The aim of the present study was to correlate the degree of the GH response to the combined arginine and GHRH (ARG+GHRH) test with clinical status in 157 adult hypopituitary patients and 35 healthy controls. On the basis of the GH response to ARG+GHRH, the 192 subjects were subdivided into 5 groups: group 1, very severe GH deficiency (GHD; 65 patients with GH peak <3 microg/L); group 2, severe GHD (37 patients with GH peak between 3.1-9 microg/L); group 3, partial GHD (25 patients with GH peak between 9.1-16.5 microg/L); group 4, non-GHD (30 patients with GH peak >16.5 microg/L); and group 5 (35 controls with GH peak >16.5 microg/L). Plasma insulin-like growth factor I (IGF-I) concentrations were lower (P < 0.001) in patients of group 1 (74.4 +/- 6.7 microg/L) and group 2 (81.4 +/- 6.8 microg/L) than in those of group 3, 4, and 5 (163.6 +/- 40.6, 185.9 +/- 21, and 188.8 +/- 11.1 microg/L, respectively). Plasma IGF-binding protein-3 concentrations were lower (P < 0.01) in group 1 (2.1 +/- 0.2 mg/L) and group 2 (2.0 +/- 0.2 mg/L) than in group 3 (3.4 +/- 0.7 mg/L) and group 5 (3.8 +/- 0.2 mg/L). In patients of group 1, total cholesterol (228.3 +/- 5.7 mg/dL) and triglycerides levels (187.4 +/- 15.3 mg/dL) were higher than those in group 3 (196.6 +/- 9.6 and 115.8 +/- 10.1 mg/dL, respectively), group 4 (176.8 +/- 11.3 and 101.4 +/- 12.5 mg/dL, respectively), and group 5 (160 +/- 6.9 and 99.3 +/- 5.4 mg/dL, respectively). High density lipoprotein cholesterol levels were lower in patients of group 1 (45.2 +/- 2.4 mg/dL) than in those of group 4 (54.7 +/- 3.5 mg/dL; P < 0.05) and group 5 (53.6 +/- 2 mg/dL; P < 0.001), whereas low density lipoprotein cholesterol levels were higher in patients of group 1 (127.3 +/- 7.9 mg/dL), group 2 (129.2 +/- 9.5 mg/dL), and 3 (133 +/- 9 mg/dL) than in those of group 5 (102.4 +/- 7.4 mg/dL; P < 0.05). Patients of group 2 had total cholesterol, high density lipoprotein cholesterol, and triglycerides levels at an intermediate level with respect to those in groups 1, 3, and 4. Among the five groups, no difference was found in fasting glucose concentrations, heart rate, or systolic and diastolic blood pressures. A significant increase in fat body mass and a decrease in lean body mass and total body water were found in all patients compared to controls. Disease duration was significantly shorter in patients of group 4 than in those of the remaining three groups (P < 0.001). A significant correlation was found between the GH peak after ARG+GHRH and disease duration (r = -0.401; P < 0.001), plasma IGF-I (r = 0.434; P < 0.001), total cholesterol (r = -0.324; P < 0.001), and triglycerides levels (r = -0.219; P < 0.05). A significant multiple linear regression coefficient was found between the GH peak after ARG+GHRH and plasma IGF-I levels (t = 2.947; P < 0.005), total cholesterol levels (t = -2.746; P < 0.01), and disease duration (t = -2.397; P < 0.05). In conclusion, the results of the present study indicate that the degree of the GH response to ARG+GHRH is correlated with the severity of lipid profile abnormalities and substantiate the reliability of the ARG+GHRH test for the diagnosis of GHD in adults. Because at present GH treatment is recommended only in adult patients with severe GHD, patients with a GH response below 9 microg/L to the ARG+GHRH test should be treated with GH, as should patients with a peak GH response to an insulin tolerance test below 3 microg/L.