Low dose aspirin in hypertensive pregnant women: effect on pregnancy outcome and prostacyclin‐thromboxane balance in mother and newborn

Objective To study the effect of daily treatment with 50 mg of aspirin (ASA) on the hypertensive pregnancy complications and on the production prostacyclin (PGI2) and thromboxane A2 (TxA2) in high risk pregnant women and their infants. Design Placebo controlled prospective study. Setting Departments of Obstetrics and Gynaecology, University of Helsinki, University of Oulu and Central Hospital of Middle Finland, Finland. Subjects Two hundred and eight pregnant women with pre-existing hypertension or a history of severe preeclampsia in their previous pregnancy. Prostanoids were studied in a subgroup of 18 women. Interventions The women were randomised to receive ASA (50 mg/day, n= 103) or placebo (n= 105) from the mean of 15 weeks gestational age to delivery. The exacerbation of pre-existing hypertension or the appearance of hypertension in previously normotensive women, the appearance of proteinuria and fetal growth were the main end points, but some other clinical characteristics were also recorded. Urinary excretion of PGI2 and TxA2 metabolites by mothers and infants and their production in umbilical arteries in vitro were also studied. Results Two women (one in both groups) had miscarriages, and one pregnancy was terminated for fetal anencephaly (ASA group). In addition, seven women discontinued the treatment due to urticaria (two women in ASA group), increased activity of aspartate amino tranferase in serum (one woman in both groups), or increased bleeding time (one woman in ASA group, two women in placebo group), and one woman in the placebo group was lost from follow-up. Thus the end points could be assessed in 97 women taking ASA and 100 women taking placebo. ASA did not diminish the rate of the rise of blood pressure without (12 vs 14, respectively) or with proteinuria (9 vs 11), but fetal haemodynamic disturbances as assessed by Doppler equipment (1/44 vs 6/45 women studied, P= 0.05) and need for treatment in neonatal intensive care unit (10 vs 21, P= 0.04) were more rare in ASA group. ASA tended to increase the birthweight of the newborn (3348 ± 707 g vs 3170 ± 665 g, mean ± SD, P= 0.07), but two perinatal deaths occurred in ASA group. ASA prolonged the bleeding time of the mother (435 s, 210–998 s (geometric mean, range) vs 349 s, 210–690 s, P= 0.02), but caused no extra blood loss during delivery, nor affected neonatal hemostasis. In a subgroup of mothers (ASA, n = 10; placebo, n = 8), ASA inhibited more than 90% of platelet TxA2-production, and caused a 65 to 80% decrease in the urinary excretion of TxA2 metabolites, but no decrease in the urinary excretion of PGI2 metabolites. Conclusions ASA did not prevent the rise of maternal hypertension, but improved fetal haemodynamic performance and reduced the need of intensive neonatal care. It inhibited strongly maternal thromboxane A2 but not PGI2 production and thus shifted the balance between PGI2/TxA2 to the dominance of the vasodilatory, anti-aggregatory side.