Because of physiologic changes during pregnancy, the risk of venous thromboembolism (VTE) is approximately five to 10 times higher than during non-pregnancy, and the postpartum risk is 15 to 20 times higher. Safely and effectively balancing the risks and benefits of anticoagulation in pregnant women is challenging, both because of the dosing complexities of the various agents in this population and the limited data available to guide treatment decisions.
ASH Clinical News invited Jean Marie Connors, MD, and Martina Murphy, MD, to debate the question: “What is the appropriate type and duration of anticoagulation for women during pregnancy?” Drs. Connors and Murphy discuss the difficulties of appropriate agent and dose selection and the importance of multidisciplinary care in this setting. Dr. Connors is medical director of the Anticoagulation Management Service at Brigham and Women’s Hospital and Dana-Farber Cancer Institute and associate professor of medicine at Harvard Medical School in Boston, Massachusetts. Dr. Murphy is assistant professor of medicine and assistant director of the Hematology/Oncology Fellowship program at the University of Florida College of Medicine in Gainesville. Disclaimer: The following positions were assigned to the participants and do not necessarily reflect ASH’s opinion, the participants’ opinions, or what they do in daily practice.
Jean Marie Connors, MD: The incidence of VTE during pregnancy is approximately one to three per 1,000 deliveries.1 The risk is increased because, during pregnancy, the balance is incredibly tipped toward a procoagulant state – both because of increased levels of natural procoagulants and decreased levels of natural anticoagulants.
Martina Murphy, MD: Absolutely, there are aspects of both pregnancy and the postpartum period that make women hypercoagulable.
During pregnancy, women experience progesterone-induced venodilation, which promotes venous stasis, venous compression by the uterus, and compression of the left iliac vein by the right iliac artery. In addition, pregnancy causes changes in the hemostatic system that create a hypercoagulable state; this includes decreased protein S activity, increased protein C resistance, and other factors that lead to increased thrombin production (higher levels of factor VIII, factor IX, and fibrinogen, for instance).
Postpartum, we have to think about the vascular damage to the vessels during delivery, and that women are usually immobilized for some period of time after delivery. These factors make for the perfect storm of hypercoagulability.
For me, decisions about anticoagulation in pregnant women must marry risks and benefits to both mother and baby which may not always align. The phase of pregnancy is also an important consideration: We approach anticoagulation differently in the antepartum period than we do as delivery nears, when a woman may desire neuraxial anesthesia or require a surgical approach to her delivery.
In the postpartum period, we need to know whether a woman taking anticoagulants intends to breastfeed, as certain anticoagulants can be detected at appreciable levels in breastmilk and are best avoided in this situation. Low-molecular-weight heparin (LMWH), unfractionated heparin (UFH), and warfarin are all safe for breastfeeding mothers.
must marry risks
and benefits to
both mother and
baby, which may
not always align.”
—Martina Murphy, MD
Dr. Connors: Whenever a patient requires anticoagulation, there is always a potential risk for bleeding; for a pregnant woman who requires anticoagulation, there’s an added level of anxiety about potential early miscarriage, excessive bleeding, placental hemorrhage, or implantation bleeding. Another fact that is not well-appreciated by people outside the coagulation world is that the risk of VTE is just as high during the first trimester as the third. Although the risk for venous compression grows as pregnancy progresses, data have shown that frequency of VTE diagnosis is almost equal in every trimester.2
Dr. Murphy: Because of the risks to both mother and baby, this is a doubly challenging patient population for the delivery of safe, effective anticoagulation. It’s worth noting that while evidence-based guideline recommendations for anticoagulation in pregnancy have been published, they are based primarily on observational studies and extrapolation of data from non-pregnant patients. All major evidence-based guidelines recommend LMWH as the preferred anticoagulant for pregnant women.
Neither LMWHs or UFH cross the placenta and, thus, are safe for the fetus. LMWH carries a lower risk of osteoporosis and heparin-induced thrombocytopenia (HIT) and is preferred for pregnant women. Dr. Connors, would you agree with that recommendation?
Dr. Connors: Generally, yes. Across the board, everyone agrees that direct oral anticoagulants (DOACs) should not be used in pregnancy. They are very small molecules that likely cross the placenta.
Dr. Murphy: Also, there are no safety or efficacy data available for pregnant women because they were excluded from all trials of DOACs.
