Objective To analyse the main characteristics, associated conditions and outcome of right aortic arch (RAA) detected in fetal life, and to assess if further investigation is required in cases of isolated RAA.
Methods Retrospective observational study of all fetuses diagnosed with a RAA between 2004 and 2012 at a tertiary centre for fetal cardiology.
Results A RAA was identified in 98 fetuses: 27 had normal intracardiac anatomy and 71 were associated with other congenital heart disease (CHD); conotruncal anomalies being the most common. An aberrant left subclavian artery was diagnosed in 18.4% of cases, a double aortic arch in 6.1%, and 12.2% had a vascular ring confirmed after birth. Overall, an extracardiac anomaly was present in 31.6% of the patients and a chromosomal anomaly in 15.3%, with half of the latter cases being 22q11.2 microdeletion. Extracardiac and chromosomal anomalies were more commonly associated with RAA with structural CHD (39.4% and 19.7%, respectively), compared to cases of RAA with normal intracardiac anatomy (11.1% and 3.7%, respectively) (p<0.05). First year mortality was 10.3%, with all deaths being in cases with associated structural CHD.
Conclusions Detailed fetal extracardiac examination should be undertaken in all cases of RAA. Isolated RAA has a good prognosis, and in the majority of the patients it is an asymptomatic vascular variant with a relatively low risk for chromosomal anomaly. The prognosis of RAA with CHD depends on the complexity of the CHD and/or the associated extracardiac anomalies. In these cases, there is a higher risk for chromosomal anomaly, particularly 22q11.2 microdeletion.
Statistics from Altmetric.com
A left or right aortic arch (RAA) refers to the position of the aortic arch in relation to the trachea. The normal embryological development of the arch is from the primitive pharyngeal arch system. The normal left aortic arch, descending on the left side of the trachea, is formed from the left fourth arch and the left dorsal aorta and the regression of a segment of the embryological RAA.1 ,2
A RAA refers to an aortic arch that courses and descends to the right side of the trachea. It is formed by regression of a segment of the embryological left aortic arch and persistence of the right fourth arch or right dorsal aorta.1 ,2 The reported incidence of individual aortic arch anomalies varies according to the study population.1 The true incidence of RAA in the general population is unknown, but a rate of 1/1000 has been suggested by previous authors in a series of low-risk pregnancies.3
A RAA can occur in isolation or it may be associated with structural cardiac anomalies, most commonly Tetralogy of Fallot (TOF), pulmonary atresia with ventricular septal defect and common arterial trunk.1 ,2 ,4–9 Extracardiac malformations and chromosomal anomalies have also been observed with RAA, particularly when associated with other forms of congenital heart disease (CHD). The most frequent chromosomal anomaly is microdeletion of chromosome 22q11.2,2 ,5–7 which is more frequently found in cases of RAA with structural CHD, with reports suggesting up to 46% of cases being affected in a fetal and postnatal series.5 ,10 The prevalence of this chromosomal anomaly in cases of isolated RAA, however, varies between studies, from 8% reported in fetal series,5 to 20–25% in postnatal series.11 ,12 This association has led some authors to recommend karyotyping and screening for deletion of chromosome 22q11.2 in all fetuses with RAA, but particularly when associated with intracardiac anomalies or extracardiac malformations.1 ,5 ,10 ,13
The aim of this retrospective study was to analyse the main characteristics, associated conditions and outcome of RAA detected in fetal life, and to quantify the risk of karyotype abnormality in cases of isolated RAA.
A retrospective study of all patients seen between January 2004 and December 2012 at a tertiary referral centre for fetal cardiology was performed. During this period, 16 332 fetal echocardiographic examinations were performed in pregnant women at increased risk of having a baby with CHD.14 Fetal echocardiography was carried out using a variety of ultrasound platforms: Toshiba Xario and Toshiba Aplio ultrasound system (Toshiba Medical Systems, Crawley, UK), Hewlett Packard Sonos 5500 ultrasound system (Philips, Andover, Massachusetts, USA) and Voluson E8 (General Electric Company, Schenectady, New York, USA). Fetuses with a RAA were identified using our fetal cardiology computerised database.
