Over the years, intracranial metal stents have been widely used for interventional treatment of atherosclerotic stenosis and stent-assisted embolization of intracranial aneurysms [3,4]. However, the visibility of implanted intracranial stents under ordinary fluoroscopy can be limited despite radioopaque markers on the proximal and distal stent ends. Kinking, incomplete expansion or poor apposition of the stent may occur in severely tortuous blood vessels or stenosed vessels with calcified plaques, resulting in complications such as acute stent thrombosis, perforator vessel occlusion, and prolonged stent endothelialization time [2,22]. The past decade has witnessed several innovations in digital subtraction angiography (DSA) equipment and scanning and postprocessing technologies that have facilitated the progress of neurointerventional therapy. In this regard, high-resolution C-arm computed tomography (CT) provides a clear visualization of the internal structure of lesions [14]. However, few studies have explored the safety and accuracy of stent apposition assessment by high-resolution C-arm CT combined with different post-scan processing techniques [5,18,20]. In this study, we aimed to determine the safety and efficacy of the combination of high-resolution C-arm CT with different postprocessing methods when evaluating the success of intracranial stent implantation in Guangxi Bama piglets.

The Ethical Committee of The First Affiliated Hospital of Zhengzhou University, authorized this study (grant number : 2022-KY-0911).

It is well-established that early 3D C-arm CT used flat-panel (FD) detectors. After replacing the image intensifier with a flat-panel detector, it was found that the ordinary DSA system could acquire CT-like images through a C-arm rotation scan. In its early stages, this technique was termed flat-panel detector CT or FD-CT. C-arm CT models currently available include the Xper-CT (Philips, Best, Netherlands) and the Dyna-CT (Siemens), while high-resolution C-arm CT models available include the Vaso-CT (Philips) and the Dyna micro-CT (Siemens). The C-arm CT is mainly used for head, neck, chest, and abdominal scanning, while high-resolution C-arm CT is indicated for scanning implanted intracranial stent and coil, providing significantly better spatial resolution. Current evidence suggests that high-resolution C-arm CT combined with metal artifact removal technology significantly reduces the metal artifacts of stents or spring coils during neuro intervention and markedly improves image quality [14]. Commercial metal artifact reduction (MAR) algorithms for spiral CT, including iterative reconstruction (e.g., iMAR) and projection interpolation techniques, have demonstrated artifact reduction rates of 22.5% to 47.3% in hip prostheses, with optimal soft-tissue performance (artifact attenuation up to 41.2% reduction in Hounsfield units [HU]) [9]. However, these techniques exhibit suboptimal efficacy in neurointerventional C-arm CT applications. This limitation arises from inherent challenges posed by 1) high-density embolization coils (attenuation >3000 HU) and 2) submillimeter-scale stent struts (strut thickness : 0.06-0.08 mm), both of which exceed the corrective capacity of conventional MAR systems optimized for orthopedic implants with bulkier metallic components (typical density <2500 HU). Compared to other MAR algorithms, the high-resolution DynaCT Micro (Siemens Healthineers, Erlangen, Germany) incorporates the SMART reconstruction algorithm by utilizing a non-binned acquisition mode, optimized exposure parameters (tube voltage : 109 kVp; tube current : 1.8 mAs/frame), and a reduced field-of-view (FOV; 12×12 cm²). This method achieves an isotropic spatial resolution of 0.15 mm, as confirmed by phantom studies, representing a 65.1% improvement over the baseline resolution of 0.43 mm achieved with other MAR algorithms [5]. A previous in vitro study demonstrated that high-resolution C-arm CT could provide a more detailed display of intracranial stents [13]. To the best of our knowledge, few studies have hitherto assessed the dose of radiation received by patients during the scanning process or the accuracy and safety of various techniques for assessing stent apposition [5,18,20].

Current evidence suggests that high-resolution C-arm CT exhibits a better performance than traditional C-arm CT for displaying different types of intracranial stents during stent-assisted aneurysm embolization and can thus help reduce the long-term recurrence rate of aneurysms and the incidence of complications [5,14]. Consistent with the literature, we found that high-resolution CT yielded clear reconstruction images of the implanted stents, with grade I quality in most cases (22/24). Indeed, a good image quality helps identify problems such as incomplete expansion, distortion, dislocation, or stent fracture during stent release. In a study where high-resolution C-arm CT was used to assess Willis covered stent placement, twisting at the “inverted V” tip of one laser-engraved stent and stent tip embedding in the vessel wall were observed and could not be detected on conventional C-arm CT and fluoroscopy [1].

The diluted contrast medium method is relatively accurate and commonly used for evaluating implanted intracranial stents, especially intracranial dense-mesh stents. However, this technique has certain limitations since the DSA technicians need to reconfigure the low-concentration contrast medium and replace the high-pressure syringe barrel, which prolongs the procedure time. At the same time, the high flow rate and prolonged injection duration (22-25 seconds) of contrast medium through a catheter into the artery increases the risk of vasospasm and dissection. Li et al. [12] reported a case of arterial dissection at the tip of the angiography catheter when the diluted contrast medium method was used following Tubridge FD implantation and treated by placement of two Neuroform EZ stents. In our experiment, mild vasospasm at the tip of the angiographic catheter was observed in four piglets using our approach, highlighting the need for extra caution when performing this technique. Moreover, the tip of the angiographic catheter should not touch the wall to avoid damage induced by the high-speed jet of contrast.

