Low molecular weight heparin for the prevention of severe preeclampsia: where next?



Low molecular weight heparin has been extensively evaluated for the prevention of preeclampsia in high-risk pregnant women; however, the results from these trials have been conflicting. This review discusses the potential mechanisms of action of low molecular weight heparin for the prevention of severe preeclampsia, how to optimize the selection of high-risk women for participation in future trials, and the importance of trial standardization.

Preeclampsia prevention

Preeclampsia is a hypertensive disorder of pregnancy, clinically diagnosed by new-onset hypertension after 20 weeks' gestation with evidence of organ injury [1]. The potential maternal and perinatal complications of preeclampsia are significant, including maternal seizures, fetal growth restriction (FGR) and stillbirth. Approximately 5% of women develop preeclampsia, with the majority of women experiencing mild preeclampsia that occurs near term and is safely managed by delivery [2]. Only 1% of pregnant women develop early-onset preeclampsia, where both the maternal and fetal implications are substantially greater, typically resulting in FGR and the need for iatrogenic preterm delivery before 34 weeks' gestation [2]. Early-onset preeclampsia is hypothesized to be a direct consequence of the interactions between a dysfunctional placenta and maternal cardiovascular system, while late-onset preeclampsia is hypothesized to be largely triggered by maternal constitutional factors [3-5]. Typically, women with early-onset preeclampsia present with an acute, severe syndrome that threatens the life of both the mother and her fetus. In contrast, late-onset preeclampsia is a less severe illness with a favourable clinical outcome in countries with well-developed maternity systems. However, in countries with limited medical resources, preeclampsia remains among the most frequent causes of maternal death [6].

The early- and late-onset subsets of preeclampsia exhibit different types and severity of placental pathologies and express divergent maternal haemodynamics. Early-onset preeclampsia is strongly associated with placental disease characterized by maternal vascular malperfusion, where the abnormal placental villi secrete excessive amounts of anti-angiogenic proteins and reduced levels of pro-angiogenic proteins into the maternal circulation [7-9]. Women with early-onset disease, typically associated with FGR, develop hypertension earlier in pregnancy that is characterized by increased systemic vascular resistance, reduced cardiac output and stroke volume along with relative bradycardia [5]. In contrast, late-onset preeclampsia is less commonly associated with severe placental disease, and consequently exhibits less dramatic alterations in the circulating levels of placental-derived angiogenic proteins [8, 10]. Late-onset preeclampsia may largely be influenced by maternal factors, including features of metabolic syndrome occurring later in gestation [11-13]. Women who develop late-onset disease typically exhibit a normal or rather exaggerated haemodynamic response to pregnancy [5]. As compared to normotensive pregnant women, they develop hypertension associated with an increase in cardiac output. Abnormal maternal haemodynamics in pregnancies complicated by preeclampsia, particularly early-onset preeclampsia, should alert clinicians to possible underlying maternal cardiovascular dysfunction, both during pregnancy and postpartum [5, 14-17].

Low molecular weight heparin (LMWH) is one of several ‘2nd line’ therapies that has been investigated for the prevention of preeclampsia in women at high risk of preeclampsia recurrence, potentially augmenting the protective therapeutic effects of low-dose aspirin (ASA) [18]. The rationale for studying LMWH in the context of preventing severe preeclampsia is based upon its wide-ranging biologic effects that could improve and normalize both placental function and maternal haemodynamics [19]. The findings from published randomized clinical trials examining the effectiveness of LMWH for preeclampsia prevention have been conflicting and have not addressed the potential mechanisms of actions of LMWH.

LMWH for preeclampsia prevention: evidence from clinical trials

Multiple clinical trials have reported that LMWH reduced the incidence of preeclampsia, as well as newborn weight < 5th percentile, FGR, major placental abruption or fetal loss after 20 weeks' gestation [20-27]; however, other trials of similar design have demonstrated no treatment effect with LMWH [28-30]. Most recently, the large, well-designed HEPEPE and EPPI trials reported that enoxaparin with ASA does not significantly reduce placental-mediated complications, including preeclampsia, when compared to aspirin alone [31, 32]. The conclusions of systematic reviews and meta-analyses summarizing this literature are conflicting. Systemic reviews based on the relatively smaller previous trials concluded that heparin significantly reduces the recurrence of preeclampsia, perinatal mortality, preterm birth and infant birth weight < 10th percentile [33-35]. A recent meta-analysis of data from 963 individual patients from eight randomized trials, which predated the recent HEPEPE and EPPI trials, concluded that LMWH did not reduce the risk of recurrent placental-mediated complications of pregnancy, as compared to no LMWH (14% vs. 22%; relative risk, 0.64; 95% confidence interval, 0.36–1.11; P = 0.11), noting significant trial heterogeneity [36].

