In addition to the haemodynamic changes in pregnancy, hormones also induce changes in the aortic wall. Women with diseases like Marfan syndrome, Ehlers–Danlo syndrome, or other aortic abnormalities, have an increased risk of complications during pregnancy. Counselling and risk assessment before pregnancy is mandatory for all women with known aortic disease. Proper information should be provided about the risks of morbidity and mortality during pregnancy and information on the risks for the fetus, including the potential recurrence of disease in the offspring. Evaluation of past medical and family history, the aortic size before conception, and any increase in size before and during pregnancy, is essential to try and estimate the risk of aortic dissection. If the aorta is dilated, prophylactic repair before pregnancy may be indicated. In some cases, elective surgery during pregnancy may be warranted. In women with a severely dilated ascending aorta, caesarean section is, at present, the advised mode of delivery.
Introduction
In young women, the occurrence of an aortic complication is fortunately rare, with a reported incidence in the general female population under 45 years of age of 0.4 per 100,000 person years. When acute aortic dissection (AOD) occurs in a person under the age of 40 years, it is frequently caused by an underlying disorder such as Marfan syndrome or a bicuspid aortic valve. In this young age group, aortic dissection has a mortality rate of 22%, which is similar to the mortality rate in people aged over 40 years.
In pregnancy, haemodynamic adaptations influence the vascular system, including the aorta. The major cause of maternal mortality in the UK between 2006 and 2008 was cardiac disease, with 2.31 deaths per 100,000 maternities. Although the incidence of aortic dissection is low, it was one of the most important causes of cardiac death. A challenging situation exists for women at risk of aortic complications who contemplate pregnancy. Moreover, some women may have AOD during pregnancy as the first presentation of their aortic disease. Women with known aortopathy reaching reproductive age who are at high risk should be counselled about the risks of pregnancy. They should be referred to a tertiary centre for the purpose of surveillance during pregnancy and management of delivery.
General recommendations and guidelines
Taskforces have made great efforts to provide recommendations on this important topic. The two most recent provide helpful advice on how to manage women with aortopathy, but prospective studies are needed to assess and provide better data to support the recommendations. The Registry of Pregnancy And Cardiac diseases is currently the largest worldwide prospective study on pregnancy in women with cardiac disease, and includes women with aortopathy. The registry is still ongoing.
Effect of pregnancy on the aorta
The aorta is subject to structural and functional changes throughout pregnancy. Cardiac output, heart rate, and circulating volume increase, causing the aortic size to increase. The aortic size is also related to haemodynamic conditions, and varies in dimension with conditions such as preeclampsia. The hormonal changes in pregnancy influence connective tissue, changing the microstructure of the medial layer. Analysis of histopathology of the tunica media reveals hypertrophy and hyperplasia of smooth-muscle cells and fragmentation of reticulin, which normally surrounds elastic fibres. Subsequently, elastic fibres lose their organised structure, changes that are most notable in late pregnancy.
The above effects of pregnancy on the aorta explain why pregnancy itself is thought to be a risk factor for acute aortic dissection, although this topic is subject to debate, and contradictory reports exist. A population-based study in Sweden reported that pregnancy-related AOD is responsible for 60% of all AOD in young women, with a 25-fold increased risk of dissection during pregnancy ; however, another population-based report in Austria reported no increase in risk of aortic dissection. In these population-based studies, the difference in reported outcomes and risk of AOD might be related to different observation periods (1987–2007 compared with 1994–2004), and the small absolute total numbers of dissections (29 compared with 15). It is, however, clear that women with an increased risk of aortic dissection are those with connective tissue diseases such as Marfan syndrome, Ehlers–Danlos syndrome, and other genetic conditions pre-disposed to developing thoracic aneurysms and aortopathy. Aortic abnormalities in the context of bicuspid aortic valves will be reviewed elsewhere.
