In this study, focusing on the ‘Venous Appearance Time (TVA)’ and the ‘Arterial Clearance Time (TAC)’ in the setting of crucial X-ray delay time during cranial venous 3D-DSA imaging, we devised a method to accurately measure the contrast agent arterial transit time using TEC from 2D-DSA. This method was reflected in the X-ray delay time settings, demonstrating its significance.

Visish et al. [9] have reported the utility of intracranial venous 3D-DSA with X-ray delay times set by visual inspection. However, our study’s findings challenge the reliability of this visual method. We found no significant correlation between visually measured aTAC and TEC-calculated eTAC (correlation coefficient R = −0.23), suggested that determining timing based on visual inspection during cerebral vein 3D-DSA imaging does not accurately reflect the optimal timing (Fig. 4). This low correlation coefficient indicates a substantial discrepancy between visual estimation and objective measurement, highlighting the potential inaccuracies in the visual method. Some previous reports [10,11,12] have pointed out challenges with the visually based inspection methods, even when standardized. Visual assessments are likely varied due to differences in experience and individual subjectivity. Our quantitative analysis supports these previous observations and provides concrete evidence of the limitations of visual assessment. Therefore, for precise venous visualization, referencing TEC is no only advantageous but arguably necessary to achieve consistent and optimal imaging results. Our method provides an objective, quantifiable approach to timing determination, potentially reducing inter-observer variability and improving the overall quality and reliability of cerebral venous 3D-DSA imaging.

TEC analysis for vein near the tumor indicates that early initiation of imaging, following the cessation of contrast agent injection and beyond the influence of arteries, is essential for visualizing tumor-adjacent veins. Kashimoto et al. [13] have proposed a novel protocol for cerebral venous 3D-DSA using low-dose contrast agents, employing TEC analysis of cerebral arteries, cortical veins, and venous sinuses. However, no previous studies have focused on measuring timing with such detail on vessels near tumors, as done in this study, making this method not only novel but also clinically valuable. Furthermore, while providing volume rendering images is useful for understanding the 3D structure of tumors and blood vessels in brain tumor resection surgeries, there is a report indicating that that accurate reproduction of shapes in 3D imaging requires the contrast to be present for more than 80% of the duration of each rotation. [14] Achieving such sufficient contrast effect is crucial, and thus it is important to start imaging only after the contrast agent has adequately filled the veins. Knowing the precise TVA of the target vessel is also essential.

For effective visualization of early enhancing veins during 3D-DSA imaging, it is essential to initiate imaging before the complete clearance of arterial contrast. This study, therefore, focuses on TAC, specifically the transit time of the contrast agent from the proximal to peripheral arterial regions. This transit time, calculated by subtracting the peak arrival times between the internal carotid artery and the M4 segment, averaged 1.16 ± 0.2 s for all patients. Figure 8 presents a conceptual model of the TEC under hypothetical 3D-DSA conditions. This model illustrates a scenario where the contrast agent was injected for 7 s until the veins near the tumor were visualized (TVA). After the cessation of contrast agent injection, the concentration of the contrast agent in the veins near the tumor and the superior sagittal sinus reached its peak before the arterial contrast was cleared. Therefore, by starting the imaging immediately after the TAC following the cessation of contrast agent injection, the peak contrast in the veins near the tumor and the superior sagittal sinus can be captured for a longer duration. The arterial contrast is cleared immediately after the imaging begins, thus avoiding any interference. Although the peak contrast in other veins occurs slightly later, the contrast concentration remains sufficient during and for some time after the imaging, thereby not significantly affecting the timing of the imaging start. Since the TAC correlates with the intracranial blood flow velocity of each case, it is a crucial factor in determining the optimal Tdelay.

A conceptual model of the TEC under hypothetical 3D-DSA conditions for a case of meningioma. In this case, the contrast agent was injected at a rate of 3.5 ml/s for 7 s (A), resulting in a total volume of 24.5 ml of contrast medium used. After the cessation of contrast agent injection, imaging is initiated based on the TAC derived from the TEC obtained from 2D-DSA images, allowing for optimal image acquisition (B)

In this study, we placed ROIs on the MCA for TEC analysis due to its accessibility and ease of measurement. Favorable results were obtained even for tumors supplied by other vascular territories, suggesting the clinical viability of this approach. However, if significant differences in clearance times are observed during TEC confirmation, using the TEC of the specific supplying artery would be appropriate. It’s important to note that placing ROIs on peripheral ACA or PCA can be challenging in some cases, which may limit the applicability of this method in certain scenarios.