Dr. Connors: Of the remaining options (warfarin or LMWH), I agree that LMWH is preferred. Unlike LMWH and UHF, warfarin crosses the placenta; when the mother receives anticoagulation, so does the baby. That can be a concern depending on the stage of pregnancy. Because warfarin is a teratogen, if the mother is taking it during embryogenesis, there is a risk of fetal anomalies. Also, warfarin use is associated with an increased risk of spontaneous miscarriage, craniofacial abnormalities, fetal bleeding, and other adverse outcomes.
Dr. Murphy: Women who require warfarin because of the presence of a mechanical heart valve are a uniquely challenging population. I spend a great deal of time with these women, offering extensive counseling to understand their personal values and preferences and having in-depth discussions about the risks to both mother and baby.
We know that, in the context of mechanical valves, warfarin continued throughout pregnancy offers the best thromboembolic protection to the mother but with higher risk for fetal loss and complications.3 I would venture to say that most women place preference on what is safest for the baby.
Expert guidelines outline several options, such as switching to an adjusted LMWH dose twice-daily once pregnancy is achieved, adjusting twice-daily UFH dose, or using one of these alternatives and then switching to warfarin after the first trimester.
Dr. Connors: The U.K. clinical guidelines recommend exactly that: Warfarin is preferred in the limited number of situations where LMWH is considered unsuitable, including pregnant women with mechanical heart valves.4 For these patients, the benefit of stroke prevention for the mother is considered to outweigh the risks to the fetus.
But, if a patient cannot tolerate or is allergic to the preferred agent and does not have a mechanical heart valve or other indication for warfarin, what are her other options?
Dr. Murphy: We’ve discussed that LMWH or UFH are the preferred agents and that warfarin and the DOACs should be avoided. We haven’t yet discussed fondaparinux. In a small study of five patients (all of whom had allergic reactions to LMWH) treated with fondaparinux, the so-called “ultra-LMWH,” researchers found that fondaparinux was not associated with adverse effects in newborns, though there was some placental transfer based on measurable anti–factor Xa (FXa) levels in the umbilical cord blood.5 These levels were small, and far below the levels needed for effective anticoagulation. Still, guidelines recommend that fondaparinux be limited to patients with HIT or other allergies to LMWH.
Dr. Connors: When it comes to preventing VTE, the question is always how much prophylaxis is enough and whether once-daily dosing of any agent that doesn’t have an extended half-life is useful. Emerging data have suggested that the standard prophylactic doses for the general population (enoxaparin 40 mg/day or dalteparin 5,000 IU/day) are insufficient for anticoagulation during pregnancy. The jury is out about the optimal dose, though.
Guidelines recommend that intermediate-intensity LMWH be used and then adjusted as the patient gains weight during pregnancy. With more experience and more knowledge of the pharmacokinetics and pharmacodynamics of LMWH in pregnant patients, we have learned that pregnant women metabolize heparin differently. The volume of distribution and the renal clearance are different, so pregnant women require higher prophylactic doses to achieve the same anticoagulant effect – especially women with higher risks for VTE (e.g., women who have had an estrogen-associated VTE related to oral contraception, or women with inherited thrombophilia and a family history of VTE).
We tend to favor enoxaparin 1 mg/kg once-daily, as opposed to enoxaparin 40 mg twice-daily, mainly because we find it hard to get people to comply with the twice-daily dosing frequency. If a patient needs full-intensity anticoagulation to treat a VTE or for a mechanical heart valve, we prescribe twice-daily enoxaparin. We follow up with patients approximately every three months and will increase the dose if they have gained between 10 and 20 pounds (5-10 kg) since the last visit, though there is a lack of data about dosing decisions.
Dr. Murphy: You make a great point about drug clearance. We know that during pregnancy, a patient gains weight and her glomerular filtration rate increases to accommodate fluid shifts. Some pharmacokinetic studies suggest that these changes may lead to lower LMWH levels, and that LMWH dose should be adjusted over the course of a pregnancy based on a patient’s changing weight or to target “therapeutic” anti–FXa levels.
However, other studies have shown that few women require a dose adjustment. This has certainly been my experience in clinical practice, but the discordance between these findings has led to some controversy. There are no good data showing that following anti–FXa levels improves outcomes, and because of concern about inter-assay reliability, the latest guidelines do not recommend routine monitoring of anti–FXa levels in this context.