Effect of pregnancy on the aorta
The aorta is subject to structural and functional changes throughout pregnancy. Cardiac output, heart rate, and circulating volume increase, causing the aortic size to increase. The aortic size is also related to haemodynamic conditions, and varies in dimension with conditions such as preeclampsia. The hormonal changes in pregnancy influence connective tissue, changing the microstructure of the medial layer. Analysis of histopathology of the tunica media reveals hypertrophy and hyperplasia of smooth-muscle cells and fragmentation of reticulin, which normally surrounds elastic fibres. Subsequently, elastic fibres lose their organised structure, changes that are most notable in late pregnancy.
The above effects of pregnancy on the aorta explain why pregnancy itself is thought to be a risk factor for acute aortic dissection, although this topic is subject to debate, and contradictory reports exist. A population-based study in Sweden reported that pregnancy-related AOD is responsible for 60% of all AOD in young women, with a 25-fold increased risk of dissection during pregnancy ; however, another population-based report in Austria reported no increase in risk of aortic dissection. In these population-based studies, the difference in reported outcomes and risk of AOD might be related to different observation periods (1987–2007 compared with 1994–2004), and the small absolute total numbers of dissections (29 compared with 15). It is, however, clear that women with an increased risk of aortic dissection are those with connective tissue diseases such as Marfan syndrome, Ehlers–Danlos syndrome, and other genetic conditions pre-disposed to developing thoracic aneurysms and aortopathy. Aortic abnormalities in the context of bicuspid aortic valves will be reviewed elsewhere.
Pre-conceptional counselling
All women at increased risk of aortic complications, including those with known Marfan syndrome, Ehlers–Danlos syndrome, Loeys–Dietz syndrome, or familial thoracic aorta dissection syndrome, should be counselled before pregnancy. Preferably, counselling should be carried out once a girl reaches reproductive age. Maternal and fetal risks should be discussed, as well as recurrence risk of the disease in the offspring and of the possible effect of medication use. It is also important that women are made aware of the value of computed tomography and magnetic resonance imaging (MRI) of the aorta before pregnancy, because the results, if abnormal, may mean that pregnancy is inadvisable or elective aortic surgery is warranted. The risk of aortic dissection is most likely a result of a combination of pregnancy effects, pre-existence of an abnormal aortic wall, and an increased aortic size at baseline. Hypertension is a significant risk factor in a healthy woman affected by an aortic dissection, but is less common in younger women suffering from aortic dissection during pregnancy.
Prenatal diagnosis
The recurrence risk of Marfan, Ehlers–Danlos and Loeys–Dietz syndrome is 50% in offspring, owing to an autosomal dominant inheritance pattern in all of these conditions. If the disease-causing mutation in a family has been identified, then prenatal testing is possible using chorion villus sampling or amniocentesis (cell culture), which provides information on the presence or absence of the mutation. In Marfan syndrome, however, the mutation does not predict the severity of the clinical disease spectrum and, moreover, sampling is associated with a 1% risk of miscarriage. Fetal echocardiography is, therefore, an alternative lower risk diagnostic tool looking for atrioventricular valve regurgitation and dilatation of the aortic root and pulmonary artery. This technique, however, is not 100% predictive.
Specific diseases
Marfan syndrome
Marfan syndrome is a connective tissue disorder with an autosomal dominant inheritance pattern, and has an incidence of 2–3 per 10,000 in the general population. The mutation affects the Fibrillin gene ( FBN-1 ), the gene encoding extracellular matrix protein or glycoprotein fibrillin-1 on chromosome 15. In women with Marfan syndrome, about 75% have a family history; the other quartile have a sporadic or de-novo mutation. Recent evidence shows that excessive signalling of transforming growth factor B (TGF-β) is also associated with Marfan syndrome, and these findings might play a key role in future treatment strategies.
