Drs Goffinet, Breart and Uzan make a number of comments about the interpretation of the results of ECPPA1 and CLASP2 which have already been dealt with previously in the reports of those studies. High risk pregnant women were included in ECPPA, as is indicated by their entry characteristics (e.g. about half with chronic hypertension, one-third with diastolic blood pressure of 90 mm Hg or more, one-fifth with previous fetal loss) and by their high rate of adverse outcomes (e.g. 6.6 % overall stillbirth and neonatal deaths, compared with 5.7 % in EPREDA). Moreover, as has been discussed1,2. it was not possible to identify in ECPPA or CLASP any particular category of pregnant woman–including those randomised early in pregnancy. or those in a substudy of CLASP who were angiotensin II-sensitive3 (as in one particularly promising small study4 which, incidentally, tested a 60 mg daily aspirin regimen similar to that used in ECPPA and CLASP)–in whom the benefits of antiplatelet therapy were as great as had been claimed from selected small studies.
The separate consideration of the larger trials (which included EPREDA) and the smaller trials, the apparent discrepancies between their results, and the meta-analysis methods of meta-analysis were discussed extensively in the CLASP report. In particular, it was noted that the trials that reported large effects of antiplatelet therapy on pre-eclampsia, IUGR or perinatal mortality tended to be small, and that at least as many women are known to have been randomised in other small trials which may not have been published because their results were less remarkable (as is, indeed, the case for the PERGAR study of antiplatelet therapy conducted by Breart and colleagues). If this is true. then the available results from the small trials would give a biased estimate of the positive effects of antiplatelet therapy, but large trials would not5.
Hence, consideration of the totality of the available evidence–rather than selective emphasis of only certain small parts of the evidence, with the inherent biases of such an approach– does not support the widespread routine use of aspirin for the prevention of pre-eclampsia or other hypertensive complications in any identifiable category of high risk pregnant women.