British Journal of Obstetrics and Gynaecology

January 1996, Vol. 103, pp. 39-47

ECPPA: randomised trial of low dose aspirin for the
prevention of maternal and fetal complications in high
risk pregnant women
ECP P A (Estudo Colaborativo para Prevenção da Pré-eclampsia

com Aspirina)

Collaborative

Group*

* Collaborators and participating centres are Iisted on pages 45-46

Objective
To determine
outcomes associated

the effectiveness of low dose aspirin
with pre-eclampsia.

Design A collaborative randomised trial comparing
placebo on pre-eclampsia and other materno-fetal
Setting

Twelve teaching maternity

hospitais

in women

at high risk of adverse

the effects of low dose aspirin (60 mg) with
complications associated with hypertension.

and 182 obstetricians'

offices in Brazil.

Subjects
One thousand and nine women considered to be at high risk for the development of preeclampsia, or its complications, entered the study between 12 and 32 weeks of gestation. They
were randomly allocated to receive aspirin (498 women) or placebo (511 women) until delivery,
and follow up was obtained for 96 %.
Resu1ts There were no significant differences between the treatment groups in the incidence of
proteinuric
pre-eclarnpsia
(6'7 % aspirin-allocated
compared with 6·0 % placebo-allocated
women), of preterm delivery (22'3 % compared with 26·1 %), of intrauterine growth retardation
(8'5 % compared with 10·1 %), or of stillbirth and neonatal death (7'3 % compared with 6·0 %),
nor were there significant differences in the incidence of proteinuric
pre-eclampsia
in any
subgroup of women studied, including those who had systolic blood pressures of 120 mmHg
or above at entry (8'5 % compared with 7·3 %) or those who were chronically hypertensive
(10'0 % compared with 7·1 %). Aspirin was not associated with a significant excess of maternal
or fetal bleeding.
ConcIusion
The results of this study do not support the routine prophylactic administration
of low
dose aspirin in pregnancy to any category of high risk women (even those who have chronic
hypertension or who are considered to be especially liable to ear1y onset pre-eclarnpsia).

INTRODUCTION

Pre-eclampsia is a major cause of maternal and
fetal morbidity and mortality', and placental
ischaemia is considered to have a central role in
the pathogenesis of these complications". Preeclampsia is associated with deficient intravascular
production of prostacyclin and with excessive
production of thrornboxane". There is also evidence of activation of the clotting system and early
involvement of platelets". This has led to the use of
antiplatelet regimens (usually low dose aspirin) in
an attempt to prevent ar delay the development
and progression of the condition.
Some small trials of antiplatelet therapy in
pregnancy have reported large reductions in the
Correspondence: Dr A. N. Atallah, Department of Medicine, D
Clinica Medica, Escola Paulista de Medicina, Rua Botucatu 740,
CEP 04023-900 São Paulo, Brazil.

© RCOG

1996 British Journal of Obstetrics and Gynaecology

incidence of pre-eclampsia with the use of \ow
dose aspirin (sometimes with the addition of
dipyridamole''").
But these findings have not
generally been confirmed by more recent large
randomised controlled trials'"!'. Despite this it has
been suggested, usually after retrospective datadependent subgroup analysis, that the benefits of
antiplatelet prophylaxis may still be of use in
certain restricted groups of women. For example,
the CLASP investigators concluded that low dose
aspirin might be beneficial for those at especially
high risk of early onset pre-eclampsiall. Similar
exploratory analyses in another study'" led to the
suggestion that low dose aspirin was efficacious in
primigravid women presenting with systolic blood
pressures of 120 mmHg ar more. Pre-eclampsia
and its sequelae are relatively common in Brazil,
and an observational study conducted there found
that about half of the chronically hypertensive
39

