Cardiovascular disease complicates around 0.2–4 % of all pregnancies, in the modern western industrialised world. In the UK, cardiac disease has been the leading cause of maternal death since the year 2000, and is now responsible for one-fifth of all maternal deaths.[2, 3, 6] The rate of cardiac deaths has increased from 1.01 (in 1985–1987) to 2.31 per 100 000 maternities in the last triennium report (2006–2008)[4-8] (Table 1).
Table 1. Maternal deaths due to cardiac causes in the UK: CEMD and CEMACH/CMACE reports (1985–2008)[4-8] showing an upward trend in maternal deaths from acquired heart disease
Congenital causes n (%)
Acquired causes (ischaemic) n (%)
Acquired causes (others) n (%)
Total number of deaths n (100%)
Mortality rate per 100 000 maternities (cardiac deaths)
Total mortality rate per 100 000 maternities in UK
The majority of maternal deaths in the UK are now from acquired heart disease and the last Centre for Maternal and Child Enquires (CMACE) report noted that 11 maternal deaths (0.48 deaths per 100 000 maternities) resulted from myocardial infarction mostly related to ischaemic heart disease.[7, 8]
Acute myocardial infarction is a rare event in women of childbearing age. However, in pregnancy the relative risk is approximately 3–4 times higher than the estimated age-specific rates of myocardial infarction per 100 000 women years in the reproductive age group.[9-11] The United Kingdom Obstetric Surveillance Study (UKOSS) noted the incidence of reported non-fatal antenatal acute myocardial infarction as 0.7 cases per 100 000 maternities (95% CI 0.5–1.1) in 2005–2010.[3, 7] However, it is recognised that the UKOSS data are likely to be incomplete because of under-reporting of cases and therefore underestimate the true incidence in the UK.[2, 12]
Profound physiological changes in pregnancy and the postpartum period can put a strain on the heart, especially in the presence of recognised risk factors for myocardial infarction. It is therefore important that all clinicians involved in the care of pregnant women are aware of the effect of pregnancy on the cardiovascular system and the risk factors for acute myocardial infarction. They should have a low threshold to consider the diagnosis and initiate early multidisciplinary management.
Physiological changes in pregnancy
There are significant physiological changes in pregnancy that affect the cardiovascular system and increase myocardial oxygen demand.[12, 13] As early as 6 weeks of gestation, an increase in plasma volume and reduction in peripheral vascular resistance occurs because of activation of the renin–angiotensin system and a mild reduction of the plasma atrial natriuretic peptide levels. The increase in blood volume continues until it plateaus at a level of 140–150% at around 32 weeks of gestation compared with the nonpregnant state. Cardiac output increases steadily until 25 weeks of gestation, initially secondary to the increase in stroke volume and later because of an increase in maternal heart rate.[15, 16]
During labour and delivery, further haemodynamic changes occur. The heart rate and blood pressure increase significantly as a result of pain, anxiety and uterine contractions. The increase in heart rate is similar to that observed during moderate-to-heavy physical exercise.[17-19] Cardiac output increases by 50% with each contraction. About 300–400 ml blood is transferred from the uterus into the circulation with each contraction. In the active phase of the second stage of labour, the Valsalva manoeuvre results in larger variations in the central venous pressures. The completion of the third stage of labour results in approximately 500 ml uterine blood being returned to the maternal circulation, with associated increases in ventricular preload, cardiac output and central venous pressures.
Approximately 48 hours after delivery, diuresis and natriuresis start, with return of cardiac output, blood volume and peripheral resistance to the pre-pregnancy state occurring over the course of 4 to 12 weeks.