The inconsistent conclusions from these clinical trials challenge the therapeutic potential of LMWH for the prevention of preeclampsia and question any future for LMWH as a preventative treatment option for preeclampsia. However, in order to conclude robustly that LMWH is unable to prevent early-onset preeclampsia, the pathways by which LMWH could prevent preeclampsia must be determined. It may be that only a subgroup of patients at risk for early-onset preeclampsia-associated placental and maternal cardiovascular dysfunction may benefit from LMWH, explaining the conflicting trial results.

Potential mechanisms of action of LMWH for early-onset preeclampsia prevention

Significant advances in knowledge regarding the pathophysiology of early-onset preeclampsia, in particular the groundbreaking discovery that placental-derived anti-angiogenic proteins released from the placenta can impair systemic maternal endothelial function, are highly relevant in the context of therapeutic drug development [7, 8, 37-39]. Restoring placental function, normalizing levels of circulating angiogenic proteins and promoting normal maternal cardiovascular function are potential strategies to prevent preeclampsia in high-risk pregnant women.

The mechanisms by which LMWH might prevent early onset preeclampsia have not been directly investigated. The majority of larger clinical studies focused on maternal and fetal clinical outcomes, and have not addressed potential mechanisms of action or documented placental pathology. Importantly, the mechanisms of action of LMWH for early-onset preeclampsia prevention could be independent of its anticoagulant actions, as heparin does not exert obvious anticoagulant activity within the placentas of women at risk of preeclampsia [40]. In vivo experiments in both pregnant women and other clinical populations have demonstrated that LMWH exerts beneficial actions directly on the maternal vasculature, resulting in lower blood pressure, improved endothelial function and modification of circulating levels of angiogenic proteins in a positive manner [25, 41-43]. Ex vivo and in vitro experiments have confirmed beneficial effects of LMWH on vascular reactivity and endothelial function [44-46]. Recently, our group demonstrated that LMWH acutely improves endothelium-dependent relaxation in pregnant women at high risk of severe preeclampsia and significantly increases circulating maternal levels of the pro-angiogenic protein, placental growth factor (PlGF) [47].

This apparent conflict and lack of translation between clinical trials and mechanistic data now challenges us to pose the question: why has LMWH not consistently shown clinical benefit for the prevention of severe, early-onset preeclampsia in high-risk women?

Patient selection

Since LMWH exerts beneficial vascular effects, it is plausible that LMWH could benefit women at the highest risk of early-onset preeclampsia, characterized by low cardiac output, endothelial dysfunction, increased maternal peripheral resistance, accompanied by abnormal levels of circulating angiogenic factors, including low levels of PlGF [47]. The criteria utilized to identify women as ‘high risk’ of developing preeclampsia have thus far differed greatly between trials. The majority of trials utilized inclusion criteria of prior preeclampsia, placental-mediated complication or pregnancy loss to identify women who are considered at high risk of preeclampsia in the current pregnancy; this was the approach in the recent HEPEPE and EPPI trials [31, 32]. The primary outcome of these clinical trials has typically consisted of a composite clinical outcome, including preeclampsia, small for gestational age, placental abruption, and maternal or perinatal death. However, the risk of recurrent severe preeclampsia even in these ‘high-risk’ populations is at most 30%, as most women develop normal placental function in a subsequent pregnancy, and the anticipated effect size was likely significantly underestimated, reducing the power of these trials [48, 49].

All phenotypes of severe preeclampsia are associated with placental pathology; however, it is not entirely clear which placental diseases were targeted in previous trials due to a lack of reported pathology [50, 51]. The main limitation of previous clinical trials evaluating the effectiveness of LMWH for preeclampsia prevention is therefore the singular diagnosis of ‘severe preeclampsia’, which is a heterogeneous syndrome rather than a single disease of different severity, and an incomplete understanding of the underlying cause of severe preeclampsia [52].

A potential strategy for improved inclusion criteria to identify pregnant women at the highest risk of developing early-onset preeclampsia with FGF into clinical trials is one that would include the assessment of placental and maternal cardiovascular function in the current pregnancy, in tandem with comprehensive assessment of clinical risk factors [5, 48]. The likelihood of placental dysfunction can be assessed using placental ultrasound as early as the first trimester in pregnant women who subsequently develop severe preeclampsia [53]. An alternative indirect method of assessing placental function is through placental-derived, circulating angiogenic proteins measured in the first trimester and at mid gestation. In a cohort of low-risk, nulliparous pregnant women, the addition of plasma PlGF to maternal clinical risk assessment significantly improved the identification of women at increased risk of severe preeclampsia, compared to clinical risk variables alone (AUC 0.84; 95% CI 0.77–0.91 vs. AUC 0.76; 95% CI 0.67–0.84) [54]. Several sites have integrated PlGF monitoring into clinical platforms, as this biomarker is more effective than blood pressure, uric acid or proteinuria for the prediction of preeclampsia and delivery within 2 weeks of triage presentation [55].

