Low dose acetylsalicylic acid in prevention of pregnancy-induced hypertension and intrauterine growth retardation in women with bilateral uterine artery notches

Authors


*: Dr M. Vainio, Department of Obstetrics and Gynaecology, Hyvinkää Hospital, Sairaalankatu 1, FIN-05850 Hyvinkää, Finland

Abstract

Objective To evaluate the efficacy of low-dose acetylsalicylic acid in the prevention of pregnancy-induced hypertension and intrauterine growth retardation in high-risk pregnancies as determined by transvaginal Doppler ultrasound study of the uterine arteries at 12 to 14 weeks of gestation.

Design Randomised, double blind and placebo-controlled trial.

Setting The Department of Obstetrics and Gynaecology, Tampere University Hospital, Finland.

Population One hundred and twenty pregnant women considered to be at high risk of pre-eclampsia or intrauterine growth retardation were screened by transvaginal Doppler ultrasound at 12 to 14 weeks of gestation.

Methods Ninety pregnant women with bilateral notches in the uterine arteries were randomised to receive acetylsalicyclic acid 0.5mg/kg/day (n= 45) or placebo (n= 45) from 12 to 14 weeks of gestation.

Main outcome measures Hypertensive disorders of pregnancy and intrauterine growth retardation.

Results Forty-three women on acetylsalicyclic acid and 43 on placebo were successfully followed up. The use of acetylsalicyclic acid was associated with a statistically significant reduction in the incidence of pregnancy-induced hypertension (11.6%vs 37.2%, RR = 0.31, 95% CI 0.13–0.78) and pre-eclampsia (4.7%vs 23.3%, RR = 0.2, 95% CI 0.05–0.86). The incidence of hypertension before 37 weeks of pregnancy was also significantly reduced (2.3%vs 20.9%, RR = 0.22, 95% CI 0.05–0.97). The reduction in the incidence of intrauterine growth retardation (2.3%vs 7%) was not statistically significant. Acetylsalicyclic acid was not associated with excess risk of maternal or fetal bleeding.

Conclusion In women rated in Doppler velocimetry waveform analysis to be at high risk of pre-eclampsia, low-dose acetylsalicyclic acid reduces the incidence of pregnancy-induced hypertension and especially proteinuric pre-eclampsia.

Introduction

The pathophysiology of pre-eclampsia and, the resulting intrauterine growth retardation involves at least in part an imbalance between vasodilating and -constricting prostaglandins1,2. Low-dose acetylsalicylic acid therapy inhibits thromboxane more than prostacyclin production, and thereby protects against vasoconstriction and pathologic blood coagulation in the placenta3,4.

A number of small prospective randomised studies have suggested significant effects of low-dose acetylsalicyclic acid on the incidence of hypertensive disorders of pregnancy and of intrauterine growth retardation5–7. In the studies in question, acetylsalicyclic acid had been used only for specific indications. Subsequent large prospective multicentre studies have tested acetylsalicyclic acid for broader indications and failed to demonstrate clinical efficacy8–11.

Selection early in pregnancy of a truly high risk group of women from a low risk population may be problematic, while a further confounding factor is the timing. The late initiation of treatment reported in several studies is probably one reason for the negative or only partially positive results obtained, as placental implantation is essentially completed by 14 to 18 weeks of gestation12,13.

It has been suggested that uterine artery Doppler measurements at 12 to 16 weeks of gestation are useful in identifying women at risk of pre-eclampsia, and in particular pre-eclampsia complicated by the delivery of a baby small for gestational age14.

The present study was undertaken to establish whether low-dose acetylsalicyclic acid treatment commenced at 12 to 14 gestational weeks could prevent pre-eclampsia and intrauterine growth retardation in high risk pregnancies identified at an early stage by transvaginal Doppler ultrasound.

Subjects

This prospective randomised placebo-controlled longitudinal trial commenced in October 1997 in the Department of Obstetrics and Gynaecology of Tampere University Hospital, Finland. The Ethics Committee of Tampere University Hospital and the Ethics Committee of the City of Tampere approved the study. The last patient was randomised in November 1999.

