Cardiac disease is the leading cause of maternal mortality in the developed world. The majority of women with heart disease are able to successfully undergo pregnancy. However, in women with severely impaired ventricular function, severe left heart obstruction, pulmonary hypertension and aortopathy, such as Marfan syndrome, with significant aortic dilatation, pregnancy is associated with a significant risk, and these women should be counselled against pregnancy if there is no option for treatment that reduces risk. Although there are increasing numbers of women with congenital heart disease who are considering pregnancy, as a result of joint expert specialist cardiac and obstetric care, maternal mortality is low. Most of the observed mortality occurs in women with structurally normal hearts who were not known to have heart disease before their pregnancy. It is therefore important that those caring for pregnant women are aware of the risk factors for and presentation of cardiac conditions in pregnancy. We review the presentation and management of both congenital and acquired heart diseases in pregnancy.
Highlights
- •
Cardiac disease is the leading cause of maternal mortality in the developed world.
- •
Most deaths occur in women who are not known to have heart disease before pregnancy.
- •
Clinicians caring for pregnant women must be aware of the presentation and causes of cardiac disease.
- •
There must be a low threshold for investigation of women with possible cardiac symptoms and referral to Cardiology.
- •
Expert joint cardiac and obstetric pre-pregnancy and antenatal management are key to improving outcomes.
Introduction
In the developed world, good antenatal and obstetric care has greatly reduced maternal mortality due to obstetric causes, and cardiac disease has now emerged as the leading cause of maternal mortality ( Fig. 1 ). Of concern, the most recent ‘Saving Mothers’ Lives’ report covering maternal mortality in the UK shows that whilst overall maternal mortality in the UK is low and continues to fall, deaths due to cardiac disease have risen .
The most recent UK maternal mortality review for which details of cardiac deaths are available is for 2006–2008 . The majority of deaths occurred in women with structurally normal hearts, many of whom were not known to have heart disease prior to pregnancy: their disease was revealed or precipitated by the hormonal and haemodynamic changes associated with pregnancy. Importantly, care was deemed to be substandard in >50% of the cardiac deaths, and in half of these cases better care may have resulted in a different outcome. The increase in cardiac mortality found between 2006 and 2008 is likely to be a reflection of the changing demographics of the overall obstetric population with increasing maternal age, obesity and smoking. All the women who died from ischaemic heart disease had at least one identifiable risk factor. The detailed review of the cases in the report identified the importance of having a low threshold for investigation and cardiological review of women who present with possible cardiac symptoms during pregnancy or the post-partum period; therefore, investigations should not be withheld because of pregnancy.
Although the numbers of patients with congenital heart disease surviving into their childbearing years are continuing to increase, and many have complex conditions, the maternal mortality in these patients continues to fall, and it accounted for only three out of the 53 cardiac deaths over the report period 2006–2008. Patients with congenital heart disease are generally under regular cardiology follow-up, and the low mortality is a testament to good pre-pregnancy counselling and the importance of the multidisciplinary team in providing joint cardiac obstetric care.
Haemodynamic and hormonal changes in pregnancy
The increased haemodynamic demands of pregnancy are significant, and they start early: cardiac output rises by 50% by the middle of the second trimester, and it must be maintained throughout the rest of the pregnancy ( Fig. 2 ) . Any condition that limits the ability to increase cardiac output will result in pregnancy being poorly tolerated. During labour, cardiac output increases further due to increased heart rate from pain and anxiety, and uterine contractions returning blood to the venous system. Blood pressure and oxygen requirements also rise during contractions. Following delivery, uterine and placental auto-transfusion and release of caval compression result in a significant increase in the venous return. The changes associated with pregnancy can take several weeks to return to normal. Patients with impaired ventricular function therefore remain at risk for several weeks after delivery, and many cases of pregnancy-associated cardiomyopathy are not revealed until several weeks later.
In addition to the circulatory changes, a variety of hormonal changes to facilitate tissue relaxation also occur. These changes may also contribute to the observed increased risk of aortic and coronary artery dissection during pregnancy and the post-partum period.