The prenatal diagnosis of RAA was made on the basis of an abnormal three-vessel view visualised in the upper mediastinum, when the transverse portion of the aorta lies and descends to the right of the trachea. As a consequence, the arterial duct and aorta form a U-shaped configuration with trachea located between these two vessels (figure 1A,B).1–3 ,15
In the rare cases of a right arterial duct, the transverse portion of the aorta and the duct lie to the right of the trachea (figure 1C). A RAA with an aberrant left subclavian artery (ALSA) was diagnosed in the presence of an aortic arch descending on the right side of the trachea with the left subclavian artery arising from the descending aorta distal to the right subclavian artery (figure 1D). A double aortic arch (DAA) was diagnosed when both aortic arches were clearly recognisable arising from ascending aorta, one on either side of the trachea, joining posteriorly to form a single descending aorta.1 ,2
A vascular ring occurs when congenital anomalies of the aortic arch and its related vessels result in encirclement of the trachea and oesophagus with the potential of causing varying degrees of obstruction to either or both. The possibility of a vascular ring was noted before birth, but the diagnosis of a definite vascular ring was made after birth by postnatal imaging (echocardiography and/or MRI).
The following variables were evaluated in all cases of RAA: maternal age, reason for referral, gestational age at diagnosis, presence of increased nuchal translucency, extracardiac malformations and karyotype anomalies, type of aortic arch anomaly, associated cardiac abnormalities and outcome.
The presence of increased nuchal translucency or fetal extracardiac anomalies was identified on the pregnancy early scan and anomaly scan reports included in maternal notes. To evaluate postnatal diagnoses and outcome, data from the postnatal period were obtained using our hospital medical records and postnatal paediatric cardiology database.
Statistical analysis was performed using the χ2 and Fisher's exact test, p<0.05 being considered significant.
This study was approved as an internal audit at our hospital. As this was a retrospective review of data already collected for clinical purposes, ethical approval was not deemed necessary.
There were 98 cases of fetal RAA diagnosed during the study period. This corresponds to 0.6% (98/16 332) of all fetal cardiac examinations preformed in this period, and to 5.0% (98/1946) of all cardiac abnormalities. Of the 98 fetuses, 27 (27.6%) had normal intracardiac anatomy and 71 (72.4%) had abnormal intracardiac anatomy. Isolated RAA cases correspond to 1.4% (27/1946) of all cardiac abnormalities observed in this period. The number of cases of RAA diagnosed per year is illustrated in figure 2.
Median maternal age was 30.5 years (range 16–40 years) and median gestational age at examination was 23.5 gestational weeks (range 14–37 gestational weeks).
The most common reasons for referral were suspected heart malformation (67.3%), followed by family history of CHD (13.2%).
Nuchal translucency was measured in 77/98 fetuses, and it was above the 95th centile for crown–rump length in 11 (14.3%) of these cases.
Associated cardiac anomalies
The type of CHD associated with RAA was divided into three major groups. The first group had the conotruncal anomalies (55/98), with TOF being the most common (n=36), of which four were tetralogy with absent pulmonary valve. The second group had septal defects (8/98), of which seven were ventricular septal defect and one atrioventricular septal defect. The third group had major complex anomalies (8/98), which included six cases of a laterality syndrome (table 1).
Associated extracardiac anomalies
A wide range of extracardiac anomalies were documented in 31/98 (31.6%) of patients: three in isolated RAA and 28 in RAA with abnormal intracardiac anatomy. Of these 31 cases, 15 had a chromosomal anomaly, of which only one case had normal intracardiac anatomy, but this baby also had Pierre Robin Syndrome. Therefore, the incidence of extracardiac anomalies and chromosomal anomalies was significantly higher in cases of RAA with abnormal intracardiac anatomy (39.4% and 19.7%, respectively), compared with RAA with normal intracardiac anatomy (11.1% and 3.7%, respectively) (p<0.05).
Of the 15 cases of chromosomal anomalies, eight patients had 22q11.2 microdeletion, all of whom had associated CHD. Karyotyping was declined or not performed in 49/98 patients, but there were no suggestive features of chromosomal abnormality in the postnatal clinical records. In the isolated RAA cases, four had normal karyotypes, one had Pierre Robin Syndrome with chromosome 3q29 duplication and 22 have not been tested, but do not have any obvious abnormality documented after birth (table 2).
Associated vascular anomalies
Among the 98 fetuses with a RAA, an ALSA occurred in 18 (18.4%), 6 (6.1%) fetuses had a DAA, and a vascular ring was confirmed after birth in 12 (12.2%).
Of the 18 patients with an ALSA, the majority had a normal intracardiac anatomy (n=12). In the other six patients with an ALSA associated with a RAA and abnormal intracardiac anatomy, four cases were TOF and two cases pulmonary atresia/ventricular septal defect. Thus, ALSA was more frequently observed in cases of isolated RAA (44.4%), than in RAA with structural CHD (8.5%) (p<0.05). Contrastingly, a mirror image branching pattern of the neck vessels was more commonly observed in RAA with structural CHD (91.5%).