The mutation of FBN-1 causes a decrease in the amount of elastin and also loss of the organised structure of elastin itself. The subsequent weakness of the supportive tissue affects the cardiovascular system (e.g. aorta, atrioventricular valves, and conduction system), skeletal system, eyes, and the dural sac of the spinal cord). As more than a 1000 different mutations are known, the diagnosis is based on a combination of clinical criteria and genetic findings, as described by Loeys et al. in the revised Ghent nosology. Other important factors to consider include a family history of AOD, aortic root aneurysm, ectopia lentis, an identifiable FBN-1 mutation, or a combination of systemic manifestations. Cardiovascular involvement is present in 80% of cases with aortic dilatation, aortic regurgitation, and mitral or tricuspid valve prolapsed, with or without regurgitation. Cardiac complications are the main cause of morbidity and mortality, although life-expectancy has improved dramatically in the past 2 decades, mainly due to the treatment with elective cardiac surgery.
Maternal risk
Numerous retrospective and prospective reports on the maternal outcome of pregnancy in women with Marfan syndrome are available. An overview of the reported rates of aortic dissection is presented in Table 1 . The outcomes vary, partly because treatment has changed over the years. More women are treated with beta blockers or undergo elective aortic root replacement before pregnancy. In addition, some studies included only women with known Marfan syndrome, whereas others retrospectively included those diagnosed during pregnancy when they presented with a complication (e.g. aortic dilatation or dissection). Subsequently, it seems that nowadays the dissection rate is relatively low.
| Study | Year | Design | Participants | Pregnancies | Dissections | Beta blocker† | Time of diagnosing Marfan syndrome |
|---|---|---|---|---|---|---|---|
| Pyeritz et al. | 1981 | Retrospective | 26 | 105 | 0 | Unknown | Unknown |
| Rossiter et al. | 1995 | Prospective | 21 | 45 | 2 | Part of the cohort | All before pregnancy |
| Lipscomb et al. | 1997 | Retrospective | 36 | 91 | 4 | Part of the cohort | Partially during pregnancy |
| Lind et al. | 2001 | Retrospective | 38 | 78 | 5 | None | Partially during pregnancy |
| Meijboom et al. | 2005 | Prospective | 23 | 33 | 0 | Part of the cohort | All before pregnancy |
| Pacini et al. | 2009 | Retrospective | 85 | 160 | 6 | None | All before pregnancy |
| Donnelly et al. | 2012 | Retrospective | 69 | 199 | 0 | Part of the cohort | Partially during pregnancy |
| Omnes et al. | 2013 | Prospective | 18 | 22 | 1 | Major part of cohort | All before pregnancy |
The first large report (retrospective study) on maternal risks was published by Pyeritz, who demonstrated a low risk of dissection during pregnancy in women with MFS. As this was the first large study of its kind, the recommendations in this article have long been accepted into clinical practice to guide management, whereby pregnancy is discouraged when the aortic diameter exceeds 40–45 mm. These recommendations were further supported by two reports in the 1990s: when the aortic diameter is smaller than 40 mm, the risk of dissection is about 1% in what was the first prospective study, whereas an aortic root diameter exceeding 40 mm was associated with a 10% risk of aortic dissection.
More recent studies, one of which was prospective, demonstrated a low risk of AOD if the aortic diameter was less than 45 mm. A low risk was also confirmed in a recent, smaller prospective study, describing women managed according to the French guidelines. From these studies, it would seem that the risk of aortic dissection is higher in women not previously diagnosed before pregnancy, and therefore were not managed according to the guidelines. This highlights the importance of recognising the disease signs and its early diagnosis, along with family screening programmes for genetic disorders.
