Intensive venous sampling paradigms disclose high frequency adrenocorticotropin release episodes in normal men

Recent studies in the rat and rhesus monkey have disclosed apparently high frequency in vivo ACTH release episodes. While the circadian pattern of plasma ACTH concentrations has been known for many years, the exact frequency of ultradian pulsatile ACTH release in man is not clear, due in part to variable intensities of blood-sampling schedules and the limited availability of sensitive ACTH assays. In this study we used a new sensitive and specific immunoradiometric assay to measure plasma ACTH concentrations in blood sampled at 1) 2-min intervals for 3 h, followed by 4-min intervals for 4 more h in six men; and 2) 10-min intervals for 24 h in eight other men. An objective peak detection algorithm (Cluster) and cosinor analyses were used to assess the episodic pulsatility and circadian rhythmicity of ACTH. Comparisons were made among the 2 min (3 h), 4, 8, and 12 min (7 h), and 10 min (24 h) time series. Mean ACTH interpulse intervals were significantly different among the five sampling groups (P < 0.00001). Sampling every 2 min yielded a mean ACTH interpulse interval of 18 min, which was significantly shorter than the mean interpulse intervals of 35, 53, 52, and 73 min resulting, respectively, from sampling every 4, 8, 10, and 12 min (P < 0.05). In contrast, maximal peak ACTH amplitudes (picomoles per L or percent increase) did not vary as a function of sampling frequency. The 24-h plasma ACTH concentration time series showed significant diurnal variation, with a mean circadian amplitude of 0.95 ± 0.15 pmol/L occurring at 1008 h (±25 min). Cosinor analysis of various ACTH pulse parameters deduced from the 24-h time series revealed significant circadian rhythmicity in the ACTH peak maxima (P < 0.05), peak increments (P < 0.05), and prepeak nadir (P < 0.05) concentrations, but not in ACTH interpulse intervals. We conclude that in men, 1) intensive sampling at 2-min intervals unmasks high frequency ACTH release episodes that cannot be detected at conventional sampling rates; and 2) ACTH peak amplitude, but not frequency, varies significantly during the course of circadian changes in the plasma ACTH concentrations.