OBSTETRICS

OBSTETRICS

Acetylsalicylic Acid for the Prevention of

Preeclampsia and Intra-uterine Growth
Restriction in Women with Abnormal Uterine
Artery Doppler: A Systematic Review and
Meta-analysis
Emmanuel Bujold, MD, MSc, FRCSC,1,2,3 Anne-Maude Morency, MD,4 Stéphanie Roberge, BSc,2
Yves Lacasse, MD, MSc, FRCPC,5 Jean-Claude Forest, MD, PhD, FRCPC,3,6
Yves Giguère, MD, PhD, FRCPC3,6
1
Département d’obstétrique-gynécologie, Faculté de médecine, Université Laval, Quebec QC
2
Département de médecine sociale et préventive, Faculté de médecine, Université Laval, Quebec QC
3
Centre de recherche, Centre hospitalier universitaire de Québec (CRCHUQ), Quebec QC
4
Département d’obstétrique-gynécologie, Faculté de médecine, Université McGill, Montreal QC
5
Centre de recherche, Hôpital Laval, Institut universitaire de cardiologie et de pneumologie, Université Laval, Quebec QC
6
Département de biologie médicale, Faculté de médecine, Université Laval, Quebec QC

Abstract (RR) with the Mantel-Haenszel method and 95% confidence

intervals.
Background: Preeclampsia is a major global cause of maternal,
neonatal and perinatal mortality. From studies of placental Results: Nine randomized controlled trials with a total of 1317
pathophysiology in women with preeclampsia, a potentially women met the inclusion criteria. ASA treatment beginning in early
important role of low-dose acetylsalicylic acid (ASA) in the gestation was associated with a greater reduction in the incidence
prevention of preeclampsia was expected, but the results from of preeclampsia than treatment beginning in late gestation: ASA
clinical trials have been disappointing. While recent evidence has treatment started at £ 16 weeks’ gestation resulted in RR 0.48
shown that uterine Doppler can predict preeclampsia as early as (95% CI 0.33 to 0.68), at 17–19 weeks RR 0.55 (95% CI 0.17 to
in the first trimester of pregnancy, most clinical trials have 1.76), and at ³ 20 weeks RR 0.82 (95% CI 0.62 to 1.09). ASA
evaluated ASA in the second and third trimesters. treatment started before 16 weeks was also linked with a
Objectives: We performed a meta-analysis to assess the influence significant reduction in the incidence of severe preeclampsia (RR
of gestational age at the time of introduction of ASA on the 0.10; 95% CI 0.01 to 0.74), gestational hypertension (RR 0.31;
incidence of preeclampsia in women at increased risk, on the 95% CI 0.13 to 0.78) and IUGR (RR 0.51; 95% CI 0.28 to 0.92).
basis of abnormal uterine artery Doppler.
Conclusion: ASA treatment initiated early in pregnancy is an
Methods: Computerized searches of randomized controlled trials efficient method of reducing the incidence of preeclampsia and its
were conducted to retrieve studies in which pregnant women at consequences in women with ultrasonographic evidence of
increased risk of preeclampsia had been identified on the basis of abnormal placentation diagnosed by uterine artery Doppler
abnormal uterine Doppler measurements. The trials compared studies.
women who received ASA with a control group. The primary
outcome was preeclampsia. Secondary outcomes included severe
preeclampsia, gestational hypertension, preterm birth, intrauterine Résumé
growth restriction, placental abruption, birth weight and gestational
Contexte : La prééclampsie est une cause globale majeure de
age at delivery. Statistical analyses used fixed effects of risk ratio
mortalité maternelle, néonatale et périnatale. On s’attendait à ce
que les études sur la pathophysiologie placentaire chez les
femmes présentant une prééclampsie indiquent que l’acide
Key Words: Preeclampsia, Doppler, intra-uterine growth restriction, acétylsalicylique (AAS) à faible dose joue un rôle potentiellement
acetylsalicylic acid, systematic review, meta-analysis, pregnancy, important dans la prévention de la prééclampsie, mais les
placenta résultats issus des essais cliniques se sont avérés décevants.
Competing Interests: None declared. Bien que de récentes données aient indiqué que la tenue d’un
Doppler utérin pouvait prédire la survenue d’une prééclampsie dès
Received on May 7, 2009 le premier trimestre de la grossesse, la plupart des essais
cliniques ont évalué l’AAS au cours des deuxième et troisième
Accepted on June 9, 2009
trimestres.

