Symptoms of tracheoesophageal compression in children are generally associated with malformations of the aortic arch (AAMs). These malformations are represented by complete vascular rings (VRs), such as the double aortic arch (DAA) and right aortic arch (RAA) with the aberrant left subclavian artery (ALSA), or non-ring anomalies of the epiaortic vessels, such as the aberrant right subclavian artery and anomalous origin of the innominate artery (IA) from the aortic arch. The most common respiratory symptoms of airway compression are wheezing and stridor, followed by recurrent respiratory infections, barking cough, and poor exercise tolerance (1–3). On the other hand, dysphagia and failure to thrive are clinical manifestations of esophageal compression, which is mostly associated with complete VRs.

The true prevalence of AAMs is unknown as not all exhibit symptoms. However, VRs are estimated to account for approximately 1% of congenital cardiovascular anomalies (4, 5). Prenatal diagnosis using two-dimensional echocardiography (2D-ECHO) has certainly increased the frequency of AAMs being diagnosed either before or after birth when previously undiagnosed or discovered accidentally or after clinical manifestations. Postnatal diagnosis of AAMs, in addition to 2D-ECHO, includes multidetector computed tomography (MDCT) with angiography, cardiac magnetic resonance imaging (MRI), and airway endoscopy (AE), and the diagnostic approach depends on the predominant symptom and/or prenatal detection (3).

The purpose of this study is to report our experience in managing airway compression in patients with AAMs. We describe the symptoms, instrumental findings, treatment, and follow-up of a group of children affected by either complete VRs or anomalous origin of the IA. Additionally, we illustrate the diagnostic and therapeutic algorithm adopted in our tertiary center.

The clinical and instrumental data of all patients with AAMs enrolled for a multidisciplinary follow-up program at the Pediatric Pulmonology and Cystic Fibrosis Unit of Bambino Gesù Children's Hospital between January 1, 2016, and December 31, 2020, were retrospectively reviewed and analyzed.

Information regarding gender, gestational age, birth weight, type of AAM, presenting symptoms, associated anomalies, diagnostic investigations (2D-ECHO, MDCT with angiography, AE), need and indication for surgical treatment, and details of performed surgical procedures were obtained from electronic medical records with approval from the institutional review board.

The AAMs were classified into three groups based on the Stewart classification (1964) (6), the revised Stewart classification (2019) (7), and Backer's Congenital Heart Surgery Nomenclature and Database Project (8), as follows: –

group 1: DAA;

MDCT with angiography was performed during the inspiration and expiration phase without general anesthesia to better define the anatomical relationship between the airway and vascular structures.

Tracheomalacia and bronchomalacia were defined as dynamic airway collapse detected during flexible bronchoscopy made under light sedation and spontaneous quiet breathing (9) or during MDCT with angiography in the end-expiration phase (10).

Renal and abdominal ultrasound, 2D-ECHO, and genetic testing (karyotype and CGH array) were considered to screen the presence of major associated abnormalities.

When surgery was indicated, the preferred surgical approach was the left posterolateral thoracotomy.

All patients underwent close medical follow-up, which included a pulmonology visit every 6 months up to 2 years of age, every 12 months up to 6 years of age, and every 24 months over 6 years of age. The patients also had multidisciplinary follow-up visits with a cardiologist and ENT surgeon every year up to the age of 6 years and every 2 years thereafter. Pulse oximetry or polysomnography and pulmonary function testing were performed based on reported symptoms and age.

The study design was approved by the local Committee Board, and written informed consent was obtained from the parents of all the children.

Statistical analysis was performed using STATA 13.1 (Lakeway Dr, TX, USA). Continuous variables were expressed as mean and standard deviation or as median and (interquartile) range, depending on the normality of the data distribution assessed by the Shapiro–Wilk test. Categorical variables were expressed as numbers and frequencies. Continuous variables were compared using the Kruskal–Wallis test and categorical variables using Fisher's exact test. A p < 0.05 was considered statistically significant.

A shared and uniform program for the management of patients with AAMs and respiratory complaints that involves pediatricians, cardiologists, pulmonologists, and both ENT and cardiothoracic surgeons is still to be implemented (1, 2, 4, 5). In light of this, we aimed to report the findings of our study that was performed in a small group of patients diagnosed with AAM, followed-up and treated, when required, in a single institution over the last five years.