Only after complete baseline information had been provided was of women studied. in the opinion of the responsible clinician. METHODS One thousand and nine women were recruited into the trial from 12 university teaching hospitaIs and 182 obstetricians' offices throughout Brazil between December 1989 and March 1993. No. 39-47 . Randomisation Entry to the study was attained by telephoning a central 24 h service at Escola Paulista de Medicina in São Paulo. renal disease. (%) in allocated treatment Aspirin (n = 498) group Placebo (n=511) Woman's age (years) < 20 20-29 30-39 ~ 40 27-5 (SD 7-4) 79 (16) 218 (44) 172 (35) 29 (6) 27-5 (SD 7-4) 84 (16) 228 (45) 174 (34) 25 (5) Estimated duration of gestation (weeks) < 12* 12 ~ 20 > 20 ~ 28 > 28 22-1 (SD 6-2) 18 (4) 186 (37) 194 (39) 100 (20) 22-4 (SD 6-0) 20 (4) 161 (32) 233 (46) 97 ( 19) Systolic blood pressure (mmHg) < 120 120-139 ~ 140 127-3 (SD 20-5) 153 (31) 171 (34) 174 (35) 126-8 (SD 20-5) 159 (31) 183 (36) 169 (33) Diastolic blood pressure (mmHg) < 90 90-109 ~ 110 81-3 (SD 15-0) 314 (63) 155 (31) 29 (6) 80-3 333 159 19 Other features of current pregnancy Proteinuria andjor facial oedema Evidence of IUGR Obstetric and medical history Primigravid Multiparous. Oxford University. allergy to aspirin. Consent to participate was sought from eligible women.40 ECPPA COLLABORATIVE GROUP pregnant women had severe materno-fetal complications attributable to hypertension'". Pre-randomisation characteristics aspmn to be contemplated. but without clear indications for or against its use. no fetal loss Multiparous. Figures in parentheses are percentages unless otherwise stated. Contraindications included an increased risk of bleeding. asthma. a history of preeclampsia or intrauterine growth retardation (IUGR) in a previous pregnancy or evidence of their presence in the current pregnancy. The study was approved by the Ethics Committee Board of Escola Paulista de Medicina. including chronic hypertension detected before or during pregnancy. Eligibility Women were eligible if they were between 12 and 32 weeks of gestation and. The present report is of the Estudo Colaborativo para Prevenção da Pré-eclampsia com Aspirina (ECPPA). gastric ulcer. such as young or old age). were at sufficient risk of preeclampsia or its sequelae for the use of low dose Table 1. © RCOG 1996 Br J Obstet Gynaecol103. Baseline details of the women (Table 1) were recorded directly on computer-generated randomisation lists prepared by the Clinical Trial Service Unit. This multicentre randomised controlled double-blind trial was designed to determine whether low dose aspirin is effective in women at particularly high risk of adverse outcomes associated with pre-eclampsia. São Paulo. diabetes. with fetal loss Chronic hypertension Diabetes or hyperglycaemia (SD 14-8) (65) (31) (4) 21 34 (4) (7) 27 28 (5) (5) 221 188 89 242 25 (44) (38) (18) (49) (5) 250 175 86 231 37 (49) (34) (17) (45) (7) * Women randomised before 12 weeks of gestation were to start treatment at 12 weeks. and placenta praevia. Women might be considered at sufficient risk for a number of reasons. primigravidity (especially with other risk factors.

Women could be randomised before an estimated gestational age of 12 weeks. 99 % for subgroup analyses14. an increment of 15 mmHg was required 11. but in such cases were instructed not to start taking the tablets before the 12th week. fetal loss or any maternal or neonatal bleeding. For the purposes of analysis. Stillbirths included ali deaths at ar after 24 weeks of gestation and neonatal deaths included all deaths after birth. The dose of aspirin was chosen to be sufficient to inhibit platelet aggregatiori'".e. ifeither had not been discharged). maximum © RCOG 1996 Br J Obstet Gynaecol103. Further exploratory analyses were conducted in response to some of the findings of CLASPl1 and other studies'". corresponds to an odds reduction of 20 % (such odds reductions are slightly larger than the corresponding risk reductions). Women were asked to take the study treatment every day until delivery. No interim analyses of ECPPA were conducted during re- . by which time 1009 women had been randomised. a small random sample ofwomen in the study were interviewed about their compliance. stillbirth and neonatal death. with paracetamol recommended when analgesia was necessary. crude birthweight. Other aspirin-containing preparations were to be avoided. Outcome measures The main prespecified outcome measures were: estimated duration of pregnancy. Standard methods were used to calculate the apparent ratio of the odds of an outcome occurring in the aspirin group compared with the odds in the control group. parity and the existence of chronic hypertension at randomisation. subsidiary comparisons were made of the results subdivided according to gestational age. 1UGR. Such an effect seemed. stillbirth ar neonatal death and any neonatal complications were also to be recorded. it should have been possible to detect reliably a reduction in the incidence of pre-eclarnpsia of about three quarters (as suggested by the results available when this study was designed) in a study of about 600 women. Efforts were made to check and correct any incomplete and inconsistent data wherever possible. the reduction (or increase) in the odds of the event in the aspirin group and its standard deviation (SD) are cited. however. to take account of the number of comparisons.8. Drug stability was confirmed at intervals throughout the study by testing a sample of the study treatment packs. up to 28 days. The duration of tablet taking was assessed crudely by recording the approximate date when study treatment was stopped and. no woman was excluded from the trial.ECPPA: a specific numbered trial treatment pack allocated. Treatment Women were assigned calendar-packed treatment with either one 60 mg film coated aspirin tablet daily ar a placebo tablet. The study outcome of proteinuricpre-eclampsia req uired thedevelopmen t of hypertension plus the detection of protein in the urine after randomisation. along with its confidence interval: 95 % for principal analyses. blood transfusion. maternal and fetal complications related to bleeding. Brief details were to be recorded of proteinuria developing during the pregnancy. identical in appearance. even after delivery. The mode of delivery. Comparisons and statistical methods The main comparison was to be of all women allocated aspirin against ali those allocated placebo. irrespective of whether treatment was dispensed ar taken. women remained in the treatment group to which they had been originally allocated (i. the highest recorded blood pressure (other than during labour).. Hypertension was defined for those with baseline diastolic pressure below 90 mmHg as a rise of at least 25 mmHg to 90 mmHg ar higher. Preterm delivery was defined as delivery before 37 weeks of estimated gestation and 1UGR as birthweight below the third centile for sex and estimated gestational maturity'". for those with initial diastolic pressure of 90 mmHg ar above. containing microcrystalline cellulose and com starch. and. to be too much to hope for and so the study was designed to detect somewhat smaller benefits. in addition. for example. Alternatively. Statistical analyses involve simple comparisons of total numbers affected. unless advised otherwise. In addition. The actual contents of the allocated study treatment were not revealed. intention to treat analyses are reported). After randomisation. 39-47 A RANDOMISED TRIAL OF LOW DOSE ASPIRIN 41 maternal blood pressure recorded after entry. birthweight. Follow up A very simple single page follow up form was completed after hospital discharge of both mother and baby (ar at six weeks postpartum. and was one that had been reported to prevent preeclampsia" while keeping side effects to a minimum. unless there was a c1ear medical reason for the treatment to be made known. whether live birth. an odds ratio of 0.15. In the high risk women to be studied in ECPPA. Resources were available to continue recruitment until March 1993 (when the study was stopped in ignorance of the results).

