1552
Doppler ultrasound and aspirin in recognition and prevention of pregnancy-induced hypertension
Screening of 1226 nulliparous women by means of doppler uteroplacental flow-velocity waveforms in early pregnancy identified 148 (12%) as being at high risk of pregnancy-induced hypertension. After exclusions and refusals, 100 women were randomly allocated to groups receiving either low-dose aspirin (75 mg daily; 48 patients) or identical placebo (52 patients) for the remainder of the pregnancy. The difference between the aspirin and placebo groups in the frequency of pregnancyinduced hypertension (13% vs 25%) did not achieve significance, but there were significant differences in the frequencies of proteinuric hypertension (2% vs 19%) and hypertension occurring before 37 weeks’ gestation (0% vs 17%). Fewer aspirin-treated than placebo-treated women had low birthweight babies (15% vs 25%), but this difference was not significant. The only perinatal death in the aspirin group followed a cord accident during labour, whereas the 3 perinatal deaths in the placebo group were all due to severe hypertensive disease. No maternal or neonatal side-effects were observed in either group.
Introduction Prevention of hypertension in pregnancy is an important clinical objective; it requires both the ability to predict accurately a high-risk group and the existence of effective interventions. More than a hundred screening tests have been suggested for pregnancy-induced hypertension,l but they are not practical, are insensitive, or are most sensitive in late pregnancy. We have lately2 confirmed the findings of previous pilot studies3-5 that doppler flow-velocity waveforms from the uteroplacental circulation in early pregnancy are an effective means of recognising a group of women at high risk of hypertensive disorders. Much of the pathogenesis of pregnancy-induced hypertension is associated with an imbalance of the prostacyclin/thromboxane systems.6,7 Manipulation of these systems by means of low-dose aspirin appears promising,8-11 especially in reducing the more severe complications of hypertensive disease of pregnancy. We designed this study to determine whether low-dose aspirin is effective in preventing pregnancy-induced hypertension and its complications in a group of women identified as being at high risk on the basis of abnormal doppler
flow-velocity waveforms.
the study was given to the patients by the ultrasound receptionist, and witnessed written consent was obtained from those agreeing to take part. The study had the approval of the local hospital ethical committee. Patients underwent doppler ultrasound examination of the uteroplacental circulation as previously described.2,s Patients with abnormal waveforms2,12 had repeat examinations at 24 weeks’ gestation, and if the waveforms were still abnormal they were invited to take part in the trial. Study participants were randomly allocated to groups receiving 75 mg aspirin daily or an identical placebo by means of a computer-generated randomisation list. The tablets were dispensed by the pharmacist in serially numbered bottles containing 16 weeks’ supply (112 tablets). The bottles were labelled "Aspirin Study" and the pharmacist took no further part in the trial. Clear written instructions were given to patients to avoid aspirin-containing compounds and non-steroidal anti-inflammatory agents. Reasons for exclusion from the study included known aspirin allergy, maternal diabetes mellitus, bleeding disorders, peptic ulceration, and systemic lupus erythematosus. Compliance was checked by random tablet counting. Patients attended for routine antenatal care but the clinicians were unaware of their treatment. Each woman’s general practitioner was informed by letter. This study had two main aims-to determine whether low-dose aspirin could reduce the incidence and the complications of hypertensive disorders of pregnancy. Our preceding study2 suggested that pregnancy-induced hypertension would develop in 25% of patients with persistently abnormal uteroplacental waveforms. We thought a reduction to 5% would be clinically significant; Wallenburg and colleagues9 had shown that such a reduction was possible. To detect this difference at a significance level of 5%, we would have to recruit about 100 women. We estimated that about 25% of women would refuse to take drugs in pregnancy and since the frequency of persistently abnormal waveforms is about 10%2 we would have to screen just over 1000 nulliparous women. Wallenburg et a19 suggested that the frequency of proteinuric hypertension could be reduced from 30% to zero by means of low-dose aspirin. Our previous study found an incidence of proteinuric hypertension of 10%;2 we could achieve abolition of this complication with approximately the same numbers. The main outcome measures used were pregnancy-induced hypertension and the development of proteinuria. The defmitions2 were those recommended by the International Society for the Study of Hypertension in Pregnancy.13
Results 106 women enrolled in the trial recruited. 52 were allocated to the aspirin group and 54 to the placebo group. 5 patients moved house during pregnancy and outcome details were not available and 1 patient withdrew after 3 weeks, leaving 100 participants (48 aspirin, 52 placebo) who completed the study. The demographic features of the groups were similar (table r). No woman recruited to the study had a blood pressure of
The
figure shows how the
were
Patients and methods All
nulliparous
women
attending
St
George’s Hospital between
October, 1987, and January, 1989, for routine ultrasound examination were offered doppler studies. The study population did not overlap with that of our previous study. A written outline of
ADDRESS Fetal Welfare Laboratory, Department of Obstetrics and Gynaecology, St George’s Hospital Medical School, Cranmer Terrace, London SW17 0RE, UK (P McParland, MRCOG, J. M. Pearce, MD, Prof G. V. P Chamberlain, MD). Correspondence to Mr J. M. Pearce
1553
TABLE II-PREGNANCY OUTCOMES
*95% CL = 95% confidence limits for difference in proportions. NS= not significant.
