The effects of insulin and insulin-like growth factors (IGFs) on ovarian androgen production were examined in ovarian stroma obtained from four women with hyperandrogenism and three women without hyperandrogenism. In incubations of stroma obtained from all four hyperandrogenic patients, insulin alone (500 ng/ml) significantly stimulated androstenedione and testosterone release. LH alone (25 ng/ml) significantly stimulated androstenedione release in incubations of stroma obtained from three of the four hyperandrogenic patients and testosterone release in incubations of stroma obtained from one of the four hyperandrogenic patients. In stromal incubations from three of the four hyperandrogenic patients, insulin alone (500 ng/ml) resulted in a significantly greater release of androstenedione and testosterone than did LH alone (25 ng/ml). Dihydrotestosterone was released in measurable quantities in incubations of stromal tissue obtained from three of the four hyperandrogenic women. In all three instances in which dihydrotestosterone was detectable, insulin alone (500 ng/ml), but not LH alone (25 ng/ml), significantly stimulated dihydrostestosterone release. Incubations of stroma obtained from three nonhyperandrogenic, normally cycling women demonstrated low levels of androstenedione release and negligible testosterone and dihydrotestosterone release. Insulin alone (500 ng/ml) and LH alone (25 ng/ml) produced no significant increase in androstenedione release. Insulin (500 ng/ml) plus LH (25 ng/ml) significantly stimulated androstenedione accumulation in stroma obtained from two of the nonhyperandrogenic women. One insulin dose-response experiment was performed using stromal tissue obtained from a hyperandrogenic woman. In this experiment, insulin, at a dose of 50 ng/ml, was as effective as insulin at a dose of 500 ng/ml in stimulating androstenedione and testosterone release. In addition to insulin, IGF-I/somatomedin C (50 ng/ml) stimulated androstenedione and testosterone release. Relaxin (1 microgram/ml) and multiplication-stimulating activity (50 ng/ml) did not stimulate androstenedione and testosterone release. These studies suggest that human ovarian stroma may be a target tissue for insulin and IGF-I, and that hyperinsulinemia may be an important factor contributing to ovarian hyperandrogenism.

Measurements of the fetal cisterna magna were obtained from a prospective sample of patients undergoing diagnostic obstetric ultrasound examinations. These normal measurements were then compared to a retrospective sample of ultrasound scans from fetuses with the diagnosis of trisomy 18 by amniocentesis. The fetal cisterna magna increases in size throughout pregnancy. The sample from fetuses with trisomy 18 was different and had a higher rate of small and large cisterna magnas. After the two samples were mixed, it was determined that a cut-off of 2.5 standard deviations for the fetal cisterna magna would give a sensitivity of 26.3%, specificity of 99.1%, positive predictive value of 50.0% and negative predictive value of 97.6% for trisomy 18. The probability that a fetal cisterna magna of abnormal size will predict trisomy 18 in an early second-trimester ultrasound examination with the 2.5 standard deviation cut-off was estimated at 0.019. A fetal cisterna magna that is of abnormal size appears to be of diagnostic value in identifying the fetus with trisomy 18, but its sensitivity is low.

No abstract

Leydig cell hyperplasia is a common histological finding in male infants of diabetic mothers. The functional correlates of this histological finding were investigated by measuring beta hCG, testosterone, androstenedione, dihydrotestosterone, and progesterone in mixed cord serum of male and female infants of diabetic mothers (n = 40) and normal mothers (n = 40) at term. Male and female infants of diabetic mothers had significantly higher cord serum beta hCG levels than male and female controls. Male infants of diabetic mothers had significantly higher cord serum testosterone concentrations than male controls, female controls, and female infants of diabetic mothers. Cord serum testosterone concentrations were similar in female infants of diabetic mothers and female controls. In the male infants of diabetic mothers, there was a significant positive correlation between beta hCG and testosterone (r = 0.64; P less than 0.01). There was no significant correlation between beta hCG and testosterone in the male controls (r = -0.15; P = NS). There was no significant difference in cord serum dihydrotestosterone in any group tested. Cord serum progesterone was significantly higher in the males than in the females. Cord serum androstenedione was lower in the infants of diabetic mothers than in the controls. This study suggests that the Leydig cell hyperplasia found in male infants of diabetic mothers is due, in part, to elevated concentrations of hCG and is accompanied by elevated testosterone concentrations in the fetal compartment.

No abstract