Most women presenting to a UK antenatal clinic with valvular lesions will have congenital heart disease, although immigrants from the Indian subcontinent and some parts of eastern Europe still have a high incidence of acquired valve disease. As with most types of heart disease, pregnancy presents a challenge to women with valvular disease due to the altered cardiovascular physiology seen. Specific concerns and management depend on how the valvular lesion functions and how the haemodynamic changes of pregnancy will be tolerated.
Women with a prosthetic heart valve present an additional challenge owing to the need for anticoagulation, and the potential catastrophic consequences of inadequate anticoagulation. Prosthetic valves may also not function optimally before or during pregnancy. Women with corrected valvular heart disease and a prosthetic valve have increased morbidity in pregnancy, especially if they are anticoagulated.
Problems and special considerations
Morbidity is related to the severity of the underlying valvular lesion and cardiac function. Regurgitant and right-sided lesions tend to be better tolerated than stenotic and left-sided lesions in pregnancy, due to the reduction in systemic vascular resistance which encourages right-to-left flow.
Mitral stenosis is a particularly high-risk lesion, and those with moderate to severe disease should be counselled against pregnancy until treated. Even if a woman is asymptomatic to start with, the increase in blood volume and heart rate in pregnancy will often lead to decompensation. Those with a mitral valve replacement have a relatively fixed cardiac output and are also at risk of developing cardiac failure during pregnancy. They have an increased risk of thromboembolism and atrial fibrillation; most will be anticoagulated throughout pregnancy, and if arrhythmia occurs it should be treated promptly, by cardioversion if necessary.
A bicuspid aortic valve is one of the most common congenital heart defects seen, affecting up to 2% of the population; it is more likely to affect women than men and results in accelerated aortic stenosis, usually presenting from the third decade. Women with aortic valve lesions or replacement tend to tolerate the physiological changes of pregnancy relatively well, unless they have severe aortic stenosis (valve area <1.0 cm2 and mean gradient > 40 mmHg) or are symptomatic before pregnancy.
Prosthetic heart valves
The relative risk of pregnancy in women with prosthetic heart valves depends not only on the underlying cardiac abnormality and residual impairment of cardiac function, but also on the type of valve replacement used. Observational data from the UK Obstetric Surveillance System (UKOSS) suggest that maternal complications in women with prosthetic valves are more common than previously thought, with a mortality of 9% and morbidity of 41%; these figures are higher than in recent reports from a multinational registry of pregnancy and cardiac disease (ROPAC; mortality 1.5%), which may reflect differences in outcomes in specialist versus non-specialist centres. Prosthetic heart valves may be mechanical, xenograft (usually porcine) or human allograft (homograft). While mechanical valves provide superior durability and haemodynamic performance to biological valves, the need for long-term anticoagulation increases maternal and fetal morbidity and mortality, so they are usually second choice for women of reproductive age.
Mechanical valves
The most important risks for women with mechanical valves are those associated with the increased thrombotic risk during pregnancy (estimated to be as high as 29%), the need for anticoagulation and the risk of endocarditis. Both warfarin and heparin therapy are associated with significant maternal morbidity and fetal morbidity and mortality. Warfarin is better for maternal health and the risk of thrombosis but worse for the fetus, while heparin is better for the fetus and worse for the mother.
Warfarin is teratogenic, causing mental retardation, short stature and multiple facial abnormalities. Its use in the second trimester of pregnancy is associated with fetal blindness, microcephaly and mental retardation. There is also an increased risk of fetal internal haemorrhage and maternal haemorrhage in women taking warfarin. The ROPAC group has reported an association between miscarriage and late fetal death and use of vitamin K antagonists.
Administration of heparin during pregnancy is also associated with increased rates of spontaneous abortion, fetal and maternal haemorrhage and stillbirth. Prolonged use of heparin may cause maternal osteoporosis and heparin-induced thrombocytopenia, though these are less common with low-molecular-weight heparins. However, the main risk with heparin is thromboembolism involving the heart valves, increasing maternal morbidity and mortality. Optimal dose of heparins can be difficult with the altered pharmacokinetics of pregnancy. Dose adjustment according to anti-Xa levels can reduce the risk of thrombosis.
Traditionally, British practice has been to convert women to heparin for the first trimester of pregnancy and then maintain them on warfarin before reverting to heparin for the last few weeks of pregnancy and for delivery, while practice in the USA has been to heparinise women throughout pregnancy. In particular high-risk cases, low-dose aspirin may be added to heparin therapy in an attempt to improve maternal outcome. It has been suggested that as there is no clear-cut optimal anticoagulation regimen, women should be informed of the risks and benefits of all options and encouraged to help choose their own therapy.
Xenograft valves
The major risks of xenograft valves are thromboembolic events and valve failure. The rate of valve degeneration at 10 years is estimated at 50-60%. There is some evidence that pregnancy accelerates the degeneration of xenograft valves, and it is therefore imperative to follow these women closely during pregnancy and to investigate immediately any possibility of deteriorating cardiac function.
Although an advantage of xenograft compared with mechanical valves is that anticoagulation is not needed routinely, women with atrial fibrillation or a history of thromboembolic events are likely to require full anticoagulation, with its attendant risks.
Homograft valves
These valves are used primarily for aortic valve replacement, and the available evidence suggests that they are associated with a significantly lower morbidity in pregnancy than either xenograft or mechanical valves. There is no need for anticoagulant therapy and there do not appear to be the same risks of degenerative change as with xenograft valves.