Aortic diameter growth during pregnancy
The aortic diameter is known to increase during pregnancy, particularly in a hypertensive state. Progressive aortic dilatation and dissection outside pregnancy is predicted by the aortic diameter itself. Several studies have focused on the potential growth of the aorta during pregnancy in women with Marfan syndrome. Rossiter et al. prospectively observed 21 women with Marfan syndrome during 45 pregnancies and compared them with a control group of non-pregnant women with Marfan syndrome. The aortic diameter did not increase in size faster in the pregnant group, but an important limitation of this study is that the reason for not embarking on pregnancy in the control group was not documented, so selection bias might be introduced with potentially more severe disease manifest in the non-pregnant group. Meijboom et al. described 33 pregnancies in women with Marfan syndrome with an aortic root diameter equal to or smaller than 45 mm. Their control group consisted of 22 age-matched nullipara with Marfan syndrome. No evident aortic root growth was reported compared with the control group, apart from a mildly increased growth of the aortic root during long-term follow up, in women who had a diameter of the root equal to or more than 40 mm at baseline. A similar comparative study divided Japanese pregnant women with Marfan syndrome into an event group and a non-event group. Aortic dilatation of more than 60 mm in any part of aorta or dissection were considered events. The aortic diameter increased by 0.41 mm per month in the event group compared with 0.05 mm ( P < 0.005) in the non-event group. The study did not, however, report on long-term outcome (i.e. whether the diameter decreased after pregnancy).In the largest study, Donnelly et al. reported on 98 nulliparous women with Marfan syndrome who experienced 199 pregnancies. An increase of 3 mm in aortic diameter was seen during pregnancy, but after pregnancy, the diameter decreased, but without full recovery during follow up, up to 5 years, after delivery.
In these four studies, the evidence for an increase in aortic diameter is contradictory. It is also difficult to reach firm conclusions because of a variable use of beta blockers and different follow-up duration. It has been suggested, however, that aortic diameter growth could be partially irreversible.
Long term outcome
Life expectancy in people with Marfan syndrome has substantially increased, and was up to 72 years in 1995. Although large studies are lacking, it been shown that use of the current guidelines confers a favourable outcome in pregnancy and Marfan syndrome; however, evidence on long-term outcome is needed. Donnelly et al. found an increase of risk over the longer term (average about 5 years) with death, aortic dissection, severe symptomatic aortic regurgitation, and need for urgent surgery defined as adverse outcomes in their study. Predictors of these outcomes are presented in Table 2 .
| Associated factors with long-term adverse outcome: | Odds ratios |
|---|---|
| Aortic size | 1.3 (1.11–1.61) |
| Number of pregnancies | 1.5 (1.15–1.97) |
| Prospective care | 0.1 (0.05–0.39) |
| Medication | 0.3 (0.14–0.92) |
| Aorta > 4 cm | 3.8 (1.11–13.3) |
| Independent correlates | |
| Initial aortic size | 1.8 (1.07–3.07) |
| Rate of aortic change (log) | 7.4 (1.32–41.22) |
Obstetric and fetal outcome
The miscarriage rate in the normal population ranges from 10–20% before 20 weeks gestation. This rate can be higher in women with Marfan syndrome, with some women experiencing habitual miscarriages. An important cause of fetal and neonatal morbidity and mortality is prematurity, which is the result of premature rupture of membranes, and occurs more often (about 5%) in women with Marfan syndrome. It has strongly been suggested that, if the fetus has connective tissue disease, the integrity of the membranes is weakened, causing a greater risk of preterm rupture. Small-for-gestational-age is also seen more frequently, with the frequent use of beta blockers implicated. It seems that the obstetric outcome of women diagnosed with Marfan syndrome after pregnancy is no different from those with known Marfan syndrome.
Ehlers–Danlos syndrome
The syndrome of Ehlers–Danlos is an autosomal dominant inherited disease. It is, however, more rare than Marfan syndrome, with an incidence of about 1in 5000. The Villefranche classification consists of six different types: classical, hypermobility, vascular, kyphoscoliosis, arthrochalasia, and dermatosparaxis Ehlers–Danlos syndrome. The disease is characterised by hypermobility of large joints, hyperelastic skin, tissue fragility, easy bruising, and poor wound healing. The vascular type is the most severe form of the syndrome. It is associated with a mutation in the gene for type III procollagen, 50% of which is a de-novo mutation. This type of Ehlers–Danlos syndrome involves the cardiovascular system, and features such as hypermobility and hyperelasticity are less frequently seen. Women suffering from the vascular type have an increased risk of aortic dissection, often not preceded by aortic dilatation. Surgery is difficult because of the features associated with the disease, such as tissue fragility. Risk of uterine rupture and haemorrhage peripartum also increases. Consequently, pregnancy is high risk, with a peripartum mortality of 12% reported owing to arterial or uterine rupture.