Interviews with a random sample of 88 women in the study supported the overall estimate of compliance. 20 weeks 16/192 8/172 53/192 49/172 > 20weeks 16/284 22/322 53/284 80/322 nulliparae 8/210 10/241 41/210 50/241 multlparae 24/266 20/253 65/266 79/253 yes 23/231 16/224 56/231 70/224 no 9/245 14/270 501245 59/270 32/ 476 (6.42 ECPPA COLLABORATIVE GROUP only after 75 % ofthe time between randomisation and delivery had been completed. however.5 Aspirin Asplrin better worse ~ ~ 19%5014 reduction (2p=0. A black square to the left of the solid verticalline suggests a benefit (but this is significant at 2P < 0·01 only if the whole Cl is to the left of the solid verticalline). cruitment and the results remained concealed until after data collection had been completed. 16 % were under 20 years of age. Ia).0 1.5 1. Post-delivery follow up forms were obtained for 96 % of the randomised women (476 allocated aspirin and 494 allocated placebo). Ia). with good balance between the treatment groups for the main pre-randomisation characteristics (Table I). RESULTS One thousand and nine women were randomised. The overall results for ali women (and 95 % Cl) are represented by diamonds. Effects of aspirin on (a) proteinuric pre-eclampsia developing after randomisation and (b) preterm delivery.Aâf'!'). 38 % were at 20 weeks of gestation or earlier.0 1. Of the women enrolled. and preterm delivery was defined as delivery before 37 weeks of estimated gestation (as in Cl. Although this represents an 11 % (SD 28) increase in the odds of developing proteinuric pre-eclampsia. this difference is not conventionally significant and is still consistent with a reduction of as much as onequarter (as well as with more than a doubling in risk). Reported compliance with study treatment was good.7) 0. and these women had 985 infants or fetallosses (482 aspirin compared with 503 placebo).. 1. with no difference between the groups allocated aspirin or placebo. inc1uding women with evidence of chronic hypertension (10'0% compared with 7'1%. X2 tests for differences between the effects observed in the different subgroups were ali nonsignificant. Odds ratios (.3%) 129/494 (26. with 88 % of the sample confirming that they had taken more than 75 % of the scheduled study tablets.1%) 106/476 (22.7%) 30/494 (6. 47 % were primigravidae. 39-47 . There was an absence of good evidence that the effect on proteinuric pre-eclampsia differed among the different subgroups of women studied. with the observed reductions or increases in the odds of the outcome developing given to the right of the solid verticalline. 90 % stopped Incidence of proteinuric pre-eclampsia Proteinuric pre-eclarnpsia in ECPPA was recorded in 6·7 % of women allocated aspirin versus 6·1 % of those allocated placebo (Fig.2) 0. Of the 967 randomised women for whom the date of stopping trial tablets was known. or those who had systolic blood pressures of 120 mmHg or over at entry: 28 (8' 5 %) of 331 such (b)PRETERM DELIVERY (a)PROTEINURIC PRE-ECLAMPSIA Entry characteristic Events/Women Aspirin Placebo Odds ratio & CI (Aspirin: Placebo) EventslWomen Aspirin Placebo Odds ratio & CI (Aspirin: Placebo) Gestation .5 1. © ReOG 1996 Br J Obstet Gynaeco/103.1%) Parity Chronic hypertension Ali women entered: 11% SO 28 increase (2p=0.5 Aspirin Aspirin better worse Fig. 47 % had chronic hypertension and 6 % had a history of diabetes mellitus or hyperglycaemia. provided to the principal investigator of ECPPA during the study. = area proportional to amount of information contributed "') and 99 % confidence intervals (Cl : horizontalline) are plotted for certain subgroups ofthe study population. The reassuring data monitoring committee reports to the steering committee of the larger CLASP study were. and 69 % stopped after 95 % of this time. Fig. The outcome of proteinuric pre-eclampsia required the development of hypertension and proteinuria after randomisation.

