Augmented frequency and mass of LH discharged per burst are accompanied by marked disorderliness of LH secretion in adolescents with polycystic ovary syndrome

The aim of this study was to quantify pulsatile LH secretion, burst frequency and mass, LH half-life, and the approximate entropy (ApEn) or (dis- ) orderliness of LH release in adolescents with polycystic ovary syndrome (PCOS), combining a high-precision immunofluorimetric LH assay with deconvolution techniques. We sampled LH concentration profiles every 20 min overnight in 12 girls with PCOS (mean ± S.E.M. age 16.4 ± 0.57 years, body mass index (BMI) 24.4 ± 1.6 kg/m²) and 11 eumenorrheic early-follicular- phase controls (mean ± S.E.M. age 16.5 ± 0.47 years, BMI 22.2 ± 1.0 kg/m²). Fasting serum levels of androstenedione, testosterone, 17- hydroxyprogesterone (17-OHP), estrone, estradiol, FSH and sex hormone- binding globulin (SHBG) were determined. Compared with euandrogenic girls, PCOS adolescents had significantly (P < 0.005) elevated serum LH/FSH ratios, 17-OHP, androstenedione, esterone and testosterone levels, decreased SHBG, and similar estradiol. PCOS subjects exhibited a 3-fold higher mean serum LH concentration with almost no overlap with controls (8.8 ± 1.2 and 2.8 ± 0.3 IU/l respectively, P < 0.001). We initially used a conventional serum hormone concentration peak analysis method (Cluster) to evaluate the characteristics of pulsatile LH release. Cluster analysis disclosed a significant increase in serum LH concentration maximal peak height, a higher LH peak frequency and a higher mean serum LH concentration in interpulse nadirs in the PCOS group. Deconvolution analysis of mechanisms underlying the foregoing showed higher frequency in the PCOS group than the controls (7.9 ± 0.4 and 5.7 ± 0.6 pulses/12 h respectively, P < 0.05). The mass of LH released per secretory event was also significantly higher in PCOS subjects than controls (5.4 ± 0.57 and 3.4 ± 0.56 IU/l respectively, P < 0.05). Since the pulsatile production rate is the product of the mean mass of hormone secreted per pulse and the number of pulses per day, we estimated a significantly higher mean pulsatile production rate of (endogenous) LH in the PCOS group (41 ± 4.2 IU/l per day in the PCOS group vs 18 ± 2.3 IU/l per day in the controls, P < 0.01). The mean estimated half-life of endogenous LH disappearance was also significantly higher in patients with PCOS than in controls (110 ± 8.5 and 77 ± 3.7 min respectively, P < 0.01). To quantify the orderliness of LH release, we used ApEn. PCOS patients had remarkably increased disorderliness (higher ApEn) of LH release (1.09 ± 0.04 vs 0.77 ± 0.08 in controls, P = 0.002). Mean serum LH concentration, mass of LH secreted per burst, and LH production rate in PCOS, but not in normal adolescents, correlated positively with androstenedione (P < 0.02, 0.02 and 0.05 respectively). The same parameters also correlated positively with 17-OHP (P < 0.05, 0.02 and 0.05 respectively). Stepwise regression analysis unmasked a negative influence of BMI in PCOS on both mass of LH secreted per burst (r = -0.77, P < 0.005) and LH production rate (r = -0.70, P < 0.01). We conclude that PCOS adolescents secrete LH molecules with amplified frequency and burst mass and with markedly disrupted orderliness. A rise in basal (non-pulsatile) LH release, more basic LH isoforms, and/or a prolongation or asymmetry of the LH secretory burst could account for the apparently prolonged LH half-life. Determining whether disorderliness of the amplified pituitary LH release process is an intrinsic abnormality in PCOS, or reflects androgen excess, may help to clarify the pathophysiology of this oligo-ovulatory syndrome in young women.