The phencyclidine derivative ketamine is a non-competitive N-methyl-D-aspartate (NMDA) receptor antagonist with the thalamo-neocortical projection system as the primary site of action. Racemic ketamine consists of the enantiomers S(+)-ketamine and R(-)-ketamine. Racemic ketamine has never been considered an adequate anaesthetic agent in neurosurgical patients since it produces regionally specific stimulation of cerebral metabolism (CMRO2) and increases cerebral blood flow (CBF) and intracranial pressure (ICP). However, recent experiments suggest that both tracemic ketamine and S(+)-ketamine may reduce infarct size in animal models of incomplete cerebral ischaemia and brain injury. This experimental protective effect appears to be related to decreases in Ca++ influx and maintenance of brain tissue magnesium levels due to NMDA and quisqualate receptor blockade by ketamine. Studies in dogs have shown that racemic ketamine (2.0 mg/kg) increases CBF in the presence of the cerebral vasodilator N2O. In contrast, studies in rats without background anaesthesia showed increases in CBF after racemic ketamine (100 mg/kg i.p.). This suggests that the cerebrovascular effects of racemic ketamine are related to the pre-existing cerebrovascular tone induced by background anaesthetics. Cerebrovascular CO2 reactivity was maintained regardless of the baseline cerebrovascular resistance. There are several mechanisms by which racemic ketamine may increase CBF. It induces dose-dependent respiratory depression with consequent mild hypercapnia in spontaneously ventilating subjects. This produces vasodilation due to the intact cerebrovascular CO2 reactivity. Racemic ketamine also induces regional neuroexcitation, which leads to stimulation of cerebral glucose consumption in the limbic, extrapyramidal, auditory, and sensory-motor systems. This regional neuroexcitation with increased CMRO2 produces increases in CBF that can be blocked by infusion of barbiturates or benzodiazepines. However, increases in CBF with racemic ketamine (1 mg/kg) may also occur during normocapnia and without changes in CMRO2. This effect is related to some additional direct cerebral vasodilating potency of racemic ketamine based on a mechanism involving blockade of Ca++ channels. The effects of racemic ketamine on CBF autoregulation have not been investigated systematically. However, studies in rats have shown that CBF autoregulation was maintained with low- and high-dose S(+)-ketamine. Infusion of racemic ketamine alters intracranial volume and ICP. Studies in spontaneously ventilating pigs with and without intracranial hypertension have shown that racemic ketamine (0.5-5.0 mg/kg) produces increases in PaCO2 and ICP. In contrast, identical experiments with mechanical ventilation and controlled PaCO2 showed no changes in ICP following racemic ketamine infusion. This implies that increases in ICP are related to inadequate ventilation with consecutive hypercapnia and increases in intracranial blood volume. However, mechanical ventilation may not be sufficient to control ICP following racemic ketamine. Experiments in mechanically ventilated dogs indicate that racemic ketamine (2 mg/kg) increases cerebral blood volume and ICP even in the presence of normoventilation, a response that is reversible by hyperventilation or the administration of diazepam. Studies in patients have shown that racemic ketamine (2.0 mg/kg) reduces CBF in the presence of cerebral vasodilators like halothane or N2O. In contrast, studies in unanaesthetised humans showed increases in CBF after racemic ketamine (2-3 mg/kg). This observation is consistent with animal studies and suggests that the cerebrovascular effects of racemic ketamine are related to the pre-existing cerebrovascular tone induced by background anaesthetics. Studies in humans with and without intracranial pathology confirm the data from animal experiments. (ABSTRACT TRUNCATED)