2P = 0'3). Symbols and conventions as in Fig. Aspirin was asso- . 2P = 0'2). pre-eclampsia occurred in 10/78 patients in the aspirin group. 39-47 without associated proteinuria was slightly more common (68 (14'3 %) compared with 56 (l1'3 %).1%) 18% 50 20 reduction (2p=0.ECPPA: A RANDOMISED (a)INTRAUTERINE GROWTH RETAROATION Entry characteristic Events/Babies Aspirin Placebo TRIAL OF LOW DOSE ASPIRIN 43 (b)STILLBIRTHS ANO NEONATAL OEATHS Odds ratio & CI (Aspirin: Placebo) Events/Babies Aspirin Placebo Odds ratlo & CI (Aspirin: Placebo) Gestation 520 weeks 13/196 9/174 121196 9/174 20 weeks 28/286 421329 231286 21/329 nulllparae 14/214 19/249 10/214 11/249 mulliparae 27/268 321254 25/268 19/254 yes 26/233 26/226 221233 17/226 no 15/249 25/2n 13/249 13/2n > Parity Chronic hypertension Ali babies: 41/482 (8.0 1. There was also a lack of support for the hypotheses generated by CLASP of a reduction in early onset pre-eclampsia: among women who were delivered before 32 weeks. while hypertension © RCOG 1996 Br J Obstet Gynaecol103. proteinuric pre-eclampsia developed in none of the 24 women allocated aspirin and in only 3 out of 36 of those allocated placebo (NS).5 Asplrln Asplrln better worse Fig. Among women who were delivered after 37 weeks. among women who were delivered between 32 and 37 weeks. There was no difference between the treatment groups in the medians ofthe highest blood pressures recorded after randomisation and before labour (140/90 mmHg among both those allocated aspirin and those allocated placebo). Duration of pregnancy The mean duration of pregnancy was about two days longer among aspirin-allocated women than among placebo-allocated women (38'08 weeks (SD 3-48) compared with 37·78 weeks (SD 3'71)).64. Effects of aspirin on (a) intrauterine growth retardation and (b) stillbirth and neonatal death. the effects on preterm delivery did not appear to differ significantly in the different subgroups studied. women allocated aspirin compared with 25 (7'3 %) of 343 allocated placebo. pre-eclampsia occurred in 3/28 patients in the aspirin group and in 3/27 patients in the placebo group.4) 0. was lower among women allocated aspirin (22'3 % compared with 26·1 %: Fig. that is before 37 weeks of estimated gestation. 2. X2 tests for differences between the effects observed in the different subgroups were all nonsignificant.0 1. As was the case for proteinuric preeclampsia. although the odds of delivering preterm was 19% (SD 14) lower among aspirin-allocated women.5 1.4) 35/482 (7. this difference was not significant (95 % CI: 40 % reduction to 9 % increase).3%) 0. 1.5%) 51/503 (10. pre-eclampsia occurred in 19/370 patients in the aspirin group and in 14/365 in the placebo group (X2 test for trend = 0.5 1. but this slight increase of 56·8 g (SD 48'7) was not statistically significant. Proteinuria without hypertension severe enough to be defined as pre-eclampsia was slightly less common among aspirin-allocated women (41 (8'6 %) compared with 52 (10'5 %). Birthweight The mean birthweight of all babies bom to women allocated aspirin was 3021·8 g (SD 763'3) compared with 2965·0 g (SD 756'0) in the placebo group. 2P = 0-4). Sixty women with a history of diabetes or hyperglycaemia were followed up. However. Intrauterine growth retardation was defined as birthweight below the third centile for sex and estimated maturity. but this difference was not statistically significant. and stillbirths and neonatal deaths as deaths at or after 24 weeks gestation and up to 28 days after birth. 1b).5 Aspirin Aspirin better worse 30/503 (6. The likelihood of preterm delivery.0%) 23%5029 Increase (2p=0. compared with 13/102 patients in the placebo group.