818 l SEPTEMBER JOGC SEPTEMBRE 2009

 Acetylsalicylic Acid for the Prevention of Preeclampsia and Intra-uterine Growth Restriction

Objectifs : Nous avons mené une méta-analyse afin d’évaluer Failure of physiological spiral artery transformation in the
l’influence de l’âge gestationnel au moment de la mise en œuvre
du traitement à l’AAS sur l’incidence de la prééclampsie chez les myometrium is now recognized as a hallmark of defective
femmes exposées à des risques accrus, en fonction de l’obtention placentation and can be found in the great majority of
de résultats anormaux à la suite d’un Doppler de l’artère utérine.
women with early onset preeclampsia.9 This failure of phys-
Méthodes : Des recherches informatisées visant les essais
comparatifs randomisés ont été menées en vue d’extraire les iological transformation is associated with resistance in
études dans le cadre desquelles les femmes enceintes courant un utero-placental blood flow that is characterized by abnor-
risque accru de prééclampsie avaient été identifiées en fonction
de l’obtention de mesures anormales à la suite d’un Doppler mal uterine artery velocity when evaluated by Doppler
utérin. Les essais comparaient les femmes recevant de l’AAS à un ultrasonography. Indeed, recent studies have shown that
groupe témoin. La prééclampsie constituait le critère d’évaluation
principal. Parmi les critères d’évaluation secondaires, on trouvait
abnormal uterine artery Doppler as early as the first trimes-
la prééclampsie grave, l’hypertension gestationnelle, ter can identify women at high risk of preeclampsia and
l’accouchement préterme, le retard de croissance intra-utérin, le IUGR.10,11 However, without the demonstration of clinical
décollement placentaire, le poids de naissance et l’âge
gestationnel au moment de l’accouchement. Les analyses benefit, it is not standard care to perform a uterine artery
statistiques faisaient appel à un modèle à effets fixes et à un Doppler study during pregnancy.
risque relatif (RR), conjointement avec la méthode
Mantel-Haenszel et des intervalles de confiance à 95 %. Acetylsalicylic acid (ASA) therapy has been implicated in
Résultats : Neuf essais comparatifs randomisés (comptant, en tout, the physiological transformation of spiral uterine arteries: it
1 317 participantes) ont répondu aux critères d’inclusion. Le
traitement à l’AAS entamé aux débuts de la gestation a été inhibits thromboxane-mediated vasoconstriction more
associé à une plus grande baisse de l’incidence de la than prostacyclin-mediated vasodilatation, and may thereby
prééclampsie que le traitement entamé aux derniers moments de offer protection against vasoconstriction and pathological
la gestation : le traitement à l’AAS entamé à £ 16 semaines de
gestation a donné lieu à un RR de 0,48 (IC à 95 %, 0,33 à 0,68); à blood coagulation in the placenta. Its use was expected to
17–19 semaines, à un RR de 0,55 (IC à 95 %, 0,17 à 1,76); et à prevent failure of physiological spiral artery transformation
³ 20 semaines, à un RR de 0,82 (IC à 95 %, 0,62 à 1,09). Le as well as the development of preeclampsia. However, sev-
e
traitement à l’AAS entamé avant la 16 semaine de gestation a
également été lié à une baisse significative de l’incidence de la eral large, prospective, multicentre studies conducted to
prééclampsie grave (RR, 0,10; IC à 95 %, 0,01 à 0,74), de establish whether or not ASA use should be a standard of
l’hypertension gestationnelle (RR, 0,31; IC à 95 %, 0,13 à 0,78) care have failed to demonstrate the clinical efficacy of ASA
et du RCIU (RR, 0,51; IC à 95 %, 0,28 à 0,92).
in preventing preeclampsia.8,9 On the other hand, the late
Conclusion : Le traitement à l’AAS entamé aux débuts de la
grossesse constitue une méthode efficace de réduire l’incidence initiation of treatment (after 18 to 20 weeks) reported in
de la prééclampsie et d’atténuer les conséquences de cette most trials, and the inclusion of low-risk patients in some,
dernière chez les femmes ayant obtenu des résultats were potential reasons for the negative or weakly positive
échographiques indiquant une placentation anormale
diagnostiquée au moyen d’études Doppler de l’artère utérine. results obtained. A meta-analysis published in 2001 sug-
J Obstet Gynaecol Can 2009;31(9):818–826 gested that ASA therapy in women with abnormal uterine
artery Doppler studies was accompanied by a significant but
INTRODUCTION modest reduction in rates of preeclampsia but not a
decrease in IUGR.12 In this meta-analysis there was no
reeclampsia is a leading cause of maternal mortality. The
P World Health Organization estimates that, worldwide,
stratification of subjects by gestational age at the beginning
of ASA prophylactic treatment.12 Therefore, we aimed to
over 100 000 women die from preeclampsia each year, and assess the influence of gestational age at the introduction of
the condition continues to be responsible for maternal ASA therapy on rates of preeclampsia and IUGR in women
deaths in developed countries.1 The prevalence of the con- identified (by abnormal uterine artery Doppler studies
dition varies between 5% and 10%, but could be as high as during pregnancy) as being at high risk of preeclampsia.
18% in developing countries.2,3 Moreover, preeclampsia is
associated with high perinatal mortality and morbidity, METHODS
including intrauterine growth restriction (IUGR) and
We performed a systematic review and meta-analysis of ran-
prematurity.4 Preeclampsia and IUGR are characterized by domized trials of ASA therapy in preventing preeclampsia.
abnormal placental implantation and inadequate utero- To identify all the trials published between 1965 and 2008,
placental blood flow. Normal placentation comprises we searched Medline, PubMed, Embase, and the Cochrane
trophoblast cell invasion of the decidual and myometrial Library with the following key words: “aspirin,”
segments of spiral arteries, which results in reversible “antiplatelet,” “salisyl,” “acetylsalicylic,” “platelet aggrega-
changes in the normal arterial wall architecture.5 tion inhibitors,” “ultrasonography,” “ultrasound,” and
Trophoblastic invasion develops from eight weeks’ gesta- “Doppler.” Trials published in any language were consid-
tion, with most of the physiological transformation of spiral ered for inclusion. The complete search strategy is available
arteries being completed between 16 and 20 weeks.6–8 on request.

