To determine the dose of acetylsalicylic acid (ASA), that inhibits the production of the vasoconstrictive, aggregatory thromboxane A2 while sparing the production of the vasodilatory antiaggregatory prostacyclin. A controlled study comparing the effects of three doses of ASA on the production of thromboxane A2 and prostacyclin. Seven pregnant hypertensive patients and five non-pregnant healthy women received 0.5, 1.0 and 2.0 mg/kg/day of ASA, each dose for 10-12 days, the treatment periods following each other immediately. Seven normotensive pregnant women served as controls and were given no ASA. Blood and urine samples were taken at baseline and after the treatment periods to determine serum thromboxane B2 and the urinary 2.3-dinor-6-ketoprostaglandin F1alpha and 11-dehydrothromboxaneB2, the major stable metabolites of prostacyclin and thromboxane A2, respectively. The urinary excretion of 11-dehydrothromboxaneB2 was significantly higher in both hypertensive (34.9+/-18.3 pg/micromol creatinine) and normotensive (39.3+/-14.4 pg/micromol creatinine) pregnant women than in non-pregnant women (14.8+/-6.4 pg/micromol creatinine). The urinary excretion of 2.3-dinor-6-ketoprostaglandinF1alpha was also higher in normotensive pregnant women (93.9+/-50.9 pg/micromol creatinine) than in non-pregnant women (18.2+/-11.3 pg/micromol creatinine). The excretion rate of 2.3-dinor-6-ketoprostaglandinF1alpha in hypertensive patients was lower than in normotensive pregnant women (44.7+/-24.2 pg/micromol creatinine). At baseline the urinary 2.3-dinor-6-ketoprostaglandin F1alpha/11-dehydrothromboxaneB2 ratio was almost the same in the hypertensive patients (1.6) and in the non-pregnant women (1.2). The ratio was 2.6 in normotensive pregnant women. In the hypertensive group, already the lowest dose of ASA inhibited urinary 11-dehydrothromboxaneB2 excretion significantly. Because none of the doses of ASA inhibited 2.3-dinor-6-ketoprostaglandinF1alpha production, the 2.3-dinor-6-ketoprostaglandinF1alpha/11-dehydrothromboxaneB2 ratio was shifted in the favor of prostacyclin at all dose levels. In the non-pregnant women, even the highest dose level of ASA failed to affect the ratio. In the dose range of 0.5-2.0 mg/kg/day, ASA has a favorable effect on the ratio of prostacyclin to thromboxane A2 in hypertensive pregnancies.