nor were there any significant differences in placental abruptions or other ante partum bleeds (Table 2). at least in part. again. Despite the high risk population studied in ECPPA. either in CLASP or in ECPPA. either overall or in any of the subgroups studied. and with some methodological problems!" in at least one small tria\. although it had been suggested from exploratory analyses of CLASPll that aspirin may be justified for those at especially high risk of early onset pre-ec1ampsia. In particular. DISCUSSION The incidence of proteinuric pre-ec1ampsia in ECPPA was similar to that reported in previous studies. and there was no difference between the treatment groups in the numbers transfused. this is not supported by the results from ECPPA. maternal hypertension ar lUGR (21 (4-4%) compared with 26 (5'2 %» or in thase associated with maternal or neonatal bleeding (50'04%) compared with 80'59%». it has not been possible. However. and similar to those of the more recent larger trials (Fig. but stillbirths and neonatal deaths were more common (6'6 % compared with 2·8 % in CLASpll. to publication bias. Values are shown as n (%). to identify any particular category of women-inc1uding those in a substudy of CLASP who were angiotensin II sensitive (as in one particularly promising small study"). but maternal transfusions were systematically sought. Two maternal deaths were reported in this study: one in the aspirin-allocated group was attributed to the HELLP syndrome and the other in the placebo group was due to a car crash at 24 weeks of pregnancy. this difference was not significant. 2b). it seems likely that this may be due. No significant differences in the incidence of intraventricular haemorrhages or other bleeds in the babies were observed. which may reflect uterine vascular lesions caused by chronic hypertension. nor were there any significant differences in the number of stillbirths and neonatal deaths associated with pre-ec1ampsia. 1·6% in the American study'" and 2·3 % in the Italian study"). It was recently suggested " that the results of the larger trials may have been diluted by their broad entry criteria and by the wide variations in care between the different participating countries. and 47 % were primigravidae (compared with 28 % in CLASP). The perinatal mortality rate among the large chronically hypertensive group in ECPPA was 8·5 %. 39-47 .6) ( 1'5) 7 8 6 7 (l A) ( 1. Moreover. Effects of aspirin on delivery type. 2a). © RCOG 1996 Br J Obstei Gynaecol103. with small trials with unpromising results being less likely to be published than those with particularly promising results. This 23 % (SD 29) increase with aspirin is nat statistically significant and the 95 % confidence interval is wide.6) ( 1'2) (l A) ciated with a slightly smaller prapartian of babies with lUGR (8'5% compared with 10·1%: Fig. Stillbirths and neonatal deaths Faur (0'8 %) fetallosses occurred before 24 weeks of gestation in the aspirin group and 6 (1·2 %) accurred in the placebo graup (Table 2). Pregnancies with data Fetal outcomes Labour and delivery Caesarean section Forceps delivery Aspirin Placebo n = 476 n = 482 n = 494 n = 503 291 (61'1) 36 (7-6) 301 (60'9) 36 (7'3) Maternal bleeding Placenta I abruption Other antepartum bleed Postpartum bleed Transfusion 5 6 3 7 Fetal bleeding Intraventricular haemorrhage Other neonatal bleeds 6 (1'2) 3 (0'6) 3 (0'6) 2 (O A) Fetal losses Losses < 24 weeks Stillbirths (~ 24 weeks) eonatal deaths « 28 days) 4 (0'8) 28 (5'8) 7 (1'5) 6 (1'2) 23 (4'6) 7 (I A) (1'1 ) (1'3) (0. Chronic hypertension was present at entry in 47% of the ECPPA patients (compared with 20 % in CLASP). Other outcomes There were no significant differences between the treatment graups in delivery by caesarean section ar forceps. the effects of aspirin on adverse outcomes appear to be much less promising than those suggested by the results of the first small trials. bleeding and fetal loss after randomisation. but. There were 28 stillbirths plus 7 neonatal deaths (35 total: 7'3%) in the aspirin group and 23 plus 7 (30: 6·0 %) in the placebo group (Fig. There was no apparent difference in the effect in the variaus subgroups of women studied. 3: updated from CLASPll). All bleeds after delivery were not explicitly recorded and were incompletely reported (overall rate of 0·9 % compared with 26 % in CLASP). Possible explanations for the discrepancy between the results of the small trials and the larger trials ha ve been discussed in detail in the report of the CLASP study.44 ECPPAcaLLABaRATIVEGRaUp Table 2. or those with e\evated blood pressure at entry (as in a post hoc subgroup analysis of one of the recent larger trials10)-in whom the reduction in proteinuric pre-ec1ampsia was as great as that reported in the previous small trials (Fig. 3a).

