Corticotropin-releasing factor-binding protein (CRF-BP) is suggested to play a role in modulating the activity of the hypothalamus-pituitary-adrenal axis during pregnancy, counteracting the actions of circulating or locally acting CRF. The aim of the present study was to evaluate whether maternal levels of CRF-BP and CRF are modified in pregnant women with a high risk of developing pregnancy-induced hypertension (n = 21). A group of nine patients developed the disease between 25-35 weeks gestation, and sequential blood samples were taken every 5 weeks throughout the pregnancy. As a control group, healthy pregnant women were studied (n = 9) using the same protocol; a group of women with pregnancy-induced hypertension (n = 5) was studied starting from the time of diagnosis. In a subgroup of patients (n = 10), CRF-BP and CRF levels were studied after 5 weeks of antihypertensive treatment. Levels of CRF-BP were determined using a specific RIA, whereas CRF was evaluated by a two-site immunoradiometric assay. In patients at risk, circulating levels of CRF-BP followed the same pattern as that in healthy controls, showing a significant decrease at term (36-40 weeks; P < 0.05). A significant and progressive increase in plasma CRF levels was observed in both groups of pregnant women; the highest values were found at term (P < 0.01). In the nine patients who developed pregnancy-induced hypertension, maternal levels of CRF-BP at the onset of signs and symptoms were lower than control values, and CRF levels were significantly higher at the onset of the disease (P < 0.01). Similarly, in these hypertensive patients studied at the time of hospitalization, CRF-BP levels were lower whereas those of CRF were higher than levels in healthy patients (P < 0.01). No effect of antihypertensive therapy on either CRF-BP or CRF levels was observed. The present study shows an inverse correlation between reduced plasma CRF-BP levels and increased CRF levels in the maternal circulation of patients with pregnancy-induced hypertension, and indicates that these hormonal changes do not occur before the onset of disease, suggesting that the measurement of these polypeptides in maternal plasma does not predict the development of hypertension.