Physiological properties of the luteinizing hormone pulse signal: Impact of intensive and extended venous sampling paradigms on its characterization in healthy men and women

Johannes D. Veldhuis, William S. Evans, Michael L. Johnson, Michael R. Wills, Alan D. Rogol

Research output: Contribution to journalArticlepeer-review

76 Scopus citations


The pulsatile nature of the gonadotropin signal is a critical determinant of physiological activation of the gonadal axis. Nonetheless, major uncertainties exist regarding the exact patterns of LH secretion that constitute normal physiological profiles in man. To assess possible bases for the discrepancies in the literature, we sampled blood at 5-min intervals for 24 h in eight normal men and eight normal women in the early follicular phase of the menstrual cycle. The constituent 5-, 10-, 15-, and 20-min immunoactive LH series and the consituent 6-, 12-, or 24-h sampling durations provided ranges of sampling intensities and durations for analysis of significant LH pulses. A technique for minimizing the influence of false positive immunoassay errors on peak detection was used to aid in estimating apparent true positive LH pulse frequency. Nonlinear curve fitting of the relationship between sampling intensity and apparent true positive LH pulse frequency revealed a stable pulse frequency estimate at intensive rates of venous sampling, with values of 19.5 ± 1.9 (±SEM) pulses/24 h (periodicity, 73.8 ± 6.5 min/pulse) in men and 20.6 ± 3.6 pulses/24 h (periodicity, 70.0 ± 10 min/pulse) in women. Further analyses indicated that sampling every 3.1 and 2.0 min for 24 h would be required to capture 90% of the LH pulses in men and women, respectively. Moreover, even with a 5-min sampling rate, the statistical counting errors of the LH pulse frequency estimates varied markedly with sampling duration; for example, in men sampled in 6-, 12-, and 24-hr sessions, the values were, respectively, 49%, 35%, and 24% of the observed pulse frequencies. Similar variations were found in women. Counting errors were 30 50% higher for conventional 20-min sampling rates than for 5-min sampling intervals. Measured interpulse intervals varied widely from 10 330 min (median, 60 min; n = 131 LH pulses) in men and from 10-340 min (median, 65 min; n = 125 LH pulses) in women. In addition, absolute LH pulse amplitudes varied from 1 28 mlU/ml (median, 4.1 mlU/ml) in men and from 1 24 mlU/ml (median, 3.6 mlU/ml) in women. These estimates were associated with a median number of points identified within each pulse of 6.0 in men and 4.0 in women. In summary, the present exhaustive sampling studies demonstrate for the first time stable estimates of physiological LH pulsations in normal men and women, and document broad ranges of normal LH pulse amplitudes and interpulse intervals. We conclude that both intensive and prolonged sampling conditions are necessary to define stable and precise estimates of physiological LH pulse patterns and to disclose the physiological ranges of these properties in normal individuals.

Original languageEnglish (US)
Pages (from-to)881-891
Number of pages11
JournalJournal of Clinical Endocrinology and Metabolism
Issue number5
StatePublished - May 1986

ASJC Scopus subject areas

  • Endocrinology, Diabetes and Metabolism
  • Biochemistry
  • Endocrinology
  • Clinical Biochemistry
  • Biochemistry, medical


Dive into the research topics of 'Physiological properties of the luteinizing hormone pulse signal: Impact of intensive and extended venous sampling paradigms on its characterization in healthy men and women'. Together they form a unique fingerprint.

Cite this