Spectrum of the Pulsatile Characteristics of LH Release in Normal Men

To assess the spectrum of LH pulse characteristics in normal men, blood samples from 36 individuals were drawn at 20‐minute intervals for 8 hours. The subsequent immunoactive LH concentrations were analyzed by computer algorithms to delineate the frequency and amplitude characteristics of pulsatile LH secretion. The absolute range for LH pulse frequency estimated by a modified threshold method was 1–6 pulses/8 hr, with a mean (± SEM) of 3.36 ± 0.17 (median − 3) pulses/8 hr. The distribution differed significantly from a Gaussian pattern. The mean LH pulse amplitude expressed as a percent increase above nadir was 92.1 ± 6.1 (median‐91.5%). When LH pulse amplitude was defined as an increment (mIU/ml) above nadir, the mean value was 5.13 ± 0.4 (median − 4.8) mIU/ml. These two expressions of amplitude were positively correlated (P < 0.01), while the incremental (mIU/ml) pulse amplitude correlated inversely with pulse frequency (P < 0.01). To examine the influence of more intensified rates of venous sampling on the spectrum of LH pulse properties, blood was sampled at 4‐minute intervals for 8 hours in a subgroup of 13 men. Under these conditions, estimated LH pulse frequency was significantly higher, with a mean of 10.31 ± 1.87 (median − 9) pulses/8 hr compared with 20‐minute sampling in the same individuals (P < 0.001). Although the estimates of LH pulse frequency at 4‐minute and 20‐minute sampling intervals were significantly correlated (P < 0.01), the dispersion of the LH pulse frequency estimates was considerably larger at more rapid rates of sampling. There was an absolute range of 2–20 pulses/8 hr for the 4‐minute sampling, and 1–6 pulses/8 hr for the 20‐minute sampling in the same individuals. This increase in LH pulse frequency and the broader dispersion of the range of frequencies estimated at 4‐minute compared with 20‐minute sampling intervals were confirmed using either another pulse detection algorithm, or separate criteria designed to adjust false‐positive error rates in relation to sampling intensity. It was concluded that eugonadal men exhibit a broad spectrum of pulsatile LH characteristics, and the range of LH pulse attributes is even greater at more intensive rates of venous sampling. The results of this study in normal men demonstrate that a wider dispersion of physiologic LH pulse characteristics must be recognized in man. Such information is particularly important whenever possible departures from normal patterns of episodic gonadotropin secretion are sought in various clinical disorders. Moreover, these observations indicate that a wide range of LH pulse signal characteristics will maintain testosterone production effectively in the normal male. 1986 American Society of Andrology