Preconception counselling and risk assessment
Patients with heart disease should be offered specialist joint cardiac and obstetric preconception assessment and counselling. This assessment should address both the maternal and foetal risks ( Tables 1 and 2 ).
| Maternal risk factors |
|---|
| Impaired ventricular function |
| Aortopathy |
| Previous arrhythmia |
| Mechanical valve |
| Pulmonary arterial hypertension |
| Myocardial ischaemia |
| Severe left heart obstruction |
| Foetal risks |
|---|
| Miscarriage/stillbirth |
| Maternal cardiovascular state leading to placental hypoperfusion and/or cyanosis leading to foetal growth restriction and prematurity |
| Teratogenicity |
| The risk of bleeding due to maternal anticoagulation |
| Inheritance risk |
| Maternal cyanosis |
Different scoring systems have been developed to assess the risk of adverse maternal outcomes during pregnancy. The factors consistently predictive of morbidity are pre-pregnancy poor ventricular function, arrhythmia or cardiac event, poor functional class (New York Heart Association (NYHA) class II or more) and left heart obstruction. Further, these risks are additive, as reflected in the Cardiac Disease in Pregnancy (CARPREG) scoring system ( Table 3 ). The ZAHARA investigators reviewed this in patients with congenital heart disease, and they identified additional predictors ( Table 3 ). Maternal risk may also be classified by lesions ( Table 4 ). Severe systemic ventricular impairment, previous peripartum cardiomyopathy with residual ventricular impairment, severe left heart obstruction, pulmonary arterial hypertension (PAH) of any cause and Marfan syndrome or other aortopathy with aorta >40 mm are associated with a significant risk of mortality, and women with these conditions should be advised that pregnancy would be associated with a high risk of morbidity and mortality. Any treatment to reduce pregnancy risk should be considered, and effective contraception should also be discussed as well as alternatives to pregnancy.
| Predictors from CARPREG study | Predictors from ZAHARA studies |
|---|---|
| Systemic ventricular impairment (EF <40%) | History of arrhythmic event |
| Prior cardiovascular event | Baseline functional NYHA Class > II |
| New York Heart Association (NYHA) Class ≥ II | Left heart obstruction (peak gradient >50 mm Hg) |
| Cyanosis (SaO 2 <90%) | Mechanical valve prosthesis |
| Severe left heart obstruction | Moderate/severe atrioventricular valve regurgitation (due to associated ventricular dysfunction) |
| Unrepaired cyanotic heart disease | |
| Use of cardiac medication prior to pregnancy |
| Class I ( No detectable increase in maternal mortality and no or mild increase in morbidity) | Class II (Small increase in maternal mortality and moderate increase in morbidity) | Classes II–III | Class III (Significantly increased risk of maternal mortality or severe morbidity) | Class IV (Extremely high risk of maternal mortality or severe morbidity) |
|---|---|---|---|---|
| Uncomplicated small or mild: | Unoperated ASD | Mild left ventricular impairment | Mechanical valve | Pulmonary hypertension of any cause |
| Repaired tetralogy of Fallot | Hypertrophic cardiomyopathy | Systemic right ventricle | Severe systemic ventricular impairment – LVEF <30% or NYHA III–IV |
| Most arrhythmias | Native or tissue valve disease not considered class IV |
| Previous peripartum cardiomyopathy with any residual ventricular impairment | |
| Marfan syndrome (or other aortopathy, e.g., EDS type IV, LDS or FTAA) without aortic dilatation |
| Severe left heart obstruction | ||
| Repaired coarctation | Post-Fontan repair | Marfan syndrome with aorta >45 mm | ||
| Successfully repaired simple lesions, e.g., | Heart transplantation | Cyanotic heart disease | ||
| Secundum ASD, VSD, PDA, Total anomalous pulmonary venous drainage | Other complex congenital heart disease | |||
| Marfan syndrome with aorta 40–45 mm |
Further evaluation to determine the haemodynamic significance of valvular lesions and ventricular function may be required, and cardiopulmonary exercise testing, exercise echocardiography, specialised cardiac imaging and cardiac catheterisation should be considered. Haemodynamically significant valve lesions may require intervention prior to pregnancy.