Women at risk of pre-eclampsia or intrauterine growth retardation were recruited from the population of pregnant women routinely attending antenatal clinics in Tampere and its envirous. Anamnestic risk factors included a history of chronic hypertension, familial risk of pre-eclampsia (mother or sister), gestational diabetes, age <20 or >40 years, previous pre-eclampsia, previous intrauterine growth retardation, or previous intrauterine death. The exclusion criteria were gestational weeks <12 or >14, asthma, allergy to acetylsalicyclic acid, previous peptic ulcer, or the use of prostaglandin inhibitors within ten days before investigation. The women were asked to attend for transvaginal Doppler ultrasound investigation at 12 to 14 weeks of gestation. Before this examination they gave informed consent and were checked for inclusion and exclusion criteria. In cases where a constant bilateral diastolic notch was found in the uterine arteries those concerned were asked to participate in a randomised pacebo-controlled trial. A total of 120 women underwent Doppler ultrasound investigation and 90 were found to have bilateral notches. Forty-five were allocated to the acetylsalicyclic acid and forty-five to the placebo group. The randomisation took place in the Pharmacy of Tampere University Hospital, and the code was not broken until after the delivery of the last woman randomised. The daily dose of acetylsalicyclic acid was 0.5mg/kg, adjusted at a follow up visit if the weight of the woman exceeded the initial weight by at least 10%. The baseline characteristics of the study population are shown in Table 1. No significant differences were found between the acetylsalicyclic acid and placebo groups with regard to age, weight, parity, diastolic or systolic blood pressure, multiple pregnancy, previous intrauterine growth retardation, previous fetus mortus, chronic hypertension, diabetes, previous pregnancy-induced hypertension or previous pre-eclampsia.

Table 1.  Characteristics of the women at trial entry. Values are given as n (%) or mean [SD]. ASA = acetylsalicylic acid. IUGR = intrauterine growth retardation. BP = blood pressure; PIH = pregnancy-induced hypertension.
 ASA (n= 43)Placebo (n= 43)
Maternal age (years)30.6 (6.3)30.0 (5.9)
Maternal weight (kg)72.2 (2.5)72.4 (2.9)
Systolic BP (mmHg)131.0 [15.6]132.4 [16.4]
Diastolic BP (mmHg)83.8 [10.9]85.4 [12.1]
Primigravid15 (34.9)10 (23.3)
Multiparous27 (62.8)33 (76.7)
Previous PIH18 (41.9)27 (62.8)
Previous pre-eclampsia14 (32.6)21 (48.8)
Previous IUGR6 (14.0)10 (23.3)
Previous stillbirth3 (7.0)1 (2.3)
Multiple pregnancy01 (2.3)
Chronic hypertension16 (37.2)13 (30.2)
Diabetes04 (9.3)

In the acetylsalicyclic acid group five women were on medication for chronic hypertension and one was taken beta-blocker medication for migraine. In the placebo group four women were on medication for chronic hypertension, one of them had Sjögren's syndrome and one had chronic glomerulonephritis. One in the placebo group had thyroxin substitution for hypothyroidism and one had Sjögren's syndrome without hypertension.

The outcome was also documented in 29 of the 30 women without bilateral notches who were screened by transvaginal sonography but excluded from randomisation.

Methods

The women were examined twice during pregnancy, at gestational weeks 24 to 26 and 32 to 34, follow up entailing ultrasound assessment of fetal growth, amniotic fluid volume, umbilical and uterine artery waveform analysis, and non-stressed fetal cardiotocography. Maternal blood pressure and urinary protein excretion were measured at the follow up study visits, during hospitalisation and when women came to labour. Blood pressure was measured in sitting position after 15 minutes rest using an automatic cuff. The uterine artery waveform analysis checked for the presence or absence of bilateral (right and left waveform) notches. At 12 to 14 weeks of gestation the women were placed in the lithotomy position and transvaginal sonography was performed using an Aloka (Aloka ECHO Camera SSD-2000) real-time colour Doppler ultrasound system with a 5MHz vaginal transducer. At 24 to 26 and 32 to 34 weeks of gestation a 3.5MHx convex transducer was used and the women were in semirecumbent position. An angle of insonation of 60 degrees or less was used to obtain waveforms acceptable for analysis.

