Se. Mirvis et al., TRAUMATIC AORTIC INJURY - DIAGNOSIS WITH CONTRAST-ENHANCED THORACIC CT - 5-YEAR EXPERIENCE AT A MAJOR TRAUMA CENTER, Radiology, 200(2), 1996, pp. 413-422
PURPOSE: To review the literature and the authors' experience with adm
ission contrast material-enhanced computed tomography (CT) in patients
admitted after blunt trauma. MATERIALS AND METHODS: From January 1990
to December 1994, thoracic CT was performed to evaluate traumatic aor
tic injury in 677 patients with positive or equivocal findings at ches
t radiography. CT scans positive for mediastinal hemorrhage or aortic
injury were retrospectively reviewed and interpreted by consensus, wit
h angiographic and surgical confirmation. RESULTS: CT findings were ne
gative for TAI in 570 (84%) of the 677 patients. Mediastinal hemorrhag
e was reported in 100 patients and was the only abnormality in 79 of t
hem. Findings at angiography were negative for traumatic aortic injury
in 77 (97%). CT signs of traumatic aortic injury in 21 patients inclu
ded contour abnormality or pseudoaneurysm (n = 19), intimal flap(s) (n
= 8), and pseudocoarctation (n = 3). Findings at angiography were pos
itive for traumatic aortic injury in 19 (90%). For aortic injury and m
ediastinal hemorrhage, respectively, specificity for traumatic aortic
injury was 99% and 87% and sensitivity was 90% and 100%; at meta-analy
sis of data from the authors and the literature, sensitivity was 97.0%
and 99.3% and specificity was 99.8% and 87.1%. Reliance on findings a
t admission CT rather than radiography to indicate suspicion for traum
atic aortic injury before angiography resulted in savings of more than
$365,000. CONCLUSION: The CT finding of mediastinal hemorrhage alone
is sensitive for traumatic aortic injury, but the finding of aortic in
jury is more specific.