Transcatheter or Surgical Aortic Valve Replacement in Pregnant Women? A Comprehensive Review of the Current Literature

© 2024 HMP Global. All Rights Reserved.
Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of the Journal of Invasive Cardiology or HMP Global, their employees, and affiliates. 

J INVASIVE CARDIOL 2024. doi:10.25270/jic/24.00065. Epub October 25, 2024.

Surgical aortic valve replacement (SAVR) has traditionally been the gold standard for managing severe aortic stenosis (AS); however, it may carry maternal and fetal risks during pregnancy. Transcatheter aortic valve replacement (TAVR) has been proposed as an alternative in this patient population, though its role during pregnancy remains largely undefined. This review aims to consolidate the current literature on the role of TAVR during pregnancy, focusing on its potential benefits and limitations. Electronic database search identified 10 case reports of pregnant women who underwent a TAVR procedure. All pregnancies resulted in uncomplicated deliveries with no fetal complications. The reported post-procedural complications included maternal left bundle branch block, premature rupture of membrane, and residual paravalvular leak. In conclusion, TAVR may be a feasible option for the treatment of aortic stenosis and failed aortic valve bioprosthesis, as demonstrated in the published case reports. Comprehensive patient care and preoperative assessment by heart and high-risk pregnancy teams remain essential in mitigating maternal and fetal risks.

Aortic stenosis (AS) is the most prevalent valvular heart disease (VHD) in developed countries.¹ It is differentiated as mild, moderate, severe, or critical based on echocardiographic measures of pressure, flow, and calculated valve area. Surgical aortic valve replacement (SAVR) on cardiopulmonary bypass (CPB) has historically been established as the gold standard treatment option for this condition, however, transcatheter aortic valve replacement (TAVR) has been proposed as an alternative treatment option to reduce maternal and fetal risks during open surgery.² While pregnancy is well tolerated in women with mild or moderate AS, women with severe AS are at a higher risk of developing cardiac complications, such as decompensated heart failure or atrial arrhythmias. In a Canadian cohort of 39 women representing 49 pregnancies, cardiac complications were observed in 10% of the women with severe AS.³ Therefore, due to the increased hemodynamic burden placed on the heart during pregnancy, aortic valve intervention may become necessary in pregnant women with severe AS.⁴ TAVR is usually indicated in moderate- to high-risk patients, who are typically of advanced age, and is rarely performed in women of reproductive age irrespective of their pregnancy status; this is attributed to the rarity of severe AS in this age group and the lack of data that support long-term outcomes in very young patients.⁵ In this review, we assess the clinical outcomes of TAVR, as an alternative to SAVR, in pregnant women with severe AS with a focus on the potential benefits and limitations of this technique.

Search strategy 

In December 2023, a literature search was conducted on PubMed (National Institutes of Health), Medline (National Library of Medicine), Embase (Elsevier), Google Scholar (Alphabet, Inc.), and Cochrane (Wiley) to identify studies that reported the use of TAVR in pregnant patients. The following keywords were used: “transcatheter aortic valve replacement” OR “TAVR” OR “transcatheter aortic valve implantation” OR “TAVI” AND “pregnancy” OR “pregnant”. We also conducted backwards snowballing by reviewing relevant references cited in the eligible primary articles.

Study selection and criteria

A study characteristic table was created (Table 1). Eligible studies included pregnant patients who underwent the TAVR procedure without restriction on the publication date. Other reviews and non-English language articles were excluded. 

Data extraction

Data extracted included the baseline demographics of the study patients, study design, and outcomes. The outcomes of interest included procedural complications, maternal and fetal post-procedural complications, and delivery mode.

