Detailed assessment of physiological and pathophysiological GH secretion has, until recently, been limited by the poor sensitivity of the available assays. We have used an ultrasensitive chemiluminescence GH assay (sensitivity, 0.002 microgram/L) to study 24-h GH profiles (20-min sampling) from 24 patients who had been treated for hypothalamic-pituitary disease with surgery and irradiation and from 24 healthy control subjects matched for age, sex, and body mass index. Twenty-three of the 24 patients demonstrated pulsatile GH secretion, determined by Cluster. The median (range) area under the curve for GH, mean pulse area, mean pulse height, average valley mean level, and mean interpeak nadir were lower in the patients than in the controls [119.25 (7.273-843.600) vs. 968.539 (227.200-4625.000) min/microgram.L (P < 0.00001); 3.777 (0.288-30.850) vs. 61.390 (12.880-224.210) min/microgram.L (P < 0.00001), 0.107 (0.010-0.958) vs. 1.408 (0.368-5.050) micrograms/L (P < 0.00001), 0.074 (0.006-0.415) vs. 0.348 (0.048-2.350) microgram/L (P < 0.00001), and 0.066 (0.003-0.270) vs. 0.205 (0.021-1.838) microgram/L (P = 0.0004), respectively]. The median (range) number of pulses, mean pulse duration, and mean interval between pulses did not differ between the patients and controls [10 (4-15) vs. 10 (7-15; P = 0.36), 96.4 (68.0-220.0) vs. 104.0 (72.0-151.4) min (P = 0.65) and 128.0 (92.8-255.0) vs. 126.2 (90.0-180.0) min (P = 0.73), respectively]. The diurnal rhythm of GH secretion was present in the controls, but there was only limited evidence of residual diurnal rhythm in the patients. This study has demonstrated that GH secretion remains pulsatile in GH-deficient patients despite the mass effect of hypothalamic-pituitary pathology, pituitary surgery, and radiotherapy. With the development of potent GH secretagogues that are active orally, our findings may have important implications for the future management of GH-deficient subjects.