The renal and systemic metabolites (the latter as 2,3-dinor derivatives) of prostacyclin and thromboxane A2 were measured, along with renal prostaglandin E2 and kallikrein, in the urine of 15 patients with pregnancy-induced hypertension, 15 normotensive pregnant women matched for both age and gestational age, and 15 normotensive nonpregnant control women. Urinary excretion of all prostaglandin and thromboxane metabolites studied proved significantly higher in normotensive pregnant women than in controls. Prostaglandin E2, 6-keto-prostaglandin F1 alpha, and 2,3-dinor-6-keto-prostaglandin F1 alpha were significantly lower in pregnancy-induced hypertensive women than in normotensive pregnant women, whereas thromboxane B2 and 2,3-dinor-thromboxane B2 showed no significant differences in the two groups. A significant negative correlation (r = -0.636, p less than 0.01) was found between urinary 2,3-dinor-6-keto-prostaglandin F1 alpha and mean blood pressure in the two groups of pregnant women taken as a whole. These data indicate that, in pregnancy-induced hypertension, there is an imbalance between vasodilator and vasoconstrictor factors, not only in the kidneys, but also at the systemic vascular level. This imbalance, which may in itself produce vasoconstriction, may also potentiate the hypertensive effect of catecholamines and angiotensin II.