Increased levels of procoagulants and reduced levels of natural anticoagulants during pregnancy induce a hypercoagulable state. While these changes aim to reduce intrapartum blood loss, they increase the risk of thromboembolism. Levels of fibrinogen, factors VII, VIII and X and von Willebrand's factor increase in pregnancy and contribute to the procoagulant state. In addition, during pregnancy the natural anticoagulants decrease as a result of lower levels of functional protein S secondary to increased levels of its binding protein, the complement component C4b and increased levels of plasminogen activator inhibitor type 1.[13, 21-24] These changes in the coagulation system do not return to prepregnancy levels until more than 8 weeks postpartum. Compared with the risk during pregnancy, the risk of thrombosis is even higher after delivery.
Pathophysiology of acute myocardial infarction
Acute myocardial infarction (AMI) is characterised by the presence of myocardial necrosis in a clinical setting consistent with myocardial ischaemia. It is classified on the basis of electrocardiogram (ECG) findings as non-ST elevation myocardial infarction (NSTEMI) and ST elevation myocardial infarction (STEMI). Both types share a common pathophysiology. The most common underlying cause is atherosclerosis, a process of plaque formation in arteries, which continues throughout life. A number of factors interact before an atherosclerotic plaque causes an acute ischaemic event. Risk factors for coronary disease such as diabetes and smoking can cause damage to the endothelium covering the atherosclerotic plaque. Dysfunction of injured endothelium leads to inflammation of the underlying plaque, which can fissure or rupture. The ruptured plaque is prothrombotic, causing platelet activation and aggregation as well as activation of the coagulation cascade. This progresses to thrombus formation and causes myocardial ischaemia. The severity of plaque rupture, degree of inflammation and the coagulability status of the patient all influence the progression of thrombus formation. Partial occlusion of a vessel with thrombus causes findings consistent with NSTEMI while total occlusion of the vessel presents as STEMI.
Impact of physiological changes on risk of myocardial infarction
Cardiovascular changes in pregnancy lead to significantly increased myocardial oxygen demand. At the same time, the physiological anaemia of pregnancy, hypercoagulability and decrease in diastolic blood pressure may reduce the myocardial oxygen supply and contribute to the aggravation of myocardial ischaemia where the coronary arterial blood supply is already compromised. Increased cardiac output in the immediate postpartum period, resulting from decompression of the inferior vena cava and transfer of blood from the contracted uterus, makes peripartum a period of particularly high risk. Up to 50% of AMI occur during the peripartum period. Coronary thrombi and dissection occur more frequently in the peripartum period than before delivery. This may be explained by the physiological changes at this time leading to increased strain on the coronary vessels resulting in greater risk of dissection.[7, 13]
Risk factors for acute myocardial infarction (AMI)
he main risk factors for AMI are similar to those in the nonpregnant population. These include advancing age (significantly increased in those over 30 years of age),[11, 25] smoking, obesity, chronic hypertension, pre-existing diabetes, hyperlipidaemia and strong family history (Box 1). In the last two CMACE reports, the impact of increasing age and lifestyle factors such as smoking and obesity is dramatic. The risk of AMI is 30 times higher in women over the age of 40 years than in women aged less than 20 years. In the last report, 64% of women who died of cardiac disease were overweight or obese.[2-4, 6] Dyslipidaemia may be worsened because high-density lipoprotein cholesterol is significantly decreased during pregnancy. However, there is no significant change in low-density lipoprotein cholesterol or triglyceride levels in pregnancy.
Risk factors identified from CEMACH/CMACE reports:
Higher parity (>3)
Increasing maternal age (>35 years)
Pre-existing ischaemic heart disease
Strong family history
Other risk factors mentioned in the literature:
The maternal death risk was greatest if the infarct occurred in late pregnancy or if the delivery occurred within 2 weeks of infarction
Most women identified with non-fatal antenatal AMI in the UKOSS study and the maternal deaths from AMI and ischaemic heart disease in the last triennium report had identifiable risk factors as described above.[3, 7] Other studies looking at AMI in pregnancy specifically identified additional risk factors such as pre-eclampsia and eclampsia, thrombophilia, postpartum infections, blood transfusions, migraine headaches and multiparity. Whether multiparity has a direct relation to AMI or it is the advancing maternal age or other risk factors associated with multiparity that play a role still needs to be established. AMI may be a result of a generalised vasospastic disorder, explaining the association noted with migraine headaches.
