Tumor necrosis factor alpha in the human ovary: presence in follicular fluid and effects on cell proliferation and prostaglandin production

OBJECTIVE

To examine the concentration of tumor necrosis factor alpha (TNF alpha) in human follicular fluid (FF) and its effects on cultured human granulosa-lutein cells.

DESIGN

The concentration of TNF alpha in FF from hyperstimulated cycles and in conditioned media from cultured granulosa-lutein cells was measured by radioimmunoassay (RIA) and bioassay using L929 cells. The effects of recombinant human TNF alpha (rTNF alpha) on proliferation and production of progesterone (P) and prostaglandin (PG, PGE2, and PGF2 alpha) by cultured human granulosa-lutein cells were assessed.

SETTING

In vitro fertilization and embryo transfer (IVF-ET) program at Reproductive Medicine Unit, The Queen Elizabeth Hospital, Woodville, South Australia, Australia.

PATIENTS

Twenty-five women undergoing IVF-ET for tubal factor infertility.

RESULTS

The concentration of immunoreactive TNF alpha in FF was 0.36 +/- 0.02 microgram/L, and there were no significant correlations between levels of TNF alpha and steroids or FF volume. Bioactivity for TNF alpha was considerably less. Immunoreactive or bioactive TNF alpha was not detected in conditioned media from granulosa-lutein cell culture. Recombinant human TNF alpha dose-dependently stimulated proliferation of cultured granulosa-lutein cells as measured by incorporation of 3H-thymidine, but in contrast to earlier reports, we were not able to demonstrate any effect of rTNF alpha on basal or human chorionic gonadotropin-stimulated P accumulation during culture periods of up to 72 hours. The accumulation of both PGE2 and PGF2 alpha was dose-dependently increased by rTNF alpha during a 48-hour incubation period. Time course studies revealed that maximal levels of both PGE2 and PGF2 alpha were reached within 12 hours of culture.

CONCLUSION

Immunoreactive and bioactive TNF alpha is present in FF. Tumor necrosis factor alpha may have a physiological role in stimulating proliferation of follicular cells and PG production at the time of ovulation and formation of the corpus luteum.

Comparison of specific immunoassays for detection of the beta-core human chorionic gonadotrophin fragment in body fluids

We have validated two new methods, one radioimmunoassay (RIA) and one immunoradiometric assay (IRMA), for the detection of beta-core hCG fragment (beta C-hCG) in body fluids. In addition, we have compared their performance with two other assays designed for beta C-hCG quantification. The RIA uses a rabbit polyclonal antibody raised against pure beta C-hCG which has a high affinity constant, is sensitive to 5 pmol/l, and has significant cross-reaction only with the free beta LH subunit. The IRMA, designed in a liquid phase, uses the same polyclonal antibody associated with a 125I-labelled mouse monoclonal antibody (32H2) raised against beta hCG, is sensitive to 1.5 pmol/l, and does not cross-react significantly with any related glycoprotein. Comparison between these two assays and two others previously published was made by measuring beta C-hCG in urine from healthy pregnant women (n = 47) and gave correlation coefficients higher than r = 0.960 with any combination. Analysis of beta C-hCG in urine of non-pregnant subjects (n = 238) showed measurable beta C-hCG in 8.8% (levels ranged from 5 to 34 pmol/l) with the IRMA and 88.3% with the RIA (n = 30; ranging from 28.4 to 228 pmol/l) (P = 0.05). We concluded that, despite different affinities of the antibody involved and different cross-reactivities with related glycoproteins, the four assays we examined may be equally employed to detect beta C-hCG in pregnancy urine. However, the IRMA appears to be more appropriate for beta C-hCG analysis in non-pregnant individuals, specifically in postmenopausal women because of the high cross-reactivity of the RIA with free beta LH or beta fragments of other glycoproteins. These studies have significance for our understanding of the physiology of beta C-hCG in cancer, pregnancy and after the menopause.

Distribution of the beta-core human chorionic gonadotrophin fragment in human body fluids

The origins of a fragment of the human chorionic gonadotrophin (hCG) molecule, beta-core (beta C-hCG) were studied by analysis of beta C-hCG concentrations in biological fluids. In addition, the ability of the placenta to produce the fragment and the metabolism of hCG to beta C-hCG by human granulosa cells was determined in tissue culture. Finally the conversion of exogenous hCG to beta C-hCG was studied in vivo. The fragment was present in pregnancy urine as well as that from premenopausal and postmenopausal subjects. The highest concentrations were found in pregnant women. Ratios of beta C-hCG to intact hCG were higher in pregnancy urine when radioimmunoassay (RIA) was used compared with immunoradiometric assay (IRMA) (0.67 and 0.37 respectively). Concentrations of beta C-hCG were higher in postmenopausal urine than in premenopausal specimens. A significant amount of a high molecular weight beta C-hCG immunoreactive material was found in serum samples after size separation, and the molar ratio of beta C-hCG/hCG was estimated as 0.019. Amniotic fluid also contained small quantities of two forms of immunoreactive beta C-hCG and the ratio of 0.01 for authentic beta C-hCG/hCG increased to 0.026 when the high molecular weight form was considered. Cultured trophoblastic tissue released material with beta C-hCG immunoreactivity in the medium and chromatographic separation revealed that the majority of this material was of higher molecular weight compared with the authentic beta C-hCG form. beta C-hCG was the principal glycoprotein found in follicular fluid after hyperstimulated folliculogenesis and intramuscular injection of 5000 IU hCG. We also demonstrated that 26% of follicular fluid samples (n = 50) were positive for beta C-hCG; levels ranged from 5.2 to 23.0 pmol/l (13.1 +/- 5.7); S.D.) when a specific IRMA was used. The RIA could detect beta C-hCG in 48 samples (96%), levels ranging from 7.0 to 28.5 pmol/l (19.4 +/- 5.2). Moreover, granulosa cells cultured in the presence of hCG were able to degrade the intact molecule to both high molecular weight and authentic immunoreactive forms of beta C-hCG. After gel filtration, material of molecular weight over a wide range and immunoreactive for beta C-hCG was present in human seminal plasma. Assaying 74 samples of this fluid by IRMA, beta C-hCG was detected in 42 (56.7%), levels ranging between 5.5 and 59.5 pmol/l (24.9 +/- 15.2).(ABSTRACT TRUNCATED AT 400 WORDS)