ln conc1usion.8%) 18 455/8151 ( 5. 3a).e. Araraquara: A Durante.ECPPA: (a)PROTEINURIC No 01 Antlplatelet Control ortrial trlals therapy TRIAL (b)STILLBIRTHS PRE-ECLAMPSIA Trial categories therapy A RANDOMISED Odds ratlo & 95°. there were no subgroups m these generally higher risk Brazilian women.ali trlals wlth data: . Details of the trials included are given in CLASpll Similarly. = 25.6%) 571/7952 ( 7.8 (p = 0.07) X~ = 6. those that included fewer than 200 women) and from larger trials were combined using standard overview methods'" and stratified odds ratios plotted. Acknowledgements The most important acknowledgement is to the hundreds of women who took part in ECPPA and to the doctors who collaborated throughout Brazil.7%) 521/7668 ( 6. The following investigators participated in the study.0%) 6 228 -I 320 10/ 289 (3. Americana: A Freitas Jr.5 Antlpletelet therapy worse 482 38/2524 1361 4821 30/ 503 7 212/ 7989 ( 2.6%) -/ 51 306 (2.5 Antlplatelet therapy worse batween: . The addition of ECPPA reduces still further the apparent size of this benefit (Fig. other antepartum haemorrhage. Symbols and conventions as in Fig. .0 1. in whom clear effects on lUGR.9) 19 218/ 8295 ( 2.6%) 214/ 8137 ( 2.tlo (Antlplatelet & 95% CI : Placebo) Small trlals 10/ 319 (3.6) X'.6%) 1%5010 reduct10n (2p:O.006) 23%506 reductlon (2p=0. ln absolute terms. L KinsuÍ. Principal Investigator: A N Atallah. with very good follow up and compliance.3) X'.9) o 0.6%) 1%5010 Inc •. 3b). even in the very high risk pregnant women of a developing country. the ECPPA results support the conclusion!' that if some special category of women exists that may benefit substantially from aspirin.7 (p < 0. J Filho. 39-47 about 2 per 100).2%) t o Test lor heterogenelty 0.4) Fig.0004) X~. As in the previous trials.5 Antlplatelet therapy batter . Consequently. and if the small 'hypothesis generating' trials are excluded. Araçatuba: R Eduardo.7%) 20417848 ( 2.•••• (2p:O. R Collins.0 35/ 17%506 reductlon (2p=0. does not support the widespread routine use of aspirin for the prevention of pre-eclampsia or other hypertensive complications. But.1%) 11 Wlth data 7 Wlthout data -/ 308 12 50/ 284 (17.3 (p = 0. it must compose a much smaller and more select group than had previously been thought to be the case. with no significant excesses of placental abruption. analysis and writing committee: A N Atallah.5%) -/ 223 Larger trials Before CLASP 5 CLASP ECPPA tOO/2697 139/2524 313/4659 3521 4650 32/ Ali larger trlals Ali trlals wlth data 476 30/ 5 • 494 7 445/7832 ( 5.00006) 1. = 10.alllarger 1.2 (p = 0. the apparent reduction in the larger trials combined is only 17 % (SD 6). overall. or mortality due to bleeding. H Handoll. Arararas: L Davolos. 1. in contrast with the reduction of about three-quarters suggested by the first small trials. transfusion. Andradina: M Amorim. a systematic overview of the available results from alI randomised trials of antiplatelet therapy indicated a 25 % reduction in the incidence of pre-eclampsia'".4 CI (Antlplatelet: Placebo) OF LOW ANO NEONATAL No 01 Antlplatelet Control trlals therapy therapy DOSE 45 ASPIRIN OEATHS Odds n. stillbirths or neonatal deaths could be demonstrated. the present randomised placebocontrolled study in 1009 women.3 (p = 0.smaller trials versus 48/ 2697 129/4810 largar trials: trials: X'" = 40. Overview of effects reported from ali randomised trials of antiplatelet therapy in pregnancy on (a) proteinuric pre-eclampsia and (b) stillbirths and neonatal deaths.5 Antlplatelet therapy batter 1. B Farrell. these more modest proportional reductions imply that antiplatelet therapy would typically prevent proteinuric preec1ampsia in about 1 woman per 100 treated (with confidence interval ranging from about zero to © RCOG 1996 Br J Obstet Gynaecol103. as was suggested by the other large trials. there is no evidence of an effect of aspirin on the incidence of stillbirths and neonatal deaths (218 (2'6 %) aspirin-allocated compared with 214 (2'6 %) placebo-allocated deaths: Fig. O Fukushima.3 (p = 0. 3. the results of ECPPA are generally reassuring as regards maternal and neonatal complications. = 11. ECPPA and some previous large scale trials!' do suggest that aspirin may prevent a few preterm deliveries per 100 women treated. Available results from smaller trials (i. Prior to ECPPA. Data management. C Vieira.000001) X~ = 1.