The causes for AMI in pregnancy can therefore also be divided into two broad categories of (a) atherosclerotic (often with cardiovascular risk factors) and (b) non-atherosclerotic, which include coronary artery dissection, thrombosis and coronary artery spasm. Box 2 lists the main causes of AMI in pregnancy. Angiography in 14 women out of the 23 women reported in the UKOSS revealed pathology in the majority: coronary atheroma (50%), coronary artery dissection (22%) and coronary arterial thrombosis (14%). Only two women (14%) had normal coronary arteries. This highlights the importance of considering AMI due to non-atherosclerotic causes in pregnant women with no known risk factors.
Coronary artery dissection must be considered as a cause of AMI in pregnant women with no cardiovascular risk factors.[13, 31] The pathophysiology behind this is thought to be related to an excess of progesterone leading to biochemical and structural changes in the vessel wall. Other suggestions include an association with the lytic action of proteases released from eosinophils and a lack of prostacyclin synthesis stimulating plasma factor and elevated lipoprotein. The risk of coronary artery dissection is highest in the third trimester and up to 3 months postpartum. Physiological changes at this time lead to increased strain on the coronary vessels which may explain the greater risk of dissection.[13, 21] In the majority of cases (80%), the left anterior descending coronary artery is affected and causes extensive AMI with an associated mortality ranging from 30% to 40%.
- Drug-induced e.g. terbutaline, ergotamine, bromocriptine and cocaine use
Another reason for AMI presentation in women without underlying atherosclerosis is coronary artery thrombosis, with a reported incidence of 8–14% of AMI cases.[13, 27] The thrombophilic state of pregnancy may contribute to this pathology or hereditary thrombosis may first manifest at this time. Hereditary thrombophilia and the procoagulant physiological changes of pregnancy appear to interact. Smoking during pregnancy may further increase the risk of thrombosis due to enhanced platelet aggregability and additional release of tissue plasminogen activator (t-PA) inhibitor.
Pregnant women can also have AMI even when the coronary vessels are normal on angiogram. Two mechanisms proposed are transient coronary spasm due to enhanced vascular reactivity to angiotensin II and noradrenaline (norepinephrine) and endothelial dysfunction or renin release and angiotensin production secondary to compromised uterine perfusion in the supine position.[34-36] In addition, drugs used in pregnancy such as terbutaline, ergotamine and bromocriptine can induce continuous coronary vessel spasm.[37, 38] Other causes of AMI in pregnancy include cocaine use, vasculitis such as Kawasaki disease, collagen vascular disease, amniotic fluid embolism and pheochromocytoma.[39-42]
Diagnosis of AMI in pregnancy may be difficult because of its low prevalence and consequent low index of suspicion. Also as the presenting symptoms and signs can be attributed to normal manifestations of pregnancy or be masked during labour, it can lead to a delay in diagnosis. The diagnostic criteria of AMI are the same as for the nonpregnant patient. In addition to chest pain, typical features of pregnancy such as epigastric pain, vomiting or dizziness, particularly in the presence of known AMI risk factors should be investigated further. A low index of suspicion is important, as two consecutive CMACE reports have shown a consistent failure to consider AMI as a cause of chest pain in women with risk factors.
Electrocardiograms (ECGs) are classically the first-line test in making a diagnosis of AMI in any patient presenting with chest pain. The most sensitive and specific ECG marker is ST elevation, which normally appears within a few minutes of onset of symptoms.[1, 43, 44] Table 2 shows the ECG changes seen. Serial ECGs are fundamental as the initial ECG can be normal with changes evolving over time. It must be noted that the sensitivity of 12-lead ECGs has been reported to be as low as 50% to identify ischaemia.[1, 44] Therefore, other markers of cardiac damage must be used in conjunction with ECGs.