Urinary concentrations of beta core fragment of hCG throughout pregnancy

OBJECTIVE

We sought to determine a reference range for urinary immunoreactive beta core fragment of hCG (beta C-hCG) in pregnancy, the ratio between beta C-hCG and intact hCG, and the earliest detectable rise of beta C-hCG in urine.

METHODS

Urine was obtained from 741 pregnant women between 6-41 weeks' gestation, as well as from women undergoing donor insemination with timed ovulation peaks.

RESULTS

The beta core fragment of hCG reached a maximum between 8-15 weeks, with a decrease between 20-29 weeks. The molar ratio of beta C-hCG to intact hCG was always greater than 1.

CONCLUSION

In pregnancy, beta C-hCG concentrations increase in the urine in parallel to intact hCG but at a higher molar ratio, suggesting either placental production of beta C-hCG or enhanced metabolism of hCG to beta C-hCG in peripheral organs.

Concentrations of immunoreactive interleukin-1 and interleukin-2 in human preovulatory follicular fluid

The presence of immunoreactive interleukin-1 (IL-1) and interleukin-2 (IL-2) in human follicular fluid obtained at the time of oocyte collection for in-vitro fertilization was ascertained by radioimmunoassay. In group I (20 fluids from 20 patients), the concentrations of IL-1 were 0.9 +/- 0.06 and 1.9 +/- 0.04 (mean +/- SEM) fmol/l in follicular fluid and plasma respectively. A positive correlation existed between IL-1 levels in follicular fluid and plasma (r = 0.56, P less than 0.01). Concentrations of IL-2 were 3.5 +/- 0.2 and 6.1 +/- 0.3 fmol/l in follicular fluid and plasma respectively. A positive correlation of IL-2 levels was also found between follicular fluid and plasma (r = 0.65, P less than 0.01). There was no association between IL-1, IL-2 and steroid levels, regardless of whether they were compared in follicular fluid or plasma. Group II was composed of a series of fluids (two to seven samples for each of seven patients) in which the follicular concentrations of IL-1 and IL-2 did not show a positive correlation with the volume of follicular fluid or the concentrations of follicular fluid steroids. It is concluded that human preovulatory follicular fluid contains immunoreactive IL-1 and IL-2. The role of IL-1 and IL-2 in ovarian physiology remains to be determined.

Distribution of leukocyte subpopulations in the human corpus luteum

Cytokines, as secreted products of leukocytes, have roles in many organs of the body via paracrine or autocrine mechanisms. In the present study, we demonstrate by immunocytochemistry the leukocytes present in the human corpus luteum in order to investigate further the relationship between leukocytes, cytokines and corpus luteum function. Ten intact corpora lutea were collected from female patients who had no apparent ovarian disease. The mean age of these patients was 37 years (range 23-55 years). Frozen and paraffin sections were subjected to analysis using monoclonal antibodies which were specific to leukocyte marker antigens. The results showed that there are macrophages, cells positive for leukocyte common antigen (LCA), T lymphocytes including T helper/inducer (T4) cells, T cytotoxic/suppressor (T8) cells and activated T (Ta) cells (interleukin-2 receptor-positive cells), monocytes and natural killer (NK) cells but not B lymphocytes present in the human corpus luteum. The distribution of the leukocytes present in the different parts of the corpus luteum was found to be in the order: theca-luteal area greater than loose connective tissue area greater than granulosa-luteal area. Macrophages and T lymphocyte subsets comprised the main components of the total leukocytes in the human corpus luteum. Ta cells were only localized in the loose connective tissue of the corpus luteum. In most cases, macrophages, LCA cells and T4 cells tended to be situated in a single cell layer on the edge of the theca-luteal area and surrounding the granulosa-luteal area. These results suggest that the leukocytes may act to a greater extent in the theca-luteal area than in the granulosa-luteal area.(ABSTRACT TRUNCATED AT 250 WORDS)

Simple quantitative measurement of serum choriogonadotropin compared with immunoradiometric, immunoenzymometric, and chemiluminescent assays

We evaluated a new simplified quantitative method (Tandem Icon QSR; Hybritech) for choriogonadotropin (hCG), which could theoretically be performed in a ward or with outpatients. The method was compared with immunoradiometric (Bioclone Australia), chemiluminescent (Amerlite; Amersham), and immunoenzymometric assays (Stratus; Dade). We analyzed by each of the methods 104 serum samples from pregnant and nonpregnant patients. For quantitative hCG values indicative of pregnancy (greater than 25 int. units/L), excellent correlation was observed between the Tandem Icon method and the other standard laboratory assays (r = 0.995, 0.990, and 0.992, respectively). Occasional problems arose because of the instability of Tandem Icon QSR reagents at room temperature but this was resolved by storing the reagents at 4 degrees C. We conclude that this simplified quantitative method for hCG is reliable and suitable for use outside of the routine immunoassay laboratory.

