html xmlns=”http://www.w3.org/1999/xhtml” xmlns:mml=”http://www.w3.org/1998/Math/MathML” xmlns:epub=”http://www.idpf.org/2007/ops”>
Practical practice points
1. When valve replacement becomes necessary, tissue valves are usually preferable to mechanical valves in young women intending to become pregnant because anticoagulation is not required, although reoperation after an average of 10 years is likely.
2. There is currently no ideal regimen for anticoagulation in women with mechanical heart valves in pregnancy. Women should be offered a choice between the higher rates of fetal loss associated with the use of warfarin and the higher risk of maternal valve thrombosis with subcutaneous heparin.
3. In pregnant women with mechanical heart valves who elect to use subcutaneous low-molecular-weight heparin, the dose should be at therapeutic levels and guided by monitoring of anti-factor Xa activity at least every 2 weeks. A peak (4 h post-dose) level of at least 1.0 iu/ml and a trough level of at least 0.5 iu/ml may be optimal.
Introduction
For a woman with a prosthetic heart valve, pregnancy poses risks for both the mother and the fetus. Selection of the most appropriate prosthetic valve for a woman who might become pregnant requires careful review and a detailed discussion of the relative risks with the woman, preferably before she becomes pregnant. A key issue is that bioprostheses (tissue valves) are much less thrombogenic than mechanical prostheses, and if the mother is in sinus rhythm with good ventricular function warfarin can be avoided. Pregnancy in the presence of a tissue prosthesis is associated with a much lower risk of maternal/fetal complications, but structural valve degeneration inevitably occurs (irrespective of pregnancy) and the woman will require reoperation, with all the attendant risks, at some point later.
Mechanical prostheses are much more problematic during pregnancy, and mothers are especially vulnerable to valve thrombosis and thromboembolism. The obvious advantage is their potential to last for a lifetime. Considerable controversy still exists about the most appropriate anticoagulant management during each of the three trimesters in order to avoid these life-threatening complications. No consensus exists as to the ideal anticoagulant strategy and no large prospective series are available to provide evidence-based advice. Whichever anticoagulant is chosen, there is the potential for an increased risk of valve thrombosis and thromboembolism (especially with heparin) and the potential for miscarriage, placental bleeding, or fetal embryopathy with the use of warfarin. Each type of prosthesis thus has a relative risk/benefit ratio in pregnancy, and these issues will be reviewed in detail along with proposed management strategies.
Tissue prostheses
Commonly used bioprostheses (heterografts) include porcine (pig) valves and human pericardial valves (homografts). For patients in sinus rhythm, anticoagulation is not required, although some are prescribed aspirin (usually at a dose of 75–81 mg; UK and US standard preparations). For women who have a normally functioning tissue prosthesis, normal ventricular function, and no significant pulmonary hypertension, pregnancy is often uncomplicated. Tissue valves, however, have a finite lifespan and develop structural valve degeneration. In general, mitral prostheses degenerate at a faster rate than aortic prostheses, and the rate of structural valve deterioration is considerably faster in younger (age <40 years) than in older patients.[1,2] As a result, a repeat valve replacement is inevitable, on average about a decade after the first operation. The mortality risks of a reoperation vary considerably among different centers: at the author’s institution, the risk is <1%, but it may be significantly higher in other institutions. Thus, the surgical results of the individual institution should be considered when counseling women of childbearing age about which type of valve replacement is preferable. In each case, the risks and benefits need to be carefully balanced; it should also be recognized that if a mother dies at reoperation, then the baby will be left without a biological mother.
It has been suggested that pregnancy accelerates structural valve degeneration, perhaps via an effect on calcium turnover. Badduke et al. reviewed 87 women with bioprostheses [3]: the need for reoperation for structural valve degeneration after 10 years was 59% for women who had had pregnancies vs 19% for those who did not. In women who had had a pregnancy, the mean interval from operation to reoperation was only 72.1 months. Other reports, however, failed to confirm accelerated valve degeneration during pregnancy.[4] Jamieson et al. evaluated the long-term performance of bioprostheses in women under 35 years of age to determine whether pregnancy influenced structural valve deterioration.[5] Between 1972 and 1992, 237 women received 255 tissue prostheses. Of the total number of operations, 53 were performed in women who subsequently had a pregnancy and 202 in women who were never subsequently pregnant. The mean interval from implantation to reoperation for the whole group was 99.6 months. The freedom from structural valve deterioration at 10 and 15 years for the nonpregnant group was 54% and 18%, respectively, and for the pregnant group was 45% and 34%, respectively (the p value was not significant). In a retrospective review of 232 New Zealand women followed for 1499 patient-years, North et al. also found that pregnancy did not influence structural valve deterioration, with a relative risk of 0.96.[6] Thus, while some reports suggest acceleration of valve degeneration by pregnancy, several large series suggest that pregnancy does not significantly influence the incidence of structural valve disease. This has also been the author’s experience at the Mayo Clinic, Rochester, Minnesota, USA. The explanation may simply be that the appearance of accelerated valve degeneration in association with pregnancy reflects the well-established deterioration of tissue valves in younger patients.[7] In my experience, most women during the childbearing years elect to have a tissue prosthesis, with the knowledge that approximately half of them will need a valve replacement after 10 years. A mechanical prosthesis is then selected at the time of reoperation.