Ten case reports that were published between 2016 to 2023 were identified. Study findings and outcomes are summarized in Table 2. The age range across the case reports was 22 to 35 years, and the gestational age at TAVR ranged from 12 to 29 weeks. Among the case reports that reported echocardiographic values, the aortic valve area ranged from 0.41 to 1.0 cm². The size of the TAVR valves ranged from 20 to 34 mm.  Fluoroscopy time pre-TAVR ranged from 10.3 to 21.3 minutes, radiation dosage (air kerma) from 16 to 548 mGy, and contrast usage from 14 to 87 mL. Six case reports did not use post-dilation.^{4,6,7,9,13,14} Three case reports reported the use of the TAVR device in a native bicuspid aortic valve (BAV),^7,11,14 while the remaining cases reported the deployment of TAVR in degenerated previously implanted bioprosthetic valves.^{4,6,8-10,12,13} There was no reporting of the coronary heights or sinus of Valsalva in any of the 10 cases.

The balloon-expandable Sapien valve (Edwards Lifesciences) was used in 6 cases,^{4,6,8,9,12,14} the first generation self-expanding CoreValve valve (Medtronic) in 3 cases,^7,10,11 and the third generation self-expanding Evolut FX valve (Medtronic) in 1 case.¹³ In all of the cases, the delivery of the infant was healthy and uncomplicated.

Ghandi et al discussed transient fetal bradycardia for 48 hours with neonatal intensive care unit observation that self-resolved eventually.⁶ Zhong et al reported mild maternal perivalvular insufficiency without any hemodynamic repercussions 1-month post-procedure.¹⁰ Authors also reported premature rupture of the membrane at 36 weeks, compromising the onset of fetal delivery;¹⁰ a follow-up transthoracic echocardiogram at 1 month demonstrated a peak velocity of 2.2 m/s, with peak and mean gradients of 20 and 12 mmHg, respectively.¹⁰ The estimated valve area was 1.26 cm², and the dimensionless valve index (DVI) was 0.40.¹⁰ Hodson et al reported a new onset of left bundle branch block (post-TAVR) without atrioventricular block and the need for permanent pacing.⁷ Berry et al reported residual paravalvular leak (PVL) without hemodynamic compromise.⁴ No other significant maternal complications were reported by any of the other case reports

Most cases reported symptom improvement post-TAVR, notably in terms of shortness of breath.^{4,6,9,10,12,14} Symptom improvement correlated with a marked reduction in both peak and mean aortic gradient post-TAVR.^4,6,8-10 Hoover et al reported stable echocardiograms at 6 months and 1-year post-TAVR with improvement of left ventricular hypertrophy and left ventricular function. In the case by Sajja et al, the patient was carefully monitored in the clinic and was switched to enoxaparin injections 3 months following the TAVR procedure due to concerns of valve thrombosis. Three months after the procedure, her aortic valve gradients remained high, with a mean of 20 mm Hg and a peak of 38 mm Hg. The heightened risk of valve thrombosis was linked to the valve-in-valve approach, under-expansion of the valve, and the hypercoagulable state induced by pregnancy. Follow-up was unremarkable in 5 cases,^6,8,9,12,14 not available in 3 cases,^7,10,13 and not conducted in 2 cases.^4,11 Post-procedural medical therapy typically included aspirin (81 mg per day)^6-8,10,12,13 and low molecular weight heparin.^6-8,10,12,13

Maternal and fetal risks associated with severe aortic stenosis

Pregnancy is typically associated with a steady increase in plasma volume and cardiac output until the end of the second trimester, during which the cardiac output plateaus at 30% to 50% higher levels than pre-pregnancy.³ In cases of severe AS, fixed stroke volumes and limited increases in cardiac output result in elevated ventricular systolic and diastolic filling pressures.¹⁵ During labor, cardiac output is further increased by 40% to 50% via sympathetic responses arising from pain or uterine contractions.¹⁵ After delivery, preload is increased as a result of decompression of the inferior vena cava.¹⁵ Maternal risks of severe AS include cardiac complications such as atrial arrhythmias, heart failure, or myocardial ischemia.^3,15 Similarly, fetal complications of severe AS include growth restriction, stillbirth, and miscarriage.¹⁵ In the study by Orwat et al, it was found that newborns of mothers with severe AS were more likely to have low Apgar scores, decreased birth weights, and smaller gestational ages, which may be due to shorter pregnancy durations and/or reduced uteroplacental blood flow.¹⁶