Simple Quantitative Measurement of Serum Choriogonadotropin Compared with Immunoradiometric, Immunoenzymometric, and Chemiluminescent Assays

We evaluated a new simplified quantitative method (Tandem Icon QSR; Hybritech) for choriogonadotropin (hCG), which could theoretically be performed in a ward or with outpatients. The method was compared with immunoradiometric (Bioclone Australia), chemiluminescent (Amerlite; Amersham), and immunoenzymometric assays (Stratus; Dade). We analyzed by each of the methods 104 serum samples from pregnant and nonpregnant patients. For quantitative hCG values indicative of pregnancy (greater than 25 int. units/L), excellent correlation was observed between the Tandem Icon method and the other standard laboratory assays (r = 0.995, 0.990, and 0.992, respectively). Occasional problems arose because of the instability of Tandem Icon QSR reagents at room temperature but this was resolved by storing the reagents at 4 degrees C. We conclude that this simplified quantitative method for hCG is reliable and suitable for use outside of the routine immunoassay laboratory.

The leuprolide flare regime for in-vitro fertilization/gamete intra-fallopian transfer and embryo cryopreservation

Gonadotrophin releasing hormone agonists (GnRHa) are now well established as adjuvant agents for in-vitro fertilization (IVF)/gamete intra-Fallopian transfer (GIFT) but several different modes of usage have been proposed. Our experience with 328 cycles of leuprolide used in a flare regime is reviewed. An endocrinologically proven flare effect was associated with a reduction of human menopausal gonadotrophin (HMG) usage (10 versus 16 ampoules) and a lower cycle cancellation/conversion rate (7.4 versus 11.3%). Overall, satisfactory rates of oocyte recovery (93%, mean number of oocytes 7.0), clinical pregnancy (24.4% per oocyte recovery) and pregnancy from frozen/thawed embryo transfers (14%) were achieved. The flare protocol appears to be a satisfactory choice for the majority of subjects but careful monitoring is required to avoid the potential for ovarian hyperstimulation.

Choriogonadotropin-mediated thyrotoxicosis in a man

A 38-year-old man with a metastatic gonadotropin-secreting tumor of unknown primary origin presented with both clinical and biochemical findings of hyperthyroidism in association with markedly increased concentrations of human choriogonadotropin (hCG) in plasma. After chemotherapy, the concentrations of both hCG and free thyroxin decreased and the patient became euthyroid. We discuss the rare occurrence of this presumably hCG-driven hyperthyroidism in men and compare it with the relatively more common eumetabolic hyperthyroidism associated with choriocarcinoma in women.

Lymphokines, including interleukin-2, alter gonadotropin-stimulated progesterone production and proliferation of human granulosa-luteal cells in vitro

The effects of human interleukin-1 (IL-1) and IL-2 on human granulosa-luteal cell progesterone production were examined with or without hCG stimulation in vitro. Human granulosa-luteal cells were recovered from follicular fluid obtained from women undergoing in vitro fertilization procedures and cultured for up to 7 days before supernatant progesterone level measurement. Lymphokine-rich conditioned medium was prepared from mitogen-stimulated human peripheral blood leukocytes (HPL-CM). The influence of HPL-CM on both granulosa-luteal cell progesterone production and cell growth was inhibitory. In contrast, supernatants of the IL-2-producing cell line MLA-144 (MLA-CM) stimulated both basal progesterone secretion and cell proliferation. Human recombinant IL-2 (from 0.1-100 IU) alone did not change progesterone levels, compared to control values, after 24 h of cell culture. However, 1, 10, and 100 IU IL-2 significantly inhibited progesterone secretion from cells stimulated by 5 IU hCG (P less than 0.01). The enhanced progesterone levels stimulated by forskolin were also significantly inhibited by 10 IU IL-2 (P = 0.01). This effect was not mediated through decreased cAMP, since the forskolin-enhanced cAMP level was not influenced by IL-2, IL-1, with or without hCG, did not show any effect on progesterone production during either 24 or 48 h of cell culture. It is concluded that 1) human recombinant IL-2 significantly inhibits progesterone production stimulated by hCG in human granulosa-luteal cells; 2) IL-2 also had a marked inhibitory effect on forskolin-induced progesterone release, but did not influence the increased cAMP level stimulated by forskolin; 3) the inhibitory influence of IL-2 on progesterone synthesis may be down-stream in the signal transduction pathway from cAMP activation; and 4) HPL-CM and MLA-CM produced inhibitory and stimulatory effects, respectively, on both basal and hCG-stimulated progesterone levels as well as on granulosa-luteal cell proliferation. These activities cannot be completely attributed to IL-2, and other mediators of leukocyte origin may, therefore, exist.

Differential effects of gonadotrophin-releasing hormone agonists administered as desensitizing or flare protocols on hormonal function in the luteal phase of hyperstimulated cycles

The incorporation of gonadotrophin-releasing hormone agonist (GnRHa) in in-vitro fertilization (IVF) stimulation protocols has led to doubt about the quality of the subsequent luteal phase. The effects of two GnRHa stimulation protocols on luteal phase concentrations of oestradiol (E2), progesterone (P), luteinizing hormone (LH) and follicle stimulating hormone (FSH) were compared with the standard clomiphene stimulation regimen. Subjects receiving clomiphene with human menopausal gonadotrophin (HMG, n = 377) showed essentially similar luteal phase P concentrations to those receiving leuprolide acetate/HMG as a desensitization protocol. Subjects receiving concomitant leuprolide and HMG from day 2 to utilize the flare effect of the GnRHa exhibited significantly lower P levels in the luteal phase compared to clomiphene/HMG and leuprolide desensitization protocols despite the addition of HCG support. This occurred despite equivalent E2 concentrations at the time of ovulation and identical numbers of oocytes recovered. LH concentrations in non-conception cycles were suppressed for at least 14 days in the luteal phase in both GnRHa protocols compared to clomiphene stimulation. Differences were less obvious in cycles where conception occurred suggesting that implantation may proceed more favourably when the luteal endocrinology was optimal. It is concluded that flare methods of GnRHa hyperstimulation are associated with significantly different luteal phases compared with clomiphene or desensitization protocols. It is proposed that the use of the flare type of stimulation may significantly influence the response of the granulosa cells to LH or HCG via gonadotrophin receptors or through altered post-receptor function.