SEPTEMBER JOGC SEPTEMBRE 2009 l 819

OBSTETRICS

Table 1. Quality of studies included in the review

Study Method of randomization Blinding Intention to treat Lost to follow-up (%)

19
Ebrashy A, 2005 Computer-generated list of random numbers None Yes 2.2
Zimmermann P, 199726 Randomized, no other information None Yes Unclear
Vainio M, 200220 Randomization in pharmacy Double Unclear 4.4
McParland P, 199024 Computer-generated randomization list Double Yes 5.7
Yu CKH, 200325 Computer-generated random number lists None Yes 1.1
18
Ferrier C, 1996 Randomized, no more details Double Unclear Unclear
Harrington K, 200023 Randomization: pre-prepared, sealed None No 2.8
envelopes, no further details
Morris JM, 199621 Randomization by taking the next in a series of Double Yes 1.9
numbered, identical blister packs
Bower SJ, 199622 By telephoning a central computerized Double Yes 4.8
randomization service

We selected randomized controlled trials that included potentially relevant or irrelevant. All potentially relevant
women identified, by abnormal uterine artery Doppler reports were independently examined in detail by two other
examination, to be at high risk of preeclampsia. The defini- investigators and categorized as relevant, possibly relevant,
tion of abnormal uterine artery Doppler was unrestricted. or irrelevant on the basis of our research criteria. Any dis-
Treatment consisted of low-dose ASA (50 to 150 mg/day), agreement was resolved by consensus. Quality and integrity
with the control group receiving either a placebo or no of the studies were assessed and reported by evaluating
treatment (i.e., usual care). Studies were excluded if more quality criteria for each study (method of randomization,
than 20% of women were lost during follow-up or excluded blinding, intention-to-treat, lost to follow-up) (Table 1). All
from the analyses. The primary outcome was the develop- possibly relevant references cited by two recent meta-
ment of preeclampsia according to gestational age at the analyses were reviewed for validation of the literature
initiation of treatment. Preeclampsia was defined by the search to confirm that no trial had been overlooked.12,14
combination of systolic blood pressure ³ 140 mmHg
and/or diastolic blood pressure ³ 90 mmHg, with ³ 300 g The analyses were performed by Review Manager 5.0.12
of urinary protein /24 h (or 1+ or more on dipstick). (RevMan. 5.0. ed. Copenhagen). Risk ratios (RR) from indi-
Secondary outcomes included IUGR and mean birth vidual studies were combined according to a fixed effect
weight. We defined IUGR as a birth weight < 10th percen- model, using the Mantel-Haenszel method. The data were
tile for gestational age. If this definition of IUGR was stratified according to gestational age at entry (beginning of
unavailable, the following definitions were also accepted: ASA therapy) in the trial (£ 16, 17–19, ³ 20 weeks’ gesta-
birth weight < 5th percentile for gestational age or < 2500 g. tion). These thresholds in gestational age were determined a
Severe preeclampsia was defined by any one of the following priori on the basis of the physiological evolution of spiral
criteria: severe hypertension (blood pressure of ³ 160 mmHg uterine artery transformation during pregnancy that usually
systolic, or 110 mmHg diastolic), severe proteinuria (³ 2, 3, ends between 16 and 20 weeks’ gestation.8,15 The analyses
or 5 g of protein in 24 h, or 3+ on dipstick), reduced urinary were also adapted according to type of variable (dichoto-
volume (< 400 to 500 mL in 24 h), neurological distur- mous vs. continuous). Each study was weighted with stan-
bances, such as headache and visual perturbations, upper dard deviation for both dichotomous and continuous data.
abdominal pain, pulmonary edema, impaired liver function The heterogeneity of treatment was evaluated by c2 and I2,
tests, high serum creatinine, low platelet count, IUGR or and was considered to be significant if P was less than 0.05
preeclampsia requiring delivery prior to 37 or 34 weeks if or if I2 was larger than 50%, according to Higgins et al.16
the prior criteria were not reported.13 Preeclampsia Publication bias was investigated by the evaluation of asym-
associated with one or more of these criteria was considered metry of a funnel plot. Agreement between reviewers was
severe. analyzed by percentage agreement and weighted kappa (ê)
statistic, with k of 0.7 as the minimum acceptable agree-
After extraction of all articles from the databases, the cita- ment.17 These data were analyzed by the SAS version 9.1
tions were reviewed by two investigators and categorized as (SAS Institute Inc., Cary NC).

