Venous reflux on carotid computed tomography angiography: relationship with left-arm injection.

Objective: We evaluated the relationship between image degradation due to the reflux of contrast agent into the major neck veins and use of a left-arm injection site during computed tomographic (CT) angiography of the carotid arteries. Methods: A total of 364 patients underwent 4-section carotid CT angiography performed with an injection into the right (n = 183) or left (n - 181) arm. We calculated the volume ofrefluxed contrast medium and measured the retrosternal distance measured as the shortest distance between the sternum and the arch or its branches. Nine patients underwent follow-up CT angiography weeks later by using the side of injection not used before for comparison. Results: The amount of refluxed contrast medium was greater with left-arm injection than with right-arm injection (P < 0.001). With left-ann injection, the volume was significantly correlated with the retrosternal distance (r = -0.514, P < 0.001), even after we adjusted for age and sex in the linear regression model (P < 0.001). Five of the 9 patients who received injections in both arms had no reflux of contrast medium, whereas 4 patients had obvious reflux after left-arm but not right-arm injection. Conclusion: Normal compression of the left brachiocephalic vein due to a developmental decreased retrosternal space may degrade carotid CT angiograms because of reflux of contrast material into the cervical veins. This degradation can be avoided if right-arm injection is used.