Lancet 1980. M Trevisan. 5 Beaufils M. Lancet 1985. Carapicuiba: D Bezerra. A Gradella. 4 Redman CWG. analysed. H Ariê. R Y oshiassu. Botucatu: I Maest. H Halbe. E de Souza. M Zugaib. A Azevedo. H Gonzalez. São Caetano: A Adans.v Bras Nefro11990. José do R. J Rebello. Early platelet consumption in pre-eclampsia. Placenta 1991. 343: 619-629. Fernandopólis: A Flumignan. G Duarte. M Mesquita. UK provided technical support and encouragement throughout the duration of the study. © RCOG 1996 Br J Obstet Gynaecol 103. C Ferraz Costa. Novo Horizonte: R Melchiori. I Silva. Doppler ultrasound and aspirin in recognition and prevention of pregnancy-induced hypertension. L Bertechini. J Neves. M Miyazawa. Porto Alegre: I Belli. D Klotzel. A de Araújo. 12: 301-308. Limeira: Z Vinhal. L de Figueiredo. Estudo prospectivo "cohort ' de gestantes com hipertansão arterial crônica. J Bencic. R Peto. R Clemente. I: 1-3. Santa Maria-RS: J da Silva Ethur. Cruzeiro: M Kisse. José Carlos Pinton. Batatais: O Alves. Bariri: C Negrão. Natal-RN: H de Oliveira. ECPPA Co-ordinating Centre: A N Atallah. M Beviláqua. Pejuçara-RS: L da Silva. C Almeida. Gregorini G. F Zanotto. Lins: J Leão. S Lustre. N Gabriel. J Tsello. I Wulkan. who also donated specially packaged aspirin and matching placebo and by Escola Paulista de Medicina. Donsimoni R. Thom E et ai. L Takano. 13 Benigini A. A Meneghetti.IC. Sertãozinho: A Sóde. Taubaté: M de Assis. CNPQ and INCLEN. L de Campos. H Lippi. 321: 351-356. Cachoeira Paulista: A Ferreira Jr. Preto: R Bertazzo. The Clinical Trial Service Unit (CTSU).46 ECPPA COLLABORA TIVE GROUP Asis: R Zambotti. Buli World Health Organ 1979. Pindamonhangaba: A Wolff. P de Godoy. 8 McParland P. J dos Santos. Mogi: J Magalhães. Guarujá: E Rimi. 9 Italian Study of Aspirin in Pregnancy. A Henrich. Lancet 1994. R de Souza Mesquita. Sorocaba: L Neto. Jundiaí: E Gennari. C dos Santos Jr. T de Oliveira. Chamberlain GVP. R Mattar. The study was principally funded by Sterling Drugs (P Tribble. 12 Atallah AN. P Franco. B Carvalho. Lowdose aspirin prevents pregnancy-induced hypertension and preeclampsia in angiotensin-sensitive primigravidae. Lancet 1990. Study Monitoring Committee: I Chalmers. X Mazzini. I: 840-842. Bonnar J. BM] 1978. G Kenji. J Kublikowski. Benedetto C. R Collins. A Allegrini. A Kataguiri. E Cavalcante. M L Duarte. R Esteves. L Primon. Belo Horizonte: C Freire. 57: 373. Franca: M Marcolini. C Antonália. A Celestini. Birigui: J Neto. A Pantoja. S Costa. de Souza Mesquita MR. Sorocaba: C Barros. Camussi G. G Paramo. Jacareí: J Neto. Guarulhos: C Simões. H Rivoire. Campinas: C Nogueira. L Sakamoto. Frusca T et ai. 2 Redman CWG. S Sanforlin. São Paulo: J de Andrade. O Ferraro. Santo André: R Serre. S Weisman). Ribeirão Preto: A Matthes. however. Makovitz JW. Low-dose aspirin in prevention and treatment of intrauterine growth retardation and pregnancy-induced hypertension. designed. M de Mello. Current topic: pre-eclampsia and the placenta. L Napp. S Filho. Itatiba: C Pavanelri. 335: 1552-1555. 341: 396-400. S. 39-47 . T Gollop. W Muniz. I Calderon. G Frehse. R de Lacerda. Uzan S. 7 Schiff E. M Scott. Epidemiology ofhypertensive disorders ofpregnancy. Etfect oflow-dose aspirin on fetal and maternal generation of thromboxane by platelets in women at risk of pregnancy-induced hypertension. M Lemos. N Engl J Med 1989. O Abujamra Jr. W de Matos Rezende. Bauru: P'Tobias. Rio Claro: G Neto. University of Oxford. J Dória. Novo Hamburgo--RS: L Jaeger. Bragança: A Neto. Sorocaba: N Bressan. The study was. I Moraes. A Mariano. Londrina: F Sobrinho. conducted. Rio Grande--RS: P Gonçalves. Mirandopólis: Z Souza. S Ximenes. A Ribeiro. Caritis SN. Rotmans P. Canguçu: D de Campos. New Engl] Med 1993. I Conceição. 321: 357-362. São Leopoldo--RS: I Plentz. C de Quadros Kroeff. J da Silva. Beilin L. J Adalaft Neto. Dekker GA. W Arlê. Nuffield Department of Medicine. 12: 113-120. Colau . The use of aspirin to prevent pregnancy-induced hypertension and lower the ratio of thromboxane A2 to prostacyclin in relatively high risk pregnancies. S Guimarães. Caxias do Sul: D Tessari. Cataguases: F Cesário. 11 CLASP (Collaborative Low-dose Aspirin Study in Pregnancy) Collaborative Group. V de Aráujo. Campinas: A de Meio. H de Carvalho. J A Grisso. W Sória. References I Davies AM. Duarte ML et ai. Inc. Jaboticabal: L Martins. and interpreted independently of the commercial sponsor. J Peraçoli. 3 Bussolino F. W Taborda. I Daniel. H Nightingale. M de Oliveira. Presidente Prudente: D Campos. M Russo. Erechim -RS: A Teixeira. C da Costa Alves. S Leung. I Machida. E Clarizia. R de Freitas. E Chaves. Lancet 1993. N Sass. P Pirozzi. A Pereira. A Andrade. 329: 1213-1218. Lancet 1986. Prevention of preeclampsia by early antiplatelet therapy. 2: 702. Osasco: E Santana. J Mottola Jr. A Elias. Prevention of pre-eclampsia with low-dose aspirin in healthy. Marilza V Rudge. S Espinosa. N Engl J Med 1989. Maternal vascular prostacyclin activity in pre-eclampsia. Piracicaba: C Negretti. C Bianco. J Masonetto. Bernardo: LC João. Ourinhos: F Cesme. Santos: Ma dos Santos. I: 467-469. Massobrio M. P David. Marília: J Prado. 6 Wallenburg HCS. D Silvestrini. O Delascio (late). F Jobim. H Paraventi. nulliparous pregnant women. S. F Simon. G Porto. L Awoke. Pearce JM. Goldenberg M el ai. Olimpia: J Minari. A Sanches. R Theodozio. D Bentivegna. Peleg E. V Freitas. C Campmany. CLASP: a randomised trial of low-dose aspirin for the prevention and treatment of pre-eclampsia among 9364 pregnant women. 10 Sibai BM.