820 l SEPTEMBER JOGC SEPTEMBRE 2009

 Acetylsalicylic Acid for the Prevention of Preeclampsia and Intra-uterine Growth Restriction

Figure 1. Summary of selection process for the systematic review of aspirin to

prevent preeclampsia

RESULTS studies (67%) a placebo was given to subjects in the control

group. The average rate of preeclampsia in the control
Through our literature search, we identified 178 published group was 22.6%, and confirmed the selection of a group of
reports, including nine randomized controlled trials that patients at high risk for preeclampsia. Finally, the funnel
met our inclusion criteria (Figure 1). Agreement between plot (Figure 2) indicated no evidence of publication bias,
reviewers in identifying the 17 studies was associated with a but was limited by the small number of studies included.27
weighted ê of 0.76. All reviewers concurred on the ade- The stringency of the inclusion criteria also allowed us to
quacy of nine trials that were finally included in the review. confirm the quality of small studies.
A review of the references from two recent meta-analyses
did not lead to additional relevant reports, except for the Maternal Outcomes
identification of an abstract that was cited differently in the In women with early abnormal uterine artery Doppler stud-
Cochrane Library.18 The nine trials involved a total of 1317 ies, there was a 52% reduction in the risk of preeclampsia,
participants (Table 2). Two of the trials had recruited compared with the control group, when ASA treatment was
women prior to 16 weeks (222 patients),19,20 one had started before 16 weeks of gestation (RR 0.48; 95% CI 0.33
recruited women between 17 and 19 weeks (102 patients),21 to 0.68) (Figure 3). The number needed to treat to prevent
and six had recruited women at or after 20 weeks’ gestation one additional case of preeclampsia, in these women with
(993 patients).18,22–26 No study had recruited pregnant early abnormal Doppler, was five. On the other hand, ASA
women in two overlapping gestational age periods. Every therapy started at 17–19 weeks (RR 0.55; 95% CI 0.17 to
trial used ASA (50 to 150 mg) daily as treatment, and in six 1.76) or at ³ 20 weeks (RR 0.82; 95% CI 0.62 to 1.09) was

SEPTEMBER JOGC SEPTEMBRE 2009 l 821

OBSTETRICS

Table 2. Characteristics of the studies included in the review

% of event in
Study Participants control group Inclusion criteria Intervention Outcome

Ebrashy A, 139 women at 22.5 abnormal uterine ASA 75 mg vs. PE; PE onset < 37 weeks; severe PE
200519 14–16 weeks artery Doppler and no treatment maternal bleeding SGA (< 10th
other risk factors for percentile); preterm birth; birthweight;
PE and IUGR neonatal bleeding
Zimmermann P, 26 women at 15.40 uterine artery bilateral ASA 50 mg vs. GH; PE; placental abruption; delivery
199726 22–24 weeks notches on Doppler no treatment < 37 weeks; IUGR (< 10th centile);
gestation at delivery; birthweight
Vainio M, 200220 90 women at 23.0 anamnestic risk factor ASA 0.5 GH; PE; Caesarean section; gestation
12–14 weeks and abnormal uterine mg/kg/day vs. at delivery (mean); IUGR (< 10th
Doppler placebo percentile); birthweight < 2500 g
McParland P, 106 women at 28.80 persistently abnormal ASA 75 mg vs. GH; proteinuria; hypertension
199024 24 weeks Doppler waveform placebo < 37 weeks’ gestation; Caesarean
section; mean gestation age at delivery
Yu CKH, 200325 560 women at 18.70 Doppler pulsatility ASA 150 mg vs. PE; early PE < 34 weeks; placental
22–24 weeks index > 1.6 (95th placebo abruption; preterm birth < 37 weeks;
percentile) very preterm birth < 34 weeks; SGA
(< 5th percentile)
Ferrier C, 199618 43 women at 5.0 placental side uterine ASA 60 mg vs. GH, PE
22–24 weeks artery resistance placebo
index > 90th
percentile, or a notch
in early diastole
Harrington F, 216 women at 8.20 abnormal uterine ASA 100 mg vs. PE; PE < 34weeks, SGA (< 10th
200023 20 weeks artery Doppler no treatment percentile; 3rd percentile); placental
(the intervention abruption; Caesarean section;
was stopped at gestational age at delivery; birthweight
24 weeks if the
Doppler was nor-
malized)
Morris JM, 104 women at 14.0 abnormal uterine ASA 100 mg vs. GH; PE; eclampsia; preterm birth; SGA
199621 18 weeks Doppler flow placebo (< 10th percentile)
(systolic/diastolic ratio
> 3.3 or S/D > 3 and
early diastolic notch)
Bower SJ, 63 women at 41. 40 abnormal uterine ASA 60 mg vs. PE; severe PE; birthweight (mean) and
199622 24 weeks artery Doppler flow placebo centile (< 3rd); gestation at delivery;
velocity waveform preterm birth (< 34, < 37weeks)
ASA: acetyl salicylic acid; PE: preeclampsia; GH: gestational hypertension; IUGR: intrauterine growth restriction; SGA: small for gestational age.