Despite these recommendations, I sometimes check anti–FXa levels in women taking LMWH, but typically only in those who are at the extremes of weight or with borderline renal function.
Dr. Connors: It’s true that for LMWH use in non-pregnant patients, anti–FXa levels have not been shown to be tied to outcomes, rates of bleeding, or recurrent events. At our institution, our standard practice is to use higher doses in pregnant women from the beginning, without checking levels.
Decisions about anticoagulation during pregnancy depend on the indication for anticoagulation before pregnancy. Some patients require full-intensity anticoagulation throughout pregnancy (such as women with mechanical heart valves or who experienced a deep vein thrombosis during their first trimester), while others fall into more nebulous categories. These patients – including women who have an inherited thrombophilia but have never had a blood clot – are the source of much of the controversy surrounding anticoagulation during pregnancy. For both groups of patients (those requiring full-intensity anticoagulation and those needing prophylaxis), we adjust doses during pregnancy, then think about how to manipulate the anticoagulation around delivery and postpartum.
Dr. Murphy: There are a several areas where we need more research. One question is whether we de-escalate anticoagulant dosing for women with pregnancy-associated VTE after an initial treatment period – particularly for women who are at a high risk of anticoagulant-related bleeding. A systematic review of four studies asking precisely this question found that it appeared to be safe, but there are concerns about the generalizability of the findings.7 I don’t think this is routinely done in clinical practice.
Anticoagulant dosing for a woman with a prior history of VTE is another controversial area. While the decision hinges, in part, on whether the prior clot was pregnancy- or estrogen-related, most guidelines agree that either prophylactic or intermediate dose LMWH or UFH should be used. A randomized trial comparing two doses of LMWH for the prevention of recurrent VTE is underway in pregnant women comparing these two doses of LMWH for the prevention of recurrent VTE.8
“There are almost
no data on anticoagulation
in this setting because
it is extremely difficult to conduct
a randomized, controlled trial of
any question related
—Jean Marie Connors, MD
Dr. Connors: Unfortunately, there are almost no data on anticoagulation in this setting because it is extremely difficult to conduct a randomized, controlled trial of any question related to pregnancy. For instance, more than a decade ago, a Canadian research team was planning a trial comparing prophylactic enoxaparin with or without aspirin, but they had to close early because of low enrollment. Another trial in Israel compared two doses of enoxaparin, rather than a placebo-controlled trial because the investigators knew it would be nearly impossible to get women to sign up to possibly receive a placebo.6 Of course, no pharmaceutical manufacturer wants to sponsor such a high-risk trial, where the outcomes could include fetal and maternal death.
Dr. Murphy: We need more data to help guide treatment decisions in so many aspects of pregnancy-related VTE, yet I worry that the challenges of enrolling pregnant women in clinical trials means data will remain limited. As mentioned before, our evidence-based guideline recommendations are derived from observational studies or extrapolation of data from non-pregnant patients; we are in dire need of more high-quality research in this area.
The lack of solid clinical trials data available to guide treatment in this population also strongly speaks to the need for multidisciplinary care throughout a patient’s pregnancy. As hematologists, we need to align ourselves and communicate closely with our obstetrics team and anesthesiologists to ensure our pregnant patients have a cohesive plan to minimize their risks for thrombosis and bleeding throughout pregnancy and the postpartum period.
Dr. Connors: I completely agree. If a patient who requires daily anticoagulation intends to get pregnant, we need to work with a multitude of providers, including the obstetrician, the cardiologist (if a patient has a mechanical heart valve), and eventually the anesthesiologist who will be involved in delivery, to plan for limited exposure of the baby to the drug, safe use of neuraxial anesthesia, and decreased bleeding during delivery.
Dr. Murphy: Managing anticoagulation for pregnant patients means navigating the challenges during pregnancy, then another set of challenges during delivery, when we have to shift gears and think about the risk-benefit ratio a bit differently than in the antepartum period. Dr. Connors, how do you adjust anticoagulation as delivery nears?
Dr. Connors: We try to switch patients on enoxaparin for VTE that developed during pregnancy to subcutaneous UFH 250 IU/kg twice-daily. Sometimes it feels like we “cave” to the obstetricians and the anesthesiologists because they are unfamiliar with the pharmacokinetics of UFH, and they don’t often see patients who require anticoagulation. In many cases, we could probably continue LMWH, but the shorter half-life of UFH and ability to monitor pharmacologic activity with partial thromboplastin time (PTT) makes obstetricians and anesthesiologists more comfortable. Only 2 percent of the total pregnant population has a VTE, so it’s not something they handle every day.
