Turner syndrome (TS) is a genetic disorder with a characteristic phenotype that occurs as a result of a structurally abnormal or absent X chromosome. TS is the only monosomy compatible with life and affects around 1 in 2000 newborn females.1,2

There are variable Turner karyotypes, such as monosomy X (45,X; the most prevalent karyotype), mosaic karyotype, isochromosome X, ring X chromosome or deletions. In all these cases, significant portions of the X chromosome are deleted.3

The diagnosis of TS requires having key clinical features, including short stature and gonadal failure (the cardinal features) and congenital cardiovascular (CV) defects as well as an abnormal karyotype.2

The most serious consequences of X-chromosome haploinsufficiency involve the CV system. Indeed, CV defects, present in up to 50% of the TS population, are the major cause of premature death and contribute to a standardized mortality ratio three times higher than in the general female population. Therefore, imaging studies (both echocardiography and cardiovascular magnetic resonance [CMR]) are crucial for the timely detection of often subclinical CV disease and should preferably be ordered before the onset of symptomatic and irreversible organ damage.2,4,5

A thorough cardiac examination is recommended for all women with TS every 5 to 10 years, at the time of transition from pediatric to adult care, before attempting pregnancy or if hypertension (HTN) develops.2,6

The aim of this paper is to review both congenital and acquired cardiovascular diseases and the role of clinical imaging in TS CV risk stratification with recommendations for CV screening. Moreover, particular attention will be given to the special risk of CV complications in pregnancy with this syndrome.

This review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, thus providing a comprehensive framework that objectively assesses quality indicators and risk of bias in the studies included.

All original studies investigating Turner syndrome and cardiovascular risk were eligible for this systematic review. Further selection criteria were: (i) publication date between January 2000 and September 2016, (ii) written in English, (iii) published in a scholarly peer-reviewed journal.

Studies were found by searching the PubMed, EMBASE and Cochrane Library electronic databases, using the following search strategy: “Turner syndrome AND Cardiovascular disease AND Risk factors AND Female”.

Studies were identified by searching relevant papers via PubMed/MEDLINE (http://www.ncbi.nlm.nih.gov/pubmed), the Cochrane Library and EMBASE using the following search strategy: “Turner syndrome AND Cardiovascular disease AND Risk factors. Finally, reference lists of the studies retrieved were manually searched in order to detect any additional relevant studies. Keywords and combinations of keywords were used to search the electronic databases (Figure 1).

Figure 1.

Flow diagram of search strategy and study selection.

(0.09MB).

After performing the initial literature searches, each study title and abstract was screened for eligibility by the first author. The full texts of all potentially relevant studies were subsequently retrieved and examined further for eligibility. The PRISMA flow diagram (Figure 1) provides more detailed information regarding the process for selecting studies. Sixty articles were included in this review and the information from the studies included was then analyzed and recorded. Table 1 shows a summary of the literature consulted.

Variations in the CV anatomy of patients with TS, the frequencies of which are summarized in Table 2, are major factors in their reduced life expectancy.7

Major defects in cardiac and aortic development during fetal life are associated with miscarriage in most cases of fetuses with 45X karyotype. Fetuses with CV failure almost always demonstrate obstructed jugular lymphatics with increased nuchal translucency or nuchal cystic hygromas, the typical intrauterine presentation of TS. The residual postnatal webbing of the neck predicts CV defects. This association suggests a pathogenetic connection between fetal lymphatic obstruction and defective CV development due to the compression or obstruction of the outflow tracts.2,8,9

However, this is just a theory regarding a pathogenetic mechanism for congenital CV defects, because the real causes are currently unknown.1

A. Valvular anomalyA.1. Bicuspid aortic valve

A bicuspid aortic valve (BAV) is a congenital valvular anomaly found in approximately 30% of TS patients and it is not often clinically apparent.14

Identifying a BAV in asymptomatic individuals is important because they are at increased risk of infective endocarditis, hemodynamically significant stenosis (promoted by accelerated valve calcification), valve regurgitation (secondary to poor leaflet coaptation) and aortic aneurysm (due to the vicious cycle of increased stroke volume that promotes aortic root dilatation). For that reason, the anomaly requires regular medical evaluation and possibly surgery to prevent aortic dissection or rupture.14,16

Indeed, the aortic valve leaflet must be clearly visualized and echocardiography is the primary test for BAV that has been reported to be quite sensitive in diagnosing this anomaly in 89% of women with TS. However, echocardiography can be inadequate to view the aortic valve and, in these cases, CMR should be the next approach.4,14

In almost all the cases, BAV results from fusion of the right and left coronary leaflets, while the fused right coronary and non-coronary leaflets variant is much less common4,14 (Figure 2).

