Pregnancy with prosthetic heart valves: navigating risks, anticoagulation, and valve function in a growing population

ABSTRACT

Pregnancy in patients with prosthetic heart valves presents complex challenges requiring multidisciplinary care and individualized decision-making. While bioprosthetic valves are often preferred in women of childbearing age to avoid anticoagulation-related fetal risks, these valves are susceptible to structural degeneration potentially accelerated by the physiologic demands of pregnancy. Conversely, mechanical valves offer durability but necessitate lifelong anticoagulation, posing substantial risks of valve thrombosis, fetal hemorrhage, and warfarin embryopathy. The Ross operation, which can be considered for individuals with advanced aortic valve disease, is an appealing option yet is offered only in highly specialized centers. In this review, we present contemporary data on maternal and fetal outcomes, valve function, and anticoagulation strategies in pregnant patients with mechanical and bioprosthetic heart valves. We highlight geographical and knowledge gaps regarding type of valve utilization, optimal anticoagulation in pregnancy, durability of modern valve designs, and long-term reoperation risk. Future research priorities include uniform global approach with improved access to contemporary surgical and medical solutions in low- and middle-income countries, safety of direct oral anticoagulants, improved surveillance of valve function during pregnancy, and comparative studies of bioprosthetic valve models. By consolidating evolving evidence, we aim to support informed decision-making and multidisciplinary management for this high-risk patient population.

Plain Language Summary

Many women with heart valve replacements (also called prosthetic valves) are now living long, healthy lives and may wish to become pregnant. However, pregnancy puts extra strain on the heart and creates special challenges for women with these valves. This is because the heart needs to pump more blood, and some valves – especially mechanical ones – require blood-thinning medications that can affect both the mother and the baby.

In this review, we summarize what is currently known about how different types of prosthetic valves behave during pregnancy. We discuss the changes that happen in a woman’s body during pregnancy and how these may impact valve function. We also review the risks and benefits of the two main types of valves – mechanical and bioprosthetic – and how they influence pregnancy planning, care during delivery, and long-term outcomes.

We highlight the importance of a team-based approach to care, including heart doctors, pregnancy specialists, and others, to help ensure the best outcomes for both mother and baby. We also emphasize how important it is for women with prosthetic valves to meet with their care team before becoming pregnant, so they can receive advice that is tailored to their specific needs.

1. Introduction and epidemiology

Cardiovascular disease is one of the leading causes of maternal mortality during pregnancy in the United States and is estimated to affect 1–4% of pregnancies worldwide with higher burden of disease in low- and middle-income countries due to factors such as rheumatic heart disease and limited healthcare access [1,2]. According to Registry of Pregnancy and Cardiac Disease (ROPAC), an international prospective observational registry of pregnant women with heart disease, valvular heart disease was observed to be the third most common cause of cardiac disease after pulmonary arterial hypertension and cardiomyopathy, with a prevalence of 28.7% (1648/5739) [2].

Data from the Global Burden of Disease in 2021 showed that the most prevalent form of valvular heart disease worldwide is rheumatic heart disease, with approximately 54.8 million cases reported globally and preferentially burdening regions with a low sociodemographic index. High sociodemographic index areas are more affected by functional and degenerative heart disease such as calcific aortic valvular disease (including calcific bicuspid aortic valvular disease) and degenerative mitral valve disease [3]. More advanced age, preexisting congenital heart disease and atherosclerotic risk factors all contribute to the prevalence of functional and degenerative heart disease in this patient population [2].

For women with moderately-severely dysfunctional native valve disease who are of childbearing age and are considering pregnancy, bioprosthetic heart valves are often recommended to reduce the known high risks of mechanical heart valves in pregnancy, yet there remains a limited understanding of how pregnancy uniquely affects the function and outcomes of prosthetic valves and women with bioprosthetic heart valves still experience higher rates of adverse events during pregnancy as compared to those without valve disease [4]. Patients with mechanical heart valves have significantly worse outcomes, with maternal mortality rates being greater than 1% and up to 9%, even in contemporary series, making it among the highest risk maternal conditions for pregnancy [5–7].

For women coming into pregnancy with prosthetic heart valves dysfunction, the hemodynamic changes of pregnancy can lead to new or worsening clinical symptoms or to decompensation [2]. Current guidelines provide general recommendations but lack robust evidence tailored to the physiologic demands of pregnancy or long-term valve durability in this context, yet valvular disease can be managed successfully in well-resourced settings with counseling and planning around the risks of various valvular interventions and anticoagulation [8].