Risk associated with the use of CPB

The reported maternal mortality rate associated with CPB use varies from 1.5% to 5%, which is similar to non-pregnant women undergoing procedures on CPB.^6,17 However, CPB may be associated with fetal mortality (rates ranging from 16%-33%) and brain injury arising from embolism, ischemia, and inflammation, resulting in impaired energy substrate delivery.^6,18 CPB may also cause fetal acidosis and greater uterine contractions as a result of altered uterine-placental blood flow.⁶ Systemic maternal cooling and rewarming on CPB increases sustained uterine contractions, resulting in fetal hypoxia and fetoplacental insufficiency.¹⁹

Potential benefits of TAVR

It is important to note that most of the data on TAVR outcomes and survival rates are derived from the general population with no available reports to reciprocate such results in pregnant women. However, TAVR may offer benefits that may be especially advantageous for pregnant women including faster recovery, early mobilization, and potentially lower rates of surgical complications.²⁰ TAVR is also associated with improvements in quality-of-life metrics such as New York Heart Association (NYHA) class and exercise performance.^21,22

Risk of fetal irradiation during TAVR

The primary concern of performing TAVR in pregnancy is the risk of fetal irradiation, which is dependent on the gestational age and conceptus radiation dose. When possible, TAVR should be conducted after the organogenesis and neuronal stem cell proliferation phase to limit radiosensitivity.¹⁷ Pre-procedural multi-disciplinary team discussion that includes the interventional cardiology and medical physics teams is highly recommended to estimate and minimize the radiation dose and to narrow the radiation fields for minimal fetal exposure.²³ Radiation doses less than 50 mGy are considered safe for the fetus in the first trimester.¹⁴ However, in the second and third trimesters, radiation doses of up to 500 mGy are considered safe.¹⁴ Fetal complications such as intellectual disability, pregnancy loss, and growth restriction are at an increased risk with radiation doses greater than 100 mGy.⁶ Implementing strategies such as an ideal table height, using fluoro-grab over cineography, shortening the frame rate, and reducing unnecessary radiation investigations can all minimize radiation exposure to the patient and fetus.⁸ While the use of a radiation shield between the abdomen and pelvis has been recommended, the inability of the internal radiation scatter to leave the abdomen might increase fetal radiation exposure.²³

Pre-procedural clinical and echocardiography assessment

In the adult population, the clinical diagnosis of severe AS is based on echocardiographic imaging and clinical presentation to reliably assess the valvular anatomy, transvalvular flow dynamics, and cardiac function.²⁴ According to the 2020 American College of Cardiology/American Heart Association (ACC/maA) VHD recommendations, severe AS is defined by a fixed valve area with a transaortic velocity of at least 4 m/s, a mean gradient of at least 40 mm Hg, and an aortic valve area less than 1.0 cm.^2,24 In pregnant patients with severe AS, valvular intervention or percutaneous balloon aortic valvuloplasty (PBAV) is recommended if there is hemodynamic compromise or symptoms of heart failure (NHYA class III or IV).²⁴ Given the potential maternal and fetal risks that aortic valve surgery during pregnancy may incur, surgery is typically reserved for pregnant patients with severe uncontrollable symptoms (such as exertional dyspnea, reduced exercise tolerance, syncope/presyncope episodes, or angina) despite maximal medical therapy.²⁴ Due to the given risk of hemodynamic deterioration during pregnancy, serial echocardiograms are indicated during the first trimester and approximately at 28 to 32 weeks (with blood volume reaching the peak) in women with known VHD preconception or unexplained cardiac symptoms.²⁵