For example, I had one patient on full-dose, subcutaneous UFH in preparation for delivery whose obstetrician suddenly wanted to induce delivery after observing some suboptimal fetal stress tests, but the patient had already taken her morning dose of UFH and her PTT was around 120 – which freaked everyone out! As hematologists, we know that this is what happens with twice-daily, full-dose UFH: There are peaks and valleys in the PTT, rather than a continuous level as seen with intravenous UFH, aiming for a PTT of 80. Eventually, the obstetrics team just waited for the enoxaparin’s effects to wear off before performing the cesarean section, and the mother and baby were healthy.
Peri-delivery management is tough. I used to try to convince the obstetrics team to stick with enoxaparin the whole way through pregnancy and up to delivery, especially for prophylactic dose and for planned induction or cesarean section. Because the obstetrician or anesthesiologist are so much more comfortable with prophylactic dose UFH, which at our institution is 10,000 IU twice-daily, it is hard to insist on sticking with LMWH.
Dr. Murphy: This reiterates what I mentioned earlier – good communication with patients and the health-care team is absolutely key! I ask all my pregnant patients at their first visit about their birth plan, including whether they want an epidural. Obviously, plans change in the moment, but I start these conversations early so we have ample time to discuss the risks and benefits of every option.
Women should be advised to stop their LMWH when contractions begin. For a planned delivery, most guidelines recommend stopping enoxaparin within 24 hours, and this is especially important if an epidural is desired. Like you, Dr. Connors, I try to continue with LMWH as opposed to switching to UFH, although I have certainly done this as well based on shared decisions with other providers.
Dr. Connors: That’s a great point. At our institution, we started a program in which patients on anticoagulation, no matter what the dose, go for an anesthesia consult prior to delivery, since the biggest concern with neuraxial anesthesia is the risk of epidural hematoma in an anticoagulated patient. Then, when a patient arrives in the middle of the night, everyone is more comfortable because there is an anesthesia plan recorded in the patient’s chart – in addition to the obstetrician’s and hematologist’s recommendations.
Dr. Murphy: That’s fantastic. We recently started an obstetric hematology clinic in which we follow every pregnant woman who has a VTE during pregnancy. We also have wonderful working relationships with our obstetric and anesthesia teams and hold multidisciplinary conferences where we discuss mutual patients and prepare for what’s going to happen during delivery. These conversations can mitigate much of the confusion and make everyone – the patient and her providers – more comfortable.
- Bates SM, Middeldorp S, Rodger M, et al. Guidance for the treatment and prevention of obstetric-associated venous thromboembolism. J Thromb Thrombolysis. 2016;41:92-8.
- Gherman RB, Goodwin TM, Leung B, et al. Incidence, clinical characteristics, and timing of objectively diagnosed venous thromboembolism during pregnancy. Obstet Gynecol. 1999;94:730-4.
- Alshawabkeh L, Economy KE, Valente AM. Anticoagulation during pregnancy: evolving strategies with a focus on mechanical valves. J Am Coll Cardiol. 2016;68:1804-13.
- Royal College of Obstetricians & Gynaecologists. Reducing the risk of venous thromboembolism during pregnancy and the puerperium: Green-top guideline No. 37a, April 2015. Accessed November 4, 2017, from https://www.rcog.org.uk/globalassets/documents/guidelines/gtg-37a.pdf.
- Dempfle CE. Minor transplacental passage of fondaparinux in vivo. N Engl J Med. 2004;350:1914-5.
- Brenner B, Hoffman R, Carp H, et al. Efficacy and safety of two doses of enoxaparin in women with thrombophilia and recurrent pregnancy loss: the LIVE-ENOX study. J Thromb Haemo. 2005;3:227-9.
- Gandara E, Carrier M, Rodger MA. Intermediate doses of low molecular weight heparin for the long term treatment of pregnancy thromboembolism. A systematic review. Thromb Haemost. 2014. 111:559-61.
- ClinicalTrials.gov. Comparison of low and intermediate dose low-molecular-weight heparin to prevent recurrent venous thromboembolism in pregnancy (NCT01828697). Accessed November 9, 2017, from https://clinicaltrials.gov/ct2/show/NCT01828697.