2. Physiologic changes in pregnancy, labor, and delivery and the impact on prosthetic valve function

The hemodynamic changes that take place during pregnancy are related to increased cardiac output (CO) and decreased systemic vascular resistance (SVR) that begin early in the first trimester and typically culminate in the end of the second trimester (Figure 1). The 30–50% change in CO is mainly driven by an increase in stroke volume in the first half of pregnancy due to increased total blood and plasma volume, partially from heightened activation of the renin-angiotensin-aldosterone system. This is followed by a gradual rise in heart rate which is driven by increased sympathetic activation [9,10] that continues to augment the CO after the first half of pregnancy. During labor, uterine contractions lead to further increases in CO through the compression of large pelvic veins, translating to increased venous and increased stroke volume.

Figure 1.

Physiologic hemodynamic changes during pregnancy.

SVR decreases throughout the first trimester and plateaus for the remainder of the pregnancy due to elevated levels of prostacyclin, estrogen, progesterone, and relaxin, which act as arterial vasodilators. Ultimately, the decrease in SVR is offset by the increased CO such that mean arterial pressure is only slightly reduced. SVR acutely increases with each contraction during labor and returns to pre-gestational values soon after delivery. Similar changes take place in the pulmonary vasculature, with a decrease in the pulmonary vascular resistance but an increase in the CO that sum up as unchanged mean pulmonary arterial pressures, and return to normal pulmonary vascular resistance values postpartum.

These changes can pertain an important hemodynamic and structural impact on preexisting native and prosthetic valvular dysfunction, as well as on the planning of valvular interventions. Patients who were previously symptom free or even undiagnosed as having valve disease or dysfunction may come to clinical attention due to new or worsening symptoms. In patients with moderate or greater valve stenosis, the increase in the CO and the drop in the SVR will typically lead to an increase in the trans-valvular gradient, which in turn can manifest as reduced exercise tolerance, lightheadedness, chest pain or even heart failure. Due to the lower pressures typically generated by the right ventricle, native or prosthetic pulmonic valve stenosis without right ventricular dysfunction is generally better-tolerated than native or prosthetic aortic valve stenosis [8]. In patients with moderate or greater valve regurgitation and ventricular dysfunction, the ventricular end-diastolic pressure and the atrial pressure increase, which will manifest as worsening dyspnea or even pulmonary congestion for left sided valve disease and as systemic congestion for right sided valve disease [11,12]. Most patients with valve regurgitation and normal ventricular function will experience minimal symptoms or may even feel slightly better due to the decrease in systemic or pulmonary resistances which can reduce the regurgitant fraction.

Prosthetic valve degeneration can ensue or progress due to the expected hypercoagulable state of pregnancy related to higher factors I, VII, VIII, IX, X, XII levels and lower plasminogen activator, protein S and Protein C as well as venous stasis, conferring to a 4 to 5-fold risk of thrombosis [13]. While this is mainly a concern for those with mechanical prosthetic valves, micro-thrombosis of bioprosthetic valves leading to tissue degeneration remains a concern.

3. Anticoagulation options and outcomes in mechanical prosthetic valves

While mechanical prosthetic heart valves are not the preferred choice for women with advanced valvular disease who desire pregnancy as mentioned above, mechanical prosthesis in the aortic, and more commonly the mitral position are not infrequently encountered throughout the world. Mechanical heart valves carry a significant risk of maternal mortality, most frequently from complications of mechanical thrombosis [4]. Long-term, uninterrupted anticoagulation is required to mitigate risk of thrombosis; however, anticoagulation therapy carries the risks of spontaneous abortion, maternal and fetal bleeding complications during pregnancy. Direct oral anticoagulants are contraindicated during pregnancy due to their potential of crossing the placenta and causing fetal bleeding and teratogenicity; therefore, warfarin and low molecular weight heparin (LMWH) are the two anticoagulation strategies employed in this group [14].

Warfarin is the most effective anticoagulant in preventing mechanical valve thrombosis in patients with mechanical heart valves during pregnancy. However, in approximately 6% of cases, warfarin portends the risk of warfarin embryopathy (characterized by nasal hypoplasia and epiphysis stippling), in addition to miscarriage, spontaneous abortion or fetal intracranial hemorrhage. Given that warfarin embryopathy develops in the first trimester of pregnancy, substituting heparin for warfarin either at or prior to 6 weeks and continued at 12 weeks, was found to eliminate the risk of warfarin embryopathy [4]. Additionally, warfarin’s fetal toxicity is dose dependent, with significantly lower risk of fetal complications (including spontaneous abortions and warfarin embryopathy) with a dose of <5 mg [15,16]. To mitigate the risks of fetal intracranial hemorrhage with full warfarin anticoagulation during delivery, guidelines recommend switching to a heparin formulation 1-week prior to a planned delivery. If a pregnant patient goes into labor while fully anticoagulated with warfarin, warfarin reversal and cesarean section is appropriate to reduce the risk of fetal intracranial hemorrhage [4].

