- Top of page
Objectives To assess whether systematic screening with an uterine artery Doppler in low risk pregnant women followed by the prescription of low dose aspirin in cases with abnormal results reduced the incidence of intrauterine growth restriction and pre-eclampsia.
Design A multicentre randomised trial.
Population 3317 low risk pregnant women. In the Doppler group, the uterine artery Doppler was performed between 20 and 24 weeks. Women with abnormal results received 100 mg of aspirin daily until the 35th week.
Main outcome measures Intrauterine growth restriction was defined as birthweight below the tenth and the third centile according to gestational age. Pre-eclampsia was defined as hypertension associated with proteinuria > 0.5g/L.
Results Intrauterine growth restriction, whether defined by the third or tenth centile, did not differ significantly between the two groups (RR = 1.22 [0.73 - 2.04] and 1.18 [0.93 - 1.51] respectively). Screening with uterine artery Doppler did not affect birthweight or any of the criteria of perinatal morbidity. There was no effect on the incidence of pre-eclampsia (RR = 1.99 [0.97 - 4.09]) or hypertensive disorders. These results were the same for nulliparae and multiparae.
Conclusions There is no justification for screening with uterine artery Doppler in a low risk population, even if abnormal results are followed by aspirin treatment and increased prenatal surveillance. Future studies must assess predictive tests that can be performed early in pregnancy and can identify populations at very high risk of pre-eclampsia and intrauterine growth restriction. Only when all of these conditions are fulfilled, aspirin or other treatments may prove its efficacy.
- Top of page
Pre-eclampsia and intrauterine growth restriction are important causes of perinatal and maternal morbidity and mortality1. Low dose aspirin has shown benefits in very high risk populations, reducing the incidence of pre-eclampsia by roughly 70% and intrauterine growth restriction by 50%2,3. Aspirin does not, however, appear efficacious among low or moderate risk populations4–7, although most patients who develop pre-eclampsia or intrauterine growth restriction are either nulliparous or multiparous without any relevant history. It is therefore essential to find a way to screen women who could benefit from this preventive treatment.
The uteroplacental artery Doppler flow velocity waveform (UAD) performed between 20 and 24 weeks of gestation has a good predictive value for intrauterine growth restriction and pre-eclampsia in a general population, especially among nulliparae. It also has the advantage of being simple to perform, reproducible and non-invasive8–10. Three trials have tested aspirin administration against a placebo for women with abnormal UAD results and two observed beneficial effects11–13.
Nonetheless, aspirin is not the only intervention that is triggered by an abnormal UAD result. Such patients are probably managed as high risk, although we do not actually know the effects, if any, of increased prenatal surveillance in such cases. Finally, if we want to be able to generalise the results of a trial, it must be performed pragmatically, in conditions like those of daily practice, and not in leading departments with several specialists.
The objective of this randomised controlled multicenter trial was to assess whether systematic screening with a UAD in a low risk population followed by the prescription of low dose aspirin in cases with abnormal results reduced the incidence of intrauterine growth restriction and of pre-eclampsia.
- Top of page
All women who came for a routine antenatal visit before 24 weeks were asked to participate in the study. We excluded women who had indications for a UAD, including chronic hypertension, diabetes, a previous fetal death, intrauterine growth, hypertensive disorders of pregnancy (i.e. women known to be at high risk before 24 weeks did not enter the trial). Women with contraindications to aspirin were also excluded.
This multicenter randomised trial took place between April 1994 and December 1997 in 17 French maternity hospitals. The study was approved by the Ethics Committee of Seine Saint-Denis. We obtained the women's consent before including them in the study. They were randomly divided into two groups. The randomisation was stratified according to centre and parity (nulliparae or multiparae). A randomisation procedure using sealed envelopes was standard in each centre, carried out by the consultant immediately after verification of inclusion criteria. The randomisation numbers were established using tables of order four permutation. The randomisation procedure was verified by checking all unused envelopes at the end of the trial and confirming that envelopes had been used in ascending order.
