Results
This study enrolled 100 consecutive patients at time of undergoing CT or MRI of chest as a part of their medical care. Participant characteristics are described in Table 1. 88% of studies were completed by an imager with less than 1 year of experience. 2% of study subjects had sedation or anesthesia prior to echocardiogram, and 6% of study echocardiograms were reported as challenging or not possible due to limited acoustic windows.
Successful imaging of the trachea by tracheal sweep was possible in 97% (95% CI 95 – 100), p < 0.01. Right aortic arch was present in 4% of patients. There were no diagnoses of double aortic arch. There were no missed diagnoses of aortic arch anomalies, and the correct diagnosis of AoA sidedness and absence of DAA was 100% when the trachea was seen. Right aortic arch was correctly diagnosed in every patient where present. Kappa coefficient was 0.9. There were 3 tracheal sweeps that were non-diagnostic with poor acoustic windows: one patient had a BSA > 2.5 m2 and BMI greater than 40, a second patient had chronic lung disease and tracheostomy dependence, and the third had a progressive mediastinal mass. 2 of 3 non-diagnostic echocardiograms for arch sidedness were performed by an imager with less than 1 year of experience. Tracheal imaging was possible in patients with various non-cardiac pathologies including chronic lung disease, cystic fibrosis, mediastinal masses, and pectus excavatum. It was also possible in patients with complex cardiac disease including heterotaxy with hypoplastic left heart syndrome, transposition of the great arteries, truncus arteriosus, tetralogy of Fallot, complete atrioventricular canal, absent pulmonary valve, and other concerns for aortopathy.
Successful imaging of trachea-to-PA relationship could rule out PA sling in 91% (95% CI 85 – 97), p = 0.12. Excluding 6 patients who had previous surgical PA intervention, success rate was 97% (95% CI 93-100), p = 0.04. There were no missed diagnoses of PA sling with correct identification of PA anatomy was made 100% of the time when the trachea was seen. Kappa coefficient was 1.0. After exclusion of patients with previous PA surgery, there were 3 studies where tracheal-PA relationship was non-diagnostic with poor acoustic windows: one patient was status-post surgical repair of interrupted aortic arch, a second patient had chronic lung disease, and the third had a progressive mediastinal mass. All non-diagnostic studies for assessing PA sling were performed by an imager with less than 1 year of experience.