Pregnancy outcome

The main outcome measures were pregnancy-induced hypertension, pre-eclampsia and intrauterine growth retardation (Table 2), duration of pregnancy and birthweight (Table 3). Pregnancy-induced hypertension was defined as a sustained blood pressure increase to levels of 140mmHg systolic or 90mmHg diastolic after 20 weeks of gestation15. Chronic hypertension was defined as hypertension developing before 20 weeks of gestation15. Pre-eclampsia was defined as blood pressure changes as above and proteinuria (defined as >300mg/24h or ≥1+ dipstick in a random urine sample)8. Superimposed pre-eclampsia was defined as proteinuria developing during pregnancy in a woman with known chronic hypertension15. Delivery before 37 weeks of estimated gestation was defined as preterm. Intrauterine growth retardation was defined as growth under the 10th centile. The intrauterine growth retardation estimate was based on postnatal infant weights. At delivery the duration of pregnancy, maternal blood pressure and blood loss, maternal serum haematocrite and urine excretion of protein were registered. Birthweight, Apgar scores, the acid–base balance of umbilical artery blood, the serum thrombocytes and haematocrite were recorded in all offspring.

Table 2.  Effect of acetylsalicylic acid (ASA) on pregnancy induced hypertension (PIH), intrauterine growth retardation (IUGR) and pre-eclampsia. Values are given as n (%).
 ASA (n= 43)Placebo (n= 43)RR95 % CI
PIH5 (11.6)16 (37.2)0.31(0.13–0.78)
Pre-eclampsia2 (4.7)10 (23.3)0.20(0.05–0.86)
Hypertension <37 weeks2 (2.3)9 (20.9)0.22(0.05–0.97)
IUGR1 (2.3)3 (7.0)0.33(0.04–3.08)
Table 3.  Effects of ASA on delivery type, pregnancy duration, birth weight, bleeding and neonatal outcome. Values are given as n (%), mean [SD], median {lower and upper quartile} and RR (relative risks) 95 % CI.
 ASA (n= 43)PlaceboRR (95 % CI)P
Spontaneous delivery 25 (58.1) 22 (46.8)1.15 (0.75–1.78) 
Induced delivery 16 (37.2) 18 (41.9) 0.9 (0.58–1.41) 
Caesarean section  7 (16.3)  4 (9.3)1.32 (0.80–2.20) 
GA at delivery (weeks)39.5 [1.7]39.2 [2.0] 0.381
Birthweight (g)3462 [604]3553 [767] 0.539
Birthweight <2500g   3 (7.0)  4 (9.3)0.85 (0.35–2.05) 
1 min Apgar scores <5   1 (1.2)0  
5 min Apgar scores <5   00  
Neonatal haematocrite0.61 [0.07] 0.61 [0.07] 0.668
Neonatal thrombocytes271 [69.7]272 [71.2] 0.940
Neonatal cord pH  7.3 [0.1]  7.3 [0.1] 0.215
Blood loss at delivery (ml) 300 {250–450}350 {250–450} 0.305
Postpartum haematocrite0.33 [0.04]0.34 [0.03] 0.227

Statistics

The intended sample size of 88 was based on ‘a priori’ sample size calculation made by Medstadt 2.11 (Astra Gruppen, Copenhagen, Denmark). The incidence of pregnancy-induced hypertension in this highly selected group of pregnant women was supposed to be 20%. The power calculation was based on having 80% power to detect that the incidence of pregnancy-induced hypertension (20%) was almost totally reduced with acetylsalicyclic acid prophylaxis and type 1 error was assumed to be 5%. The most important factor limiting the sample size was the small amount of pregnant women that could be recruited in the time available.