2.1Management in each group
Women allocated to the control group did not have a UAD on the day of the second-trimester abdominal ultrasound examination. In the Doppler group, the UAD was performed between 20 and 24 weeks with the aid of colour Doppler. The ratio of diastole to systole (D/S) was calculated, and any notch was diagnosed visually. An abnormal result was defined as a diastole: systole < 35% or a notch on at least one of the uterine arteries, according to the definition of Uzan et al.14. Photographs were required. The patients’ obstetrician was to prescribe 100mg of aspirin daily until the 35th week to patients with abnormal results after verifying normal bleeding time, which was defined as less than 10 minutes.
No other particular recommendations were given to the obstetricians in either group. The women with abnormal Doppler results were treated according to the standard practices of their antenatal centre. Because all participating centres routinely used Doppler velocimetry for high risk pregnancies, we assumed that they had a management policy for cases with abnormal Doppler (i.e. this was conceived as a pragmatic trial15). In the control group, the obstetricians were allowed access to Doppler velocimetry if clinically indicated.
2.2Outcome criteria for assessing the Doppler's utility
The principal outcome was intrauterine growth restriction, defined as birthweight below the tenth and third centile according to gestational age16.
The other outcome was pre-eclampsia, defined as hypertension associated with proteinuria >0.5g/L. Gestational hypertension was defined by systolic pressure greater than 140 mmHg or diastolic pressure greater than 90 mmHg.
Episodes of uterine bleeding, oligohydramnios and suspected intrauterine growth restriction (both diagnosed by ultrasound), and abnormal prelabour cardiotocography (CTG) patterns were recorded.
Pregnancy care was evaluated by the number of antenatal consultations, days of antenatal hospitalisation, CTG measurements, and ultrasound and Doppler tests. We also collected the following perinatal outcomes: peri- and neonatal death, fetal distress defined by abnormal CTG patterns resulting in intervention (caesarean or instrumental delivery), the newborn's clinical state at birth (one and five-minute Apgar scores, neonatal resuscitation, neonatal transfer).
2.3Data collection and analysis
The number of women necessary to show a reduction of one-third in the incidence of intrauterine growth restriction (estimated at 8%, for a low risk population) was 1325 in each group (two-tailed test; α= 0.05, β= 0.20). This would also show a 60% reduction of pre-eclampsia (estimated at 2%, for a low risk population, half nulliparous). Such a study would not be large enough, however, to assess any differences between the two groups for perinatal neurological complications or mortality. We used Student's t test to compare means and a χ2 test to compare percentages. Results were presented using relative risks and their 95% confidence intervals17.
- Top of page
In this study 3317 women were randomly assigned to one of two groups. Shortly after randomisation, follow up ceased for 69 women (39 women in the Doppler group and 30 in the control group). For 48 (27 and 21, respectively), follow up ended when verification of admission criteria indicated that they should not have been randomised. Four women refused further participation after randomisation and six had miscarriages before ultrasound. Two centres stopped inclusions a few months after the beginning of the study and did not send the records of the 11 women they had included. For these reasons, data for these 69 women were not available.
There were 115 other women for whom follow up data were missing but inclusion information was available. The characteristics collected at inclusion were comparable in both groups and comparable to those of the women who remained in the study (data not reported). In the end, we were able to analyse the complete data of 3133 women, 1572 in the Doppler group and 1561 in the control group.
Table 1 summarises the general characteristics at inclusion of the women who initially entered the study and for whom we have available data. These characteristics are comparable for the two groups. The groups available for the final analysis (3133 women) were also comparable. The UAD was performed between 20 and 24 weeks in 90.4% of cases, with 3.0% performed before 20 weeks and 6.6% after 24 weeks (Table 1). The gestational age at which the ultrasound was performed was similar for both groups. Only 3.0% of women in the control group had a uterine Doppler, compared with 94.2% in the Doppler group (Fig. 1).