Evaluation of renal function and fluid homeostasis during recovery from exercise-induced hyponatremia

Renal function including fluid and electrolyte balance was studied during recovery in eight subjects who developed symptomatic hyponatremia (HN; plasma sodium concentration less than 130 mM) during an 88-km ultramarathon footrace and compared with results for normonatremic runners [NN; n = 18, mean postrace plasma sodium concentration, 138.2 +/- 1.2 (SE) mM]. Estimated fluid intake during the race for HN was 12.5 +/- 1.6 (SE) liters over 9 h 41 min (+/- 28 min). HN excreted a net fluid excess of 2.95 +/- 0.56 (range 1.2-5.9) liters compared with a fluid deficit of 2.7 +/- 0.3% body weight in NN. The sodium deficit was 153 +/- 35 mmol in HN and 187 +/- 37 mmol in NN. Despite the fluid overload, plasma volume was decreased by 24.1 +/- 5.0% in HN compared with 8.2 +/- 2.6% in NN. Serum renin activity (5.1 +/- 2.0 ng.ml-1.h-1), aldosterone concentrations (410 +/- 34 ng/l), creatinine clearances (174.8 +/- 28.2 ml/min), and urine output (6.4 +/- 1.0 ml/min) were markedly elevated in HN during recovery. Thus the hyponatremia of exercise results from fluid retention in subjects who ingest abnormally large fluid volumes during prolonged exercise.

Decrease in luteal gonadotropin concentration in conception cycles after in vitro fertilization/gamete intrafallopian transfer

The concentrations of the gonadotropins, luteinizing hormone (LH), and follicle-stimulating hormone (FSH) were measured in the luteal phase of the cycle in patients undergoing ovarian hyperstimulation. In nonconception cycles, FSH and LH were increased in the late luteal phase compared with conception cycles in which both gonadotropins were suppressed. Estradiol (E2) and progesterone concentrations increased in pregnancy cycles and may be the sole cause for the decreased gonadotropin concentrations as shown by equivalent concentrations of LH and FSH in both pregnancy and nonpregnancy cycles after matching for E2 concentrations. Subjects who subsequently had twin pregnancy or a spontaneous abortion were compared with those with a successful ongoing singleton conception. There were no significant differences relative to LH and FSH between the three groups, although in twin pregnancy FSH tended to be lower at day 16 from oocyte recovery. It is concluded that suppression of LH and FSH in hyperstimulated pregnancy cycles occurs after the time of the rising human chorionic gonadotropin concentrations in plasma.

Metabolic profiles and lipoprotein lipid concentrations in non-obese and obese patients with polycystic ovarian disease

Clinical parameters, androgen status and lipoprotein lipid profiles were assessed in 10 non-obese and 10 obese patients with polycystic ovarian disease (PCOD) and reference subjects matched for age, height and weight. Both obese and non-obese women with PCOD had significantly higher androgen levels when compared to the reference groups. When comparison of lipoprotein lipid profiles were made between groups, non-obese women with PCOD had significantly higher total cholesterol, triglycerides and LDL-cholesterol levels than non-obese reference subjects. Obese PCOD women manifested significantly higher total cholesterol, LDL-cholesterol, cholesterol/HDL, and LDL/HDL values than did obese reference subjects. Correlations between serum androgens and lipoprotein lipid concentrations in PCOD and normal women were unhelpful. Both non-obese and obese patients with PCOD had significantly higher systolic and diastolic blood pressures (BPs) than the reference groups. Thus, both non-obese and obese women with PCOD manifest hyperandrogenaemia which may result in a male pattern of lipoprotein lipid concentrations.

The effect of a brain lesion on bone healing

Using a feline model, the effects of stereotaxic brain lesions producing unilateral spasticity on bilateral fibular osteotomies were observed. Healing in spastic limbs occurred with more new bone formation than on the contralateral (control) side. Union appeared to be slightly faster in spastic limbs. Histomorphometric analysis of resected fibular specimens following union revealed little difference between spastic and non-spastic limbs. When these brain-lesioned animals were compared with a group with identical osteotomies but no brain lesions in respect of various systemic biochemical and endocrine parameters, no differences were found.

Insulin levels and 125I-insulin binding to monocytes in a non obese pregnant woman with polycystic ovarian disease

Insulin responses to oral glucose, 125I-insulin binding to monocytes and androgen profiles were studied in a 25-year-old non obese woman with polycystic ovarian disease pre pregnancy, during pregnancy and in the post partum period. The patient was hyperinsulinaemic both pre and post pregnancy, however, her insulin resistance worsened during pregnancy. 125I-insulin receptor binding to monocytes during pregnancy decreased primarily due to a decrease in receptor number. This decreased binding was attributed to the down regulatory effect of the basal hyperinsulinaemia. Androgen levels were elevated pre and post pregnancy and suppressed during pregnancy.

Increased and intermittent prostaglandin release from amnion detected by a new superfusion technique for full thickness fetal membrane

Prostaglandin E2 (PGE) and F2 alpha (PGF) release by the intact fetal membranes is described using a novel superfusion technique allowing for the independent assessment of prostaglandin release from the amnion and chorio-decidua whilst maintaining the anatomical integrity of the fetal membranes. The effect of labour on prostaglandin release is described. Using this system it was confirmed that the amnion is a major site of prostaglandin release and possibly production. Labour resulted in a significant increase of both PGE and PGF release from the amnion side only (Pre-labour: PGE 918 pg/cm2/3h, PGF 370 pg/cm2/3h; Labour: PGE 2993 pg/cm2/3h, PGF 662 pg/cm2/3h). No change in either PGE or PGF release from the chorio-decidual side was observed in relation to labour. In addition a change in the pattern of prostaglandin release from the amnion was observed in tissues obtained after the onset of labour. In 6 of 8 samples obtained after spontaneous labour an intermittent or pulsatile release of both PGE and PGF was observed from the amnion side as compared to the steady state of prostaglandin release from all 10 samples obtained before labour.