Cardiac-specific troponin I and troponin T are the biomarkers of choice for diagnosing myocardial infarction.[13, 45] Different hospitals will use either troponin I or troponin T and recommended sampling times vary depending on the assay, so clinicians should check their local hospital guidelines. A negative troponin at presentation does not exclude cardiac damage as it can take 12 hours for the level to peak.
Troponin is never increased above the upper limit of normal in healthy pregnant women and is not affected by anaesthesia, a prolonged labour or caesarean section, and therefore is the investigation of choice.[45, 46] In contrast, other cardiac markers – myoglobin, creatinine kinase, creatinine kinase isoenzyme MB – can be increased significantly in labour. The troponin levels can be raised in pre-eclampsia, gestational hypertension and pulmonary embolism in the absence of significant coronary disease.[13, 47, 48] It is important to note that in pre-eclampsia the troponin level is never above standard threshold set for MI.
Transthoracic echocardiogram is a useful method of excluding other conditions, which may produce symptoms similar to AMI such as aortic dissection. The use of echocardiogram to diagnose AMI is limited but can be an adjunctive technique to examine left ventricular function and wall motion abnormalities. It can be undertaken safely in pregnancy.
Coronary angiography aids in the diagnosis and potential treatment in AMI. It is reassuring to note that there is no evidence to suggest an increased risk of congenital malformations, intellectual disability, growth restriction or pregnancy loss at doses of radiation of less than 50 mGy to the pregnant woman.[1, 44, 49] Most diagnostic procedures will not cross this threshold (Table 3).[1, 49] Also, fetal exposure to a low dose of about 1 mGy is associated with a very low risk of childhood cancer. Therefore, women and clinicians should feel reassured and confident to undertake all indicated imaging in pregnant women with a differential diagnosis of AMI. If undertaken, coronary angiography by radial access is recommended with the abdomen shielded and fluoroscopic time minimised. Figure 1 show the use of angiography in the diagnosis and treatment of AMI in a postpartum patient.
Table 3. Estimated fetal and maternal effective doses for diagnostic and interventional radiology procedures in AMI
AMI = acute myocardial infarction; CT = computed tomography; PA = posterior–anterior
Exposure depends on the number of projections or views.
Percutaneous coronary intervention (PCI) or radiofrequency catheter ablationa
Treatment of AMI in pregnancy follows the same principles used in nonpregnant patients and aims to re-establish normal coronary blood flow. Prompt restoration of blood flow limits myocardial damage and reduces mortality. However, in pregnancy there are added restrictions and considerations. A low index of suspicion of AMI and reassurance regarding the safety of interventions will allow for early initiation of treatment and prevent inappropriate delays.
ST elevation myocardial infarction (STEMI)
Decisions regarding the mode of reperfusion therapy need to be made in consultation with a cardiologist and are influenced by local factors such as available resources. With STEMI, coronary angiography and primary percutaneous coronary intervention (PPCI) is the treatment of choice. There is growing literature on successful treatment with PPCI in pregnancy.[43, 50-53] The European Society of Cardiology recommends the use of bare metal stents rather than drug-eluting stents because of the lack of safety data for the latter. Even in pregnancy, thrombolysis is a suitable alternative if a significant delay is expected in accessing PPCI. The thrombolytic agent of choice is intravenous t-PA and this has the benefit of not crossing the placenta. In a review of 28 case reports of thrombolysis in pregnancy, Leonhardt et al. found no evidence to suggest a risk of teratogenesis. However, one has to bear in mind that there is an associated risk of maternal haemorrhage (8%), which in turn can lead to fetal compromise.
Non-ST elevation myocardial infarction (NSTEMI)
The first line of management in NSTEMI is antiplatelet drug treatment aimed at preventing further thrombus formation, facilitating clot dissolution and increasing blood flow to the myocardium. Coronary angiography with a view to revascularisation by stenting is considered if the symptoms of coronary ischaemia continue despite medical treatment and/or in the presence of haemodynamic instability. Patients whose symptoms settle on medical treatment but have high-risk features may also be considered for angiography following discussion of the risks and benefits at multidisciplinary team meetings.
