Evidence over hope for pregnancy complications

Ian A. Greer

Low molecular weight heparins are widely used to try to prevent pregnancy complications. In this issue of Blood, Martinelli and colleagues report a critical randomized trial that demonstrates no efficacy from such treatment.1

Women with primary composite outcomes during the observation period according to treatment arm. See Figure 2 in the article by Martinelli et al on page 3269 of this issue.1

Obstetricians are frustrated by the lack of effective interventions to prevent late complications of pregnancy such as preeclampsia, fetal growth restriction (FGR), and abruption. Despite considerable research over many years, no definitive cause has been identified for these problems, which are likely to be heterogeneous in origin, but manifest through common pathophysiologic processes; specifically, disturbance of the hemostatic system and inadequate placentation. For example, in preeclampsia we have known for almost half a century that there is coagulation activation, thrombin generation, microvascular fibrin deposition, endothelial dysfunction, disturbed trophoblast invasion of the maternal circulation, and placental infarction.2 Indeed this knowledge led to antiplatelet therapy with low-dose aspirin being introduced in the 1980s with a modest effect (∼ 15%) in preventing preeclampsia and FGR.3 However, these conditions remain major challenges affecting approximately 10% of pregnancies, with major contributions to both maternal and perinatal morbidity and mortality.

In antiphospholipid syndrome similar hemostatic changes and placental infarcts are seen. This is manifest clinically not only by the late pregnancy problems of preeclampsia, FGR, and abruption, but also by recurrent miscarriage. The latter problem is responsive to antithrombotic intervention with low-dose aspirin and heparin.3 Further, women with acquired or heritable thrombophilia are more likely to develop preeclampsia and FGR, although the risk may be overestimated from retrospective case-control and cohort studies as prospective investigations have not confirmed these findings.3 Nonetheless, a logical conclusion from these data was that antithrombotic interventions may prevent late pregnancy complications. The increasing awareness of the association between thrombophilia and late pregnancy complications, and the lack of alternative treatment, led obstetricians to use low molecular weight heparins (LMWHs), which are safe in pregnancy,3,4 for prevention and treatment of these conditions. This practice, driven by the lack of effective therapy, was based on extrapolation, with the hope and anticipation that subsequent supportive evidence would emerge. Trials were therefore urgently required to test the hypothesis that such treatment was actually effective.

Martinelli et al report the first large, well-designed, multicenter, prospective, randomized trial to examine Heparin in pregnant women with Adverse Pregnancy outcome to improve the rate of successful PregnancY (the HAPPY trial).1 They compared event recurrence in 135 women, after screening 250, considered at increased risk because of previous late pregnancy complications. The women were randomized to treatment with a LMWH (nadroparin) in addition to medical surveillance, or to medical surveillance alone. The difficulty in conducting these trials should not be underestimated given the demand from women for an active intervention, the frustration felt by obstetricians because of the lack of an effective therapy, and the now widespread nonevidenced-based use of LMWH for pregnancy complications. Despite the commendable perseverance of the researchers, after 3 years only 50% of the planned study participants had been recruited and the trial was stopped by reason of futility. Overall, 21% of participants randomized to active treatment developed a combined end point compared with 18% of controls. This is a nonsignificant event risk difference of 2.2 (95% CI: −11.6 to 16.0). The distribution of the single components of the composite end point (preeclampsia, eclampsia, HELLP [Hemolysis Elevated Liver enzymes and Low Platelets] syndrome, FGR, intrauterine fetal death, and abruption) was also similar. There were no serious adverse events associated with LMWH.

These data show that nadroparin has no impact in preventing these conditions (see figure). This is in agreement with other reports including a systematic review of several heterogeneous studies of LMWH for women with late pregnancy complications,5 and is also consistent with 2 recent large, randomized trials that showed no benefit from LMWH and low-dose aspirin in women with recurrent pregnancy loss.6,7 The results of these large, methodologically sound trials may differ from smaller, methodologically limited or pilot studies,3,8 thus emphasizing the critical importance of an adequate evidence base to avoid the premature adoption of potential new interventions into clinical practice. Such interventions might still prove effective in specific subgroups, such as those with thrombophilia, but this cannot be assumed and specific trials are required, some of which are under way.

In the meantime we should learn the lesson of premature acceptance of hypothetically beneficial treatment into routine clinical practice. This is important, not only to reduce cost in already challenged health care services, but also to protect our patients from unnecessary risk, and specifically to protect women suffering such devastating pregnancy complications from iatrogenic false hope.


  • Conflict-of-interest disclosure: The author has received honoraria for lectures and ad hoc advisory boards from Leo and Sanofi. ■