Statistical analysis was made by SPSS 9.0 for Windows statistical software. Differences were assessed by independent samples t test (confidence interval 95%) or Mann–Whitney test for quantitative data. Qualitative data were assessed by Pearson's χ2 test or Fisher's exact test. P values of <0.05 were considered statistically significant.

Results

A total of 120 pregnant women considered to be at high risk of pregnancy-induced hypertension or intrauterine growth retardation were screened by transvaginal sonography at 12 to 14 weeks of gestation. Outcome data were obtained on 29 of those 30 women without bilateral notches who were excluded from the randomisation.

Ninety women with bilateral notches were randomised and forty-three in both groups were successfully followed up. Two women in the acetylsalicyclic acid group discontinued before the second visit; one had urticaria and the other had itching in her throat. In the placebo group one woman moved to another town after the second visit and a second gave no reason for dropping out after the first visit.

Fifteen women (34.9%) in the acetylsalicyclic acid group and eleven (25.6%) in the placebo group needed dose adjustment. The difference was not statistically significant (P= 0.348). The average increase in the daily dose of acetylsalicyclic acid was 4.3mg (2.6mg–9.2mg). The dose of acetylsalicyclic acid was adjusted at 24 to 26 weeks in nine cases and at 32 to 34 weeks in six. Three women needed adjustment at both visits. Three of those requiring an acetylsalicyclic acid dose adjustment had pregnancy-induced hypertension. There was no statistically significant difference in the rate of adverse outcome between those requiring dose adjustment and those not, 20% and 10.7%, respectively.

There were six women in either group with other medical conditions. One of those in the placebo group had pre-eclampsia and one in the acetylsalicyclic acid group intrauterine growth retardation. The outcome of the remainder was normal. The incidence of adverse outcome was no greater in this subgroup (16.7%) vs the total incidence of adverse outcome (25.6%) in the acetylsalicyclic acid and placebo groups.

Hypertensive disorders of pregnancy

Pregnancy-induced hypertension developed in five women (11.6%) allocated to acetylsalicyclic acid as compared with 16 (37.2%) of those receiving placebo (relative risk 0.31, 95% CI 0.13–0.78) (Table 2). Pregnancy-induced hypertension was proteinuric in two pregnancies (4.7%) on acetylsalicyclic acid and in 10 (23.3%) on placebo (RR = 0.20, 95% CI 0.05–0.86). The hypertension set in or was exacerbated before 37 gestational weeks in two women (2.3%) randomised to acetylsalicyclic acid and in nine (20.9%) randomised to placebo (RR = 0.22, 95% CI 0.05–0.97). The recurrence rate of previous pregnancy-induced hypertension was 22.2% in the acetylsalicyclic acid and 48.1% in the placebo group (RR = 0.47, 95% CI 0.19–1.20) and of pre-eclampsia 14.3% and 33.3% (RR = 0.48, 95% CI 0.13–1.75), respectively. In the acetylsalicyclic acid group there was one new pregnancy-induced hypertension and no new pre-eclampsia cases, in the placebo group three new cases of pregnancy-induced hypertension and two of pre-eclampsia; the differences between the groups are not statistically significant.

Four (13.8%) of the women excluded had pregnancy-induced hypertension and one of them (3.4%) had proteinuric pre-eclampsia. Three of those who had pregnancy-induced hypertension had unilateral uterine artery diastolic notching. The incidence in hypertensive disorders here was lower than in the placebo group but did not differ from the acetylsalicyclic acid group.