LMWH is the other frequently utilized anticoagulation strategy employed in pregnant patients for its low risk of fetal adverse effects, with similar rate as low-dose (<5 mg) warfarin. While popular in many centers, it requires a meticulous follow up of trough anti-Xa levels, as insufficient dosing is associated with higher incidences of valve thrombosis (4–9%). Additionally, even with meticulous monitoring, valve thrombosis remains a significant risk; patients with therapeutic peak anti-Xa levels were found to have subtherapeutic trough levels [16–18]. It is unclear whether this discrepancy mediates this risk, or whether peak levels are not sufficient to prevent mechanical valve thrombosis. Hence, LMWH is recommended mainly for consideration in patients with warfarin doses greater than 5 mg, given the maternal risks of thrombosis through the use of LMWH balances the fetal risk of warfarin that comes with higher doses [4]. Figure 2 summarizes the fetal complication rates across anticoagulation regimens. Balancing fetal and maternal health effects with required anticoagulation for prosthetic mechanical valves is thus challenging and requires a multidisciplinary approach with the involvement of experienced Cardiologists, Maternal-Fetal Medicine experts, Hematologists, and Anti Coagulation Centers.

Figure 2.

Comparison of fetal complication rates across anticoagulation regimens (based on data from references [15–18]).

4. Maternal and fetal outcomes in patients with bio-prosthetic heart valves

Although bioprosthetic heart valves are preferred in patients of childbearing age desiring pregnancy given the lack of need for systemic anticoagulation [8], data on valve longevity and maternal and fetal adverse outcomes is scarce.

Pregnancy is purported to accelerate the rate of structural valve degeneration (SVD) in bioprosthetic valves, primarily through increased wear-and-tear due to the increase in cardiac output during pregnancy [19,20] (Figure 3). One study showed that in a cohort of young women with bioprosthetic valves, structural valve degeneration was present in 27% at the first antenatal visit and negatively affected pregnancy outcomes with higher rates of adverse outcomes present in left-sided bioprosthetic valves [21]. In another small retrospective study of 34 pregnant patients and matched controls conducted by Price et al., the authors reported that pregnancy was associated with an almost 4-fold increase in rate of aortic, mitral and pulmonary valve degeneration as measured through structural valve degeneration scores [20]. While the exact cause of this is unknown, the increased shear stress from elevated cardiac output and increased blood volume is purported to cause more physical wear-and-tear. The formation of microthrombi in the setting of hypercoagulability and activation of immune response causing proteolytic degradation are also possible physiologic contributions to this phenomenon.

Figure 3.

Suggested mechanisms for structural valve degeneration during pregnancy.

Studies measuring long-term clinical outcomes, however, such as time to reoperation, show conflicting results. Badduke et al. found that freedom from reoperation at 10 years parallels freedom from SVD (20.3% for the pregnant cohort versus 64% for the non-pregnant cohort, p < 0.05) [22], but other studies with long-term follow up between 5 and 10 years did not find any differences in rates of structural valve degeneration or time to reoperation [19,23–25].

5. Rheumatic valve disease and valve replacement strategies in low- and middle-income countries

Rheumatic heart disease remains highly prevalent in low- and middle-income countries (LMICs), where it is a leading cause of cardiovascular morbidity and mortality, particularly among young adults and children. Recent data from sub-Saharan Africa and other low- and middle-income countries show that surgical outcomes can be acceptable, but major adverse valve-related events remain common, especially with bioprosthetic valves [26]. Mechanical valve replacement is generally more cost-effective and associated with better long-term outcomes if anticoagulation can be managed, but both options are suboptimal compared to valve repair, which is technically demanding and not widely available.

Access to prosthetic valve implantation and cardiac surgery is severely limited in most LMICs. Only a minority of patients with severe RHD receive surgical intervention, and the availability of cardiac surgical facilities is restricted to a few countries, often relying on visiting mission teams rather than sustainable local programs [27–30]. Even when surgery is available, the choice between mechanical and bioprosthetic valves is complicated by socioeconomic factors: mechanical valves require lifelong anticoagulation, which is challenging due to poor access to monitoring and high rates of nonadherence, while bioprosthetic valves degenerate rapidly in young RHD patients, often necessitating reoperation that is rarely feasible.