Mechanical prosthetic heart valves
Pregnancy induces a hypercoagulable state and is therefore associated with an increased risk of valve thrombosis and thromboembolism; this is especially important with mechanical prosthetic heart valves that already cause a risk of clotting. During pregnancy, there is an increased concentration of clotting factors, decreased fibrinolysis, and increased platelet adhesiveness. Anticoagulation management is thus critical, and one should not extrapolate the same therapies used for patients with mechanical prostheses having noncardiac surgery to those in the pregnant state. Appropriate anticoagulation, however, is challenging because anticoagulation with warfarin can be associated with important fetal adverse effects, while heparin is a less effective anticoagulant and therefore associated with important maternal complications, particularly valve thrombosis and thromboembolism In many series, valve thrombosis and thromboembolic events occur in approximately 10% of cases and are often fatal.[8] Mitral tilting disk valves and older generation valves such as the Bjork–Shiley and Starr–Edwards valves appear to confer the highest risk of thrombotic complications. Thus, the anticoagulant strategy selected poses a difficult clinical dilemma for both women and the clinicians caring for them, and requires a detailed and informed discussion before pregnancy is contemplated. What is considered to be an “acceptable risk” to mother and baby may differ from one clinician to another, and may be very different between one woman and the next.[7]
Unfractionated heparin
Heparin is a large molecule that not cross the placenta to affect the fetus. It has a relatively short circulating half-life, however, and may be given either subcutaneously or intravenously. The use of heparin can be associated with osteoporosis [9] and an immune immunoglobulin G mediated thrombocytopenia. In early series, it was often the anticoagulant of choice in pregnancy, particularly in the first trimester, for women with mechanical prostheses in an effort to avoid the potential complications of fetal embryopathy due to warfarin. The laboratory control of anticoagulation, however, is challenging and there is wide variability in the sensitivity of activated partial thromboplastin time (APTT) reagents for monitoring the dosage and considerable variability in response to standard dosing. In addition, the APTT may vary by as much as 50% in response to the same continuous intravenous dose of heparin, with a higher anticoagulant effect at night-time. This diurnal variation may relate to a circadian rhythm in the pharmacokinetics of heparin or, more likely, to a spontaneous circadian variation in coagulation also seen in normal controls.[10] During pregnancy, the APTT response to heparin is often attenuated because of increased levels of factor VIII and fibrinogen.[11] It is thus imperative that heparin should be given at an adequate intensity, with frequent measurements of trough and peak levels. Contemporary APTT reagents are more sensitive to the anticoagulant effect of heparin, and an APTT ratio of 1.5 is inadequate. The APTT ratio should be at least 2, which correlates with an anti-factor Xa (“anti-Xa”) level of >0.55 IU/ml in approximately 90% of women.[12] Because of the variability in laboratory sensitivities for monitoring heparin, APTT alone may be insufficient and anti-Xa assays may periodically be more useful in guiding therapy. If this is done, it is recommended that the anti-Xa level be maintained between 0.35 and 0.7. It may be that suboptimal target APTT ratios account for some of the reported treatment failures.
In the author’s experience, unfractionated heparin is difficult to manage via the subcutaneous route because of wide peaks and troughs. In accordance with both the American Heart Association and American College of Cardiology Valvular Heart Disease Guidance [13] and the European Society of Cardiology Guidelines,[14] if unfractionated heparin is used, the intravenous route is preferable. The dose must be adjusted to achieve an APTT of at least two times the control (recommendation class IIa).