Birthweight and intrauterine growth retardation

There was no significant difference between the groups in the mean birthweight; 3462g (SD 604) with acetylsalicyclic acid and 3553g (SD 765) with placebo (Table 3). Intrauterine growth retardation occurred slightly more often in the placebo group (3/43 versus 1/43, respectively), but the difference was not statistically significant (Table 1). There was no significant difference between the acetylsalicyclic acid and the placebo groups in the rate of birthweights <2500g. The rate of recurrence of previous intrauterine growth retardation was 0% in the acetylsalicyclic acid group and 10% in the placebo group. Two women in the placebo group had pre-eclampsia and one had pregnancy-induced hypertension concomitant with intrauterine growth retardation. In the acetylsalicyclic acid group there was no growth restriction concomitant with pregnancy-induced hypertension. The difference between the groups is not statistically significant (P= 0.241). The mean birthweight, 3574g (SD 508), and the incidence of intrauterine growth retardation (3.4%) in the excluded group of women without bilateral notching did not differ significantly from those in the acetylsalicyclic acid or the placebo group.

Duration of pregnancy

There was no difference in the mean duration of pregnancy; it was 39.5 weeks (SD 1.7) among the women on acetylsalicyclic acid and 39.3 weeks (SD 1.8) among those on placebo.

Other maternal outcomes

There were no statistically significant differences between the acetylsalicyclic acid and placebo groups in the rate of induction of labour, spontaneous labour or caesarean section. In the acetylsalicyclic acid group 16 deliveries were induced; pregnancy-induced hypertension was the reason for induction in five, essential hypertension in four, intrauterine growth retardation in one pregnancy, other reasons being oligohydramnion, hepatosis, post-term pregnancy, fetal macrosomy, hydrocephalus of the fetus and premature rupture of membranes. In the placebo group there were 18 inductions: pregnancy-induced hypertension being the reason in nine pregnancies, pregnancy-induced hypertension and intrauterine growth retardation in three, essential hypertension in three, fetal macrosomia in two and post-term pregnancy in one. There were 25 spontaneous deliveries in the acetylsalicyclic acid group and 22 in the placebo group. The acetylsalicyclic acid group had seven and the placebo group four caesarean sections, two and three of them being elective, respectively.

There were also no significant differences in postpartum haemorrhage: 300ml (250–450) among women receiving acetylsalicyclic acid and 350ml (250–450) among those on placebo. The postpartum haematocrite was 0.33 and 0.34, respectively (Table 3).

Other outcomes of the newborn

There were no significant differences between the groups in one- or five-minute Apgar scores or in cord pH values. The postpartum haematocrite of the infants was 0.61 in the acetylsalicyclic acid and 0.61 in the placebo group, and the blood thrombocyte counts of the infants were 271 (SD 69.7) and 272 (SD 71.2) in acetylsalicyclic acid and placebo groups, respectively (Table 3). The differences were not statistically significant. Four neonates in the acetylsalicyclic acid group and six in the placebo group were treated after birth in the neonatal intensive care unit. There were no stillbirths or neonatal deaths in either group. One neonate without intrauterine growth retardation in the acetylsalicyclic acid group had hydrocephalus and meningomyelocele. This infant was included in the analysis. In the placebo group one neonate developed pulmonary hypertension. There was no evidence of cerebroventricular haemorrhage or other haemostatic abnormalities in neonates.

Data on uterine artery notching in Doppler studies

Data on 86 randomised women's uterine artery Doppler measurements were collected at gestation weeks 12 to 14, 24 to 26 and 32 to 32. The data from uterine artery Doppler recordings were also collected in those 29 excluded cases without bilateral notches at 12 to 14 weeks of gestation.

Bilateral notching evidenced sensitivity, specificity and positive and negative predictive values of 84%, 50%, 29.6% and 90.6%, respectively, in predicting pregnancy-induced hypertension and 83%, 45%, 27.4% and 92.8%, respectively, in predicting pre-eclampsia; in predicting intrauterine growth retardation the figures were 75%, 41%, 24% and 86.8%, respectively.