Table 1. General characteristics of patients. Values are given as n (%) or mean [SD]
| ||Doppler group (n= 1 633)||Control group (n= 1 615)|
|Mother's age (years)*||32.5 [5.7]||32.6 [5.8]|
|Geographic origin||n (%)||n (%)|
| France||1038 (69.8)||992 (67.2)|
| Southern Europe||84 (5.7)||87 (5.9)|
| French Overseas Territories||73 (4.9)||70 (4.7)|
| North Africa||121 (8.1)||151 (10.3)|
| Sub-Saharan Africa||83 (5.6)||81 (5.5)|
| Other||88 (5.9)||95 (6.4)|
|Working||955 (67.4)||951 (67.9)|
|Nulliparae||797 (48.8)||773 (47.9)|
|First pregnancy||536 (33.8)||505 (32.1)|
|Gestational age at randomisation (weeks)||19.5 [3.1]||19.6 [3.0]|
|Gestational age at ultrasound|
| Weeks of gestation||22.3 [1.3]||22.2 [1.4]|
| ≤ 19 weeks of gestation||46 (3.0)||62 (4.1)|
| 20-24 weeks of gestation||1386 (90.4)||1344 (89.6)|
| ≥ 25 weeks of gestation||101 (6.6)||94 (6.3)|
In the Doppler group, 232 women (15.7%) had an abnormal result: 11.2% (n= 166) had a diastole: systole < 35% for at least one uterine artery, and 10.5% (n= 153) had a notch visible on at least one uterine artery (Fig. 2). Aspirin was prescribed to 64.2% (n= 149) of these women: 14.1% (n= 21) did not comply with the treatment, and inadequate information is available for 4.0% (n= 6). Aspirin was prescribed for 81.3 % of patients with a bilateral notch or a bilateral D/S < 35%, but for only 55.3 % of women with other abnormal UAD results and 1.0 % with a normal UAD (P < 0.01). The mean gestational age at which aspirin treatment began was 24.5 ± 2.8 weeks.
3.1Effects of the Doppler
UAD screening did not lead to a significant increase in the number of ultrasound and Doppler examinations or to changes in any other aspects of pregnancy management. We found no increase in antenatal hospitalization in the Doppler group (Table 2), no difference in the incidence of post-randomisation complications (Table 2), and no differences in preterm delivery, onset of labour, or mode of delivery (Table 3).
Table 2. Pregnancy management and antenatal complications after randomisation. Values are given as n (%) or RR [95% CI]. IUGR = intrauterine growth restriction.
| ||Doppler group (n= 1 572) n (%)||Control group (n= 1 561) n (%)||RR [95% CI]|
|Prenatal consultations ≥9 *||345 (22.3)||340 (22.1)||1.00 [0.88 - 1.15]|
|Prenatal hospitalisation||408 (35.9)||419 (37.3)||0.96 [0.86 - 1.07]|
| ≥1 CTG before labour||1154 (74.9)||1127 (73.7)||1.02 [0.97 - 1.06]|
| 0||59 (4.0)||60 (4.1)||1|
| 1||762 (51.9)||735 (49.9)||1.00 [0.98 - 1.03]|
| ≥2||647 (44.1)||667 (46.0)||1.00 [0.97 - 1.03]|
| 0||803 (54.1)||765 (51.3)||1|
| 1||464 (31.3)||477 (32.0)||0.95 [0.86 - 1.06]|
| ≥2||217 (14.6)||250 (16.7)||0.86 [0.74 - 1.01]|
|Medications after randomisation|
|Betamimetics||191 (12.5)||191 (12.6)||0.99 [0.82 - 1.19]|
|Antihypertensives||23 (1.5)||31 (2.0)||0.74 [0.43 - 1.26]|
|Anticoagulants||8 (0.5)||8 (0.5)||0.99 [0.37 - 2.65]|
|Antenatal sick time for pregnancy-related reasons||337 (35.7)||348 (36.8)||0.97 [0.86 - 1.09]|
|Antenatal complications after randomisation|
|Haemorrhage||36 (2.3)||31 (2.0)||1.16 [0.72 - 1.86]|
|Oligohydramnios||51 (3.4)||45 (3.0)||1.14 [0.76 - 1.68]|
|Suspected IUGR||64 (4.1)||45 (2.9)||1.41 [0.97 - 2.06]|
Table 3. Labour and delivery. Values are given as n (%) unless otherwise shown.
