To ask if the determination of central-nervous-system-derived catecholamine metabolites in peripheral circulation could be a useful index of brain dysfunction after subarachnoid hemorrhage, 3-methoxy-4-hydroxyphenylethyleneglycol and homovanillic acid concentrations in plasma, together with those of free catecholamines (noradrenaline, adrenaline, and dopamine), were serially measured for up to 3 weeks after the initiation of symptoms in 23 patients with aneurysmal subarachnoid hemorrhage as compared to 17 healthy and 9 patient controls. Catecholamines and their metabolites were determined by using high-performance liquid chromatography with electrochemical detection. Plasma 3-methoxy-4-hydroxyphenylethylenegtycol concentrations were markedly elevated in subarachnoid hemorrhage patients with coma compared to those without, and the maximal concentrations observed in comatose patients never occurred in normal subjects or in patients with other neurological disorders. The mean maximal plasma concentrations of free catecholamines did not differ significantly between the comatose and noncomatose groups. Combining 3-methoxy-4-hydroxyphenylethyleneglycol with homovanillic acid level data more clearly discriminated between the comatose and noncomatose subarachnoid hemorrhage groups. The results suggest that plasma concentration of 3-methoxy-4-hydroxyphenylethyleneglycol, a major metabolite of brain noradrenaline, can be a prognostic discriminator for patients with subarachnoid hemorrhage and its discriminating power can be strengthened by combining it with homovanillic acid data. (Stroke 1987;18:223-228) D URING the past decade, convincing evidence I has accumulated indicating that free 3-methoxy-4-hydroxyphenylethyleneglycol (MHPG), a major metabolite of brain noradrenaline, in the systemic circulation reflects a functional activity of noradrenergic neurons in the central nervous system (CNS).1 Free MHPG concentration in plasma is considered a reliable index of central noradrenergic turnover from the highly significant relation between plasma and brain levels in primates treated with drugs altering the central noradrenaline metabolism.2 In humans, a centrally acting a-adrenergic agonist, clonidine, is reported to reduce plasma MHPG level.3 Maas et al 4 have demonstrated that approximately 60% of total MHPG production in humans is derived from the brain, while others have indicated that the proportion may be lower because of the possible peripheral conversion of MHPG to vanillylmandelic acid.3 As well as of MHPG, plasma concentration of a major dopamine