Intraventricular dexmedetomidine decreases cerebral blood flow during normoxia and hypoxia in dogs

We tested the hypothesis that a centrally administered α₂-receptor agonist could alter the cerebrovascular response to hypoxia, without evidence of systemic absorption of the drug. Beagle dogs were anesthetized with 1.4% isoflurane and exposed to hypoxic hypoxia (PaO₂ approximately 22 mm Hg) before and after ventricular-cisternal perfusion with mock cerebrospinal fluid (CSF group, n = 5) or dexmedetomidine (100 μg/mL; total dose 300 μg; DEX group, n = 6). Cerebral perfusion pressure, PaCO₂ and arterial oxygen content were controlled and regional cerebral blood flow (CBF; microspheres) and global cerebral metabolic rate for oxygen consumption (CMRO₂) were measured. In another group (n = 5), drug distribution under the experimental conditions was assessed by ³H-clonidine administered by ventricular- cisternal perfusion. In the mock CSF group, flow to the cerebral hemispheres increased during hypoxia under baseline conditions and after CSF infusion: 66 ± 8 to 170 ± 15 mL · min^-1 · 100 g^-1 (265% ± 24% of baseline value), 83 ± 9 to 154 ± 14 mL · min^-1 · 100 g^-1 (201% ± 54% of post-CSF infusion value). DEX decreased normoxic flow in the cerebral hemispheres from 76 ± 6 to 44 ± 4 ml · min^-1 · 100 g^-1 with decreases in other regions of similar magnitude. After DEX, the absolute flow in all regions during hypoxia was 52%-55% of that prior to DEX (P < 0.05). However, because DEX also decreased normoxic CBF, the percent increase in flow during hypoxia was similar before and after DEX. CMRO₂ was not affected by hypoxia prior to DEX. However, after DEX, hypoxia caused a marked reduction in cerebral oxygen delivery (5.2 ± 1.0 vs 13.7 ± 2.3 ml · min^-1 100 g^-1 for the CSF group) and CMRO₂ (2.5 ± 0.6 vs 3.9 ± 0.6 ml · min^-1 · 100 g^-1). Regional accumulation of intraventricularly administered ³H-clonidine was greatest in periventricular brain structures (e.g., caudate nucleus, dorsal brainstem), and the concentration in the cerebral cortex was approximately 1% of the concentration in the ipsilateral caudate nucleus. We conclude that centrally administered DEX reduces CBF during normoxia and prevents adequate oxygen delivery during hypoxia. The mechanism of DEX-induced CBF reduction is not metabolically mediated, since CMRO₂ is maintained at control values during normoxia despite the significant blood flow reduction. We believe that the reduction in CMRO₂ during hypoxia in DEX-treated dogs is the result of a reduction of oxygen delivery rather than the underlying mechanism for the observed reduction in CBF during hypoxia.