Purpose: To investigate the influence of (1) an additional side port for flushing the hollow pusher in Zenith thoracic stent-graft delivery systems and (2) additional carbon dioxide flushing on the amount of air released during stent-graft deployment. Methods: Twenty thoracic stent-grafts with an additional flush port to fill the hollow pusher were separated into 2 equal groups (C and D). Both groups were flushed with 20 mL of normal saline through the extra side port connected to the pusher and with 60 mL of saline through the regular flushing port. One group of grafts (group D) was additionally flushed with carbon dioxide through the regular flushing port prior to saline. All grafts were deployed into a curved plastic pipe attached to the bottom of a water-filled container. The released gas was recorded and measured using a calibrated setup. To evaluate the influence of the extra side port irrespective of the carbon dioxide flushing technique, group C was compared with a previously published reference group A without an extra side port that was flushed with the standard 60 mL of saline. Results: Volumes of gas were released in various amounts from the stent-grafts during deployment. The average amount of released gas was 0.51 mL in group C and 0.07 mL in group D (p<0.001). The mean amount of gas from group C samples (0.51 mL) was also significantly lower (p=0.002) compared with the reference group (0.79 mL). Conclusion: Thoracic endografts release air during deployment. Reducing the air-filled space inside the pusher of the catheter assembly using an additional side port can significantly reduce the amount of released air. Using the extra side port in combination with the carbon dioxide flushing technique reduces gas release further to small volumes. In a clinical setting this could be a promising approach to lower the risk of air embolism and stroke during thoracic endovascular aortic repair.