Our results show contrast and image quality improvement in low-keV VMI reconstructions in the evaluation of carotid vessels in comparison with other VMI energy levels and standard images in PCCT. In particular, 40 keV VMI showed the highest mean attenuation, CNR and SNR (HU, 1362.32 ± 457.81; CNR, 33.19 ± 12.86; SNR, 34.37 ± 12.89) followed by 55-keV VMI series (HU, 736.94 ± 150.09; CNR, 24.49 ± 7.11; SNR, 26.25 ± 7.34). Furthermore, the low-keV VMI series (40–85 keV) received a superior subjective rating compared to both higher keV images and the standard polychromatic 120 kV. Notably, 55-keV VMI reconstructions exhibited the highest values for subjective parameters, demonstrating an overall image quality of 4.6 ± 0.3. This was closely followed by the 70-keV VMI, with an overall image quality of 4.2 ± 0.7 with not statistically significance (p > 0.05). In contrary, subjective image quality assessment yielded superior ratings for 70-keV VMI, possibly due to the visual perception of high noise levels at 40-keV.

As already demonstrated in the literature, DECT-derived VMI + provides numerous advantages in cardiovascular imaging, improving image quality and diagnostic accuracy [17,18,19,20,21]. In particular, low-keV levels (40–75 keV) close to the iodine K-edge (33.17 keV), increase the iodine attenuation, providing superior quantitative image quality parameters, such as SNR and CNR [22]. Previous studies evaluated DECT VMI for vascular assessment. In this context, Leithner et al. [23] demonstrated increased suitability of 40-keV VMI + images for carotid and intracranial artery assessment using a second-generation dual-.source DECT system. Aside from this, low-keV VMI + showed to reduce the contrast medium dose needed to obtain good image quality with high diagnostic value [24, 25].

There are pioneering PCCT studies evaluating VMI for vascular assessment. Dillinger et al. [26] evaluated PCCT VMI on the visualization of abdominal arterial vessels, demonstrating that 60–70 keV VMI provides best results in qualitative and quantitative image quality assessment, which is slightly different compared to our results showing optimal image quality and suitability for carotid artery assessment at lower keV (40–55 keV energy series). Accordingly with our findings, Sartoretti et al. [27] demonstrated the value of PCCT VMI in the evaluation of coronary arteries, with optimal image quality at an energy level of 40 keV. Reason for these results may be the invention of photon-counting detectors, allowing the reduction of electronic noise by counting the number of pulses greater than a preset threshold; secondly, photons of different energy are equally weighted, unlike in DECT technologies with energy-integrating detectors (EID), where high-energy photons contribute to more than low-energy photons, consequently decreasing CNR [28].

In addition, the increased iodine attenuation at low keV (40–55 keV) could lead to a reduction of contrast medium amount, still maintaining high image quality for clinical practice. Previous studies conducted with DECT have been presumed to enable a reduction of contrast media amount [22, 29]. This may be very useful in patients with comorbidities, such as mellitus diabetes and renal kidney disease [12, 30]. Although our study did not specifically investigate this aspect, our findings showed a substantially improved quantitative image quality at low keV, far greater than 300 HU average enhancement defined as diagnostic for vascular structures [31, 32]. This improvement suggests the possibility to reduce contrast medium amount, while preserving image quality comparable to that of standard CTA. This significant improvement concerning iodine attenuation could open up possibilities for optimizing acquisition parameters with the aim of reducing radiation dose during CTA scans. Further studies are needed to comprehensively evaluate the feasibility and effectiveness of this future perspective (Fig. 6).

A 78-year-old female with aortic valve stenosis and comorbidities including diabetes mellitus type 2, and chronic kidney disease, undergoing PCCT angiography pre-TAVI, using 80 ml of 350 mgI/mL Xenetix at an infusion rate of 5.0 ml/second (Xenetix 350 mgI/ml, Guerbet, Villepinte, France). PCCT angiography reconstructions in paracoronal and axial plane (A–B) at low-VMI energy levels 40 keV show less amount of hypoattenuating artifacts and a regular opacification of right ICA, clearly indicating a regular perfusion, compared to paracoronal and axial reconstructions (C–D) in standard 120 kV. ICA Internal Carotid Artery; PCCT Photon-counting CT; TAVI Transcatheter Aortic Valve Implantation

Moreover, PCCT maintains high time resolution with no risk of compromising vascular imaging [33]. Furthermore, PCCT provides spectral imaging in every CT scans. As a result, the integration of PCCT applications into routine clinical practice holds the potential to modify current protocols, allowing reduction of contrast medium dose through VMI, as well as radiation dose, also because of the availability of VNC substituting the native scan in selecting cases. In this context, VNC demonstrated the capability to differentiate between intracranial hemorrhage and extravasation of iodinated contrast media after ischemic stroke therapy with high diagnostic accuracy [34, 35]. This benefit could enable radiologists to use dedicated applications, even for the evaluation of incidental findings, where the lack of a dedicated protocol in conventional CT would otherwise render a definite diagnosis impossible, or in case of insufficient intravascular enhancement (e.g., in case of a missed contrast medium bolus) [36].

This study has limitations that have to be addressed. First, the retrospective study design may have influenced our results. Second, the ROI drawing was difficult in some cases due to vessel disease (calcifications and soft plaques) and may have affected the analyses. Third, we arbitrarily chose energy levels from 40 to 100 keV, with an increment of 15 keV; results may be different at unrated energy levels. Higher energy levels were not included because our study is focused on the benefit of VMI in the improvement of iodine attenuation and image quality in carotid CTA, provided by low-energy level, according to the literature. Nevertheless, it should be noted that in the case of calcified plaques or stenosis, higher VMI energy levels may reduce blooming artifacts. Indeed, we did not evaluate the diagnostic performance in the assessment of carotid pathologies, such as stenosis and atheromatic plaques (both soft and calcified). In this context, standard 120 kV reconstructions play a crucial role in minimizing artifacts arising from extensive calcified components within the plaque. Forth, the relatively small sample size based on available patient data may affect the generalizability of our results; further studies with larger prospective cohorts are needed to provide more robust evidence in this field. Finally, we have not conducted a direct comparison between PCCT technology and older DECT technologies; consequently, the assessment of VMI performance obtained from these two distinct technologies remains unknown.

In conclusion, PCCT 40-keV VMI showed the highest values of SNR and CNR in quantitative evaluation. Conversely, regarding subjective evaluation of vessel assessment, 55-keV VMI received the best scores. Thus, our results suggest that these VMI PCCT reconstructions (40–55 keV) could be feasible to optimize the suitability for carotid arteries assessment, improving image quality, compared to standard 120 kV reconstructions. Moreover, in clinical practice, the ever-available spectral dataset might improve vascular visualization in all studies, even those performed with non-vascular protocols, thereby leading to a reduction in contrast medium amount.