Medications used in AMI
Medical management is an important component in the treatment of AMI and prevention of further coronary artery events.
Aspirin is one of the first-line medications given to nonpregnant patients and is crucial in the prevention of subsequent cardiovascular events.[12, 44] Low-dose aspirin must be commenced in pregnant women. The CLASP (Collaborative Low-dose Aspirin Study in Pregnancy) study confirmed the safety of low-dose aspirin (60–150 mg) in pregnancy. This study showed only a slight increase in the use of blood transfusion post-delivery, otherwise there were no significant safety issues for the mother, fetus or newborn when using low-dose aspirin. This was supported by an earlier meta-analysis by Imperiale and Petrulis. In the absence of evidence, high-dose aspirin should be avoided in pregnancy.
Low-molecular-weight heparin and unfractionated heparin are safe to use in pregnancy but anticoagulants should be stopped 24 hours prior to commencing induction of labour (effects may last up to 24 hours). A major benefit of both is that they do not cross the placenta.
Nitrates, labetalol (a non-selective beta-blocker with vascular alpha-1 receptor blocker) and nifedipine (a calcium channel blocker) can also be used safely in pregnancy.[57, 58] However, nifedipine should be avoided after an acute coronary event as it has been shown to increase mortality. Nitrates have been used in pregnant women who have experienced AMI as well as those who have severe hypertension, pulmonary oedema and can also be used as a tocolytic agent. Reports of fetal distress such as bradycardia and loss of beat-to-beat variability with intravenous nitroglycerine have been reported. Labetalol is currently the beta blocker of choice in pregnancy and has been shown to be safe when administered orally to women with chronic hypertension.[1, 60]
If indicated, for example, after stenting, antiplatelet drugs such as thienopyridine derivatives like clopidogrel can be used in pregnancy.[1, 57] In the absence of safety data, its use should be limited to the shortest duration possible in pregnancy. In a case report by Arimura et al. 75 mg clopidogrel was used for 2 weeks after insertion of a bare metal stent at 21 weeks of gestation with no adverse outcomes.
Angiotensin receptor blockers and angiotensin-converting enzyme (ACE) inhibitors are contraindicated in pregnancy. In the second and third trimester, their use is associated with renal dysgenesis, oligohydramnios, calvarial and pulmonary hypoplasia, fetal growth restriction and fetal demise. In the first trimester, ACE inhibitors can cause malformations of the cardiovascular and central nervous system.[57, 60] In the absence of safety data, statins are not recommended in pregnancy. Table 4 provides a summary of drugs considered in the management of AMI.
Pregnant women with AMI must be cared for in a high-dependency/intensive care unit with provision for fetal monitoring and comprehensive obstetric care as delivery must be considered if the maternal condition deteriorates in the presence of a potentially viable fetus.[19, 21]
Timing and mode of delivery and AMI
There are no standardised guidelines for obstetric management because of the lack of prospective data and the impact of the variation of individual patient characteristics in this setting. Collaborative management involving cardiologist, obstetric physician, obstetrician, obstetric anaesthetist and neonatologist will guide further management. Interventions need to be individualised based on maternal cardiac status and gestational age. If there is a risk of preterm delivery, maternal steroid injections must be administered at the earliest opportunity. If possible, delivery should be delayed by 2–3 weeks following AMI because of the increased risk of maternal mortality during this time. Delivery should take place in high-risk obstetric units with intensive care expertise.
