Within an appropriate clinical context, GH deficiency (GHD) in adults can only be diagnosed biochemically by provocative testing. The evaluation of IGF-I, IGFBP-3 and even of spontaneous GH secretion do not establish the diagnosis of adult GHD. In fact, remarkable overlaps between normal and GHD adults have been reported for all these parameters. On the other hand, it is well known that even short-term fasting stimulates GH secretion in normal subjects. The aim of our study was to determine the effects of 36 h fasting on 8-h diurnal GH, insulin and glucose levels as well as on basal IGF-I, IGFBP-3, acid-labile subunit (ALS), IGFBP-1, GHBP and free fatty acid (FFA) levels. We studied 9 GHD adults (GHD, 8 males, 1 female; age, mean +/- SEM: 37.6 +/- 2.3 years, body mass index (BMI): 24.5 +/- 1.0 kg/m2) and 20 age-matched normal subjects (NS) as controls (13 males, 7 females; age: 28.9 +/- 0.6 years, BMI: 21.6 +/- 0.4 kg/m2). In all subjects we studied the effects of 36 h fasting on 8-h daytime GH, insulin and glucose levels (assay every 30 min from 0800 h to 1600 h) as well as on basal IGF-I, IGFBP-3, ALS, IGFBP-1, GHBP and FFA levels. Before fasting, basal mean IGF-I, IGFBP-3 and ALS levels in GHD were lower (P < 0. 0001) than in NS. IGFBP-1, GHBP and FFA levels were similar in both groups. Before fasting mean GH concentration (mGHc) in GHD was lower (P < 0.05) than in NS (0.4 +/- 0.2 vs. 2.2 +/- 0.6 mu/l) but with a clear overlap between the 2 groups (range 0.4-0.8 vs. 0.4-6.8 mu/l). After fasting, both in GHD and NS basal IGF-I, IGFBP-3, ALS and GHBP levels did not change significantly. On the other hand, in both GHD and in NS IGFBP-1 was increased (P < 0.0001) to a similar extent, while FFA increased in NS more (P < 0.01) than in GHD. Fasting significantly increased mGHc in NS (12.0 +/- 1.2 mu/l, P < 0.0001) but not in GHD (0.6 +/- 0.2 mu/l). After fasting, no overlap was present between GHD and NS (0.4-1.6 vs. 2.4-20.8 mu/l, respectively). Mean glucose and insulin concentrations over 8 h in GHD and NS in basal conditions were similar and were reduced to the same extent in both groups. Our findings demonstrate that after short-term fasting, the study of spontaneous GH secretion distinguishes between GH-deficient adults and normal subjects; this phenomenon occurs before significant changes in IGF-I and IGFBP-3 levels. These results suggest that the assessment of spontaneous GH secretion could be useful for the diagnosis of adult GH deficiency only after short-term fasting.