In this study, we found excellent consistency between the dual-volume reconstruction and the diluted contrast medium methods for evaluating stent apposition evaluation, although lower radiation exposure and contrast agent dose were associated with the diluted contrast medium method. It should be pointed out that the dual-volume reconstruction method has strict requirements regarding the body posture during the two scans since even slight displacement can lead to the mismatch of merged images and result in false-positive results for stent malapposition. In addition, the quality of the processed images obtained by the dual-volume reconstruction method is determined by the postprocessing staff’s expertise and patience. In recent years, the high-resolution C-arm CT scanning technique (Vaso-CT) has emerged, requiring an injection of a small amount (9.6 mL) of undiluted contrast agent, with the flow rate adjusted to allow dilution by the blood flow [7]. This technique can provide images with resolution comparable to the diluted contrast medium method and improve the efficiency of high-resolution flat-panel CT for assessment of FD stent apposition.

The quality of imaging yielded by the diluted contrast medium method is related to the C-arm CT device’s physical parameters and the contrast agent’s concentration. Excessively high or low concentrations affect the assessment of the relationship between the stent and the vessel wall. In such cases, the procedure may be repeated. It should also be borne in mind that the concentrations of contrast agents used for high-resolution scanning largely depend on the clinical experience of the operator and the angiography device. The concentration of contrast agents tends to be 10-25% on Siemens devices [5,8] and 10-20% on Philips devices [18,20]. The contrast agent concentrations and injection times required to achieve optimal visualization of the stent, vessel lumen, and vessel wall on different angiography devices must be determined in future studies.

It should be noted that, due to significant differences in vascular anatomy between the experimental animals and humans (average vessel diameter ratio of 1 : 2.8), as well as variability in radiation beam angles during surgery (standard deviation up to 12.7°) caused by positioning challenges, we did not directly compare the absolute radiation doses between the dual-volume reconstruction and the diluted contrast medium methods. Instead, we measured the additional radiation dose relative to the standard surgical procedure. Our experimental data indicate that the dual-volume reconstruction method incurs an approximately 26% higher additional radiation dose than the diluted contrast medium method (Table 1). Nevertheless, this extra dose provides valuable guidance for clinicians when selecting imaging techniques (Table 2). Specifically, the 26% increase is primarily due to : 1) increased X-ray tube loading during the dual-phase acquisition and 2) the additional fluoroscopy time required for 3D-DSA.

While based on an animal model, the multiplanar reconstruction capability of high-resolution C-arm CT effectively overcomes limitations inherent to conventional 2D fluoroscopy, demonstrating superior performance in evaluating stent apposition within calcified vascular segments (simulating diabetic/hypertensive vasculopathy). Furthermore, the technique shows broad applicability for assessing complex vascular pathologies encountered clinically, including atherosclerotic plaque-induced stenosis and congenital tortuosity.

Our research has been extended to human cohorts with protocol optimization [15]. Given the anatomical disparities between Bama minipigs and humans (vascular diameter : 2.8±0.3 mm vs. 4.1±0.5 mm; p<0.001), dual injection protocols were implemented [16] : dual-volume reconstruction : undiluted contrast (5 mL/s; total 14 mL) and low-concentration protocol : 12% hyperosmolar (iopromide 370 mg I/mL) or 16% iso-osmolar (iodixanol 320 mg I/mL) contrast at 3.0 mL/s (total 66 mL over 20 seconds acquisition).

Some limitations were found in the present study. The sample size was small, with only 24 experimental animals and the findings of this study are inherently linked to the imaging parameters of the Artis Zeego system. Furthermore, it should be emphasized that although our study systematically evaluated stent expansion morphology and wall apposition using high-resolution C-arm CT, the study protocol did not include the next-generation Pipeline Flex FD (Ev3, Plymouth, MN, USA). Since this advanced device was not yet integrated into our institution’s clinical practice at the time of the experiments, our findings primarily apply to the technical assessment of conventional stents and current flow-diverter platforms. To address this evidence gap, we plan to conduct a prospective, multicenter registry study with independent, blinded DSA core laboratory readings, comparing three categories of devices—1) conventional stents/first-generation FDs; 2) FRED/Pipeline Flex FD; and 3) FRED Jr/FRED X—using standardized protocols. This study will focus on evaluating stent apposition relative to aneurysm morphological parameters (e.g., dome-to-neck ratio, vessel angulation at the aneurysm neck) and its impact on long-term aneurysm occlusion. Future studies should prioritize multicenter comparisons of imaging equipment, standardized postprocessing protocols, systematic animal cohort validations, and confirmatory, hypothesis-driven clinical trials to ensure reproducibility and generalizability of results across biological scales and clinical contexts.

High-resolution C-arm CT enables clear visualization of the microstructure and expansion of intracranial stents. When combined with different postprocessing techniques, the degree of stent apposition can be clearly assessed. However, the advantages and disadvantages of each postprocessing technique should be considered before implementation during clinical practice.