Intracerebroventricular (i.c.v.) administration of CDP-choline (0.25, 0.5, 1 and 2 μmol) induced prompt, dose- and time-dependent increase in blood pressure in normotensive rats. Equimolar dose of CDP-choline (1 μmol; i.c.v.) and choline (1 μmol; i.c.v.) caused similar increases in blood pressure while cytidine (1 μmol; i.c.v.) failed to produce any pressor effect. In haemorrhagic shock, CDP-choline (0.1, 0.25, 0.5 and 1 μmol; i.c.v.) increased blood pressure dose- and time-dependently. The complete reversal of hypotension was observed with the i.c.v. injection of CDP-choline (1 μmol) and choline (1 μmol). Cytidine (1 μmol; i.c.v.) produced small, but significant (P<0.05) increase in blood pressure in haemorrhaged rats. Dose-related bradycardia was observed with the injection of CDP-choline in normotensive rats, but the changes in heart rate were not significantly different (P>0.05) in hypotensive conditions. Choline levels in lateral cerebral ventricle and hypothalamus increased about nine- and fivefold, respectively, after CDP-choline (1 μmol) administration in normotensive rats. In haemorrhagic shock extracellular choline levels in hypothalamus increased sevenfold after an i.c.v. administration of CDP-choline (1 μmol). Hemicholinium-3 (20 μg; i.c.v.), a neuronal high affinity choline uptake blocker, and mecamylamine (50 μg; i.c.v.), nicotinic receptor antagonist, pretreatment abolished the pressor effect of CDP-choline in normal rats. The increase in blood pressure was also attenuated by atropine (10 μg; i.c.v.) pretreatment. Atropine blocked the bradycardic response observed after CDP-choline. In haemorrhaged rats, the pressor effect of CDP-choline was attenuated by hemicholinium-3 and mecamylamine while atropine failed to alter the pressor response to CDP-choline. I.c.v. CDP-choline increased plasma adrenaline and vasopressin levels in normal rats. Haemorrhage, itself, increased plasma catecholamines and vasopressin levels. CDP-choline (1 μmol) produced additional increases in the elevated plasma levels of these hormones. An α1-adrenoceptor blocker, prazosin (0.5 mg/kg; i.v.), or vasopressin V1 receptor antagonist, [β-mercapto, β,β-cyclopenta-methylenepropionyl1, O-Me-Tyr2-Arg8]-vasopressin (10 μg/kg; i.v.), pretreatments partially blocked the pressor response to CDP-choline (1 μmol; i.c.v.). Simultaneous administration of these two antagonists completely blocked the pressor effect of CDP-choline in haemorrhagic shock. These results show that the exogenous administration of CDP-choline increases blood pressure and reverses hypotension in haemorrhagic shock. In normotensive conditions, increase in blood pressure appears to be due to the activation of both nicotinic and muscarinic central cholinergic receptors through the activation of presynaptic cholinergic mechanisms. In hypotensive rats, activation of nicotinic cholinergic receptors is solely involved in the pressor effect. Increase in plasma vasopressin and adrenaline mediates the pressor response of CDP-choline in both normotensive and hypotensive conditions.