The mode of delivery (vaginal or elective caesarean section) needs to be based on the obstetric and maternal factors as neither method of delivery is associated with a higher mortality.[1, 13, 17] Vaginal delivery is associated with a lower risk of vaginal blood loss, infection and thromboembolism, while an elective caesarean section avoids the haemodynamic fluctuations associated with labour and allows for a timely delivery with all appropriate multidisciplinary team members present.[3, 21]
The timing of vaginal delivery (spontaneous versus induction of labour) will depend on the favourability of the Bishop score and fetal wellbeing as a long labour can negatively influence the myocardial workload with higher increases in the preload from uterine contractions.
If vaginal delivery is attempted, an epidural analgesia is recommended to reduce the pain-associated increase in sympathetic activity and the urge to push. Women must labour in the left lateral position to attenuate the impact of uterine pressure on haemodynamic circulation. In addition to supplementary oxygen, continuous maternal cardiovascular monitoring, which also includes pulse oximetry and ECG, must be undertaken. Continuous fetal monitoring is also essential. If left ventricular function is significantly impaired or in women with a recent cardiac event, invasive monitoring with an arterial catheter may be appropriate.[1, 17, 21] Shortening the second stage with instrumental delivery is recommended to avoid the unwanted haemodynamic effect of the Valsalva manoeuvre and to shorten the labour. Slow intravenous infusion of oxytocin (<2 U/min), which avoids systemic hypotension, is administered after placental delivery to reduce the risk of postpartum haemorrhage. Ergometrine is contraindicated because of the increased risk of coronary artery spasm secondary to its vasoconstriction effect.
After delivery, maternal monitoring should take place in a high-dependency/intensive care unit for at least 24–48 hours as significant haemodynamic changes and fluid shift occur in this period. Irrespective of the underlying pathology for the AMI, thromboembolic risk assessment and appropriate management should be undertaken during each admission.
For prevention or treatment of myocardial ischaemia during delivery, intravenous nitroglycerin, beta blockers and calcium antagonists can be used. However, nitroglycerin and calcium antagonists do have tocolytic properties and may prolong labour.
It is vital that the multidisciplinary team is involved in designing and implementing individualised birth plans and communicating this with the women and all personnel involved in their care. An explicit instruction on escalation of concerns needs to be included.
The general practitioner should be updated at all stages and arrangements should be made for postnatal follow-up with cardiologists. Contraception and recommendations about future pregnancies will also depend on the underlying aetiology and cardiac function.
Some groups recommend screening asymptomatic women with multiple risk factors for coronary disease. However, the proportion of pregnant women with risk factors who have a cardiac event is very small. No previous randomised study has ever demonstrated the value or cost-effectiveness of widespread screening for coronary artery disease in an asymptomatic population group and therefore it cannot be recommended at this stage.
Nonetheless, we recommend that consideration be given to grouping maternal risk factors, such as age, body mass index, smoking, hypertension and so on, during antenatal booking, such that the ‘cardiovascular disease risk’ is specifically highlighted if three or more factors are present. This may prompt earlier AMI investigations if chest symptoms or non-specific symptoms of concern are present and may also motivate women in addressing lifestyle issues, for example, giving up smoking or limiting weight gain, while still pregnant.
Although rare, the incidence of myocardial infarction is increasing in the UK in pregnant women. This can probably be explained by the changing demographics of the obstetric population, women delaying pregnancies, increasing incidence of obesity, diabetes, pre-existing hypertension, smoking and a family history of coronary artery disease.
There is no evidence to support the screening of all asymptomatic pregnant women with risk factors for AMI. However, explicit identification of risk factors and recommendations to reduce modifiable risk factors such as smoking and excessive weight gain while achieving good control of hypertension must be supported.
The majority of women with cardiovascular risk factors have an uneventful pregnancy. In the absence of population screening, clinicians can improve outcomes by having a low index of suspicion of AMI in pregnant women. Appropriate investigations and treatment must not be delayed because of fetal concerns regarding radiation exposure or the use of thrombolytic treatment. Treatment outcome is dependent on the time to treatment and early involvement of the cardiologist. Multidisciplinary team management is vital for a good outcome although this may not always be possible in an acute emergency setting.