Detection of a small molecular species of human chorionic gonadotropin in the urine of patients with carcinoma of the cervix and cervical intraepithelial neoplasia: comparison with other assays for human chorionic gonadotropin and its fragments

A low-molecular-weight glycoprotein containing sequences of the beta subunit of human chorionic gonadotropin (hCG) has been found in the urine of patients with carcinoma of the cervix using an immunoradiometric assay. This fragment has chromatographic and immunological identity with hCG beta core. This molecule was present in 52 to 77% of all patients with invasive disease, while between 11 and 27% of patients with cervical intraepithelial neoplasia (CIN) also exhibited significant hCG beta-core immunoactivity. Few patients had either a positive assay for intact hCG or a positive assay directed at an epitope on the beta subunit (beta-hCG radioimmunoassay) in serum. However, between 17 and 40% of patients with invasive disease were positive for free beta-subunit immunoactivity in the blood. The origin of the beta-core immunoactivity in the urine is uncertain; while tumor production cannot be excluded, it is possible that the molecule originates from renal metabolism of small quantities of the beta subunit of hCG. Regardless of the source of the molecule, hCG beta core is a far more sensitive marker of hCG production by tumors than is serum hCG.

Measurement of human chorionic gonadotrophin (hCG): indications and techniques for the clinical laboratory

The structure of human chorionic gonadotrophin (hCG) is so similar to that of luteinizing hormone (LH) that a variety of assay techniques have been devised to differentiate between these two hormones. The principal indications for measurement of hCG using these methods have not changed greatly over the past decade but the improvements in the sensitivity, specificity and the development of assays for free subunits and metabolic fragments have expanded the use of hCG assays. The review discusses the use of hCG measurement in a routine clinical immunoassay laboratory and emphasizes different requirements for clinical situations.

Luteal function associated with single, multiple and ectopic embryo implantation in natural cycles or after ovarian hyperstimulation for in-vitro fertilization/gamete intra-fallopian transfer

Levels of reproductive steroids and gonadotrophins were analysed retrospectively during the peri-implantation period following non-conceptional and conceptional natural cycles and in cycles associated with ovarian hyperstimulation for in-vitro fertilization or gamete intra-Fallopian transfer. In cycles not associated with conception, the luteal phase of hyperstimulated cycles (n = 100) was characterized by higher serum progesterone and oestradiol levels (P less than 0.01) and with an earlier decline in steroids than in natural cycles (n = 21). On day 11 (day of oocyte recovery = day 0), the level of progesterone in twin (n = 59) and triplet (n = 13) pregnancies was higher than singleton pregnancies (n = 176) (P less than 0.006, P less than 0.006 respectively) while those destined to abort (n = 66) had lower progesterone levels (P less than 0.01). Ectopic implantation (n = 11) had the lowest progesterone concentrations on day 11 (P less than 0.01) and this may imply a delay in corpus luteum rescue or a later implantation time than intrauterine conception.

The incidence of hyponatremia during prolonged ultraendurance exercise

Recent studies have shown that potentially fatal hyponatremia can develop during prolonged exercise. To determine the incidence of hyponatremia in athletes competing in ultradistance events, we measured serum sodium levels in 315 of 626 (50%) runners who were treated for collapse after two 90 km ultramarathon footraces (total starters 20,335; total finishers 18,031) and in 101 of 147 (69%) finishers in a 186 km ultratriathlon. In both races the athletes drank fluids with low sodium chloride content (less than 6.8 mmol.l-1). Hyponatremia (serum sodium level less than 130 mmol.l-1) was identified in 27 of 315 (9%) collapsed runners in the 90 km races and in none of the triathletes. In response to diuretic therapy, the runner with the most severe hyponatremia (serum sodium level = 112 mmol.l-1) excreted in excess of 7.5 l dilute urine during the first 17 h of hospitalization. These data suggest that, although symptomatic hyponatremia occurs in less than 0.3% of competitors during prolonged exercise even when they ingest little sodium chloride, it is found in a significant proportion (9%) of collapsed runners. A regulated contraction of the extracellular fluid volume would explain why the majority of athletes maintain normal serum sodium levels even though they develop a significant sodium chloride deficit during prolonged exercise. Alternatively, sodium chloride losses during prolonged exercise may be substantially less than are currently believed. Physicians treating collapsed ultradistance athletes need to be aware that as many as 10% or more of such patients may be hyponatremic.

125I-insulin binding to circulating erythrocytes, monocytes and cultured fibroblasts in non-obese patients with polycystic ovarian disease

The insulin receptor status of circulating erythrocytes, monocytes and cultured fibroblasts were studied in non-obese Indian women with polycystic ovarian disease (PCOD) with no clinical evidence of acanthosis nigricans and age, height and weight matched reference subjects. The women with polycystic ovaries had decreased mean maximum specific binding to erythrocytes (PCOD 7.4 +/- 0.6%, normal women 11.5 +/- 0.3%; p less than 0.0001) and monocytes (PCOD 2.4 +/- 0.3%, normal women 4.1 +/- 0.4%; p less than 0.003) when compared to the normal women. This decreased binding was attributed to a change in both receptor number and affinity. 125I-insulin binding to cultured fibroblasts revealed similar mean maximum specific binding and affinity values in both groups studied. Although further work is necessary to exclude genetic or post-receptor defects, it is likely that an environmental factor is implicated in the decreased 125I-insulin binding to erythrocytes and monocytes.