Chan et al. performed a meta-analysis from 1966 to 1997, looking at 976 women with 1234 pregnancies and comparing three management strategies [15]: (1) unfractionated heparin throughout pregnancy; (2) subcutaneous unfractionated heparin in the first trimester and warfarin thereafter until just prior to delivery when intravenous heparin was administered; and (3) warfarin throughout pregnancy. The risks of miscarriage and fetal loss for those on warfarin throughout pregnancy were 25% and 34%, respectively, and (perhaps surprisingly from a theoretical point of view because heparin does not cross the placenta) the use of heparin in the first trimester yielded similar results with a miscarriage rate of 25% and a fetal loss rate of 27% (Table 11.1). They reported that the use of warfarin throughout pregnancy was associated with embryopathy in 6.4% of live births but that if heparin were substituted at or prior to 6 weeks of gestation and continued up to 12 weeks then the risk of embryopathy was eliminated (Table 11.2). The overall risks of fetal loss (miscarriage, stillbirth, and neonatal death) were similar in women treated with warfarin throughout compared with women treated with heparin in the first trimester. The overall maternal mortality in this series was 2.9% and major bleeding events occurred in 2.5% of all pregnancies, most commonly at the time of delivery. The maternal risk, however, was quite different among the various anticoagulant strategies: the regimen associated with the lowest maternal risk of valve thrombosis was the use of oral anticoagulants throughout pregnancy, with a risk of 3.9% (Table 11.3). While the use of heparin between 6 and 12 weeks of gestation eliminated the risk of fetal embryopathy, the risk of maternal valve thrombosis was more than doubled, at 9.2% (Table 11.3). It thus appears that the window between 6 and 12 weeks is pivotal, since the continued exposure to warfarin resulted in an increase in the risk of fetal loss by >50%. The apparent advantages of reduced fetal loss and embryopathy gained from the use of heparin, however, appeared to be counterbalanced by an increase in maternal complications and an increased risk of thromboembolism.
| Regimen | Miscarriage | Fetal anomalies | Fetal loss |
|---|---|---|---|
| Warfarina | 196/792 (25%) | 35/549 (6%) | 266/792 (34%) |
| Heparin/warfarinb | 57/230 (25%) | 6/174 (3%) | 61/230 (27%) |
| Heparinc | 5/21 (24%) | 0/17 (0%) | 9/21 (43%) |
| No anticoagulantsd | 10/102 (10%) | 3/92 (3%) | 20/102 (20%) |
| Early anticoagulant strategy | Miscarriage | Fetal anomalies | Fetal loss |
|---|---|---|---|
| Regimen 2: heparin for <6 weeks | 19/129 (15%) | 0/108 (0%) | 21/129 (16%) |
| Heparin for >6 weeks | 19/56 (34%) | 4/36 (11%) | 20/56 (36%) |
a Comparing the results of discontinuing warfarin in the first 6 weeks and substituting heparin vs those who continued on warfarin beyond 6 weeks into the more vulnerable period before heparin was substituted
| Regimen | Thromboembolic complications | Death |
|---|---|---|
| Warfarin | 31/788 (4%) | 10/561 (2%) |
| Heparin/warfarin | 21/229 (9%) | 7/167 (4%) |
| Heparin | 7/21 (33%) | 3/20 (15%) |
| No anticoagulants | 26/107 (24%) | 5/106 (5%) |
a As described in Table 11.1
These findings were confirmed by Meschengieser et al. in a study of 92 pregnancies in 59 women.[16] They reported that miscarriage or fetal losses were similar in women exposed to oral anticoagulants in the first trimester (15/61 [25%]) compared with those who received adjusted subcutaneous heparin (6/31 [19%]; p=0.5717). Embolic events, however, were more common in women receiving heparin. The authors concluded that heparin did not offer a clear advantage over oral anticoagulants in the pregnancy outcome.
These studies highlight the dilemma that caregivers face when managing pregnant women with mechanical heart valves. While the risk and severity of warfarin embryopathy may not be as high as previously reported, the continued use of warfarin until close to term can result in a high risk of fetal loss. The use of warfarin throughout pregnancy, however, confers the greatest protection to the mother against valve thrombosis and death. Substituting heparin in the first trimester as early as possible avoids fetal embryopathy but exposes the woman to a period of increased risk of thrombosis.[15]
Intravenous unfractionated heparin is the appropriate treatment of choice during the peripartum because it can be discontinued several hours before planned delivery and its anticoagulant effects disappear within 6 h. This helps to avoid complications of hemorrhage with epidural anesthesia and bleeding complications around the time of delivery. Early reinstitution of anticoagulation postpartum is important, but almost certainly contributes to high rates of primary and secondary postpartum hemorrhage.[7] Heparin can be restarted approximately 6 h after delivery, depending on the individual situation and bleeding complications.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