not associated with a significant decrease in the incidence of was greater in the subjects treated with ASA than in the con-
preeclampsia compared with the control group. In addition, trol group, but was reported only in three trials (all recruit-
when ASA was initiated before 16 weeks, a significant ing at ³ 20 weeks), and the difference was not statistically
decline in the rate of severe preeclampsia (RR 0.38; 95% CI significant (RR 2.34; 95% CI 0.91 to 6.01).
0.15 to 0.98) and GH (RR 0.61; 95% CI 0.40 to 0.93) was
seen, whereas no significant reduction was observed in Neonatal Outcomes
comparison to the control group when ASA was started Statistically significant decreases in adverse neonatal out-
between 17 and 19 weeks or at ³ 20 weeks of gestation comes occurred only in the subgroup £ 16 weeks of gesta-
(Table 3). Detailed outcomes according to gestational age at tion and in the overall effect (Table 3). The IUGR rate was
the beginning of treatment are reported in Tables 3 and 4. reduced by 49% in the subgroup of women treated £ 16
Use of ASA was not associated with significant changes in weeks (2 trials involving 222 women; RR 0.51; 95% CI 0.28
the rate of preterm births, regardless of gestational age at to 0.92) compared with only 18% in the studies overall (8
the initiation of treatment. The rate of placental abruption trials involving 1274 women; RR 0.82; 95% CI 0.68 to 1.00).

822 l SEPTEMBER JOGC SEPTEMBRE 2009

 Acetylsalicylic Acid for the Prevention of Preeclampsia and Intra-uterine Growth Restriction

Figure 2. Funnel plot of comparison for outcome: preeclampsia

DISCUSSION studies. However, our investigation was limited by the small

number of studies and participants in each gestational age
We evaluated the effect of gestational age at the time of ran-
subcategory, especially in the subgroup of women recruited
domization in trials of ASA for the prevention of
between 16 and 20 weeks’ gestation. Therefore, few conclu-
preeclampsia in women with abnormal uterine artery
sions can be drawn for that particular subgroup. Moreover,
Doppler studies. While ASA treatment was associated with
the confidence intervals were very wide for rare outcomes,
overall reductions of preeclampsia, severe preeclampsia,
such as severe preeclampsia. While we found a significant
gestational hypertension, and IUGR, a significant effect was
reduction in the incidence of severe preeclampsia when
observed only in the subgroup of women randomized prior
ASA treatment began before 16 weeks’ gestation, it was dif-
to 16 weeks’ gestation. More importantly, ASA therapy
ficult to estimate if the expected decrease was two-fold or
reduced the risk of developing these outcomes by one half
ten-fold. Our study emphasizes the need for a large investi-
in this subgroup of women. The importance of these find-
gation in this area.
ings is magnified by the very low cost, the relative safety,
and the wide availability of ASA throughout the world, as Another limitation of our study was the definition of
well as the very low number needed to treat to avoid one “abnormal uterine artery Doppler,” which varied among
adverse outcome. authors. However, all definitions employed have been pre-
viously published as predictors of preeclampsia. In this
In a similar review, Coomarasamy et al. reported a compara- regard, we were reassured by the fact that the preeclampsia
ble decrease in adverse outcomes in the overall population, rate was similar across the different studies. Therefore, we
but did not stratify the analysis according to gestational age believe that the variable definition is unlikely to have led to
at randomization.12 Other reviews evaluating the effect of biases in our results.
ASA in the prevention of preeclampsia also did not stratify
the analysis for women recruited at £ 16 weeks,11,14,28,29 Our findings confirm original fundamental research that
eliminating the possibility of making conclusions about suggested major potential benefit from ASA treatment in
potential benefits from early ASA prophylaxis. the prevention of placental insufficiency, IUGR and
preeclampsia. We remain hopeful that an inexpensive and
The strengths of this study include the extensive literature readily available treatment such as ASA could eventually
review without language restrictions, respect of the decrease the rate of preeclampsia, a leading cause of mater-
QUOROM statement,30 and the homogeneity of the nal mortality, in a high-risk population. Such findings will

