Left-Sided Bioprosthetic Valves Often Dysfunctional in Pregnant Women

Among women of reproductive age who have bioprosthetic valves (BPVs), structural dysfunction is common and increases risk for cardiac events in pregnancy compared with normally functioning BPVs, particularly if the dysfunction is in a left-sided valve, according to data from the Canadian Cardiac Disease in Pregnancy (CARPREG) study.

Earlier studies have shown that young patients have reduced durability of BPVs and higher rates of reoperation than older patients. Mechanical prosthetic valves are typically considered first choice in younger individuals with no contraindication to anticoagulation because they are associated with better longevity. However, due to the need for anticoagulation and increased risk for valve thrombosis during pregnancy, guidelines recommend considering bioprosthetic valves in young women despite sparse data on pregnancy outcomes, Barbara Wichert-Schmitt, MD (Kepler University Hospital, Linz, Austria), and colleagues report in the Journal of the American College of Cardiology.

“We showed for the first time that structural valve dysfunction (SVD) of bioprosthetic valves is common at the time a young patient gets pregnant, as 40% of all aortic bioprosthetic valves had SVD and 29% of mitral BPVs, resulting in 44% of women with any left-sided BPV . . . [having] a dysfunctional BPV at the beginning of pregnancy,” Wichert-Schmitt said in an email.

Having a dysfunctional BPV was associated with a high risk for adverse maternal cardiac and fetal events (47% and 60%, respectively), including serious cardiac events. Other adverse events, including premature birth, low birth weight, and respiratory distress syndrome, also were more common in women with valvular dysfunction of a left-sided BPV.

The researchers say earlier degeneration of left-sided compared with right-sided BPVs likely relates to higher functional demand in high-pressure versus low-pressure circulation.

“For those patients who have decided to have a BPV implanted, it is absolutely necessary to have regular cardiology surveillance with an echocardiogram,” Wichert-Schmitt noted, adding that valve function should be checked prior to pregnancy in patients contemplating having a child.

In an editorial accompanying the study, Deirdre J. Mattina, MD (Cleveland Clinic Hillcrest Hospital, Mayfield, OH), and colleagues say multidisciplinary care coordinated by cardio-obstetric clinics or tertiary referral centers with experience in high-risk pregnancies “should be considered” for patients with structural valve dysfunction of a left-sided BPV. Although there are no data to support improvements in pregnancy outcomes for women who receive such care, they note that preconception counseling and access to timely prenatal and postpartum care “have been recognized as critical components of pregnancy outcomes, and these resources may not be readily available or feasible for general community cardiologists.”

They say the study findings “will enrich informed, individualized, shared decision-making when choosing the type of prosthetic valve as well as expectant pregnancy management for those patients with BPVs placed before pregnancy.”

CARPREG Outcomes

The study consisted of 125 women (mean age 31 years) with one or more BPVs who were observed from their first antenatal visit through 6 months after delivery at one of two Canadian tertiary care cardio-obstetric centers. All were enrolled in the CARPREG study’s subset of women with preexisting heart disease. The mean time from valve implantation to pregnancy was 6 years.

In the 34 women with left-sided BPVs, 17 had aortic valves, nine had mitral, and eight had both aortic and mitral valves. The majority of women in the study had right-sided BPVs, with 86 of those being pulmonary and five being tricuspid. BVPs were the result of acquired or simple congenital cardiac disease in the majority of women with left-sided valve disease, while most women with right-sided BPVs had complex congenital cardiac disease, with tetralogy of Fallot accounting for more than half of the cases.

At 6-month follow-up, 13% of women had a primary adverse maternal cardiac event (a composite of maternal cardiac death, cardiac arrest, left-sided or right-sided heart failure [HF], supraventricular or ventricular arrhythmia requiring treatment, cardiac thromboembolism, and stroke or transient ischemic attack). When valve dysfunction was present in those with left-sided BPV, the rate of any cardiac event was 47% compared with just 5% when the valve was functioning normally (P = 0.01). However, the situation was different for women with right-sided BPV, where there was no difference in cardiac event rates between those with and without valve dysfunction (P = 0.67).

Cardiac events occurred in 24% of women in the left-sided BVP group and in 9% in the right-sided group (P = 0.03), driven by higher rates in women with SVD. Factors associated with greater risk of cardiac events included age over 35 years, a prior cardiac event, and NYHA functional class III or IV.

One cardiac death occurred in the left-sided group and none in the right-sided group. Congestive HF occurred at a rate of 12% for left-sided versus 5% for right-sided BVPs (P = 0.25), while the rate of arrhythmia was 9% and 3%, respectively (P = 0.34).

Fetal events were not significantly higher in pregnancies with SVD versus no SVD, or in women with left-sided versus right-sided BPVs. However, there was a much higher rate of fetal events in women who had left-sided BPVs and SVD versus left-sided BPVs with no SVD (60% vs 16%; P = 0.012). Two women with BPV dysfunction required urgent valve replacement during pregnancy that resulted in the death of both fetuses.

Secondary adverse cardiac events, which consisted of urgent invasive cardiac procedures during pregnancy or within 6 weeks after delivery, occurred in 9% of the left-sided BVP group and in none in the right-sided group (P = 0.01).

Making the Right Choice

“The right prosthesis choice for young women of childbearing age with a significant left-sided valvular lesion still remains a difficult decision, particularly in very young patients, who might not even consider pregnancy within the following 10 or even 20 years, and therefore might require recurrent valve replacement surgeries due to SVD,” Wichert-Schmitt said. “However, for aortic valve lesions, I do think a Ross operation is a reasonable option in the hands of an experienced cardiac surgeon.”

As the study showed, on the first antenatal visit, only one woman who had a Ross operation, or pulmonary autograft, was found to have dysfunction of the autograft in the aortic position. According to Wichert-Schmitt and colleagues, the better outcomes for Ross patients are in line with long-term outcome studies showing improved survival for the Ross procedure versus bioprosthetic or mechanical valves.

Mattina and colleagues say that the study is a good start, but that data are needed in more-diverse populations of pregnant women with valvular heart disease.

“Although we know that maternal mortality is disproportionately higher in non-Hispanic Black women, pregnancy outcomes in ethnically diverse women with BPVs are not well known,” they write. “Further data are also needed on percutaneous treatment options during pregnancy and medical management of SVD beyond 6 months postpartum.”