| ||Doppler group (n= 1 572)||Control group (n= 1 561)||RR [95% CI]|
|Gestational age at delivery (weeks): mean (SD)||39.0 (2.0)||39.1 (2.0)|| |
|Gestational age at delivery (weeks)|
| ≤28||7 (0.4)||8 (0.5)||0.88 [0.32 - 2.41]|
| 29-33||19 (1.2)||9 (0.6)||2.01 [0.95 - 4.62]|
| 34-36||51 (3.3)||50 (3.2)||1.02 [0.69 - 1.50]|
| ≥37||1490 (95.1)||1491 (95.7)||1|
|Onset of labour|
| Elective caesarean||117 (7.5)||111 (7.2)||1.05 [0.82 - 1.35]|
| Induced labour||374 (24.1)||369 (23.8)||1.01 [0.89 - 1.15]|
| Spontaneous labour||1061 (68.4)||1070 (69.0)||0,99 [0.94 - 1.04]|
|Mode of delivery|
| Spontaneous vaginal delivery||1036 (67.6)||1036 (68.0)||0,99 [0.95 - 1.04]|
| Instrumental delivery||265 (17.3)||275(18.1)||0.96 [0.82 - 1.12]|
| Caesarean section||232 (15.1)||212 (13.9)||1.09 [0.91 - 1.29]|
| ||(n= 1562)||(n= 1548)|| |
|Caesarean before labour due to fetal distress||15 (1.0)||19 (1.2)||0.79 [0.40 - 1.54]|
|Intervention for fetal distress during labour||56 (3.7)||38 (2.5)||1.46 [0.97 - 2.19]|
|Induced labour for maternal indication||38 (2.5)||43 2.8)||0.88 [0.57 - 1.35]|
|Induced labour for fetal indication||35 (2.3)||40 (2.6)||0.87 [0.56 - 1.36]|
|Induced prematurity||16 (1.0)||24 (1.6)||0.66 [0.35 - 1.24]|
Intrauterine growth restriction, whether defined by the third or the tenth centile, did not differ significantly between the treatment groups (RR = 1.22 [0.73 - 2.04] and 1.18 [0.93 - 1.51] respectively). UAD screening did not affect birthweight or any of the criteria of perinatal morbidity (Table 4). There was no effect on the incidence of pre-eclampsia (RR = 1.99 [0.97 - 4.09]) or hypertensive disorders. These results were the same for nulliparas and multiparas (Tables 5 and 6).
Table 4. Perinatal and maternal outcomes.
| ||Doppler group (n= 1572) n (%)||Control group (n= 1561) n (%)||RR [95% CI]|
|Pre-eclampsia||22 (1.4)||11 (0.7)||1.99 [0.97 – 4.09]|
|Non proteinuric hypertension||61 (3.9)||70 (4.5)||0.87 [0.62 – 1.21]|
|Isolated proteinuria||32 (2.0)||23 (1.5)||1.38 [0.81 – 2.35]|
|Neonatal outcomea||(n= 1562)||(n= 1548)|| |
|Birthweight (g)||3282 ± 503||3316 ± 486||[-68.3; 1.25]b|
|IUGR (<3th)||32 (2.0)||26 (1.7)||1.22 [0.73 – 2.04]|
|IUGR (<10th)||130 (8.3)||109 (7.0)||1.18 [0.93 – 1.51]|
|Apgar score <7 at 1 minute||84 (5.4)||70 (4.5)||1.19 [0.87 – 1.62]|
|Apgar score <7 at 5 minutes||12 (0.8)||11 (0.7)||1.09 [0.48 – 2.45]|
|Neonatal resuscitation||127 (8.1)||134 (8.7)||0.94 [0.74 – 1.18]|
|Neonatal transfer||142 (9.1||121 (7.8)||1.16 [0.92 – 1.47]|
|Perinatal deaths||13||14||0.92 [0.44 – 1.96]|
|Perinatal deaths excluding lethal malformation||5||8||0.62 [0.20 – 1.90]|
|Stillbirth||10||13||0.76 [0.34 – 1.74]|
|Neonatal death||3||1||2.98 [0.31 – 28.6]|
Table 5. Perinatal and maternal outcomes amongst nullipara.
