Technical difficulties of selective venous blood sampling in the differential diagnosis of female hyperandrogenism.

To determine glandular steroid release of adrenals and ovaries in female hyperandrogenism, a standardized method for percutaneous transfemoral venous blood sampling was developed. In eight volunteers and 67 patients, catheterization was performed during the early follicular phase (days 3-7; between 8 and 10 a.m.) to reduce interference from cyclic and circadian variations of secretion. Serial samplings reduced the episodic effluent changes. Anatomical variations and collateral flow as well as stress effects and the dosage of contrast media were studied. During catheterization, peripheral cortisol levels did not differ significantly from control groups. Collaterals had no effect on hormone levels. Contrast media increased cortisol effluent levels only when they were sampled following venography. Four-vessel venous sampling was found to be indicated if peripheral testosterone was more than 1.5 ng/ml and/or dehydroepiandrosterone sulfate more than 6,700 ng/ml. If an ovarian (adrenal)/peripheral gradient of testosterone exceeded 2.7 ng/ml, surgical intervention for tumor removal at the site of hormone excess was felt to be necessary.