39-47 A RANDOMISED TRIAL OF LOW DOSE ASPIRIN 47 16 Atallah AN. 1: 1267. myocardial infarction. 17 Beilin L. Received 27 April 1995 Accepted 13 September 1995 . Armitage P et ai. LanceI 1990. BMJ 1994. 308: 1200-1251. Shinar R. 308: 81-106. and stroke by prolonged antiplatelet therapy in various categories ofpatients. © RCOG 1996 Br J Obstet Gynaecol103.ECPPA: 14 Peto R. Br J Cancer 1977. Design and analysis of randomized clinical trials requiring prolonged observation or each patient l l. BMJ 1994. 35: 1-39. 15 Antiplatelet Trialists' Collaboration. Pike MC. Analysis and examples. Pre-ec1ampsia and prostaglandins. Collaborative overview of randomised trials of antiplatelet therapy-I: Prevention of death. Aspirin and pre-ec1ampsia.

** Department of Obstetrics and Gynaeeology. 50 % of preterm deliveries". North Manehester General Hospital. low socio-economic status and cervical incompetence predicts. J. The positive predictive value was 50 % and the nega tive predictive value was 73 %. Vol. Baldwin Registrar * Department of Obstetrics and Gynaeeology. There are more than 20 isoforms of the molecule. tt Department of Obstetrics and Gynaeeology. Manchester M 13 9PT. tissue repair and coagulation+". Saint Mary's Hospital. a woman was counted as positive only if the final swab was positive and preceded delivery by not more than 14 days. * A.British Journal of Obstetrics and Gynaecology January 1996. A substantial morbidity accompanies preterm delivery as well as iatrogenic morbidity associated with prolonged neonatal intensive care. Mahmood Registrar. twin pregnancy. Gloueester Royal Hospital Objective The study was designed to determine whether fetal fibronectin before 37 weeks in women at high risk of preterm delivery. Saint Mary's Hospital. tt N. Hawkes SHO. Fetal fibronectin is part of a family ofubiquitous dimeric glycoproteins present predominantly in plasma and extracellular matrix and which influence cell adhesion. The sensitivity of an individual fibronectin swab in predicting preterrn delivery within 14 days of testing was 71 % and the specificity was 93 %. Women were counted as positive if any swab in the sampling sequence was positive. Results Results were calculated by swab and by subject. screening for reduced fetal breathing movements on ultrasound scanning or screening for recurrent contractions with external tocography provides little additional sensitivity". 64 % and 79 %. * M. Correspondence: Dr S. the positive predictive value and the nega tive predictive value were 85 %. Quantitative assays of fetal fibronectin were obtained from sequential high vaginal swabs taken fortnightly from 24 to 34 weeks of gestation. For a woman who has had a positive swab the sensitivity in predicting delivery before 37 weeks was 54 %. respectively. motility. Using a history of previous preterm delivery. would predict delivery Study methods Forty-three women considered at risk ofpreterm delivery were recruited antenatally into a blind longitudinal study. 103.000 daltons and is produced © RCOG 1996 Britisli Journal of Obstetrics and Gynaeeology . Maresh Consultant. Leeson. Hathersage Road. 48-53 Detection of fetal fibronectin as a predictor of preterm delivery in high risk asymptomatic pregnancies * s. For a woman who has had a positive swab the sensitivity in predicting preterm delivery within 14 days of testing was 80 % and the specificity was 83 %. Such testing should be performed every two weeks. Fetal fibronectin has a molecular weight of about 450. The use of serial vaginal examination. UK. Fibronectin swabbing when calculated by patient did predict preterm delivery within 14 days of testing (P = 0'01) and before 37 weeks (P = 0'01). at best. Long term neurological problems are more difficult to evaluate. Analysis of the accuracy of predicting delivery from 7 to 28 days after sampling revealed that the best prediction for delivery was within the following 14 days. t Department of Obstetrics and Gynaeeology. 48 Prediction ofpreterm labour is unreliable. Martindale Registrar. The overall positive predictive value was 31 % and the negative predictive value was 99 %. Leeson Senior Registrar. Apart from the human cost. Muotune SHO. C. The specificity. Most of these deaths occur in pregnancies ending before 29 weeks'. The sensitivity of the fibronectin swab in predicting delivery before 37 weeks was 17 % and the specificity was 93 %. Manehester. J. Fibronectin concentrations of 0·05 Ilg/ml or more were considered as positive. The positive predictive value was 36% and the negative predictive value was 97%. the financial cost of providing inpatient care for preterm babies and providing support in cases of lifelong handicap is great. pp. A. uterine abnormality. Bolton General Hospital. ** E. C. tT. INTRODUCTION Preterm delivery is the leading cause of newborn infant mortality in the developed world. Conclusion Serial fetal fibronectin assessment from 24 to 34 weeks of gestation anticipated preterm delivery within 14 days oftesting and before 37 weeks for high risk asymptomatic women. A. Department of Obstetrics and Gynaecology. * K.