SEPTEMBER JOGC SEPTEMBRE 2009 l 823

OBSTETRICS

Table 3. Relative risks of outcomes associated with low-dose ASA in women with abnormal uterine artery Doppler
studies according to gestational age at the beginning of treatment

Outcomes and gestational age Number Number of Prevalence in

at randomization, weeks of trials participants control groups RR 95% CI P NNT (95% CI)

Preeclampsia
£ 16 2 222 47.20 0.48 (0.33 to 0.68) 0.001 5 (3 to 8)
17–19 1 102 14.00 0.55 (0.17 to 1.76) 0.31
³ 20 6 993 17.37 0.82 (0.62 to 1.09) 0.18
Overall 9 1317 21.97 0.68 (0.55 to 0.85) 0.001 16 (10 to 34)
Severe preeclampsia
£ 16 1 73 14.29 0.1 (0.01 to 0.74) 0.02 8 (5 to 25)
17–19 0 0 NA NA NA NA
³ 20 3 824 8.53 0.53 (0.24 to 1.17) 0.12
Overall 4 960 9.30 0.38 (0.15 to 0.98) 0.04 21(13 to 50)
Gestational hypertension
£ 16 1 86 37.21 0.31 (0.13 to 0.78) 0.01 4 (3 to 13)
17–19 1 102 28.00 0.82 (0.42 to 1.60) 0.57
³ 20 3 169 20.00 0.72 (0.36 to 1.42) 0.34
Overall 5 357 26.40 0.61 (0.40 to 0.93) 0.02 10 (6 to 50)
Preterm birth
£ 16 1 136 19.05 0.93 (0.46 to 1.90) 0.85
17–19 1 102 10.00 0.58 (0.15 to 2.29) 0.43
³ 20 3 640 25.94 0.83 (0.63 to 1.10) 0.20
Overall 5 878 23.10 0.83 (0.65 to 1.08) 0.16
Intrauterine growth restriction
£ 16 2 222 22.64 0.51 (0.28 to 0.92) 0.02 10 (5 to 50)
17–19 1 102 22.00 1.22 (0.62 to 2.43) 0.56
³ 20 5 950 27.18 0.85 (0.68 to 1.06) 0.14
Overall 8 1274 25.99 0.82 (0.68 to 1.00) 0.05
Placental abruption
£ 16 0 0 NA NA NA NA
17–19 0 0 NA NA NA NA
³ 20 3 790 1.20 2.34 (0.91 to 6.01) 0.08
Overall 3 790 1.20 2.34 (0.91 to 6.01) 0.08

NNT: number needed to treat; NA: not available or not applicable

encourage research into early predictive biomarkers of studies would have to be performed, even if Doppler
preeclampsia, including Doppler and serum markers, in screening is not likely to be offered widely. There is an
order to better identify high-risk women who could benefit increasing body of evidence suggesting that the diagnosis of
greatly from ASA prophylaxis. First trimester ultrasound abnormal placentation by Doppler screening and the pre-
examination is becoming standard practice in many regions diction of placental insufficiency leading to preeclampsia
and countries throughout the world, to obtain accurate ges- and IUGR is more accurate when combined with serum
tational age and to assess fetal nuchal translucency, and the markers.31,32 It remains unclear whether or not ASA treat-
addition of Doppler evaluation of the uterine arteries to ment should be recommended in this high-risk population,
routine screening processes is potentially useful in predict- or if uterine artery Doppler should be adopted as a screen-
ing pregnancy complications. However, it is clear that ing or diagnostic tool for the diagnosis of abnormal
expertise would have to be developed and reproducibility placentation. Urgent confirmation of our findings in larger

824 l SEPTEMBER JOGC SEPTEMBRE 2009

 Acetylsalicylic Acid for the Prevention of Preeclampsia and Intra-uterine Growth Restriction

Figure 3. Meta-analysis of randomized controlled trials evaluating the effect of ASA on the rate of
preeclampsia according to gestational age at the initiation of treatment

Table 4. Mean difference of outcomes associated with low-dose ASA in women with abnormal
uterine artery Doppler studies according to gestational age at the beginning of treatment

