Limitations
The study defined aortic arch anomalies in patients undergoing CT scan
chest for variety of indications at a single institution. Most of these
indications were for non-cardiac/vascular reasons. While this eliminates
investigator bias, we noticed that CT scans read by radiologists who are
not fellowship trained in cardiovascular imaging underreport the
presence of arch anomalies, particularly the “bovine arch” variant.
The study included all types of CT scans (contrast CT
scans=49.3%(n=10,508) and EKG gated scans=2.5%(n=540)) which allowed
for a larger sample size, but might have resulted into significantly
lower rates of the reported anomalies, which are known to be better
visualized using contrast studies. We used a size threshold for
diagnosing TAA (4 cm as cut-off value for defining the root, ascending
and arch dilatations/aneurysms and 3 cm for the descending aorta) as
aortic size continues to be the initial screening tool for TAA by most
radiologists and physicians who can further refer the patients for
appropriate consultation. Although, the impact of normalization to body
characteristics (age, sex, BSA) in this study is unknown and might
influence the prevalence of TAA. The study did not address the
prevalence of arch anomalies in patients with thoracic aortic dissection
as the number of dissection patients in this database was too low to
analyze (n=32).