At 24 to 26 weeks of gestation 13 women (30.2%) in the acetylsalicyclic acid group and 15 (34.9%) in the placebo group had bilateral notching (P= 0.645). Three of them (23.1%) had an adverse outcome (two pre-eclampsia and one intrauterine growth retardation) in the acetylsalicyclic acid group and seven (46.7%) in the placebo group (seven had pregnancy-induced hypertension, two of them pre-eclampsia and two pre-eclampsia with intrauterine growth retardation), (P= 0.194).

At 32 to 34 weeks of gestation four women (9.3%) had bilateral notching in the acetylsalicyclic acid group and ten (23.3%) in the placebo group, the difference being not statistically significant, (P= 0.08). Two (50%) in the acetylsalicyclic acid group had pre-eclampsia and seven (70%) in the placebo group had pregnancy-induced hypertension; two cases were proteinuric pre-eclampsia and two pre-eclampsia with intrauterine growth retardation. The difference between the groups is not statistically significant (P= 0.455).

The rate of adverse outcome was 64.3% among those who had bilateral notching at 32 to 34 weeks of gestation and 18.5% among those without. The difference is statistically significant (P= 0.000). Fourteen women had bilateral notching persisting at the 32 to 34 weeks of gestation; nine of them had pregnancy-induced hypertension and six of them had pre-eclampsia. Of the 72 women without bilateral notching 12 had pregnancy-induced hypertension and six of them were pre-eclampsia. The difference between the groups is statistically significant (P= 0.000 and P= 0.01). In both groups there were two cases of intrauterine growth retardation. The difference is not statistically significant (P= 0.122).

Discussion

The main finding in this study was that low-dose acetylsalicyclic acid given to women at high risk of gestational hypertension significantly reduced the incidence of pregnancy-induced hypertension and especially proteinuric pre-eclampsia. Gestational hypertension without proteinuria is not associated with increased perinatal mortality and in many cases very likely constitutes a different disease from pre-eclampsia16–18. Another finding was that the incidence of hypertension before 37 gestational weeks was lower in the acetylsalicyclic acid-treated group than in the placebo-treated group; this is an important finding in that a woman with gestational hypertension diagnosed after 36 weeks of gestation has only some 10% risk of developing pre-eclampsia19.

There are considerable disparities among the results of randomised trials involving acetylsalicyclic acid treatment. The discrepancies have been attributed to differences in subjects enrolled into trials (low risk vs high risk), gestational age at randomisation, and the dose of acetylsalicyclic acid. In previous trials aspirin has been used at doses between 50 and 150mg/day for the prevention of pregnancy-induced hypertension and intrauterine growth retardation. The dose of acetylsalicyclic acid in this present study was much lower, 0.5mg/kg/day, and was adjusted at a control visit if a mother's weight had increased by at least 10% from baseline. There was no statistically significant difference in adverse outcomes between those who needed dose adjustment and those who did not. The dose of acetylsalicyclic acid selected was based on the results of our previous study, where in pregnant women with pre-eclampsia acetylsalicyclic acid at a dose range of 0.5–2.0mg/kg/day significantly inhibited the production of thromboxane but left prostacyclin production unaffected20. Patrignani and associates also concluded that a daily dose of 0.45mg/kg aspirin produced a cumulative and complete inhibition of platelet thromboxane without affecting prostacyclin production21.

In this present trial we sought to establish whether 0.5mg/kg/day is also an appropriate dose in clinical practice to prevent pregnancy-induced hypertension or intrauterine growth retardation. There was a trend in both the CLASP8 and EPREDA22 trials suggesting that low-dose acetylsalicyclic acid may be more effective if commenced before 20 weeks of gestation, and targeted at a group of women likely to develop complications requiring preterm delivery23. In the present study, treatment commenced at 12 to 14 gestational weeks. The importance of gestational age is obvious, since placental implantation is essentially completed by 14 to 18 weeks of gestation13. The late commencement of treatment reported in several publications is probably one reason for the negative or only partially positive results observed11.

