscopy (NIRS) continuously measures regional cerebral oxygen saturation (rSO₂) noninvasively and has been shown to detect even small changes in cerebral oxygen supply-demand balance. Although widely used, only the effect of midazolam on cerebral blood flow has been studied in humans and evidence is lacking about its effect on cerebral metabolic rate. We therefore evaluated the effect of midazolam on cerebral oxygen supply-demand balance with NIRS.

Sixty patients undergoing elective coronary artery bypass graft surgery were randomly allocated into either midazolam (n = 30) or propofol (n = 30) group. rSO₂ was recorded before induction while patients were breathing room air as baseline, after pre-oxygenation with 100% oxygen, after administration of either midazolam or propofol, after completion of administration of sufentanil and after tracheal intubation. Hemodynamic variables including cardiac index and mixed venous oxygen saturation were recorded at the same time points.

rSO₂ and hemodynamic variables were similar between the groups throughout the study period. After pre-oxygenation, rSO₂ significantly increased compared to baseline in each group, and did not show any additional increase after administration of either midazolam or propofol and sufentanil in both groups.

Midazolam preserves cerebral blood flow-metabolism coupling to a similar degree to propofol as assessed by near infrared spectroscopy.