Outcomes and gestational Number Number of Mean in Mean

age at randomization of trials participants control group difference 95% CI P

Birth weight, g
£ 16 weeks 2 222 3193 20 (-178 to 219) 0.42
17–19 weeks 1 102 3049 96 (-139 to 331) 0.27
³ 20 weeks 4 396 2929 81 (-62 to 223) 0.86
Overall 7 720 3025 67 (-37 to 171) 0.86
Gestational age at
delivery (weeks)
£ 16 weeks 1 86 39.2 0.3 (-0.5 to 1.1) 0.45
17–19 weeks 0 NA NA NA NA NA
³ 20 weeks 4 396 38.3 0.4 (-0.3 to 1.0) 0.27
Overall 5 482 38.5 0.3 (-0.2 to 0.9) 0.89

NA: not available or not applicable

studies is necessary, including the assessment of the poten- ACKNOWLEDGEMENTS

tial use of serum, biophysical, or genetic markers, for the
This work was funded by the Jeanne and Jean-Louis
identification of high-risk populations who could be tar- Lévesque Chair for Perinatology Research at the Faculty of
geted for ASA prevention.31,33–36 Our findings justify fur- Medicine, Université Laval, Québec. Emmanuel Bujold
ther investigation of the role of gestational age at the time of holds a Clinician-Scientist Award from the Canadian Insti-
introduction of other treatments, such as anti-oxidants, tutes of Health Research (CIHR) and a Clinician-Scientist
folic acid, and other therapies that have been suggested for Award from Fonds de la recherche en santé du Québec
the prevention of preeclampsia.39 (FRSQ). Yves Giguère is a research scholar from the FRSQ.

SEPTEMBER JOGC SEPTEMBRE 2009 l 825

OBSTETRICS

REFERENCES 20. Vainio M, Kujansuu E, Iso-Mustajarvi M, Maenpaa J. Low dose

acetylsalicylic acid in prevention of pregnancy-induced hypertension and
1. WHO. Risking death to give life. Geneva: World Health Organization; intrauterine growth retardation in women with bilateral uterine artery
2005.
notches. BJOG 2002;109:161–7.
2. Helewa ME, Burrows RF, Smith J, Williams K, Brain P, Rabkin SW. Report
of the Canadian Hypertension Society Consensus Conference: 1. 21. Morris JM Fay RA, Ellwood DA, Cook C, Devonald KJ. A randomized
Definitions, evaluation and classification of hypertensive disorders in controlled trial of aspirin in patients with abnormal uterine artery blood
pregnancy. CMAJ 1997;157:715–25. flow. Obstet Gynecol 1996;87:74–8.
3. Geographic variation in the incidence of hypertension in pregnancy. World 22. Bower SJ, Harrington KF, Schuchter K, McGirr C, Campbell S. Prediction
Health Organization International Collaborative Study of Hypertensive of pre-eclampsia by abnormal uterine Doppler ultrasound and modification
Disorders of Pregnancy. Am J Obstet Gynecol 1988;158:80–3. by aspirin. Br J Obstet Gynaecol 1996;103:625–9.
4. Khan KS, Wojdyla D, Say L, Gulmezoglu AM, Van Look PF. WHO
23. Harrington K, Kurdi W, Aquilina J, England P, Campbell S. A prospective
analysis of causes of maternal death: a systematic review. Lancet
2006;367:1066–74. management study of slow-release aspirin in the palliation of uteroplacental
insufficiency predicted by uterine artery Doppler at 20 weeks. Ultrasound
5. Brosens I, Robertson WB, Dixon HG. The physiological response of the Obstet Gynecol 2000;15:13–8.
vessels of the placental bed to normal pregnancy. J Pathol Bacteriol
1967;93: 569–79. 24. McParland P, Pearce JM, Chamberlain GV. Doppler ultrasound and aspirin
6. Pijnenborg R, Dixon H, Robertson W, Brosens I. Trophoblastic invasion of in recognition and prevention of pregnancy-induced hypertension. Lancet
human decidua from 8 to 18 weeks of pregnancy. Placenta 1980;1:3–19. 1990;335:1552–5.
7. De Wolf F, DeWolf-Peeters C, Brosens I, Robertson W. The human 25. Yu CK, Papageorghiou AT, Parra M, Palma Dias R, Nicolaides KH.
placental bed: electron microscopic study of trophoplastic invasion of spiral Randomized controlled trial using low-dose aspirin in the prevention of
arteries. Am J Obstet Gynecol 1980;137:58–70. pre-eclampsia in women with abnormal uterine artery Doppler at 23 weeks’
8. Khong T, De Wolf F, Robertson W, Brosens I. Inadequate maternal gestation. Ultrasound Obstet Gynecol 2003;22:233–9.
vascular response to placentation in pregnancies complicated by
pre-eclampsia and by small-for-gestational age infants. Br J Obstet 26. Zimmermann P Eiriö V, Koskinen J, Niemi K, Nyman R, Kujansuu E, et al.
Gynaecol 1986;93:1049–59. Effect of low dose aspirin treatment on vascular resistance in the uterine,
uteroplacental, renal and umbilical arteries—a prospective longitudinal study
9. Robertson WB, Brosens I, Dixon HG. The pathological response of the
on a high risk population with persistent notch in the uterine arteries. Eur J
vessels of the placental bed to hypertensive pregnancy. J Pathol Bacteriol
1967;93:581–92. Ultrasound 1997;5:17–30.