| ||Doppler group n (%)||Control group n (%)||RR [95% CI]|
|Pre-eclampsia||14 (1.9)||8 (1.1)||1.71 [0.72 - 4.06]|
|Arterial hypertension||32 (4.3)||39 (5.4)||0.80 [0.51 - 1.27]|
|Neonatal outcomea||(n= 739)||(n=717)|| |
|Birthweight (g)||3218 ± 532||3249 ± 493||[-83.4; 22.1]b|
|IUGR (<3th)||19 (2.6)||16 (2.2)||1.15 [0.60 - 2.22]|
|IUGR (<10th)||88 (11.9)||75 (10.5)||1.14 [0.85 - 1.52]|
|Apgar score <7 at 5 minutes||8 (1.1)||7 (1.0)||1.11 [0.41 - 3.05]|
|Neonatal resuscitation||80 (10.8)||88 (12.3)||0.88 [0.66 - 1.17]|
|Neonatal transfer||84 (11.4)||78 (10.9)||1.04 [0.78 - 1.40]|
|Perinatal deaths||5||8||0.61 [0.20 - 1.86]|
|Perinatal deaths excluding lethal malformation||1||5||0.20 [0.02 - 1.67]|
|Stillbirth||3||7||0.42 [0.11 - 1.61]|
|Neonatal death||2||1||1.95 [0.18 - 21.5]|
Table 6. Perinatal and maternal outcomes amongst multipara.
| ||Doppler group (n= 830) n (%)||Control group (n=837) n (%)||RR [95% CI]|
|Pre-eclampsia||8 (1.0||3 (0.4)||2.70 [0.72 - 10.1]|
|Arterial hypertension||29 (3.5)||31 (3.7)||0.94 [0.57 - 1.55]|
|Neonatal outcomea||(n= 823)||(n=831)|| |
|Birthweight (g)||3340 ± 469||3374 ± 472||[- 79.0; 11.70]b|
|IUGR (<3rd)||13 (1.6)||10 (1.2)||1.31 [0.58 - 2.98]|
|IUGR (<10th)||42 (5.1)||34 (4.1)||1.25 [0.80 - 1.94]|
|Apgar score <7 at 5 minutes||4 (0.5)||4 (0.5)||1.01 [0.26 - 4.04]|
|Neonatal resuscitation||47 (5.7||46 (5.5)||1.03 [0.69 - 1.53]|
|Neonatal transfer||58 (7.0)||43 (5.2)||1.36 [0.93 - 1.99]|
|Perinatal deaths||8||6||1.34 [0.47 - 3.86]|
|Perinatal deaths excluding lethal malformation||4||3||1.35 [0.30 – 5.99]|
|Stillbirth||7||6||1.18 [0.40 - 3.49]|
|Neonatal death||1||0|| |
Of the 13 infants who died in the Doppler group, there were six terminations of pregnancy for fetal malformation, three sillbirths (including one intrauterine growth restriction), two neonatal deaths due to multiple malformation syndrome, one neonatal deaths after placental abruption (without evidence of intrauterine growth restriction) and one neonatal death after delivery at 26 weeks (without evidence of intrauterine growth restriction).
Of the 14 infants who died in the control group, there were three terminations of pregnancy for fetal malformation, ten stillbirths (three fetal malformations, one severe intrauterine growth restriction, three intrapartum fetal deaths and three unexplained stillbirths without evidence of intrauterine growth restriction), and one neonatal death after delivery at 31 weeks (severe intrauterine growth restriction).
- Top of page
Our study has shown that the screening of a low risk population with UAD, followed by the prescription of low dose aspirin for patients with abnormal results, did not reduce the incidence of intrauterine growth restriction or pre-eclampsia. This was the case in both nulliparous and multiparous women.
It is very unlikely that the missing data affected the results, because of their paucity; and also because they were equally distributed, both numerically and by reasons, across both groups. Moreover, their general characteristics were comparable in both groups and comparable to those of the women who remained in the study.