Prostaglandins in dysfunctional labour; evidence for altered production of prostaglandin F2 alpha

Dysfunctional labour was studied in relation to prostaglandin concentrations in amniotic fluid and production by fetal membranes. Initial clinical validation of the model established the presence of hypokinetic labour with no evidence of obstruction to the fetal progress. Prostaglandin F2 alpha (PGF2 alpha) and 13, 14 dihydro-15-keto-prostaglandin-F2 alpha concentrations in the amniotic fluid were low despite relatively normal concentrations of prostaglandin E2. Membranes removed from patients with the condition released very low concentrations of PGF2 alpha from the amniotic side with no alteration on the choriodecidual side of the membrane. Studies of free and phospholipid-associated arachidonic acid indicated normal release of arachidonic acid in dysfunctional labour. No changes in amniotic fluid-related inhibitors and stimulators of prostaglandin synthetase were detected. It is suggested that PGF2 alpha production is impaired in dysfunctional labour and that this prostaglandin is primarily involved in the progress of labour.

Changes in the binding characteristics of oestrogen and progesterone in uterine and intrauterine tissue with the progression of human pregnancy

The binding characteristics of oestrogen and progesterone in intrauterine tissue were studied in the non-pregnant state and throughout pregnancy using sucrose density gradient centrifugation and titration analyses. High affinity receptors for these steroids which were present in high concentrations in tissue from non-pregnant patients and patients in early pregnancy, decreased with the advance of pregnancy to undetectable levels at term. Non-saturable binding to progesterone detected in term tissue was identified as albumin and cortisol binding globulin.

Insulin and C-peptide secretion in non-obese patients with polycystic ovarian disease

Plasma glucose, immunoreactive insulin (IRI) and C-peptide responses during an oral glucose tolerance test (oGTT) were assessed in 11 non-obese patients with polycystic ovarian disease (PCOD) and 11 reference subjects matched for age, height and weight. Also, 6 patients with PCOD and 6 normal women were subjected to intravenous glucose tolerance testing (ivGTT) On oGTT, all subjects exhibited normal glucose tolerance; however, PCOD patients had significantly higher mean plasma glucose levels at 30, 60, 90 and 120 min and higher mean incremental glucose areas. In addition the patients with polycystic ovaries showed higher mean basal IRI and C-peptide levels, higher mean glucose stimulated IRI and C-peptide levels and higher mean incremental IRI and C-peptide values. The molar ratios of C-peptide/IRI were significantly lower in the PCOD group at all time intervals after glucose stimulation when compared to the normal women. During ivGTT, there were significantly higher mean glucose levels at 5, 40, 50 and 60 min in the PCOD group when compared to the reference group. The IRI response to intravenous glucose in the PCOD women was similar to the reference group. The findings on oGTT suggest that non-obese patients with PCOD have increased pancreatic IRI secretion as well as impaired hepatic extraction of the hormone.

Attempts at dietary alteration of prostaglandin pathways in the management of pre-eclampsia

It has been suggested that dietary supplementation with prostaglandin precursors may enhance the synthesis of PGE which lowers vascular sensitivity to increased levels of angiotensin II in pregnancy. Therefore the effect of dietary supplementation with evening primrose oil (linoleic acid and gamma-linoleic acid) in African primigravidae with established pre-eclampsia was studied. Patients were randomly allocated to one of two groups. Group A (23 patients) received 8 capsules/day of evening primrose oil and group B (24 patients) received 8 capsules of placebo. No significant differences were found between the groups in respect to perinatal outcome, blood pressure lowering effect and haematological indices.

Termination of pregnancy with mifepristone after intra-uterine death. Clinical and hormonal effects

Mifepristone was used in a dosage of 400 mg/d in a double-blind study to induce labour in patients with intra-uterine fetal death in late pregnancy. Eight of 12 patients who received the drug delivered within 72 hours while only 2 of 12 patients treated with placebo delivered during a similar period. No adverse effects, viz. excessive vaginal bleeding and abnormal biochemical or haematological parameters, were associated with the use of this drug.

Comparison of human chorionic gonadotrophin concentrations in the sera of patients with normal and abnormal pregnancy measured by radio-immunoassay and immunoradiometric assay

Human chorionic gonadotrophin (HCG) concentrations were measured in the serum of 40 patients with normal pregnancy, 18 with ectopic pregnancy, 36 with incomplete abortions and 24 patients with gestational trophoblastic disease using a radio-immunoassay (RIA) and immunoradiometric assay (IRMA). There were no significant differences in HCG concentrations measured by the two assays with regard to the first three groups but the patients with gestational trophoblastic disease had consistently higher results on RIA, probably reflecting increased circulating free beta-subunit of HCG. It was concluded that measurement of HCG using IRMA for the intact molecule gives equivalent results to RIA in patients with normal and ectopic pregnancy as well as after abortion. However, monitoring of patients with choriocarcinoma and hydatidiform mole requires an assay that detects free beta-HCG.

The role of prostaglandins in the inhibition of cultured carcinoma cell growth produced by gamma-linolenic acid

The growth of the cultured human breast carcinoma cell line NUB 1 as well as that of other cultured malignant cells has been shown to be inhibited by addition of gamma-linolenic acid (GLA) to the culture medium. It has previously been suggested that these findings may be attributed to correction of a GLA deficiency in malignant cells, with supplementation of this fatty acid leading to increased prostaglandin (PG) production and consequent growth inhibition. To test this hypothesis the effect of 50 micrograms/ml concentrations of GLA and its sequential metabolite dihomo-gamma-linolenic acid (DGLA) and cell growth, morphology and prostaglandin (PGE and PGF) production by NUB 1 cells was investigated. GLA increased PGE and PGF production, inhibited cell growth and caused accumulation of lipid containing cytoplasmic granules. While treatment with DGLA increased PG production to a significantly greater extent than GLA administration it had no apparent effect on cell growth of morphology and did not inhibit cell growth. These findings suggest that some action other than the ability to increase PG production may be responsible for the inhibitory effects produced by GLA in malignant cells.