TABLE III-CLINICAL DETAILS OF PERINATAL DEATHS
Recruitment of trial
participants.
140/90 mm Hg or higher during the first 20 weeks of pregnancy. Random tablet counting showed that 12 (25%) women took all their aspirin tablets and 14 (27%) all their placebo tablets. The median number of missed tablets was 2 in both groups with a range of 0-10 for the placebo group and 0-9 for the aspirin group. The incidence of pregnancy-induced hypertension in the placebo group (table II) was exactly the same as that reported in our previous study.2 Although it was greater than that in the aspirin group (25% vs 13%), the difference was not statistically significant (X2 =1.79). However, proteinuric hypertension was significantly more frequent in the placebo than in the aspirin group (19% vs 2%; X2 = 5-85, p < 0 02). Hypertension arising before 37 weeks’ gestation occurred in 17% of the placebo group but in none of the aspirin group There was no significant difference (X2=7.14, p
Doppler ultrasound and aspirin in recognition and prevention of pregnancy-induced hypertension
Screening of 1226 nulliparous women by means of doppler uteroplacental flow-velocity waveforms in early pregnancy identified 148 (12%) as being at high risk of pregnancy-induced hypertension. After exclusions and refusals, 100 women were randomly allocated to groups receiving either low-dose aspirin (75 mg daily; 48 patients) or identical placebo (52 patients) for the remainder of the pregnancy. The difference between the aspirin and placebo groups in the frequency of pregnancyinduced hypertension (13% vs 25%) did not achieve significance, but there were significant differences in the frequencies of proteinuric hypertension (2% vs 19%) and hypertension occurring before 37 weeks’ gestation (0% vs 17%). Fewer aspirin-treated than placebo-treated women had low birthweight babies (15% vs 25%), but this difference was not significant. The only perinatal death in the aspirin group followed a cord accident during labour, whereas the 3 perinatal deaths in the placebo group were all due to severe hypertensive disease. No maternal or neonatal side-effects were observed in either group.
Introduction Prevention of hypertension in pregnancy is an important clinical objective; it requires both the ability to predict accurately a high-risk group and the existence of effective interventions. More than a hundred screening tests have been suggested for pregnancy-induced hypertension,l but they are not practical, are insensitive, or are most sensitive in late pregnancy. We have lately2 confirmed the findings of previous pilot studies3-5 that doppler flow-velocity waveforms from the uteroplacental circulation in early pregnancy are an effective means of recognising a group of women at high risk of hypertensive disorders. Much of the pathogenesis of pregnancy-induced hypertension is associated with an imbalance of the prostacyclin/thromboxane systems.6,7 Manipulation of these systems by means of low-dose aspirin appears promising,8-11 especially in reducing the more severe complications of hypertensive disease of pregnancy. We designed this study to determine whether low-dose aspirin is effective in preventing pregnancy-induced hypertension and its complications in a group of women identified as being at high risk on the basis of abnormal doppler
flow-velocity waveforms.
