Impaired ability to excrete a water load occurs in a substantial number of patients with advanced cirrhosis and in animals with experimental cirrhosis. The nonosmotic stimulation of arginine vasopressin release from the pituitary has been implicated as an important factor in the abnormal water excretion in patients and animals with cirrhosis. In this study, arginine vasopressin hypothalamic gene expression was studied in cirrhotic rats. Cirrhosis was induced by a combination of phenobarbital treatment in drinking water and weekly intragastric administration of carbon tetrachloride for 13 to 15 wk. Severe cirrhosis was confirmed by morphological analysis and the presence of ascites. Plasma arginine vasopressin was also significantly higher in rats with cirrhosis (control = 1.77 f 0.16 and cirrhotic rats = 4.14 ? 0.62 pglml, n = 9, p < 0.002). Hypothalamic arginine vasopressin messenger RNA was also significantly higher in cirrhotic rats (control = 762.1 +. 132.3 and cirrhotic rats = 1,834.2 f 271.9 pflypothalamus, n = 9, p < 0.005). Pituitary arginine vasopressin content was significantly lowered in cirrhotic rats (control = 3.69 f 0.98 and cirrhotic rats = 1.57 f 0.09 pg/pituitary, n = 9, p < 0.05). No difference was seen in hypothalamic arginine vasopressin content between the two groups (control = 4.64 f 0.34 and cirrhotic rats = 4.23 ? 0.33 nghypothalamus, n = 9, NS). Oxytocin messenger RNA in the hypothalamus was also not significantly different between the two groups (control = 8.61 +. 0.68 and cirrhotic rats = 9.33 * 0.65 unit of density, n = 9, NS). No differences were seen in plasma sodium (control = 145.7 +. 1.62 and cirrhotic rats = 145.9 -+ 1.09 mmolb), plasma osmolality (control = 296.6 f 2.6 and cirrhotic rats = 297.1 f 4.2 mOsm/kg H,O), creatinine clearance and blood urea nitrogen between the two groups. These results therefore provide support for increased hypothalamic biosynthesis and pituitary release of arginine vasopressin in experimental cirrhosis in the rat. (HEPA-