Spectrum of thyroid function abnormalities in gestational trophoblastic neoplasia

The relative influences of the putative thyroid stimulator, pregnancy state and non-thyroidal illnesses, were assessed in 52 patients with gestational trophoblastic neoplasia (GTN). In this study biochemical thyroid status was assessed with the aid of thyrotrophin (TSH) measurements by a sensitive immunoradiometric assay and measurement of total and free thyroid hormone concentrations. The reference group consisted of 63 patients with normal pregnancy. Thyroid function in the GTN patients was also compared with that in 36 female patients with Graves' disease and 34 female patients with non-thyroidal illnesses. We found the major influence on thyroid functions in Group I ('Hyperthyroid') patients with GTN to be the presence of the thyroid stimulator, while non-thyroidal illnesses played an important role in Group II ('High T4 syndrome') and III ('euthyroid') patients with GTN. Thyroid function was also stimulated in Group II patients. The influence of pregnancy was seen in all three groups of patients with GTN, who had higher TBG concentrations than non-pregnant patients. We conclude that there is a spectrum of thyroid function abnormalities in GTN, and that thyroid function in an individual patient is determined by the relative influence of the thyroid stimulator, non-thyroidal illnesses and the pregnancy.

Serum progesterone in the diagnosis of ectopic pregnancy: a valuable diagnostic test?

The value of a single serum progesterone (P) assay in the diagnostic work-up of suspected ectopic pregnancy was investigated in 89 patients with ectopic pregnancy and 27 patients with incomplete abortion. Reference values for P in the blood were obtained from 77 patients with normal intrauterine pregnancies in the first trimester. With the use of a discriminatory level of 63 nmol/l (20 ng/ml), sensitivity of 92%, specificity of 84%, positive predictive index of 90%, and negative predictive index of 87% were achieved. Addition of pelvic ultrasound further improved the diagnostic accuracy. It is concluded that serum P measurement offers a valuable adjunct to existing methods of diagnosis of ectopic pregnancy.

Serial oxytocin levels in amniotic fluid and maternal plasma during normal and induced labour

Oxytocin was measured by a specific and sensitive radioimmunoassay in plasma and amniotic fluid after extraction with Sep-Pak cartridges in patients undergoing elective caesarean sections, normal labour and labour induced with oxytocin infusion or prostaglandins. In maternal plasma, levels of oxytocin correlated with the period of gestation; concentrations at term were significantly higher than in the first two trimesters. Maternal plasma levels of oxytocin before the onset of contractions were not significantly different from those at the onset of spontaneous labour or at full cervical dilatation. Levels of oxytocin in patients induced with oxytocin were not statistically different from levels observed in spontaneous labour. Amniotic fluid oxytocin levels were not significantly different between the groups either at the onset of labour or immediately before delivery. Umbilical arterio-venous differences in oxytocin were present in all groups except the prostaglandin-induced group. Increased prostaglandins associated with the onset of normal labour are probably not a consequence of changes in oxytocin concentrations.

Hypogonadotrophic variant of Klinefelter's syndrome. A case report

A 16-year-old boy with 47,XXY chromosomal complement (Klinefelter's syndrome) presented with delayed puberty and apparent gonadotrophin deficiency. Despite an inadequate growth hormone response to insulin-induced hypoglycaemia and to L-dopa administration, his somatic growth was appropriate for his delay in pubertal development, increasing markedly on testosterone treatment. Patients with Klinefelter's syndrome may have abnormalities of neuro-endocrine regulation.

Intrauterine pressure studies in multigravid patients in spontaneous labour: effect of oxytocin augmentation in delayed first stage

Uterine activity was assessed by measuring the uterine activity integral (UAI) in 116 black multigravid patients in spontaneous active labour at term. Although individual patients showed a wide range of mean UAI levels, normal labour (group I, n = 54) was associated with a mean UAI level of 1640 kPas/15 min which was higher than the mean level observed in patients in whom delay in cervical dilatation occurred in active labour. The patients who failed to progress in labour were treated with oxytocin infusion and 32 of them made good progress in labour and achieved vaginal delivery (group IIa: mean UAI pre-oxytocin treatment 1040 (SD 424) kPas/15 min, post-oxytocin 1890 (SD 559) kPas/15 min). The other 23 patients required operative delivery (group IIb) despite correction of uterine activity after oxytocin treatment (pre-oxytocin mean UAI 1230 (SD 570) kPas/15 min, post-oxytocin 1815 (SD 650) kPas/15 min). The rate of oxytocin infusion varied between patients from 2 to 16 mU/min but in 75% uterine activity was corrected to normal levels by a dose of less than or equal to 8 mU/min. It is concluded that delay in progress in labour is associated with lower intrauterine pressures than are present in normal labour, and that management of patients presenting with inefficient uterine action may benefit from the use of uterine activity integral measurements.

The SimulTRAC FT4/TSH assay evaluated as a first-line thyroid-function test

We evaluated the SimulTRAC FT4 57Co/TSH 125I dual-isotope assay for the simultaneous measurement of free thyroxin (FT4) by radioimmunoassay analog techniques and of thyrotropin (TSH) by immunoradiometry. Inter- and intra-assay CVs were less than 10% over the entire range tested except for 15.9% at the lowest FT4 concentration. Results obtained by the SimulTRAC assay allowed complete differentiation of 85 hyperthyroid patients and 35 hypothyroid patients from normal subjects. However, such estimations of FT4 or TSH concentrations occasionally were misleading for assessing thyroid status in various clinical conditions. We conclude that the SimulTRAC assay has the same inherent disadvantages possessed by FT4 analog and TSH immunoradiometric assays; however, where results of one of the simultaneous assays may be misleading, the results provided by the other may indicate the underlying pathology without requiring an additional assay.