Although computed tomographic angiography is the gold standard pre-TAVR imaging modality, it is not recommended during pregnancy because of the risks of fetal irradiation and potential maternal breast cancer in women with metabolically active breast tissue.⁷ Thus, transoesophageal echocardiogram (TEE) and intervascular ultrasound (IVUS) are alternative options for the evaluation of the aortic annulus/root and vascular access. The safety of intravenous contrast dyes during pregnancy is not very well established and, thus, is not generally recommended. However, some studies have suggested that gadolinium and iodinated contrast mediums may not be associated with teratogenic or mutagenic adverse effects.⁷ In addition, some other investigations may be considered for pre-procedural risk stratification, such as exercise stress testing, cardiopulmonary stress testing, and B-type natriuretic peptide levels.¹⁴

Benefits, limitations, and considerations of PBAV

In patients with favorable valvular anatomy, PBAV may be considered as an initial treatment option or as a repeat intervention to provide symptomatic relief till delivery.²⁶ Specifically, an aortic valve without heavily calcified commissures and no significant valvular regurgitation may allow for a small increase in valve area with an undersized balloon.²⁶ Of note, most native aortic valves in young individuals are pliable rather than heavily calcified, which may allow for successful balloon dilation with significant reduction in gradients.²³ At present, there are no large series of PBAV in pregnancy, however, several case reports show favorable results in this population.²³ Specifically, most patients have successful induction of delivery at term, with no significant intra- or post-procedure fetal complications.²³ However, maternal complications may include mild-to-moderate aortic regurgitation or its progression, hemodynamic instability, and recurrence of stenosis.^12,23,27 Other less frequent intra-procedural complications include bleeding, stroke, mitral valve injury, arrhythmia, iliac-femoral arterial complications, and mortality.²⁷ These factors, in addition to the generally scarce data on PBAV in pregnancy, may limit the use of this procedure in this population.²⁶

If PBAV is being considered, thorough investigations are required to exclude any associated aortopathy or aortic coarctation, and the ascending aorta size must be taken into consideration due to the risk of intra-procedural aortic dissection.²⁶ Like TAVR, TEE is used to determine the aortic balloon sizing and aortic annulus measurement while limiting the dose of radiation. The use of a balloon larger than the AV annular diameter and sinotubular junction should be avoided to minimize iatrogenic injury. Balloon sizes 2 to 3-mm smaller than the maximal size are recommended, and the procedure should be aborted if new 1 to 2+ aortic regurgitation develops.²³

Long-term outcomes after TAVR

One of the major long-term concerns after TAVR implantation is prosthetic valve degeneration, especially in the context of TAVR in a previous SAVR device and repeat pregnancies. This was highlighted in the case by Zhong et al, who reported a case of TAVR in a degenerated porcine aortic root prosthesis.¹⁰ Authors argued that a repeat intervention from accelerated TAVR valve deterioration in a repeat pregnancy could be challenging because of the limitations in suitable treatment options.¹⁰ From a technical perspective, the TAVR-in-SAV or TAVR-in-TAV procedures may impose additional risks such as transcatheter valve migration or embolization, paravalvular leak, and patient-prosthesis mismatch due to high residual trans-valvular gradients.⁶ This highlights the importance of proper pre-procedural planning in these patients to precisely determine the optimal transcatheter valve type, size, and implantation height, taking into account the varying anatomy and fluoroscopic appearances of the different valves.⁶

In a study by Matta et al, authors reported that, in comparison to native valve TAVR (NV-TAVR), valve-in-valve-TAVR (ViV-TAVR) patients were likely to experience fewer post-TAVR complications, such as vascular complications, stroke, and bleeding events, in addition to lower mortality rates in the early postoperative period.²⁸ Also, ViV-TAVR was associated with lower rates of postprocedural permanent pacemaker, which may be explained by the fact that the TAVR device is inserted within the annulus of the failed bioprosthetic valve, resulting in less contact with the myocardium and less strain on the cardiac conduction system.²⁸ Similar findings were reported by Ahmad et al, who demonstrated that, in a large cohort of TAVR patients, ViV-TAVR was a safe and feasible strategy for the management of patients with failed bioprosthetic valve failure and had comparable long-term outcomes to NV-TAVR.²⁹ Of note, most of the case reports in our review included pregnant women with failing aortic bioprostheses who underwent ViV-TAVR, which may explain the high procedural success rates and the low rates of major complications in this population.