Exercise echocardiography can refine risk stratification for some women with significant disease who wish to conceive, for example, the asymptomatic woman with severe aortic stenosis (valve area ≤1 cm 2 ) and good ventricular function. Such a patient is likely to tolerate pregnancy well if she has a normal exercise capacity, normal blood pressure response to exercise, no exertional electrocardiographic (ECG) changes, the increase in Doppler-derived aortic valve gradient at peak exercise is <18 mm Hg, left ventricular ejection fraction increases on exercise (indicating contractile reserve) and there is no exercise-induced PAH (estimated pulmonary artery systolic pressure <60 mm Hg). Conversely, pre-pregnancy aortic valve replacement should be considered if exercise capacity and the pressor response to exercise are reduced, there are exertional ECG changes, exercise is associated with no increase in the left ventricular ejection fraction, there is a rise in Doppler-derived aortic valve gradient of >20 mm Hg or pulmonary artery systolic pressure rises to ≥60 mm Hg . Women with impaired ventricular function should also be made aware that ventricular function may deteriorate with pregnancy and may not fully recover.
Women with aortopathy and aortic dilatation >40 mm should be considered for prophylactic surgery, and care should be discussed with an aortic multidisciplinary team. It is important to remember that prophylactic aortic root replacement does not eliminate the risk of dissection in the descending aorta.
Many drugs have potential teratogenic effects, and therefore medication should also be reviewed prior to pregnancy ( Table 5 ).
| Type of drug | Drug | FDA class | Breast feeding |
|---|---|---|---|
| Antiplatelet agents | Aspirin (low dose) | N but considered safe | Small amounts excreted |
| Aspirin (high dose) | Considered D | ||
| Clopidogrel | B | No data | |
| Anticoagulants | Warfarin | X (except for mechanical valves, considered category D in this setting) | Considered safe |
| Heparins | C but human reports show no adverse effects | Considered safe | |
| Anti-arrhythmics | Calcium channel blockers | C | Small amounts excreted, consider risks versus benefits, monitor for neonatal bradycardia |
| Flecainide and Propafenone | C | As above | |
| Adenosine | C | Safe, short half-life | |
| Amiodarone | D | Excreted in breast milk, not advised | |
| Beta blockers | Atenolol | D | Small amounts excreted, consider risks versus benefits, monitor for neonatal bradycardia |
| Sotalol | B (All have a risk of causing intrauterine growth restriction) | ||
| All others | C | ||
| Diuretics | Furosemide | C | Excreted but no adverse effects seen |
| Bumetanide | C | Excreted but adverse effects not expected | |
| Thiazides | Bendrofluazide C Hydrochlorothiazide D | As above | |
| Amiloride | B | ||
| Spironolactone | C | Excreted in breast milk, use with caution | |
| Vasodilators | ACEI/ARB | D | Captopril/Enalapril not associated with adverse effects |
| Hydralazine | C | Excreted but no adverse effects seen | |
| Nitrates | C | As above | |
| Lipid-lowering agents | Statins | X | Not advised |
Ventricular dysfunction
Pregnancy places a prolonged stress on the cardiovascular system, and heart failure is an important cause of morbidity as well as mortality. With improvements in medical care, increasing numbers of women with impaired ventricular function are well enough to consider pregnancy. Ventricular impairment may be due to congenital heart disease, previous chemotherapy, cardiomyopathy or ischaemic heart disease. However, ventricular impairment often presents for the first time during pregnancy or the post-partum period because pregnancy either unmasks previously undiagnosed heart disease or precipitates peripartum cardiomyopathy. Breathlessness and mild peripheral oedema are common during pregnancy, but marked symptoms or signs should raise the suspicion of heart failure ( Table 6 ), particularly in women with risk factors such as hypertension, obesity, diabetes, advanced maternal age, multiparity, smoking, previous chemotherapy or human immunodeficiency virus (HIV) infection.