We studied a very low risk population because all cases with an indication for UAD were excluded. Thus, the frequency of hypertension (4.3%) was lower than the prevalence in pregnant women in France (9.3% to 14.5%)18,19. Similarly, the perinatal mortality was 6.4‰ (20/3 144), compared with 7.4‰ in France in 199520.
We calculated our sample size assuming that the incidence of intrauterine growth restriction is 8% (< 10th centile) and expecting a reduction by one-third. This was consistent with the reported trials and a first meta-analysis of aspirin among high risk populations that showed a 50% reduction in intrauterine growth restriction2,3. We found intrauterine growth restriction in 7.0% of the control group and in 8.3% of the Doppler group. It is thus unlikely that a type II error is responsible for the lack of observed effect in this study. It is difficult to reach any definitive conclusion regarding the impact of parity. Nonetheless, the confidence intervals for the relative risks (Tables 5 and 6) are similar for the nulliparous and multiparous women.
Protocol compliance for UAD was very good: 94.2% of the women in the Doppler group had a UAD, compared with 3.0% in the control group. Protocol compliance concerning aspirin treatment was not as good: only 64.2 % of the patients with an abnormal UAD received aspirin prescriptions, and 14.1% of them did not comply correctly with the treatment. This level of compliance may have been influenced by the publication of the CLASP study and other similar reports that suggest little or no benefit from the use of aspirin in low or moderate risk populations4–7. Although protocol variations are difficult to avoid in pragmatic multicentre trials (16 centres, 45 participating physicians, 30 ultrasound examiners), non-compliance could partially explain the absence of benefits. Nonetheless, if the routine use of UAD were recommended in the general population, the prescription rate and patient compliance would probably not be very different than they were here. Finally, the compliance was higher for women with bilateral abnormal UAD results than other abnormal results, which indicates that the patients at the highest risk received the treatment.
The observation of an abnormal UAD did not result only in the prescription of aspirin; it was also followed by a different management. Women with abnormal UAD were hospitalised more often (50.3 vs 32.5 %, P< 0.01), had more CTG measurements (82.2 vs 74.4 %, P=0.01), Doppler examinations (≧ 2 Doppler. 24.8 vs 13.0 %, P< 0.01) and pregnancy-related sick leave (45.7 vs 34.3 %, P=0.01). These data make it clear that the abnormal results were considered in the overall management. Despite this, we found no evidence of a beneficial effect.
Two main hypotheses may explain the absence of an observed effect in this study. The first is that the UAD may lack predictive value. The predictive value of the UAD in our study is comparable to that reported in studies of low risk populations21. The risk of pre-eclampsia doubled in the case of an abnormal UAD (RR = 2.1 CI [0.76 - 5.8]) with a sensitivity of 28%. The risk of intrauterine growth restriction increased by 2.3 (CI [1.6 - 3.5]) with a sensitivity of 28%. Thus, even if aspirin halved the number of cases of pre-eclampsia and intrauterine growth restriction, there would be only a moderate 14% reduction in the number of these disorders for the overall population.
It is unlikely that changing the definition of an abnormal uterine Doppler would substantially change its predictive value22. The percentage of abnormal Doppler results reported was similar or a little higher in our study than in other large prospective studies in the same type of population at the same gestational age8,22,23. We cannot hope, however, to improve the test's sensitivity and increase the number of at-risk patients treated with aspirin simply by reducing the number of abnormal uterine Doppler results by using a stricter definition, even if the positive predictive value would then be higher. In addition, as we have seen, most of the patients with the most abnormal Doppler findings did receive aspirin treatment.
The second hypothesis involves the efficacy of aspirin for this indication. In the meta-analysis published in the Cochrane database, Knight et al.24 reported a 15% reduction in the risk of pre-eclampsia (32 trials with 29,331 women; relative risk (RR) 0.85, 95% CI (0.78, 0.92)) and a 14% reduction in baby deaths in the antiplatelet group (RR 0.86, (0.75, 0.98). Although the data should be interpreted with caution, the reduction appeared to be greatest for women at high risk and for those randomised before 20 weeks. It is thus possible that, in our study, the aspirin may lack of efficacy because of its belated administration and in part by the fact that this population was at too low a risk of pre-eclampsia, despite the UAD screening.