Vascular rings were diagnosed postnatally by echocardiography in 12 patients, seven of whom had CHD. To obtain further clarification of aortic arch and/or intracardiac anatomy, a postnatal MRI was performed in seven patients. Of the 12 patients with a vascular ring, 10 had associated documented vascular anomalies (six had an ALSA and four had a DAA).
The only two patients with DAA that did not develop a vascular ring had a DAA with an atretic left-sided component.
Of the seven patients with a vascular ring and abnormal intracardiac anatomy, 2/7 underwent complete correction of TOF, 2/7 have not required any treatment as yet, and 3/7 died in the context of a complex CHD. All five patients with a vascular ring and normal intracardiac anatomy had symptoms of tracheo-oesophageal compression; two have had surgical repair of the vascular ring, and the other three have surgery planned.
The outcome is unknown in one patient. In six cases, the pregnancy was interrupted and four fetuses died in utero. Of the remaining 87 patients, 10 died postnatally in the context of complex CHD (six neonatal deaths and four infant deaths). Thus, first-year mortality was 10.3%. We report no mortality in isolated RAA (table 3).
During the study period, there were four missed diagnoses of RAA, all in the context of complex abnormal intracardiac anatomy. Our unit policy is to try and obtain follow-up on all our patients from the mothers themselves, the general practitioner and the women's local hospital. We were able to obtain follow-up in 75–80% of our patients, but we accept that, potentially, there may have been more missed diagnosis during the study period. There was one incorrect diagnosis of RAA in a fetus with normal intracardiac and extracardiac anatomy.
The diagnosis of a RAA can be made prenatally by fetal echocardiography. The increasing documentation of RAA noted over time in our series is likely to be related to a more detailed assessment of the aortic arch position during fetal cardiology assessment, as well as to a more accurate imaging of the transverse three-vessel view during obstetric screening at the fetal anomaly scan in more recent years. This increase in detection rate, particularly in cases of isolated RAA, raises the dilemma of whether further investigation is required following the diagnosis of a RAA in a fetus. Certainly, associated cardiac and extracardiac malformations should be excluded, but the decision to proceed with karyotyping in isolated cases remains controversial.
Associated structural cardiac anomalies
The majority of cases in our fetal series had an associated structural CHD. In all reported series,6 ,7 ,9 the type of CHD most commonly associated with RAA are the conotruncal anomalies. The results from our fetal series concur with this, with conotruncal malformations accounting for 76% of cases of CHD with RAA. Therefore, when a RAA is identified during an obstetric fetal anomaly scan, referral should be made for a detailed fetal echocardiography.
Associated extracardiac anomalies
A wide range of extracardiac anomalies can be associated in a significant proportion of the patients (31.6% in our study), so that a detailed extracardiac evaluation should also be undertaken in all cases. Extracardiac anomalies are more frequent in fetuses with RAA associated with other CHD, but did also occur in 11.1% of our isolated RAA cases, suggesting that these cases should have a detailed extracardiac examination as well.
The association of RAA with chromosomal anomalies, specifically 22q11.2 microdeletion, has been documented in different studies, mostly with concomitant cardiac and/or extracardiac anomalies.10 ,16 ,17 However, it has been reported that even in patients with apparently isolated RAA or DAA, up to 24% have 22q11.2 microdeletion postnatally.10 These findings have led some authors to recommend fetal karyotyping in all cases of prenatal diagnosis of RAA. The overall incidence of chromosomal anomalies in our study was 15.3%, with 22q11.2 microdeletion being responsible for approximately half the cases. However, all patients with known 22q11.2 microdeletion had associated structural CHD, and two of the cases also had an extracardiac structural anomaly. In our series, there is only one case of an isolated RAA with a chromosomal anomaly; this baby had Pierre Robin Syndrome with chromosome 3q29 duplication. We would therefore advocate that a chromosomal anomaly should be ruled out when a RAA coexists with structural CHD and/or with an extracardiac anomaly. The presence of a RAA as an isolated finding in a fetus may not justify karyotyping, as the risk for a chromosomal anomaly is relatively low. This information is important to assist parental counselling during pregnancy.
Some authors have suggested that the identification of thymus hypoplasia or aplasia in the fetal scan may help to identify the patients at higher risk for 22q11.2 microdeletion.18 We did not perform this assessment in our retrospective study, though this may justify further prospective investigation.