The SimulTRAC FT4/TSH assay evaluated as a first-line thyroid-function test

We evaluated the SimulTRAC FT4 57Co/TSH 125I dual-isotope assay for the simultaneous measurement of free thyroxin (FT4) by radioimmunoassay analog techniques and of thyrotropin (TSH) by immunoradiometry. Inter- and intra-assay CVs were less than 10% over the entire range tested except for 15.9% at the lowest FT4 concentration. Results obtained by the SimulTRAC assay allowed complete differentiation of 85 hyperthyroid patients and 35 hypothyroid patients from normal subjects. However, such estimations of FT4 or TSH concentrations occasionally were misleading for assessing thyroid status in various clinical conditions. We conclude that the SimulTRAC assay has the same inherent disadvantages possessed by FT4 analog and TSH immunoradiometric assays; however, where results of one of the simultaneous assays may be misleading, the results provided by the other may indicate the underlying pathology without requiring an additional assay.

Impaired corpus luteum function in ectopic pregnancy cannot be explained by altered human chorionic gonadotropin

We studied the cause of the low serum progesterone, 17 beta-estradiol, and 17-hydroxyprogesterone levels that occur in women with an ectopic pregnancy. Only women who had been amenorrheic for less than 8 weeks were studied in order to assess corpus luteum rather than placental biosynthesis of these steroids; each woman with an ectopic pregnancy was matched to a woman with a normal intrauterine pregnancy on the basis of serum intact hCG levels within 10% of one another to obviate the influence of different levels of this luteotropic hormone. Every woman with an ectopic pregnancy had lower serum progesterone, estradiol, and 17-hydroxyprogesterone levels than her matched normal pregnant pairmate (median values: progesterone, 27.9 vs. 83.5 mmol/L; estradiol, 0.36 vs. 1.79 nmol/L; 17-hydroxyprogesterone, 4.95 vs. 22.1 nmol/L, respectively; all P less than 0.002). The ratios of intact hCG, measured by immunoradiometric assay, to hCG, measured by a hCG beta-specific RIA, were similar in the two groups. Serum hCG bioactivity was assayed by measuring the ability of serum to stimulate testosterone secretion from mouse Leydig cells. The mean biological to intact immunological hCG ratios were 2.06 +/- 1.39 (+/- SD) for ectopic pregnancy and 1.91 +/- 0.81 for normal pregnancy (P greater than 0.05). The biological hCG to immunoreactive hCG beta ratios were 1.98 +/- 0.75 and 2.02 +/- 0.82, respectively. Serum hCG from both groups of women stimulated cAMP generation by testicular cells similarly. We conclude that the lower serum steroid levels in women with ectopic pregnancy cannot be explained by altered hCG bioactivity. The lower steroid levels may thus reflect a primary defect of the corpus luteum, absence of another stimulator of ovarian steroid biosynthesis, or more subtle alterations in hCG glycosylation which are important in vivo but not assessed by the in vitro bioassay.

Endotoxaemia in exhausted runners after a long-distance race

The extent to which plasma endotoxin concentrations increased was measured in 89 randomly selected exhausted runners who required admission to the medical tent for treatment in the 1986 Comrades Marathon (89,4 km). Eighty-one per cent had concentrations above the upper limit of 0,1 ng/ml ('endotoxaemic'), including 2% above 1 ng/ml (the reported lethal level in humans), and only 19% had normal levels. There was a negative correlation between plasma endotoxin and plasma anti-endotoxin IgG concentration (P less than 0,025). Those runners completing the race in less than 8 hours had a significantly lower average endotoxin value than those taking longer than 8 hours (P less than 0,025). Also 80,6% of runners (58/72) with high plasma endotoxin values reported nausea, vomiting and/or diarrhoea, compared with 17,7% (3/17; P less than 0,001) with low endotoxin values. Elevated plasma endotoxin concentrations of 32 randomly selected endotoxaemic runners had returned to normal 1-3 weeks later, and most of them (25/32) had increased anti-endotoxin IgG concentrations (P less than 0,02). Fifty-nine runners randomly selected in a short run (21,1 km) 3 weeks after the 89,4 km run completed the race without problems and none showed any increase in endotoxin levels. Further studies in this field are warranted, especially the measurement of endotoxin and anti-endotoxin values from commencement of training to full fitness. It is possible that these measurements may prove useful as predictors of an athlete's or combat soldier's performance.

Blood or urine measurement of human chorionic gonadotropin for detection of ectopic pregnancy? A comparative study of quantitative and qualitative methods in both fluids

One hundred seventy-five patients with features of ectopic pregnancy were studied, of whom 95 were subsequently shown to have an ectopic pregnancy. Paired blood and urine samples were assayed for human chorionic gonadotropin (hCG) by two radioimmunoassays, one immunoradiometric assay for intact hCG and an immunoradiometric assay for free beta subunit that also detects the "beta core" of hCG in urine. Qualitative testing was also performed using the Tandem Icon method of immunoconcentration on a membrane. The quantitative results for intact hCG showed an approximately unitary relationship between concentrations in both fluids, with a close correlation (r = 0.84, gradient = 1.01). The qualitative tests for hCG in both serum and urine were positive in all patients with ectopic pregnancy. The Tandem Icon is equally useful in blood and urine, whereas quantitative assays are more reliable in the blood. Quantitation of urinary hCG is not recommended because of the variable dilution of the glycoprotein in this fluid.