Also, substantial uncertainty exists on how women who have undergone a previous TAVR procedure will tolerate subsequent pregnancies. In their case report, Rasmussen et al described a patient who developed acute coronary syndrome late in the third trimester. The authors hypothesized that this could be attributed to the formation of a microthrombus on the TAVR device due to the hypercoagulable state of pregnancy, or calcific embolization in a deteriorated TAVR prosthesis.⁵

Timing for TAVR procedure

It is generally accepted that the most appropriate timing of TAVR in pregnancy is during the second trimester. This timing reduces the associated likelihood of spontaneous abortions and/or teratogenesis in the first trimester, spontaneous labor in the third trimester, and increased heart rate and stroke volume at 24 to 28 weeks of gestation.⁹

Limitations of the TAVR procedure

According to guidelines, active endocarditis is a contraindication to TAVR.³⁰ Additionally, TAVR for pure severe native aortic valve regurgitation remains challenging and is associated with a notable risk of transcatheter valve embolization or migration because of the lack of calcification in the native valve.³¹ Furthermore, patients with severely calcified BAV commonly have increased rates of adverse outcomes following TAVR, such as paravalvular regurgitation, aortic root rupture, valve leaflet thrombosis, increased permanent pacemaker implantation rates, and difficult coronary access and durability.³² Also, unfavorable anatomies such as small native vessel size, severe peripheral artery disease, extensive vessel and aortic tortuosity, horizontal aortic root, and concomitant aortic aneurysm may significantly impede accurate TAVR delivery implantation.³³

Pregnant patients undergoing TAVR face additional, unique limitations because of hormone-mediated changes in the aortic wall. These changes increase the risk of ascending and descending thoracic aorta dissection or perforation, particularly when stiff wires or catheters inadvertently injure the aortic wall. Additionally, despite the evolution of valve technologies and deployment techniques, the need for a permanent pacemaker (PPM) remains a risk, particularly with self-expanding valves and in patients with pre-existing electric conduction abnormalities. This dependency on PPM may lead to long-term complications such as device-related complications, tricuspid valve injury, biventricular dysfunction and dyssynchrony, increased risk of heart failure-related hospitalizations, and delayed mortality. Furthermore, pregnancy accelerates the deterioration of bioprosthetic valves, which could similarly impact THVs. If a TAVR valve deteriorates in young patients, the complexity of subsequent SAVR is increased because of the adhesion of the TAVR valve to surrounding tissues, such as the anterior leaflet of the mitral valve. This situation often necessitates aortic root replacement, coronary reimplantation, and potentially mitral valve repair or replacement, which can result in less favorable outcomes compared with initial SAVR.²³ Finally, depending on the size and anatomy of the ilio-femoral access, device-related vascular complications may occur, adding complexity to the procedure.¹³

Post-TAVR medical therapy

One of the major benefits of TAVR is the lack of need for long-term post-procedural oral anticoagulation therapy, such as warfarin, which may potentially be associated with a risk of teratogenicity, miscarriage, stillbirth, and neurodevelopmental deficits.^7,32 For short-term anticoagulation, unfractionated heparin (UFH) and low molecular weight heparin (LMWH) can be safely administered for thrombo-embolic risk prevention.³⁴ However, LMWH is preferred over UFH because of its superior safety profile and lower incidence of bleeding, heparin-induced thrombocytopenia, and heparin-related osteoporosis. However, UFH is recommended for pregnant patients with significant renal impairment.³⁵