The first articles recommending acetylsalicyclic acid treatment for the prevention of pre-eclampsia, intrauterine growth retardation or both, all stressed that their results concerned only patients presenting with a specific indication5–11. Subsequent trials tested acetylsalicyclic acid for broader indications and reported negative results8–11. In this present series treatment focused on a high risk group of pregnant women with abnormal Doppler flow in uterine arteries, defined as bilateral notches at 12–14 weeks of gestation. Harrington et al.14 reported that there are differences in uterine and umbilical artery Doppler blood flow indices at 12 to 16 weeks in pregnancies with normal or complicated outcome. Bilateral notching was associated with a 22-fold increase in the odds of developing pre-eclampsia. If bilateral notches alone were used to predict pre-eclampsia, the test achieved a sensitivity of 93% but a specificity of only 69%. The problem with most Doppler screening studies has been the relatively low positive predictive value of abnormal waveforms. As gestational age advances the specificity and positive predictive values increase significantly, but sensitivity decreases24.

The results of the present study would also indicate that the definition of bilateral notches of uterine arteries at 12 to 14 weeks of gestation is a sensitive screening test for pregnancy-induced hypertension but has rater low specificity.

The relatively high rate of pregnancy-induced hypertension (37.2%) and pre-eclampsia (23.3%) in the placebo group confirms the high risk nature of the pregnancies here. The incidence of pregnancy-induced hypertension (13.8%) and pre-eclampsia (3.4%) in non-randomised women without bilateral notches was much lower and the difference between these women and the placebo group was statistically significant (P= 0.03 and P= 0.022). The incidence of pregnancy-induced hypertension and proteinuric pre-eclampsia did not differ significantly between women in the acetylsalicyclic acid group and the women without bilateral notches who were excluded. This result suggests the protective effect of acetylsalicyclic acid in reducing the rate of pregnancy-induced hypertension.

All the women in this study receiving acetylsalicyclic acid or placebo delivered at term and there were no pregnancies with early severe pre-eclampsia or intrauterine growth retardation necessitating delivery before 32 weeks of gestation. The size of the study population was based on calculation of the reductive effect of acetylsalicyclic acid on the incidence of pregnancy-induced hypertension, which was the primary endpoint of the study. The population should have been much larger to bring out any significant reducing effect of acetylsalicyclic acid on very early severe pre-eclampsia; in practice this would probably require a large multicentre study. The reduction in the incidence of intrauterine growth retardation in the acetylsalicyclic acid group was not significant when compared with the placebo group. The small number of women recruited into the study restricted the possibility of conclusions as to the prophylactic effect of acetylsalicylic acid on intrauterine growth retardation. There were only four pregnancies with intrauterine growth retardation defined as growth under the 10th centile; one in the acetylsalicyclic acid group and three in the placebo group.

There were no significant differences between two groups in uterine artery Doppler measurements at the control visits. At 32 to 34 weeks of gestation persisting notching in uterine arteries were seen in 9.3% in the acetylsalicyclic acid group and in 23.2% in the placebo group. Persistent bilateral notches in high risk pregnancies selected a group with 50% and 70% incidence of pregnancy-induced hypertension in the acetylsalicyclic acid and placebo group, respectively.

The use of acetylsalicyclic acid during pregnancy is potentially associated with maternal or fetal haemostatic problems. Fortunately, no such adverse effects were noted during the present study. There were no differences in one- or five-minute Apgar scores, in umbilical arterial pH levels, or postpartum haematocrite in neonates. There were no pregnancy losses in either group, acetylsalicyclic acid treatment did not increase maternal bleeding or lower maternal haematocrite.

Conclusion

Acetylsalicylic acid treatment reduces the risk of pregnancy-induced hypertension in general, and in particular proteinuric pre-eclampsia if the treatment is initiated early, at 12 to 14 weeks of gestation in high risk women. Transvaginal Doppler ultrasound of the uterine arteries appears to be a useful screening method for the selection of pregnancies involving a high risk of hypertensive disorders.

Acknowledgements

The study was supported by the Medical Research Fund of Tampere University Hospital.

Ancillary