To validate our theory, we defined the A/ETDR and divided the cases into two groups: Group A, where the aTAC used during imaging was close to or shorter than the eTAC calculated from the TEC, and Group B, where it was longer. The threshold was set to A/ETDR = 1.2–1.5 for comparison. The superior sagittal sinus, internal cerebral vein, and veins near the tumor, all of which enhanced early, showed significantly higher signal values in Group A compared to Group B. Additionally, as the threshold decreased, the p-values between the groups decreased, suggesting the significance of this method. While no statistically significant difference was observed for the superior cerebral vein, this may indicate that the limited number of cases influenced the results. However, for the cavernous and sigmoid sinuses, where the contrast agent peak appears later and the concentration decreases gradually, leading to no significant difference between Groups A and B.

By using our proposed TEC-based timing measurement method, it is possible to perform imaging at a timing that approaches an A/ETDR of 1.0, allowing for detailed venous imaging with a high level of information. Our proposed method for determining the timing of cerebral venous 3D-DSA imaging, based on TEC calculated from 2D-DSA images, can be applied to any angiography system with the same rotation time. Although this study focused solely on the veins of tumors, we believe that this method may potentially be applied to certain cases of arteriovenous malformations (AVMs) and dural arteriovenous fistulas (dAVFs), which can enhance early similar to tumor veins. However, it is important to note that in cases where there is minimal or no time lag between arterial and venous enhancement, such as in high-flow AVMs or dAVFs, this method may not be applicable. The effectiveness of this technique in these conditions would depend on the presence of a discernible time difference between arterial and venous phases. This method might also be useful in the detection of occlusion sites in certain cases of cerebral venous thrombosis, where normal venous drainage patterns are altered. Future studies are needed to validate the applicability and limitations of this method in these diverse pathological conditions.

Figure 9 presents actual case images for Groups A and B. In Group A, visualization of peripheral arteries was observed immediately after the initiation of 3D-DSA imaging; however, visualization of the rapidly disappearing veins near the tumor was also confirmed, with no complete disappearance of the contrast agent in the veins during imaging. The 3D images showed no arterial delineation and clear visualization from the entire cerebral vein to the veins near the tumor (Fig. 9a). Conversely, in Group B, there was no initial visualization of peripheral arteries. During imaging, some venous vessels were almost entirely devoid of contrast agents, resulting in insufficient venous visualization in many cases, leading to poor image quality (Fig. 9b).

Actual case presentation. In both cases, the contrast agent was injected at 3.5 mL/s for 7.0 s, totaling 24.5 mL. a Images obtained from a 72-year-old man with a right petrous part meningioma. (aTAC: 1.2 s, eTAC: 1.34 s, A/ETDR = 0.90) In this case, the area near the tumor to the peripheral vein is clearly depicted without arterial contamination. b Images obtained from a 26-year-old man with a left fornix meningioma. (aTAC: 2.3 s, eTAC: 1.33 s, A/ETDR = 1.73) In this case, the veins near the tumor were not visible, and the peripheral veins were poorly delineated

This study has several limitations. The timing investigation for cerebral venous 3D-DSA imaging was conducted retrospectively using past image data and was solely reliant on TEC analysis. Factors such as patient background (including tumor type, location, size, intracranial pressure status, sex, and age) were not considered. Furthermore, obtaining high-quality venous images prerequisite optimized contrast agent conditions (such as injection rate, volume, catheter type, and vessel positioning). While this study focused on optimizing imaging timing, it is essential in clinical practice that both timing and contrast agent parameters are optimized in tandem. As timing optimization may also lead to the optimization of contrast agent injection volume, future studies should investigate both imaging timing and contrast agent conditions simultaneously. Additionally, this study excluded cases with injections from the vertebral artery. It is necessary to verify whether this method can be applied to cases with injections from the vertebral artery. A more detailed analysis of these factors may enable more precise timing settings in the future.