When the protocol was being developed, two trials had reported a reduction in the occurrence of pre-eclampsia when aspirin was prescribed after an abnormal UAD finding11,25. These results suggested that aspirin could be efficacious in women with an abnormal UAD. It should be noted that one of those trials was retracted by the Board of Editors of the British Journal of Obstetrics and Gynaecology because original data could not be retrieved during an audit26. The debate has remained open, with two subsequent trials reporting contradictory results12,13.
Other screening tests for pre-eclampsia and intrauterine growth restriction have been proposed, but they usually have a lower diagnostic value (various laboratory assays) or are less reproducible and sometimes more invasive (e.g. roll-over test, angiotensin II receptor density)27. Similarly, other preventive treatments for pre-eclampsia and intrauterine growth restriction have been proposed in recent years. Calcium supplementation might be beneficial for women at high risk of gestational hypertension and for those living in communities with low dietary calcium intake, but optimum dosage requires further investigation28. Antioxidants and nitric acid have shown promising preliminary results, but require further trials before their introduction into clinical practice29–31. When our trial protocol was written, we had available almost no data from randomised trials of these treatments.
We do not think that the efficacy of aspirin is challenged by our results or by the previous negative trials. The reasons for the absence of any observed effect may be related to the modality of prescription (dose and timing) and to the difficulty in identifying pregnancies at very high risk of pre-eclampsia and intrauterine growth restriction with the screening methods currently available.
- Top of page
There is no proof justifying the recommendation of a systematic UAD in a low risk population, even if abnormal results are followed by aspirin treatment and increase in prenatal surveillance. Future studies must assess predictive tests that can be performed early in pregnancy and can identify populations at very high risk of pre-eclampsia and intrauterine growth restriction. Only when all of these conditions are fulfilled aspirin or other treatments may prove its efficacy.
Investigators and participating centres
Hôpital Européen de Paris. “la roseraie”. Aubervilliers (Dr M.A. Rozan, Dr M.C. Auvray, Dr G. Giliberti). Centre Hospitalier Général d'Aulnay sous bois (Dr P. Caubel, Dr Y. Chitrit). Hôpital Privé de la Seine Saint-Denis. Le Blanc-Mesnil (Dr Y. Debuire , Dr G. Bajer, Dr M. Cannameras). Clinique de la ferme. Bobigny (Dr P. Gobert). Centre Hospitalier Universitaire Jean Verdier. Bondy (Pr M. Uzan, Dr L. Carbillon, Dr C. Largilière). Clinique Vauban. Livry Gargan (Dr G. Kamoun, Dr B. Consile, Dr S. Haber). Centre Hospitalier de Montfermeil (Dr J-F. Ropert, Dr F. Rideau). Centre Hospitalier de Montreuil (Dr S. Saint-Léger, Dr L. Monthe). Centre Hospitalier de Saint-Denis (Dr D. Rotten, Dr O. Bourdet, Dr G. Ekoukou). Clinique Hoffman. Rosny sous Bois (Dr G. Montefiore). Clinique du Vert Galant. Tremblay en France (Dr J. Mina). Centre Hospitalier Régional Universitaire de Caen. Caen (Dr P. Hamel). Centre Hospitalier de Périgueux. Périgueux (Dr G. Origet, Dr A. Darwiche). Centre Hospitalier de la Côte Basque. Bayonne (Dr P. Guerre, Dr C Poumier Chabanier). Centre Hospitalier Intercommunal de Poissy. Poissy (M.A. Soula, Dr M. Perdu, Dr F Goffinet). Maternité Port-Royal. Paris. (Dr G. Grangé, F. Razafindramiadala. J. Favreau).
D Aboulker,. J Paris-Llado, L Roux, F De Moegen, P Senanedy, C Bortuzzo, M Bucourt, F Goffinet,
Scientific committee M Uzan, E Papiernik, F Goffinet, G Bréart
Writing committee F Goffinet, G Bréart, D Aboulker, E Papiernik