the study was given to the patients by the ultrasound receptionist, and witnessed written consent was obtained from those agreeing to take part. The study had the approval of the local hospital ethical committee. Patients underwent doppler ultrasound examination of the uteroplacental circulation as previously described.2,s Patients with abnormal waveforms2,12 had repeat examinations at 24 weeks’ gestation, and if the waveforms were still abnormal they were invited to take part in the trial. Study participants were randomly allocated to groups receiving 75 mg aspirin daily or an identical placebo by means of a computer-generated randomisation list. The tablets were dispensed by the pharmacist in serially numbered bottles containing 16 weeks’ supply (112 tablets). The bottles were labelled "Aspirin Study" and the pharmacist took no further part in the trial. Clear written instructions were given to patients to avoid aspirin-containing compounds and non-steroidal anti-inflammatory agents. Reasons for exclusion from the study included known aspirin allergy, maternal diabetes mellitus, bleeding disorders, peptic ulceration, and systemic lupus erythematosus. Compliance was checked by random tablet counting. Patients attended for routine antenatal care but the clinicians were unaware of their treatment. Each woman’s general practitioner was informed by letter. This study had two main aims-to determine whether low-dose aspirin could reduce the incidence and the complications of hypertensive disorders of pregnancy. Our preceding study2 suggested that pregnancy-induced hypertension would develop in 25% of patients with persistently abnormal uteroplacental waveforms. We thought a reduction to 5% would be clinically significant; Wallenburg and colleagues9 had shown that such a reduction was possible. To detect this difference at a significance level of 5%, we would have to recruit about 100 women. We estimated that about 25% of women would refuse to take drugs in pregnancy and since the frequency of persistently abnormal waveforms is about 10%2 we would have to screen just over 1000 nulliparous women. Wallenburg et a19 suggested that the frequency of proteinuric hypertension could be reduced from 30% to zero by means of low-dose aspirin. Our previous study found an incidence of proteinuric hypertension of 10%;2 we could achieve abolition of this complication with approximately the same numbers. The main outcome measures used were pregnancy-induced hypertension and the development of proteinuria. The defmitions2 were those recommended by the International Society for the Study of Hypertension in Pregnancy.13
Results 106 women enrolled in the trial recruited. 52 were allocated to the aspirin group and 54 to the placebo group. 5 patients moved house during pregnancy and outcome details were not available and 1 patient withdrew after 3 weeks, leaving 100 participants (48 aspirin, 52 placebo) who completed the study. The demographic features of the groups were similar (table r). No woman recruited to the study had a blood pressure of
The
figure shows how the
were
Patients and methods All
nulliparous
women
attending
St
George’s Hospital between
October, 1987, and January, 1989, for routine ultrasound examination were offered doppler studies. The study population did not overlap with that of our previous study. A written outline of
ADDRESS Fetal Welfare Laboratory, Department of Obstetrics and Gynaecology, St George’s Hospital Medical School, Cranmer Terrace, London SW17 0RE, UK (P McParland, MRCOG, J. M. Pearce, MD, Prof G. V. P Chamberlain, MD). Correspondence to Mr J. M. Pearce
1553
TABLE II-PREGNANCY OUTCOMES
*95% CL = 95% confidence limits for difference in proportions. NS= not significant.
TABLE III-CLINICAL DETAILS OF PERINATAL DEATHS
Recruitment of trial
participants.
140/90 mm Hg or higher during the first 20 weeks of pregnancy. Random tablet counting showed that 12 (25%) women took all their aspirin tablets and 14 (27%) all their placebo tablets. The median number of missed tablets was 2 in both groups with a range of 0-10 for the placebo group and 0-9 for the aspirin group. The incidence of pregnancy-induced hypertension in the placebo group (table II) was exactly the same as that reported in our previous study.2 Although it was greater than that in the aspirin group (25% vs 13%), the difference was not statistically significant (X2 =1.79). However, proteinuric hypertension was significantly more frequent in the placebo than in the aspirin group (19% vs 2%; X2 = 5-85, p < 0 02). Hypertension arising before 37 weeks’ gestation occurred in 17% of the placebo group but in none of the aspirin group There was no significant difference (X2=7.14, p
