PurposeTo propose an optimization method of contrast medium injection for the split-bolus protocol based on the contrast medium pharmacokinetics and investigate the utility of the optimized split-bolus protocol in 3D CT angio-venography for laparoscopic gastrectomy.Materials and methodsA pharmacokinetic relationship between injection duration and time to the peak enhancement was taken into account in the protocol design. The first 20 consecutive patients underwent a multi-phase scan with a single-bolus injection (single-bolus protocol), and the next 20 underwent the proposed split-bolus protocol. CT attenuations of the arteries and veins and dose-length products (DLPs) were compared between the two protocols. Two radiologists visually assessed arterial and venous depictions and the misregistrations.ResultsMean arterial CT attenuations were not significantly different between the two protocols. Though mean venous CT attenuations for the split-bolus protocol were 7–11% lower than those of the single-bolus protocol, they were visually evaluated as similar. The mean DLP of the split-bolus protocol was 46% lower than that of the single-bolus protocol. Misregistration between the arteries and veins occurred 35–80% during the single-bolus protocol, but was not indicated in the split-bolus protocol.ConclusionThe split-bolus protocol with optimized timing was more effective for providing improved image quality with reduced radiation dose compared with the single-bolus protocol in 3D CT angio-venography for laparoscopic gastrectomy.