Figure 2.

Systolic cine magnetic resonance image of a 36-year-old TS patient, with fusion of the right and left coronary leaflets, with an anterior raphe (white arrow).

(0.03MB).

B. Aortic abnormalitiesB.1. Coarctation of the aorta

Coarctation of the aorta (CoA) affects around 12% of women with TS11 and is usually diagnosed in infancy, often with congestive heart failure in critical cases17 (Figure 3).

Figure 3.

Coarctation of the aorta (white arrow) in a 24-year-old TS woman in different projections: (a) sagittal-oblique cine magnetic resonance image, (b) volume-rendered magnetic resonance aortography (posterior-oblique view), (c) magnetic resonance aortography maximum intensity projection (sagittal-oblique view).

(0.12MB).

Indeed, in contrast to BAV, which is detected by screening, CoA is usually diagnosed based on clinical grounds – HTN and brachial-femoral delay are common features. However, since many cases are detected later in life, any woman with suspected CoA should have a CMR or computed tomography (CT) angiography with three-dimensional reconstruction of the thoracic aorta.1,11

Concentric narrowing of the aortic lumen leads to the development of pressure gradients across the lesion, with the possibility of arterial collaterals becoming a detour for oxygenated blood traveling to the lower segments of the body.7

If left untreated, complications such as HTN, congestive heart failure, aortic dissection and aortic rupture may occur. Surgical intervention is required when peak-to-peak coarctation gradient is ≥20 mmHg, or <20 mmHg with radiological evidence of significant collateral flow. Even after surgical repair, there is an increased risk of HTN, coronary artery disease (CAD), cerebrovascular disease, aortic dissection and restenosis.18

B.2. Elongated transverse arch

Elongated transverse arch (ETA) is a silent anomaly, detected by CMR, defined as an increased distance between the origin of the left common carotid and left subclavian arteries, with flattening of the arch and kinking along the inferior curvature. This distinct anatomy, sometimes referred to as “pseudocoarctation”, is embryologically similar to coarctation and may predict aortic complications, such as aortic dilation, aortic dissection or progression to overt coarctation.1,11 However, ETA differs from true coarctation in that there is no true luminal narrowing, nor are there pressure gradients or collateral circulation.7

ETA has been significantly associated with BAV, CoA and also with aortic sinus dilatation.10

B.3. Aortic dilatation

Aortic dilatation is reported in 23% of women with TS.12

Risk factors for this and, secondarily, dissection include HTN, BAV, CoA, 45,X karyotype and pregnancy. Aortic dilatation may less commonly occur alone in approximately 5% of women with TS. This indicates that all women with TS should be periodically monitored.5,19–21

In TS patients with enlargement of the ascending aorta, a general aortopathy is present, which is accelerated by the presence of a BAV.22

It has been suggested that patients with an enlarged aortic root should undergo echocardiographic evaluation every year and that patients with normal aortic root dimensions should be evaluated every 2 to 3 years.23

Aortic diameter, corrected for age and body size, is accurate for predicting aortic events, and is the principal risk marker for aortic dissection.1,5 An ascending to descending aortic diameter ratio above 1.5 signifies ascending aorta dilatation if descending aorta diameter is normal. There is another alternative to identifying aortic dilatation that correlates more closely with aortic diameter, because it adjusts aortic dimensions for body surface area (BSA); this is important due to the relatively short stature of patients with TS. This latter method involves estimating the aortic size index (ASI): aortic diameter (at the ascending aorta or at the site with the largest dimension)/BSA. It has been proposed that an ASI ≥2 cm/m2 identifies those who require close monitoring and an ASI ≥2.5 cm/m2 (as in Figure 4) requires aortic surgery to prevent aortic dissection.4,21,24,25

Figure 4.

Long-axis three-chamber-view CMR of a 24-year-old patient with TS, showing dilatation of the ascending aorta, with ASI >2.5 cm/m2 indicating a high risk of dissection.

(0.02MB).