Management options
Pre-pregnancy counselling should be offered to women with valvular heart disease, firstly to advise those with congenital heart disease of the increased risks of congenital heart disease in their offspring, and secondly to advise those who are dependent on anticoagulants of the risks of such therapy in pregnancy. Valve function and cardiac status should also be assessed before pregnancy if possible.
All women with valvular heart disease, and in particular those with prosthetic heart valves, should be regarded as having high-risk pregnancies and should be delivered in large maternity units, preferably in or near to centres with facilities for cardiac surgery. Multidisciplinary management should involve the obstetric, cardiology, haematology and anaesthetic teams, with input from midwifery and critical care staff as appropriate. Cardiac function should be assessed early in the first trimester of pregnancy and at regular intervals throughout the pregnancy. It is particularly important for women with prosthetic valves to receive regular dental care during pregnancy, and any dental treatment should be preceded by prophylactic antibiotics. Similarly, any intercurrent infection during pregnancy should be aggressively treated. Most units continue to give prophylactic antibiotics to cover delivery in women with prosthetic heart valves, despite recent guidelines suggesting this is unnecessary.
In general, management and monitoring will depend on the severity of residual cardiac disease and the underlying lesion, and the requirements of each woman must be determined on an individual basis. Aortocaval compression and the consequent reduction in preload must be avoided in all types of valvular heart disease. An arterial line may be useful to monitor changes in blood pressure and allow a rapid response.
In the presence of a stenotic valvular lesion, tachycardia, hypovolaemia and a reduction in afterload are poorly tolerated and should be avoided. If anaesthesia is required, general anaesthesia is often chosen over spinal anaesthesia. However, pain-mediated tachycardia may increase flow across the valve and precipitate cardiac failure; carefully titrated regional analgesia may be beneficial, and neuraxial anaesthesia is not contraindicated.
In the presence of a regurgitant lesion, forward flow increases with a drop in systemic vascular resistance, so regional analgesia/anaesthesia is usually advised. Normal blood pressure and a high-normal heart rate should be maintained. Most anaesthetists would avoid spinal anaesthesia in favour of a more gradual epidural or combined spinal-epidural technique.
Sinus rhythm should be maintained if possible. Particular attention to fluid balance is necessary in the postpartum period.
Prosthetic heart valves
The presence of a prosthetic heart valve is not in itself an indication for operative delivery. There are obvious advantages in planned induction of labour in anticoagulated women to allow appropriate management of anticoagulation over the period of induction and delivery; an individualised plan for labour and delivery should be agreed in advance by the multidisciplinary team. This would usually involve stopping low-molecular-weight heparin 24 hours before planned induction and converting to unfractionated heparin, which can then be stopped at an appropriate time before delivery. This enables regional analgesic and anaesthetic techniques to be used (following laboratory assessment of coagulation status) if appropriate. If regional analgesia is contraindicated, patient-controlled intravenous opioid analgesia is the most appropriate alternative for labour, and also for provision of postoperative analgesia if general anaesthesia has been used for caesarean section.
Key points
• Women with prosthetic heart valves and valvular heart disease are not a homogeneous group. They differ in their underlying cardiac disease, degree of impairment of cardiac function and type of prosthetic valve.
• Pre-pregnancy counselling is recommended whenever possible.
• Antenatal care and delivery should be undertaken in a hospital with facilities for high- dependency care.
• Regular assessment of cardiac function during pregnancy is important.
• Anaesthetic goals include maintaining normal cardiovascular parameters, sinus rhythm and fluid status.
Further reading
Dennis AT. Valvular heart disease in pregnancy. Int J Obstet Anesth 2016; 25: 4-8.
European Society of Cardiology. ESC guidelines on the management of cardiovascular diseases during pregnancy: the Task Force on the Management of Cardiovascular Diseases during Pregnancy of the European Society of Cardiology (ESC). Eur Heart J 2011; 32: 3147-97.
Lawley CM, Lain SJ, Algert CS, et al. Prosthetic heart valves in pregnancy, outcomes for women and their babies: a systematic review and meta-analysis. BJOG 2015; 122: 1446-55.
McLintock C. Thromboembolism in pregnancy: challenges and controversies in the prevention of pregnancy-associated venous thromboembolism and management of anticoagulation in women with mechanical prosthetic heart valves. Best Pract Res Clin Obstet Gynaecol 2014; 28: 519-36.
Thorne S. Pregnancy and native heart valve disease. Heart 2016; 102: 1410-17.
van Hagen IM, Roos-Hesselink JW, Ruys TP, et al.; ROPAC investigators. Pregnancy in women with a mechanical heart valve: data of the European Society of Cardiology Registry of Pregnancy and Cardiac Disease (ROPAC). Circulation 2015; 132: 132-42.
Vause S, Clarke B, Tower CL, Hay C, Knight M; UKOSS. Pregnancy outcomes in women with mechanical prosthetic heart valves: a prospective descriptive population based study using the United Kingdom Obstetric Surveillance System (UKOSS) data collection system. BJOG 2017; 124: 1411-19.
Windram JD, Colman JM, Wald RM, et al. Valvular heart disease in pregnancy. Best Pract Res Clin Obstet Gynaecol 2014; 28: 507-18.