Chronic liver disease is associated with raised basal and TRH-stimulated PRL and GH levels. In a recent study we found the kidney to be the main site of prolactin elimination in patients with liver disease. In order to determine whether this is specific for PRL or a more general mechanism for polypeptide removal, we studied the elimination of GH, which resembles PRL in molecular weight and primary amino acid sequence, in 5 patients with portal hypertension and hepatic cirrhosis and 5 patients with noncirrhotic portal hypertension. Plasma GH levels were measured before and after TRH in peripheral, hepatic and renal vein samples, taken during diagnostic hepatic vein catheterization. An excessive paradoxical increase of GH after TRH stimulation was found in 4 out of 5 cirrhotic patients but in none of the noncirrhotic individuals (p < 0.025). After TRH the mean hepatic venous levels were significantly lower than the peripheral venous levels in 4 out of 5 noncirrhotic patients but in only 1 of the 5 cirrhotic patients (p < 0.05). The mean renal vein GH levels were significantly lower than the peripheral levels in 3 out of 5 noncirrhotic patients and in none of the cirrhotic patients. In 2 patients in whom renal and hepatic plasma flow was measured, renal extraction of GH was found to be 0 and 6.4 µg, while liver extraction amounted to 22.1 and 34.7 µg of GH during the same 60-min period. Despite the similarity in molecular weight and primary amino acid sequence between PRL and GH, GH appears to be mainly taken up by the liver while PRL is mainly eliminated by the kidney in this group of patients with portal hypertension. This suggests that the renal elimination of prolactin is not solely dependent on glomerular filtration. The selective hepatic removal of growth hormone is probably related to a specific action of growth hormone on liver metabolism.