B.4. Aortic dissection

Although rare, aortic dissection is a major concern in TS. It has an estimated incidence of 40 cases per 10000 patients, but is often fatal. It occurs at a much earlier age; in the general population, the peak incidence of dissection occurs between ages 50 and 80 and in TS population median age is 35, with higher incidence rates at ages 20–29 and 30–39.2,26

Up to 90% of aortic dissections have predisposing CV risk factors (BAV, aortic dilatation and CoA) that can be identified with CMR imaging. HTN and pregnancy are associated with a higher risk of dissection, the former due to an association with increased aortic root diameter and the latter due to physiological adaptation, which imposes a higher cardiac workload.4,19

However, up to 25% of cases have no apparent risk factors other than TS itself, suggesting that TS phenotype may also include a primary aortopathy, which can be present as early as 9 years of age.18,26–28

Indeed, histological evidence of cystic medial necrosis in aortic tissue taken from patients with BAV, Marfan syndrome and TS indicates a shared etiology despite genetically diverse backgrounds.29

Aortic dissection is sometimes overlooked at onset because it could include apparently minor complaints such as abdominal pain, heartburn, back or shoulder pain or a voice change due to traction of the recurrent laryngeal nerve. Therefore, patients should be advised to go to the emergency department soon after experiencing sustained chest pain (>30min), regardless of the severity of symptoms.17,29

C. Venous anomaliesC.1. Partial anomalous pulmonary venous connection

In TS, partial anomalous pulmonary venous connection may be right-sided, which is the most typical in the general population. Nevertheless, it frequently involves the left upper pulmonary vein, which makes echocardiographic detection more challenging.2,30

When the pulmonary veins connect to the right atrium or with one of its venous tributaries, blood from the pulmonary circulation returns to the right atrium and this may become clinically significant in adult life. Indeed, if sufficiently severe, pulmonary HTN and right ventricular volume overload can lead to congestive heart failure, which has been related to death in middle-aged TS patients secondary to undiagnosed partial anomalous pulmonary venous return lesions. Echocardiography and ideally CMR imaging should therefore be performed to assess pulmonary vein drainage patterns.30,31

C.2. Persistent left superior vena cava

The venous return from a persistent left superior vena cava (PLSVC; Figure 5) has a prevalence of around 13%. In approximately 80% to 92% of cases, the PLSVC drains into the right atrium via the coronary sinus, resulting in no hemodynamic consequences. However, in about 10% to 20% of cases, it drains into the left atrium and results in venous blood returning to the left atrium.11,32

Figure 5.

CMR angiography of a PLSVC (white arrow) in a TS patient.

(0.04MB).

It is important to always report this irregular anatomy, as it needs to be taken into account during interventional procedures.32

D. Other congenital cardiovascular disorders

Cardiac conduction defects are also common, caused by left ventricular hypertrophy, myocardial ischemia, previous myocardial infarction and congenital cardiac malformations.33

Other associated lesions can be found in TS, such as ventricular and atrial septal defects, hypoplastic left heart syndrome, single ventricle, mitral valve abnormalities, coronary artery abnormalities and aberrant right subclavian artery4 (Figure 6).

Figure 6.

Volume-rendered magnetic resonance aortography (posterior-oblique view) showing an aberrant right subclavian artery (white arrow) in a 37-year-old TS woman.

(0.05MB).

In addition to potentially harmful CV malformations, women with TS display evidence of adverse CV risk profiles that expose them to greater risk of CAD with a relative risk (RR) of 2-11, cerebrovascular disease with a RR of 2-7, HTN and types 1 and 2 diabetes mellitus.16,18,34,35

The precise mechanisms of increased CV risk in TS are unclear, but estrogen deficiencies (ensuing from premature ovarian failure due to gonadal dysgenesis) and haploinsufficiency of the X chromosome seem to be major factors that distinguish this metabolic phenotype.35–37

Awareness of considerable CV risk requires effective preventive medicine from the earliest years of life.35 Therefore, an appropriate and rigorous CV evaluation with the spectrum of CV issues encountered in TS should be performed (Table 4). The best cardiac imaging modalities for visualizing each of the congenital CV anomalies are shown in Table 5.

Table 4.

