Prevalence and type of MI in pregnancy

Elkayam et al reviewed 150 cases of AMI in pregnancy between 2006 and 2011,6 finding that cases with documented ST-segment elevation MI (STEMI) were three times more common than non-ST-segment elevation MI (NSTEMI). Furthermore, the majority of cases of MI occurred either in the third trimester or post partum; this is in keeping with UK Obstetric Surveillance System  (UKOSS), although in this study, NSTEMI was slightly more common.7 Table 1 shows the underlying causes of AMI in pregnancy. Atherosclerotic disease is the single largest cause of AMI in pregnancy accounting for about 40% of cases.8 However, in pregnancy, other causes are more prevalent, such as coronary thrombus without evidence of atherosclerotic disease, which accounts for almost 10% of cases of AMI, and spontaneous coronary artery dissection (SCAD).

SCAD is an uncommon cause of AMI outside of pregnancy, but is increasingly well recognised. It is a non-traumatic, non-atheromatous cause of dissection, with formation of intramural haematoma. An intimal tear is thought to occur with resulting blood in the endoluminal space, with compression of the true coronary artery lumen by this haematoma. Registry data have shown SCAD disproportionately affects younger women9 and maybe the common mechanism for pregnancy-associated MI. In a series of 120 women with SCAD in pregnancy, Havakuk et al showed that three-quarters of women presented with ST elevation, STEMI, with almost two-thirds having an anterior infarct.10 Rarer underlying causes for AMI in pregnancy also need to be considered such as chronic inflammatory conditions like lupus, or thrombotic causes like antiphospholipid syndrome. Table 1 underlines some of different underlying causes of AMI in pregnancy.

Pathophysiology

The underlying pathophysiology of increased risk of AMI in pregnancy remains poorly understood, but several hypotheses have been proposed, including: (1) the haemodynamic changes associated with pregnancy include both increased stroke volume and heart rate during pregnancy, both of which increase myocardial oxygen demand11; (2) blood volume expansion resulting in a physiological anaemia coupled with the fall in diastolic blood pressure may reduce myocardial oxygen supply, contributing to the development of myocardial ischaemia particularly if coronary blood supply is already compromised12; (3) hormonal alterations, particularly in the last months of pregnancy and the postpartum period, may alter the integrity of coronary vessels leading to increased rates of dissection; (4) during pregnancy the risk of thrombosis is higher. Degeneration of the ground substance of the connective tissue in the intima and media of the coronary arteries has been proposed as a mechanism for higher rates of peripartum MI.13 It has also been suggested that repeated pregnancies may expose the coronary circulation to progressive progesterone and oestrogen-mediated damage, weakening the vessel wall and explaining the higher incidence of SCAD.14

Risk factors for IHD in pregnancy

Taking a complete detailed history is critical when assessing pregnant women with suspected IHD as most studies report that women with AMI in pregnancy actually have at least one underlying risk factor for coronary artery disease (CAD) and that many actually have multiple risk factors.15 In their prospective study, Bush et al showed that risk of AMI increased with maternal age, even after adjusting for confounders.7 In their review, Elkayam et al showed over 40% of women with AMI were over 35 years old.6 Parous women are also more likely to develop AMI compared with nulliparous women.8

Despite an overall decline in smoking rates in the developed world, in women of childbearing age, smoking rates are typically 15%. In their review of over 100 cases of MI in pregnancy, Roth et al reported 45% of smokers. A similar percentage was reported in the more recent systematic review by Lameijer et al.8 15 It has been postulated that the negative cardiovascular effects of smoking may be attenuated by pregnancy, through the increased vascular reactivity induced by progesterone and oestrogen.5

Pre-eclampsia and hypertension have long been associated with cardiovascular disease in later life16 17; recent data suggest an increased risk of AMI associated with pre-eclampsia which is particularly marked in the puerperium.18 Mulla et al showed in their retrospective cohort study that 16% of cases of AMI in pregnancy were associated with pre-eclampsia19 Hypertension is also strongly associated with AMI and pregnancy as it may exacerbate damage to blood vessels that have already undergone adverse remodelling due to the haemodynamic stress of pregnancy and endothelial activation.

