The GHRH test has been proposed to replace conventional stimuli in the diagnosis of GH deficiency. However the reliability of GHRH in discriminating between normal and GH-deficient children is still uncertain. The aim of this study was to compare the GH-releasing effect of GHRH (1 microgram/kg i.v.) with that of two neuroactive drugs, clonidine (CLON, 150 micrograms/m2 orally), an alpha 2-receptor agonist, and pyridostigmine (PD, 60 mg orally), a cholinergic agonist that inhibits cholinesterases, in 23 children and adolescents with normal and familial short stature. The plasma GH peak (mean +/- SEM) after GHRH (20.3 +/- 2.5 ng/ml), CLON (17.0 +/- 2.1 ng/ml) and PD (14.9 +/- 1.5 ng/ml) did not significantly differ. According to the conventional limit (less than 10 ng/ml), a false negative response was present in 6, 5 and 6 subjects after GHRH, CLON and PD, respectively. In conclusion, GHRH, CLON and PD have a similar GH-releasing effect. A similar percentage of false negative responses was observed with all tests and this evidence reduces their diagnostic ability.