Grouping our patients, we distinguished DAA as a separate category due to its tendency to cause greater and progressive compression of the trachea and esophagus, leading to earlier clinical manifestations compared to other vascular causes of airway compression (11–13). Our analysis showed that RAA with ALSA was the most common type of VR, as recently confirmed by other studies (14, 15), often associated with a deletion of 22q11.2 (16, 17).

In agreement with other series (4, 18–23), respiratory symptoms were predominant in our series, including mainly stridor and wheezing. We observed a significantly higher prevalence of respiratory symptoms at presentation in group 1 compared to group 2 as DAA exerts stronger compression on the airways and digestive tract compared to other complete vascular rings.

In our cohort, the median age at the onset of symptoms was 2 months. However, no significant difference in the age of symptoms onset among the three groups was detected, contrary to findings reported by other authors (21).

The median age of patients in our series diagnosed after birth aligned with previous reports (23–26). A different issue concerns prenatal diagnostics (21, 25), which supported us in cases of DAA, where proactive treatment was initiated even in asymptomatic patients to prevent the long-term effects of prolonged tracheal compression (27). On the other hand, for cases of RAA with ALSA, fetal diagnosis supported us in targeted patient monitoring, anticipating postnatal diagnostics, and allowing for the early detection of symptoms that might have otherwise gone unnoticed. Finally, with reference to the anomalous origin of IA, prenatal diagnostics had no role as this condition presents symptoms only after birth.

MDCT imaging supported by angiography was the primary diagnostic examination at our center for postnatal diagnosis. Indeed, it is able to accurately delineate the anatomy of AAMs and associated tracheal or bronchial pathology in a short acquisition time (28) during free breathing and without general anesthesia. Dynamic MDCT has also proven to be a diagnostic non-invasive exam for trachea-bronchomalacia (29). Consequently, children with suspected AAM underwent MDCT with angiography either at the onset of symptoms or routinely 6 months after birth in the case of prenatal diagnosis.

The combination of imaging and airways endoscopic evaluation allowed us to detect tracheomalacia and bronchomalacia in 42.4% and 13.6% of patients, respectively. This agrees with what has already been reported by Yubbu and colleagues, who described complete vascular rings as the principal cardiovascular cause of tracheomalacia (53.6%) and bronchomalacia (10.7%) (30).

Surgery remains the only effective treatment for symptomatic AAMs (19). In our entire cohort, the median age at surgery was 6 months, earlier than reported in other studies (13, 21, 31), probably due to the impact of prenatal diagnosis, particularly in asymptomatic patients. The surgical approach predominantly includes thoracotomy, as previously described (13, 32, 33), There was no intra- and postoperative mortality but only a limited number of minor complications, proving the safety of surgery in the case of AAMs (18, 21).

Our study has several limitations. Firstly, it had a retrospective design without a control group. Secondly, the small number of patients in group 3 may explain the low strength of our statistical analysis. Finally, the short follow-up period prevented us from defining clear indications for surgery that predict the resolution of symptoms. Intuitively, we speculate that significant airway compression (>50% reduction in airway lumen) and/or the presence of symptoms may serve as significant predictive factors. Despite these speculations, we acknowledge the relatively weak strength of our conclusions.

On the other hand, the strength of our study lies in the exclusive inclusion of patients with AAMs related to vascular compression and the exclusion of those with associated tracheo-bronchial and pulmonary malformations. This approach minimized confounding factors that could influence the onset of respiratory symptoms.

In summary, our study describes our comprehensive approach to patients with AAMs, as shown in Figure 5. This approach involves: (1) the early diagnosis of vascular abnormalities; (2) timely surgical intervention in the case of DAA and/or respiratory or digestive symptoms impacting on the child's quality of life and/or airway stenosis exceeding 50%; and (3) close monitoring of asymptomatic patients and/or cases where vascular compression leads to less than 50% reduction in airway lumen (3). However, further prospective studies with long-term follow-up, ideally including a control group, are needed to establish clear guidelines for the diagnosis, treatment, and follow-up of patients affected by AAMs.