BACKGROUND: To improve computed tomographic (CT) prediction of local irresectability and to correlate preoperative CT findings with patient outcome. METHODS: Eighty-five patients with esophageal carcinoma underwent CT in supine, left lateral decubitus, and prone positions. CT signs that were indicative of local irresectability included (1) an angle of contact >45 degrees with the aorta; (2) obliteration of triangular fat pad between the tumor, aorta, and spine; (3) tumor contiguous with the aorta in all three positions; and (4) indentation of the airway in all three positions. RESULTS: All CT signs indicative for local irresectability concerning the aorta had comparable percentages of false-positive scans (75%) when correlated with surgical findings. When correlated with pathologic findings, >45 degrees angle of contact with the aorta yielded the fewest false-positive cases (9%). Concerning the airway, additional positions changed the staging correctly in 1 of 18 cases. Median survival was 21 and 8 months, respectively, for tumors considered CT resectable or irresectable. CONCLUSION: Additional patient positions do not improve the CT prediction of aortic invasion. Predicted resectability correlates with a significant longer life expectancy.
BACKGROUND: To improve computed tomographic (CT) prediction of local irresectability and to correlate preoperative CT findings with patient outcome. METHODS: Eighty-five patients with esophageal carcinoma underwent CT in supine, left lateral decubitus, and prone positions. CT signs that were indicative of local irresectability included (1) an angle of contact >45 degrees with the aorta; (2) obliteration of triangular fat pad between the tumor, aorta, and spine; (3) tumor contiguous with the aorta in all three positions; and (4) indentation of the airway in all three positions. RESULTS: All CT signs indicative for local irresectability concerning the aorta had comparable percentages of false-positive scans (75%) when correlated with surgical findings. When correlated with pathologic findings, >45 degrees angle of contact with the aorta yielded the fewest false-positive cases (9%). Concerning the airway, additional positions changed the staging correctly in 1 of 18 cases. Median survival was 21 and 8 months, respectively, for tumors considered CT resectable or irresectable. CONCLUSION: Additional patient positions do not improve the CT prediction of aortic invasion. Predicted resectability correlates with a significant longer life expectancy.