6. Discussion and future perspectives

Significant advances in valve design, surgical techniques, and anticoagulation strategies over the past few decades have improved outcomes for pregnant patients with prosthetic valves. However, addressing the risks of valve thrombosis and fetal/maternal morbidity in cases of mechanical prosthetic valves and increasing durability of prosthetic heart valves remain critical challenges in further advancing this field. Optimizing anticoagulation strategies is critical in improving outcomes in patients with mechanical valves; warfarin carries the risk of fetal teratogenicity and bleeding, and LMWH is an ineffective option in preventing complications of mechanical valve thrombosis. Closing these gaps in research would allow for less maternal and fetal burden of risk in women requiring >5 mg of warfarin to maintain therapeutic anticoagulation. There also remains a pressing need for prospective studies to evaluate the role of direct oral anticoagulants, which have been avoided in pregnancy due to limited safety data, and to further study the mechanisms by which LMWH is ineffective prophylaxis for valve thrombosis.

Updated, multicenter, prospective studies on short- and long-term outcomes in patients with bioprosthetic valves are necessary given the changing landscape of prosthetic valve models over the past decade and emerging bioprosthetic valves with enhanced durability. More transparency into the timeline of risks and benefits of models would enhance discussions around risks and benefits of mechanical versus bioprosthetic valve implantation in patients of child-bearing age, and potentially avoid the complex risk-benefit discussions around lifelong anticoagulation during the intrapartum period.

Pre-conception counseling is a critical component of care for women with prosthetic heart valves who are considering pregnancy. It allows for a comprehensive assessment of maternal and fetal risks, evaluation of valve function, and optimization of medical therapy prior to conception and also facilitates shared decision-making regarding valve type, anticoagulation strategies, and timing of pregnancy. Early counseling done in the Cardiolo-Obstetrics clinic for adult patients, or in the setting of transition programs by pediatric or adult Cardiologists or even by primary care providers in the absence of the former for young adult or teenagers can ensure that young women are informed about the potential complications and can engage in multidisciplinary planning to support safer maternal and neonatal outcomes. In LMIC’s, there remains an urgent need to expand access to affordable, high-quality surgical care and to develop context-appropriate solutions, such as low-cost prosthetic valves and improved health system infrastructure. Addressing these gaps through research and multidisciplinary collaboration will be crucial in improving maternal and fetal outcomes, providing more informed guidance and optimizing care delivered to patients who are pregnant or of child-bearing age.

7. Conclusion

Pregnancy in patients with prosthetic heart valves presents a complex interplay of hemodynamic demands, anticoagulation challenges, and long-term valve considerations. Mechanical valves offer durability but introduce significant maternal and fetal risks due to the need for anticoagulation, while bioprosthetic valves mitigate those risks but may face accelerated degeneration. The physiologic changes of pregnancy can unmask or exacerbate valvular dysfunction, requiring proactive surveillance and multidisciplinary planning. Optimal outcomes hinge on individualized risk assessment, timely patient counseling, and coordinated management by cardiology, maternal-fetal medicine, hematology, and anesthesia teams. Ongoing research into safer anticoagulation strategies, enhanced valve durability, and long-term maternal outcomes is critical to guide future care and improve decision-making for women with prosthetic valves considering pregnancy.

Funding Statement

This paper was not funded.

Article highlights

• Pregnancy in women with prosthetic heart valves poses significant maternal and fetal risks, especially with mechanical valves requiring anticoagulation.

• Bioprosthetic valves are preferred in women of childbearing age to avoid the fetal risks of anticoagulation but may undergo accelerated degeneration during pregnancy.

• Physiologic changes in pregnancy – including increased cardiac output and hypercoagulability – can unmask or worsen valvular dysfunction, necessitating close surveillance.

• Mechanical valves offer durability but carry high risks, including maternal thrombosis and fetal complications, making anticoagulation strategy selection critical.

• Low-molecular-weight heparin is safer for the fetus but less reliable for thrombosis prevention in mechanical valves compared to warfarin, highlighting a key clinical dilemma.

• Pregnancy may accelerate structural valve degeneration in bioprosthetic valves, though data are mixed regarding long-term clinical impact and reoperation rates.

• Women in low- and middle-income countries face disproportionately high risks, due to limited access to surgery, anticoagulation monitoring, and valve repair expertise.

• Pre-conception counseling is essential to assess risks, discuss valve choice and anticoagulation plans, and optimize outcomes through multidisciplinary care.

• There is an urgent need for better research on valve durability, safe anticoagulation options during pregnancy, and maternal-fetal outcomes in diverse settings.

• Multidisciplinary management and individualized planning are crucial for achieving optimal maternal and fetal outcomes in this high-risk population.

Author contributions

Ishani Joshi:

- Made a significant contribution to the work reported, whether that’s in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas.

- Have drafted or written, or substantially revised or critically reviewed the article.

Jonathan Buber:

- Made a significant contribution to the work reported, whether that’s in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas.

- Have drafted or written, or substantially revised or critically reviewed the article.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.