In terms of post-procedural anti-platelet therapy, aspirin is considered safe to administer in low doses (80-160 mg/d).²³ High-dose aspirin is not recommended because it increases the risk of congenital defects, miscarriage, and fetal bleeding.⁷ Earlier reports recommended the use of dual anti-platelet therapy (aspirin and clopidogrel) after TAVR to reduce leaflet thrombosis.³⁶ However, more recent evidence showed that dual anti-platelet therapy (DAPT), as compared with aspirin alone, may be associated with an increased risk of bleeding with no additional benefits.^29,37

Anesthetic considerations during TAVR

Another major advantage of TAVR is the ability to perform the procedure under conscious sedation, thus avoiding general anesthesia (GA).³⁸ While GA does have benefits, including secure airway and ventilation control, reduced patient movement, and simpler management of hemodynamic changes in addition to the facilitation of TEE use,³⁹ it may be associated with some technical and procedural challenges. Pregnant patients are 3 times more likely to develop hypoxia as a result of increased oxygen consumption and decreased functional residual capacity; therefore, adequate preoxygenation is suggested.⁴⁰ Pregnant patients are also more susceptible to regurgitation and aspiration during induction and emergence. Increased intra-abdominal pressure in the second trimester reduces the lower esophageal sphincter tone, necessitating rapid-sequence induction along with cricoid pressure.⁴⁰ Regardless of the chosen anesthetic technique, a 15-degree left uterine displacement needs to be maintained to prevent aortocaval compression and maternal supine hypotension syndrome.⁹ The decision to perform fetal monitoring must be individualized to each patient based on gestational age, available facilities, and expected procedural complexity, as well as the American College of Obstetricians and Gynecologists guidelines.⁴¹

Multi-disciplinary team discussion

Finally, to ensure both fetal and maternal well-being, the 2020 ACC/AHA guidelines for the treatment of VHD patients suggested that a team of cardiologists, anesthesiologists, obstetricians, surgeons, neonatologists, and other providers should evaluate the available treatment options for each pregnant woman with aortic valve disease.²⁴ These discussions are essential to providing patient-tailored care that considers the peculiar clinical situation of each individual and guarantees the best possible clinical outcomes. 

Limitations

All 10 cases included in this narrative review were single case reports. Therefore, causality cannot be established because of the uncontrolled observations and unplanned nature of the descriptive reports. Also, the small sample size restricts the generalizability of the findings. Finally, all 10 cases were retrospective reports, which, by nature, are subject to recall bias, information bias, and selection bias. 

TAVR can be a feasible treatment option in pregnant women with failing aortic valve bioprostheses and, to a lesser extent, those with native aortic valve stenosis. The main limitations of this technique include the risk of fetal irradiation, potential exposure to general anesthesia, and the lack of long-term data on prosthetic valve durability in these patients. PBAV may be an alternative option in patients with favorable anatomy, however, major limitations include the development of aortic regurgitation and the recurrence of valvular stenosis. A multi-disciplinary team approach is essential to provide patient-tailored care and to mitigate maternal and fetal risks.

Aparna Kuchibhatla¹; Nazanin Soghrati¹; Yazan Saleh, BHSc²; Greggory Rushing, MD³; Mohamed Abdel Halim, MD, PhD³; Marc Pelletier, MD³; Cristian Baeza, MD³, Mohammad El-Diasty, MD, PhD, FRCS, FACC³

From the ¹School of Medicine, Queen’s University, Kingston, Ontario, Canada; ²Department of Biomedical and Molecular Sciences, Queen’s University, Kingston, Ontario, Canada; ³Cardiac Surgery Department, University Hospitals Cleveland Medical Centre, Cleveland, Ohio.

Disclosures: The authors report no financial relationships or conflicts of interest regarding the content herein.

Data availability: All data is available upon request to the corresponding author.

Corresponding author: Mohammad El-Diasty, MD, PhD, FRCS, FACC, University Hospitals Cleveland Medical Centre, 11100 Euclid Ave, Cleveland, OH 44106, USA. Email: cardiac.science.lab@gmail.com