An additional screening tool to identify pregnant women at high risk of severe preeclampsia may be maternal haemodynamic assessment. Studies have demonstrated that women who develop early-onset preeclampsia associated with FGR, late-onset preeclampsia with normally grown or large fetuses, or have a normotensive pregnancy present strikingly different haemodynamic profiles in the second trimester of pregnancy, leading to the hypothesis that these two preeclampsia subsets may evolve from diverse haemodynamic adaptations to pregnancy [5]. To the best of our knowledge, the identification of women at high risk of preeclampsia based on early pregnancy maternal haemodynamics has not been a strategy used in previous clinical trials. As non-invasive haemodynamic monitoring devices become more available for clinical use, haemodynamic screening may be an important adjunct screening tool to identify women at risk of the rare phenotype of early-onset, severe preeclampsia [47].

The combination of placental and haemodynamic assessments in the index pregnancy, combined with clinical risk assessment, may refine the identification of pregnant women at the highest risk of developing severe, early-onset preeclampsia and provide a more appropriate population to assess the effectiveness of LMWH for preeclampsia prevention. The recent HEPEPE and EPPI trials, based solely on pre-pregnancy risk assessments, concluded that enoxaparin is an ineffective drug for the prevention of preeclampsia in women with a history of placental-mediated complication or pregnancy loss in a prior pregnancy. A limited investigation of LMWH in a more precisely defined population of women at high risk of early-onset preeclampsia is therefore justified.

Standardization of research approach

Recognizing many challenges in preeclampsia research, a plea has been issued to standardize the design of clinical and translational studies investigating the prediction, prevention and treatment of preeclampsia [52]. The significant heterogeneity of clinical trials investigating LMWH therapy for the prevention of preeclampsia largely void the validity of combining these studies using meta-analysis [36, 52]. By contrast, in single-centred trials, where patient selection may be more meticulous, LMWH seems to significantly reduce the primary composite outcome compared to no LMWH therapy (8% vs. 27%; P < 0.0001) and was particularly beneficial in women who had a history of preeclampsia, pregnancy loss, small for gestational age child and placental abruption [36]. Discussion is required amongst the clinical and scientific community to establish standardization amongst preeclampsia prevention trials, where LMWH or other therapies are being investigated, in order to optimize the impact of the findings. At this point, we believe that it is too premature to discard LMWH as a preventative pharmacological approach to prevent severe, early-onset preeclampsia.

Where next?

A significant amount of time and resources have been utilized to define the extent of benefit of LMWH for the prevention of preeclampsia. A patient population that is at high risk of severe preeclampsia is low, approximately 1% [2]. Importantly, LMWH is a costly therapy. In order to conclusively determine the benefit of heparin, we propose a multi-centred, international, sufficiently powered trial that enrols pregnant women early in pregnancy using a combination of clinical risk factors, combined with evidence of both placental dysfunction and an abnormal maternal haemodynamic profile that is associated with the subsequent development of severe, early-onset preeclampsia resulting in preterm delivery (Figure 1). The trial should then evaluate the primary outcome of incidence of severe preeclampsia, as well as secondary outcomes of placental pathology and maternal and fetal health in a standardized manner, as previously outlined [52]. Although a daunting undertaking requiring a network of collaborative, high-risk pregnancy research centres, we believe that this approach will conclusively determine whether LMWH is effective for the prevention of early-onset preeclampsia in high-risk pregnant women.

Figure 1.

Proposed algorithm to recruit pregnant women at the highest risk of early-onset preeclampsia to assess LMWH for its prevention. Women with a high-risk clinical background, including history of preeclampsia, maternal age >35 and BMI >30, should then be further evaluated through placental ultrasound, circulating PlGF levels and haemodynamic assessment to improve clinical trial patient selection


The results from clinical trials evaluating LMWH for the prevention of preeclampsia are conflicting. In order to conclusively determine any benefit of LMWH, a standardized, well-powered trial investigating LMWH for prevention of early-onset preeclampsia in diligently selected women at the highest risk of this disease is required.

Nomenclature of ligands

Key ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY [56].

Competing Interests

There are no competing interests to declare.

J.C.P.K. is supported by the Alva Foundation.