Associated vascular anomalies
Although all types of RAA may be associated with CHD, cases of RAA with mirror image branching are associated with CHD in more than 90% of cases.1 ,7 ,8 A RAA with mirror image branching is commonly found with conotruncal malformations, with a reported incidence of 15–35% in TOF and common arterial trunk, and 30–35% in pulmonary atresia/ventricular septal defect. It is less common in transposition of the great arteries, tricuspid atresia and isolated ventricular septal defect.9 By contrast, cases of RAA with an ALSA are associated with CHD in fewer than 20% of cases, with ventricular septal defect, atrial septal defect and transposition of the great arteries being the most common associations.7 ,9 The findings in our series concur with this, as ALSA was associated with structural CHD in 22.2% of cases, contrasting with RAA with mirror image branching which was associated with structural CHD in 91.5% of cases (p<0.05).
The presence of an ALSA has been described in the literature as an important marker for 22q11.2 microdeletion in patients with conotruncal malformations.10 ,19 In our series, we did not find a link between ALSA and 22q11.2 microdeletion, though the numbers are small and further evaluation is required.
Potential for a vascular ring
A RAA can be clinically important due to the potential for a vascular ring, which can lead to mechanical compression of the airway and/or oesophagus. The most common abnormalities of the aortic arch causing tracheo-oesophageal compression are DAA and RAA associated with an ALSA and left ductus arteriosus.1 ,6 In these cases, the possibility that a baby may develop a vascular ring is discussed at the time of prenatal counselling. The identification, before birth, of cases at risk of developing a clinically significant vascular ring can avoid delay in postnatal diagnosis and treatment. Some forms of vascular rings can be easily identified in fetuses, while others are more difficult to predict. A meticulous search for an ALSA or a DAA may help in the identification of patients at risk, as these findings are much more commonly associated with vascular rings. In cases of RAA with ALSA, the trachea and oesophagus are usually entrapped on the U-shaped vascular loop formed by the RAA, left descending aorta and left ductus arteriosus, connecting the left subclavian artery to the left pulmonary artery, completing a vascular ring. In DAA, there are two aortic arches, one on each side of the trachea and the oesophagus causing airway and oesophageal compression.
The indication for an MRI scan in every patient with a RAA is not clear. An MRI is indicated in symptomatic patients but may also be helpful, if available, in patients with aberrant vessels (ALSA or DAA) or in whom the anatomic details of the arch remain uncertain postnatally.
This study has limitations. First, we have a high-risk referral population and the incidence rate of 0.6% of a RAA observed in our study cannot be used to predict the rate of a RAA in a low-risk population. Second, we do not have a karyotype result for all patients, particularly in patients with isolated RAA, as it was not our policy to perform an amniocentesis or postnatal karyotyping in all patients with a RAA in the absence of any other abnormal finding.
Detailed fetal cardiac and extracardiac examination should be undertaken in all cases of RAA. Our study shows that an isolated RAA has a good prognosis. If completely isolated, a RAA is, in most cases, an asymptomatic vascular variant, and the risk for chromosomal abnormalities is relatively low. This information is important during prenatal parental counselling and for decision making with respect to karyotyping. The prognosis of a RAA with structural CHD depends on the complexity of the CHD and on the associated extracardiac anomalies. In these cases, there is a higher risk for a chromosomal anomaly, particularly 22q11.2 microdeletion. We would therefore advocate that karyotyping should be considered and discussed when a RAA coexists with structural CHD.
What is already known about this subject
The association between chromosomal anomalies and right aortic arch (RAA), although well reported, has been variable between different studies. Some authors have recommended karyotyping and screening for deletion of chromosome 22q11.2 in all fetuses with RAA. While this is widely accepted in cases associated with intracardiac anomalies or extracardiac malformations, controversy remains regarding management of cases with an isolated RAA.
What this study add
Our study shows an increasing prenatal detection rate of RAA, including cases of isolated RAA. The dilemma of whether further investigation is required following the diagnosis of a RAA in a fetus is further evaluated. Our data shows that an isolated RAA has a good prognosis and, in most cases, is an asymptomatic vascular variant with a relatively low risk for chromosomal abnormalities. By contrast, cases with associated congenital heart disease have a higher risk of associated chromosomal and extracardiac abnormalities, so that the prognosis in these cases is more variable.
How might this impact on clinical practice
The understanding that a completely isolated RAA has a good prognosis, and the risk for chromosomal abnormalities is relatively low, is important to assist prenatal parental counselling and for decision making with respect to prenatal karyotyping.
Centro Hospitalar São João, Porto—Portugal, for supporting Joana O Miranda's training in the Fetal Cardiology Unit, Evelina London Children's Hospital.
Contributors JOM and GS have contributed with the conception, design, writing and review of the manuscript. JOM and NC obtained and summarised all the data. OM and JS contributed with critical opinion and revision of the manuscript.
Competing interests None.
Patient consent Obtained.
Ethics approval Internal audit at Evelina London Children’s Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.