Screening and monitoring algorithm for cardiovascular imaging in TS girls and women. Adapted from Marin et al.4

ASI: aortic size index; BAV: bicuspid aortic valve; CMR: cardiovascular magnetic resonance; CoA: coarctation of the aorta; Echo: echocardiography; ECG-gated CTA: electrocardiogram-gated computed-tomography aortography; HTN: hypertension

At diagnosis, all infants and children with TS should undergo comprehensive CV evaluation, including clinical examination, electrocardiogram (ECG) to detect conduction abnormalities, BP measurement and echocardiography, which are currently the standard of care to assess the cardiac anatomy of girls with TS.2,55

CMR is recommended in TS children from the age of 12 (when it can be performed without sedation), even if no cardiac anomalies are detected on echocardiography. Indeed, CMR is the gold-standard imaging test for the diagnosis and monitoring of morphological anomalies of the thoracic aorta in TS, since echocardiography underestimates the size of both the ascending and the descending aortas.5,18

For outpatients, alternating echocardiography and CMR may be helpful. Aortic diameter in TS patients with no risk factors for aortic dissection (BAV, CoA and/or increased aortic dimensions) should be re-evaluated every 5-10 years. In patients considered to be at high risk, CMR should be repeated sooner.21Table 4 summarizes a reasonable imaging approach for these patients.

In the acute setting, CMR is not useful due to the relatively long examination time and difficulty in monitoring unstable patients. ECG-gated CT aortography is a valuable alternative for obtaining images of the entire aorta and aortic branch vessels including coronary arteries.5

As mentioned above, ovarian failure is a typical feature in TS with no or incomplete development of secondary sexual characteristics and primary amenorrhea. The X chromosome locus responsible for primary ovarian insufficiency in TS is probably on the short arm (Xq). However, 2-8% experience spontaneous pregnancy, and advances in assisted reproductive medicine with donated oocytes have been used, enabling successful pregnancies to be achieved.3,56

These pregnancies carry substantial fetal and maternal risks, exerting significant stress on the heart and aorta (with an increase in stroke volume and rise in estrogen level) and hypertensive disorders such as pre-eclampsia that may directly promote vascular damage and aneurysm formation. If uncontrolled, this can lead to aortic rupture and dissection.3,18

The risk of death during pregnancy from aortic dissection and rupture may be 2% or higher in women with TS.57

Thus, careful follow-up, including echocardiography and CMR performed by trained specialists familiar with TS, is essential for patients who intend to become pregnant.58

According to the 2008 publication of American Society of Reproductive Medicine, TS is a relative contraindication for pregnancy, but an absolute contraindication if there is a documented cardiac anomaly. However, the French task force 2011 recommendations, summarized in Table 6, do not list an isolated BAV as a contraindication, but rather a risk factor.59,60

If the ASI is less than 2.5 cm/m2 or the aortic diameter is less than 3.5 cm, and if there is no associated CoA, pregnancy can be authorized although careful CV monitoring is required, with an echocardiography suggested at the end of the first and the second trimesters and monthly during the third. If a patient is pending oocyte donation and the aorta increases by 10% or more in the annual echocardiography examination, said increase must be confirmed via a second imaging technique (CMR, cardiac CT or transesophageal echocardiography) and, if confirmed, pregnancy is contraindicated.59

In TS, the detection of congenital CV anomalies and their continuous monitoring with the use of cardiac imaging methods, particularly CMR (now recommended for optimal cardiac risk assessment), is of vital importance.

Aortic dissection, an extremely common event in these women, can occur at a remarkably young age and in pregnancy. Therefore, special attention to prevention and treatment should be given by monitoring the standard CV risk factors in TS. This approach is crucial to avoid life-threatening complications.

Pregnancy, which is not often spontaneous and can be achieved through assisted reproduction, increases risk in women with TS. Contraindications to pregnancy are defined, but even if none are present, patients should be carefully monitored before and during pregnancy.

To sum up, it is clear that TS is associated with several CV anomalies and risk factors present at an early age, which contribute to high TS mortality and morbidity. However, there are numerous unresolved issues that require further research.

Firstly, studies are warranted to clarify the pathophysiology of CV risk and the development of congenital CV anomalies. TS aortopathy has been postulated as a connective disorder, a theory that has not yet been confirmed. Determining the etiology will enable offering potential therapies to prevent aortic dissection. It will also provide information about potential risk stratification and, therefore, help determine the frequency of cardiovascular screening (which is currently not appropriately addressed).

Furthermore, the cut-off for aortic dilatation that puts an individual at high risk of aortic dissection in TS remains unclear. As such, the relationship between aortic dimensions and aortic dissection needs to be defined, and the adjustment of aortic dimensions for younger patients and those with lower BSA needs clarification.

Finally, the predictive values of aortic dimensions for prophylactic aortic surgery to prevent aortic dissection are undefined, as are the efficacy and safety of the procedure.

The authors have no conflicts of interest to declare.