10. Vainio M, Kujansuu E, Koivisto A, Mäenpää J. Bilateral notching of uterine 27. Sterne JA, Egger M, Smith GD. Systematic reviews in health care:
arteries at 12–14 weeks of gestation for prediction of hypertensive disorders Investigating and dealing with publication and other biases in meta-analysis.
of pregnancy. Acta Obstet Gynecol Scand 2005;84:1062–7. BMJ 2001;323:101–5.

11. Cnossen JS, Morris RK, ter Riet G, Mol BW, van der Post JA, 28. Askie LM, Duley L, Henderson-Smart DJ, Stewart LA. Antiplatelet agents
Coomarasamy A, et al. Use of uterine artery Doppler ultrasonography to for prevention of pre-eclampsia: a meta-analysis of individual patient data.
predict pre-eclampsia and intrauterine growth restriction: a systematic Lancet 2007;369:1791–8.
review and bivariable meta-analysis. CMAJ 2008;178:701–11. 29. Ruano R, Fontes RS, Zugaib M. Prevention of preeclampsia with low-dose
12. Coomarasamy A, Papaioannou S, Gee H, Khan KS. Aspirin for the aspirin—a systematic review and meta-analysis of the main randomized
prevention of preeclampsia in women with abnormal uterine artery controlled trials. Clinics 2005;60:407–14.
Doppler: a meta-analysis. Obstet Gynecol 2001;98:861–6. 30. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving
13. Meher S, Duley L. Prevention of Pre-eclampsia Cochrane Review authors. the quality of reports of meta-analyses of randomised controlled trials: the
Interventions for prevention of pre-eclampsia and its consequences: generic QUOROM statement. Quality of Reporting of Meta-analyses. Lancet
protocol. Cochrane Database Syst Rev 2005(2) CD005301. 1999;354:1896–900.
14. Duley L, Henderson-Smart DJ, Meher S, King JF. Antiplatelet agents for 31. Than NG, Romero R, Hillermann R, Cozzi V, Nie G, Huppertz B.
preventing pre-eclampsia and its complications. Cochrane Database Syst Prediction of preeclampsia—a workshop report. Placenta 2008;29(Suppl
Rev 2007(2):CD004659. A):S83–5.
15. Pijnenborg R, Dixon G, Robertson WB, Brosens I. Trophoblastic invasion 32. Gagnon A, Wilson R. Obstetrical complications associated with abnormal
of human decidua from 8 to 18 weeks of pregnancy. Placenta 1980;1:3–19. maternal serum markers analytes. J Obstet Gynaecol Can 2008;30:918–32.
16. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency 33. Bernard N, Giguere Y. Genetics of preeclampsia: what are the challenges?
in meta-analyses. BMJ 2003;327:557–60. J Obstet Gynaecol Can 2003;25:578–85.
17. Landis J, Koch G. The measurement of observer agreement for categorical 34. Becker T, Vermeulen MJ, Wyatt PR, Meier C, Ray JG. Maternal obesity and
data. Biometrics 1977;33:159–74. the risk of placental vascular disease. J Obstet Gynaecol Can
18. Ferrier C NR, Kincaid-Smith P. Low dose aspirin delays the onset of 2008;30:1132–6.
pre-eclampsia in pregnancies with abnormal uteroplacental circulation.
35. Crane JM, White J, Murphy P, Burrage L, Hutchens D. The effect of
Proceedings of the 10th World Congress of the International Society for the
gestational weight gain by body mass index on maternal and neonatal
Study of Hypertension in Pregnancy; 1996 August 4–8; Seattle, Washington,
outcomes. J Obstet Gynaecol Can 2009;31:28–35.
USA; 1996:151.
36. Arendas K, Qiu Q, Gruslin A. Obesity in pregnancy: pre-conceptional to
19. Ebrashy A, Ibrahim M, Marzook A, Yousef D. Usefulness of aspirin therapy
postpartum consequences. J Obstet Gynaecol Can 2008;30:477–88.
in high-risk pregnant women with abnormal uterine artery Doppler
ultrasound at 14–16 weeks pregnancy: randomized controlled clinical trial. 37. Bilodeau JF, Hubel CA. Current concepts in the use of antioxidants for the
Croat Med J 2005;46:826–31. treatment of preeclampsia. J Obstet Gynaecol Can 2003;25:742–50.

826 l SEPTEMBER JOGC SEPTEMBRE 2009