Diagnosis

As is the case outside of pregnancy, not all patients with AMI present with typical features of central chest pain that radiates to the jaw or arms. Indeed, in pregnancy women may present with atypical features, which healthcare professionals ascribe to symptoms of dyspepsia and reflux. Pregnancy can cause a physiological tachycardia, typically an increase in maternal heart rate of 10%–15% from non-pregnant state.17 Normal pregnancy-induced ECG changes include 15° left axis deviation as a result of diaphragmatic elevation, T wave inversion in lead III and aVF and the presence of supraventricular and ventricular ectopic beats.20

Pregnant women presenting with possible cardiac chest pain should have a 12-lead ECG and cardiac troponin levels measured. A diagnosis of NSTEMI in pregnancy should be based on elevated cardiac troponins, with a documented rise and fall, in the context of pain or shortness of breath. While some studies have suggested that, in the context of severe pre-eclampsia, there may be an isolated troponin rise,21 a recent systematic review failed to show a clear correlation between pre-eclampsia and raised troponin levels and suggested pre-eclampsia alone would not cause an increase in troponin levels above the diagnostic threshold for MI.22

Unstable angina can present in a similar fashion to NSTEMI, but without raised troponins. In pregnant women presenting with such features where the index of suspicion for cardiac cause of pain is high, then non-invasive investigations should be undertaken, including exercise testing, stress echocardiography or MRI. If these are suggestive of underlying IHD, then coronary angiography can be performed during pregnancy if clinically indicated.

Pregnancy in women with pre-existing CAD

Studies of women with known CAD are scarce. Burchill et al published a series of 43 women from a multicentre study who had either established CAD or previous acute coronary syndrome (ACS).23 Nearly half of the women had atherosclerotic disease and more than 80% of women had at least more than one risk factor for CAD. The Registry of Pregnancy and Cardiac Disease (ROPAC) investigators reported 20 pregnancies in women with known IHD.24 While maternal outcomes were generally very favourable, they showed higher rates of preterm delivery and lower birth weights. However, ROPAC reported on a large number of cardiac diagnoses and established IHD only made up a small proportion of these 20/1321 women. Estimating the incidence of pregnancy in women with known IHD is challenging as most studies report the incidence of AMI and not pregnancies in women with established IHD.6 Recent guidance from the European Society of Cardiology suggests that in women with prior CAD presenting with impaired ventricular function or with residual ischaemic pregnancy should be discouraged.25

Acute myocardial infarction

Relatively more is known about the incidence of AMI in pregnancy. Data from a retrospective nationwide North American inpatient sample showed the rate of AMI was 6.2 per 100 000 pregnancies.5 This figure is similar to a Californian study showing the incidence of AMI to be 1 in 35 700 deliveries.4 Data from a UK prospective population-based study (UKOSS) showed the rate to be lower at 0.7 per 100 000 pregnancies.7 The authors concluded that the lower reported rate of pregnancy-related MI may in fact be more accurate as previous studies have failed to directly examine medical notes and patients in prior studies may have been cared for at multiple institutions so may have been counted more than once. The rate of maternal mortality from AMI in pregnancy varies hugely from 0% to 7.3%5 7; but even this figure is lower than previous historical data where mortality was reported to be almost 20%.26 In their study, Bush et al only included cases of AMI until 1 week post partum whereas Ladner et al included cases until 6 weeks post partum.4 7 Nevertheless, all maternal deaths reported in the study by Ladner et al’s study occurred during pregnancy. Interestingly, data from National Registry of AMI that included over 384 000 patients between the ages of 30 and 89 showed that AMI in younger women was associated with a higher mortality than in men of the same age.27 Women in this study were less likely to be given a diagnosis of AMI and were less likely to receive thrombolytic therapy.27 These findings echo the UK Confidential Enquiry into maternal death where pregnant women were often given an incorrect diagnosis, were not properly investigated or had treatment withheld or delayed.28

Management of acute coronary events in pregnancy

The principle management of AMI does not differ in the pregnant and non-pregnant situations (figure 1). However, it is advisable that women are managed in a collaborative manner involving cardiologists, obstetricians, obstetric anaesthetists and, where necessary, cardiothoracic surgeons. Aspirin is safe in pregnancy when used at doses below 150 mg/day. There are no large data series describing the use of clopidogrel in pregnancy, but published reports do not appear to suggest fetal toxicity,29 but women need counselling about the scarcity of data. At present, there is a paucity of data to confirm safety of other antiplatelets such as ticagrelor. In cases of STEMI, the most effective treatment is primary percutaneous coronary intervention with stent implantation. In order to reduce the exposure of radiation from coronary angiography to the fetus, some authors advise maternal lead shielding. However, the evidence for this is lacking and lead shielding is associated with an increased risk of scatter radiation.30 Low dose acquisition times have been shown to be effective in radiation reduction exposure.31 Data from the UK from the Health Protection Agency suggest the incidence of childhood cancer attributable to fetal procedural exposure is 1 in 8000; notably the background risk of childhood cancer is 1 in 500.32 Despite this, there is a reluctance to investigate pregnant women because of the perceived fetal risk, as demonstrated by the UKOSS paper, which highlighted the fact that only 60% of women diagnosed with AMI underwent coronary angiography.7 In the majority of cases of ACS in pregnancy it is imperative that coronary angiography is considered, unless it is a small infarct with decreasing troponins, as it will help elucidate the underlying cause and define the appropriate treatment modality.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Diagram for acute coronary syndrome (ACS) and pregnancy management.