| Normal symptoms and signs | Abnormal symptoms and signs |
|---|---|
| Breathlessness on exertion | Extreme breathlessness |
| Palpitations due to atrial and ventricular ectopics | Marked peripheral oedema |
| Mild peripheral oedema | Pleural effusion |
| Heart rate rise by 10–20 bpm | Persistent tachycardia >100 bpm |
| Jugular venous pressure visible by up to +2 cm | Jugular venous pressure >2 cm |
| Third heart sound | Fourth heart sound |
| Soft systolic murmur | Diastolic murmur |
The risk of major morbidity and mortality rises with the severity of ventricular dysfunction, and there is an unpredictable risk of a permanent deterioration in ventricular function. Thus, women with pre-existing moderate or severely impaired systemic ventricular function with an inability to increase cardiac output on stress testing should be advised that pregnancy is associated with a high risk of morbidity and mortality. If ventricular impairment is milder, then the main risk is of progressive permanent decline in function.
Although treatment of heart failure should follow the same principles as in the non-pregnant state, drug therapy has to be modified. Cardioselective beta blockers (bisoprolol and metoprolol) can be used with appropriate monitoring of foetal growth. Hydralazine and oral nitrates may be used as vasodilator agents instead of angiotensin-converting enzyme inhibitors (ACEI) and angiotensin receptor blockers (ARB). The latter are not recommended because of foetal toxicity, but they may be used in individual cases following specialist advice if the maternal benefits are considered to outweigh the foetal risks. Similarly, loop diuretics should be used instead of aldosterone antagonists unless the maternal benefits of spironolactone outweigh the foetal risk.
Admission for bed rest and intravenous diuretics and anticoagulation should be considered if there is an evidence of overt heart failure and early delivery planned. Mechanical and inotropic support as a bridge to transplantation may be required.
Peripartum cardiomyopathy
Peripartum cardiomyopathy develops during the last trimester or the first 5 months post partum. The precise aetiology remains unclear, but pre-eclampsia, multiparity and advanced maternal age are consistently identified as risk factors . Autoimmune processes and oxidative-stress-mediated cleavage of prolactin have been purported as possible underlying causes, and immunoglobulins and bromocriptine have been used as treatment, but their efficacy has not yet been established. Therefore, at present, medical therapy is the same as for other causes of heart failure. These patients also frequently develop intracardiac mural thrombus, which may embolise and so should be anticoagulated. Ventricular function may decline rapidly, and mortality is 15–30% . If ventricular function remains impaired, a subsequent pregnancy carries a 44% risk of recurrence and a 19% risk of mortality . Even if ventricular function recovers fully, a further pregnancy is associated with a 20% risk of recurrence.
Hypertrophic cardiomyopathy
If systolic ventricular function is preserved, pregnancy is generally well tolerated as the vasodilatory effects of pregnancy are beneficial in patients without significant left ventricular outflow obstruction . However, atrial arrhythmias may occur as a result of increased atrial stretch and should be treated promptly as they are not well tolerated because of diastolic dysfunction. The incidence of ventricular arrhythmias and sudden cardiac death is not increased by pregnancy. Patients with significant outflow tract obstruction, a history of ventricular arrhythmias and severe hypertrophy are at an increased risk of decompensation. Beta blockers may be used to reduce outflow tract obstruction and in the management of arrhythmias. Patients with implantable defibrillators may undergo pregnancy provided that the ventricular arrhythmias are adequately controlled. The highest risk period is around the time of delivery and the first 48 h post partum as the rapid fluid shifts may precipitate heart failure and arrhythmias.
Ventricular dysfunction
Pregnancy places a prolonged stress on the cardiovascular system, and heart failure is an important cause of morbidity as well as mortality. With improvements in medical care, increasing numbers of women with impaired ventricular function are well enough to consider pregnancy. Ventricular impairment may be due to congenital heart disease, previous chemotherapy, cardiomyopathy or ischaemic heart disease. However, ventricular impairment often presents for the first time during pregnancy or the post-partum period because pregnancy either unmasks previously undiagnosed heart disease or precipitates peripartum cardiomyopathy. Breathlessness and mild peripheral oedema are common during pregnancy, but marked symptoms or signs should raise the suspicion of heart failure ( Table 6 ), particularly in women with risk factors such as hypertension, obesity, diabetes, advanced maternal age, multiparity, smoking, previous chemotherapy or human immunodeficiency virus (HIV) infection.