Turner syndrome
Turner syndrome occurs in 1 in 2000 female births, and is caused by complete or partial monosomy for the X chromosome. It is characterised by short stature and premature ovarian failure, but also associated with a variety of congenital heart diseases and aortic pathology. Cardiovascular and aortic complications are the main reason for premature mortality in women with Turner syndrome. The highest prevalence of AOD is seen in women aged 20–39 years. No information on previous gravidity was available in this study. The risk of aortic dissection seems to be elevated about 100-fold in Turner syndrome, which might be partly related to the higher incidence of a bicuspid valve present in 20–30% of people, as well as coarctation of the aorta (found in 12% of cases). In addition, aortic arch abnormalities are described in up to 50% of people. A recent cardiac resonance imaging study of 102 women with Turner syndrome found an association between the aortic diameter and aortic growth with coarctation and bicuspid aortic valve. The aortic dilatation in Turner syndrome tends to involve the ascending aorta. Specific echo views are needed to image this region, but cardiovascular magnetic resonance imaging or computed tomography can also assess this region along with the whole aorta, so are useful assessment tools. It is important that all aortic measurements are adjusted for body surface area. One study of 158 women with Turner syndrome, with a follow-up duration of 3 years, showed three aortic dissections occurring in women all with a pre-existing aortic diameter of larger than 25 mm/m 2 and an abnormal aortic valve.
Infertility caused by premature ovarian failure is an important clinical feature in women with Turner syndrome, although some women with mosaic Turner syndrome may be fertile. Subsequently, women often seek assisted conception with oocyte donation. Data on the long-term cardiovascular risks associated with these treatments are limited. A review of reported cases of AOD revealed a 58% mortality rate in women with Turner syndrome. Specifically seven AOD during pregnancy have been described, causing six maternal deaths. Two retrospective national studies reported on mortality outcome in pregnancy. A French study showed two AOD, both fatal, in 93 pregnancies resulting from oocyte donation. The women in this study had not been treated using the recommended guidelines. More recently, no AOD were reported in 124 deliveries in women with Turner syndrome in Denmark. The exact incidence of AOD in pregnancy is not accurately defined, but mortality rate is definitely increased. Overall maternal mortality during pregnancy is estimated at 2%.
Other aortic syndromes
Other less common syndromes with a potential high risk for aortic dissection are Loeys–Dietz syndrome, aneurysm–osteoarthritis syndrome and familial thoracic aortic aneurysm and dissection. The number of women with these syndromes contemplating pregnancy is unknown, but they are all associated with a high risk of dissection.
Loeys–Dietz syndrome is caused by mutation of the genes TGFBR1 or TGFBR2 , and shows some overlap with other syndromes of thoracic aortic aneurysm and dissection, such as vascular Ehlers–Danlos and Marfan syndrome. It is characterized by arterial tortuosity, hypertelorism, and bifid uvula, along with a large range of other features. The mean age of death is 26 years, owing to the vascular complications. Pregnancy has been reported in a few case reports. In addition, Loeys et al. described 21 pregnancies in 12 women with Loeys–Dietz syndrome. During pregnancy or postpartum, four AOD and two uterine ruptures occurred.
Thoracic aneurysms resulting from SMAD3 mutations have been recently described, often presenting with early onset osteoarthritis, called aneurysm–osteoarthritis syndrome. It is a spectrum consisting of arterial aneurysms, dissections and tortuosity, accompanied by skeletal, craniofacial and cutaneous features. The mutations are located on chromosome 15q22.2-24.2 and are allied with the TGF-β pathway.
Aneurysm–osteoarthritis syndrome seems to be an aggressive phenotype, with AOD occurring in vessels with even smaller diameters than Marfan syndrome. In women at increased risk of AOD, AOD can occur even with normal aortic rood dimensions, similar to Loeys–Dietz syndrome. Pregnancy outcomes have not yet been reported. Familial thoracic aortic aneurysm and dissection syndrome is associated with a variety of gene mutations, and few data have been published on pregnancy outcomes in this condition; they are, however, high risk, and should be managed in the same way as other fragile aortic syndromes.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