Percutaneous treatment permits the use of stents, with the choice of stent should be made by the treating cardiologist and time of event during the pregnancy. Drug-eluting stents have a lower risk of restenosis but typically require longer periods of antiplatelet medications, which may increase the risk of bleeding around the time of delivery. In settings where primary percutaneous coronary interventions are not available, then thrombolysis with recombinant tissue plasminogen activator may be undertaken. Pregnancy used to be a relative contraindication to thrombolysis, but emerging data taken mainly from cases of stroke or thromboembolism suggest that it is relatively safe, with complications appearing to be no greater than in the non-pregnant state.33 Nevertheless, caution should be undertaken with regard to the use of thrombolysis around delivery and early postpartum because of the risk of maternal haemorrhage.34

Medical therapy

Many medications used to treat either AMI, or for secondary prevention of cardiovascular disease, can be continued in pregnancy and breast feeding. Table 1 outlines which medications are considered safe. When there is uncertainty about the use of medications because of a lack of safety data in pregnancy, which, if given may have significant benefit, then a detailed discussion is needed with the mother. For instance, while statins appear to be generally contraindicated in pregnancy there are emerging data that some types may be given without any evidence of fetal harm.35 Women who are taking statins for conditions such as familial hypercholesterolaemia are usually advised to stop statins in pregnancy although these women are at risk of AMI in pregnancy or the puerperium should they stop them.

Most women with IHD during pregnancy require cardiac medication. However, not all cardiac drugs can be safely administered due to teratogenicity or other unwanted effects for the fetus. Furthermore, the haemodynamic changes accompanying pregnancy can influence pharmacokinetics, sometimes requiring dose adjustments. Table 2 summarises recommendations for use during pregnancy of the most important drugs for IHD.

Management of labour and delivery

Figure 2 outlines some of the principles of labour management in women with IHD. In their systematic review of cases of AMI in pregnancy and the puerperium, Lameijer et al reported an overall caesarean section rate of 56%, with a rate of 62% if the AMI had occurred in the index pregnancy.15 This is more than double the background rate of caesarean section for most European regions.36 Burchill et al showed a relatively lower rate of delivery by caesarean section in women with a history of CAD disease at 38%.23 The ROPAC investigators have previously demonstrated that planned caesarean section fails to confer any benefit for mothers with heart disease, and, as such, delivery by caesarean section should be reserved for obstetric indications only, with some exceptions including dilated aorta and the presence of oral anticoagulation with spontaneous labour.37 Delivery by caesarean section is associated with greater rates of maternal haemorrhage, infection and thromboembolism all of which may increase the risk of myocardial ischaemia. For women with a recent MI, it may be best to defer delivery for several weeks, if possible, to allow for myocardial healing. This may reduce the risk of immediate complications associated with MI as labour itself is associated with increased cardiac stress, but this may be ameliorated by the use of epidural analgesia to reduce the pain of contractions. Passive descent of the fetal head at full dilatation should also be employed as this has been shown to reduce the duration of maternal pushing.38 Limiting maternal pushing with an earlier assisted delivery has been proposed for some women with cardiac disease although the benefits of this for women with heart disease in general and IHD specifically remain unclear.39 Active management of the third stage (early cord clamping and administration of uterotonic agent) is usually recommended for women with heart disease as it reduces the risk of postpartum haemorrhage by 40%.40 However, ergometrine, an ergot alkaloid, should be avoided due to its powerful vasopressor activity that can result in coronary artery spasm.41 Cardiac monitoring both during delivery and for at least 24 hours post partum should be undertaken, especially in the context of recent AMI.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Delivery management for women with ischaemic heart disease.

Maternal and neonatal outcomes

While maternal mortality for women suffering AMI has improved over the last few decades it should be remembered that women are still at greater risk of suffering other complications. Cardiac arrest, heart failure and ventricular tachycardia are the most common subsequent cardiac complications reported for women with AMI in pregnancy. Data from the UKOSS study showed that over a quarter of women with AMI were admitted to either the intensive care unit or high dependency unit.9 For women diagnosed with SCAD, cardiogenic shock is reported in a quarter of the cases, with a similar proportion of women requiring mechanical assist devices.42

Perinatal outcomes are related to maternal outcomes, but overall a perinatal mortality rate of 4% is reported in women with AMI. Most of these cases are associated with a poor maternal outcomes, but most studies show a higher rate of prematurity which is largely iatrogenic.23 This is at least three times greater than the background rate of prematurity for healthy women.43 There does not appear to be a strong correlation between IHD and babies born small for gestational age (<10th centile), as has been shown for women with other forms of heart disease.24

Maternal outcomes