The aim of this study was to examine the hypothesis that hypersecretion of ovarian androgens in polycystic ovary syndrome results from an intrinsic abnormality of androgen biosynthesis by thecal cells. Steroid accumulation by human thecal cells from normal and polycystic ovaries (PCO-theca) was examined under basal and LH-stimulated conditions. A method for dispersing and culturing human thecal cells as primary monolayers in serum-free medium was developed. LH increased androstenedione (A), progesterone (P), 17 alpha-hydroxyprogesterone, dehydroepiandrosterone, and estradiol accumulation in the overlying medium in a dose-dependent manner at a maximum effective dose of 2.5 ng/mL. The principal variables affecting the magnitude of steroid accumulation were plating density, duration of incubation, and follicle size. Using only theca from follicles less than 10 mm and keeping plating density constant, 48-h steroid production by theca from five normal ovaries was compared to that from nine polycystic ovaries isolated from both anovulatory and ovulatory women. There was a significant increase in both basal (median, 32.1 pmol/1000 cells.48 h; range, 18.7-250) and LH-stimulated (56 pmol/1000 cells; range, 40.7-406) A accumulation by PCO-theca compared to basal (1.7 pmol/1000 cells; range, 1.1-4.3) and LH-stimulated (2.8 pmol/1000 cells; range, 2.0-8.1) A accumulation by normal theca, with no overlap in values between the two. Although P production was also increased in the PCO-theca, the A to P ratios under both basal and LH-stimulated conditions were significantly higher in the PCO-theca [A/P ratio normal; PCO basal, 0.1 and 0.53 (P < 0.01); LH-stimulated, 0.04 and 0.65 (P < 0.001)], suggesting increased conversion of P to A. The steroid response to LH was similar in both groups. This is the first report of a difference in thecal androgen production between normal and polycystic ovaries and supports the hypothesis that there is a primary abnormality in the regulation of androgen production in PCOS.

In-vitro studies in both rodents and man suggest that GH can stimulate ovarian steroidogenesis, but it is not clear whether this effect is mediated by changes in circulating concentrations of insulin-like growth factor-1 (IGF-1) or whether it is a direct action on the ovary (or, indeed, both). In this study the effects of biosynthetic human GH (hGH) on the production of oestradiol and IGF-1 by human granulosa cells in culture were examined using ovarian tissue (from both normal and polycystic ovaries) which had not previously been exposed to exogenous gonadotrophin therapy. Addition of hGH (1 or 10 ng/ml) to the incubation medium resulted in a significant (1.7 to 3.6 fold) increase in oestradiol accumulation after 48h in culture. Human GH also had a significant additive effect on the dose-related responsiveness of granulosa cell oestradiol production to hFSH. Concentrations of IGF-1 in the medium were undetectable in each of these experiments. These studies demonstrate that hGH has a potent, direct stimulatory effect on production of oestradiol by the human ovary which is independent of the effect of FSH. These findings have important implications for understanding the physiological role of hGH in human ovarian function as well as for therapeutic use of biosynthetic hGH for induction of ovulation.

No abstract

The underlying cause of anovulation in polycystic ovary syndrome is unknown. Circulating levels of immuno- and bioactive FSH are within the normal range, and the follicles contain measurable levels of bioactive FSH. The aim of this study was to compare estradiol (E2) production in response to FSH by granulosa cells from normal ovaries with those from polycystic ovaries derived from both anovulatory (anovPCO) and ovulatory subjects (ovPCO). Intrafollicular levels of immunoactive FSH, E2, and androstenedione in follicles of less than 12 mm were also measured. Follicular fluid steroid concentrations were obtained from 41 pairs of normal ovaries and 23 pairs of polycystic ovaries (8 anovPCO and 15 ovPCO). In size-matched follicles from each group there were no significant differences in follicular fluid FSH or E2 concentrations, but androstenedione levels were significantly higher in 5- to 11-mm follicles from ovPCO than in corresponding follicles from normal ovaries. Dose responses to FSH were determined in granulosa cells derived from 9 pairs of normal ovaries, 7 anovPCO, and 8 ovPCO. Cells from anovPCO produced 6- to 10-fold more E2 in response to FSH than normal cells, although there was no significant difference in the ED50 values. The response in cells from ovPCO was reduced compared to normal, but this difference did not reach significance. In summary, as judged by their FSH and E2 contents, polycystic ovaries do not have a higher proportion of atretic follicles than normal. Indeed, cells from anovPCO are hyperesponsive to FSH in vitro. This could be explained by stimulation of aromatase in vivo by either paracrine or, more probably, by endocrine factors, of which insulin is an arguable candidate.

Summary. An on‐line maternity data collection system has been designed to provide the information required for perinatal audit and to allow many of the letters and forms required for effective communication in pregnancy to be produced automatically. The system meets the requirements of the Korner Committee on Health Services Information and has been approved by the Computer Policy Committee. Pregnancy is followed prospectively from the first antenatal clinic visit until the file is closed 28 days after delivery. Data are entered by midwives and secretaries onto a network of microcomputers placed at convenient points in the maternity unit. The system has been fully operational with no significant problems since the beginning of 1984 and has led to improved communication between hospital and the community. Analysis of 253 consecutive case notes showed a high level of accuracy of the data recorded on computer. Reports and clinical audit are readily available both from the system locally, from standard programmes on the regional mainframe computer and via a mainframe computer at London University