Dual-Layer Spectral CT of Pancreas Ductal Adenocarcinoma: Can Virtual Monoenergetic Images of the Portal Venous Phase Be an Alternative to the Pancreatic-Phase Scan?

Objectives: To determine the performance of virtual monoenergetic images (VMIs) of the portal venous phase (PVP) compared with the pancreatic-phase image for pancreatic ductal adenocarcinoma (PDAC) evaluation. Materials and methods: This retrospective study enrolled 64 patients with PDAC who underwent pancreatic CT with dual-layer spectral CT between February 2018 and January 2020. A polychromatic pancreatic-phase image and VMIs at 40 (VMI40), 55 (VMI55), and 70 keV (VMI70) of the PVP were generated. The tumor-to-pancreas contrast-to-noise ratio (CNR), attenuation difference, peripancreatic vascular signal-to-noise ratio (SNR), and CNR were compared among the four images. Subjective image analysis was performed for tumor conspicuity, heterogeneity, size, and arterial invasion. Results: VMI40 and VMI55 demonstrated higher tumor-to-pancreas CNR, attenuation difference, and higher peripancreatic vascular CNR and SNR than the pancreatic-phase image and VMI70 (p < .001). On subjective analysis, VMI55 showed the best tumor conspicuity. Moreover, the inter-reader agreement for arterial invasion in VMIs from the PVP was not inferior to that in the pancreatic-phase image. Conclusion: For evaluating PDAC, the VMI55 of the PVP was superior to the pancreatic-phase image in terms of tumor conspicuity and peripancreatic vascular enhancement. Therefore, the VMI55 of the PVP could be an alternative to the pancreatic-phase scan in patients suspicious of PDAC.

The flow diagram of study participants were demonstrated in the Fig. 1. The inclusion criteria were PDAC patients who underwent pancreatic CT scan with dual-layer spectral CT. For patient inclusion, the CT reports of 2,848 consecutive patients who underwent pancreatic CT using dual-layer spectral CT between February 24, 2018 and January 29, 2020 were reviewed. Among CT reports of 2,848 patients, primary pancreatic cancer is mentioned in 77 patients. Sixty-five patients were included who were subsequently diagnosed with PDAC. The PDAC were pathologically diagnosed after pancreatic resection (n=8), EUS guided pancreatic biopsy or FNA (n=40), biopsy from hepatic or lung metastasis (n=7), or clinically diagnosed based on radiologic findings (MRI, follow-up CT), PET-CT, and elevated cancer antigen 19-9 (n=10). Twelve patients were not included because of the following reasons; final diagnosis of other tumors (n=4) or focal pancreatitis (n=2), normal pancreas on MRI and PET-CT (n = 1), follow-up loss (n = 5). The exclusion criteria were poor tumor conspicuity on CT scan for image anlysis, and one patient was excluded because of the tumor was not visualized on CT image owing to pancreatitis. This patient was diagnosed with PDAC after a pancreatectomy.

CT Protocol and Image Reconstruction
All CT scans were acquired using a dual-layer spectral CT (IQon Spectral CT; Philips Healthcare, Cleveland, OH). Pancreatic CT comprised a non-enhanced phase, a pancreaticphase, and a PVP. The tube voltage was 120 kVp, and automated tube current modulation (three-dimensional dose modulation) was applied. Intravenous contrast medium (Iomeron 350; Bracco, Milan, Italy) was injected at a rate of 3 mL/s for a total amount of 2 mL/kg body weight using a power injector system. The bolus tracking method was applied, and the pancreatic-phase scan started 27 s after the abdominal aorta reached 150 HU. The portal venous scan started 50 s after the abdominal aorta reached 150 HU. The upper abdomen was scanned in the non-enhanced and pancreatic-phases, and the whole abdomen and pelvis were scanned in the PVP. After the CT evaluation, the conventional polychromatic images of the pancreatic-phase were reconstructed at 120 kVp, and VMIs of the PVP were reconstructed at 40 keV (VMI 40 ), 55 keV (VMI 55 ), and 70 keV (VMI 70 ). VMI 40 is the lowest energy setting provided by the software; VMI 55 is equivalent to an 80-kVp (low kVp) image, and VMI 70 is equivalent to a 120-kVp (standard kVp) image. A VMI reconstructed at >70 keV is not expected to show a better image contrast than a 120-kVp polychromatic image. An iterative reconstruction algorithm (iDose level 4) was used for the conventional polychromatic image of the pancreatic-phase and VMIs of the PVP. All conventional and VMI images were reconstructed axially with 3-mm slice thickness and 3-mm reconstruction intervals.

Image Analysis
For objective analysis, one radiologist with 5 years of experience in interpreting abdominal images performed image analysis using a workstation based on a picture archiving and communication system (INFINITT PACS version 3.0; Infinitt Healthcare Co., Seoul, South Korea). The investigator was aware of the patients' PDAC diagnoses. Circular ROIs were drawn in the normal pancreatic parenchyma, tumor, celiac trunk, portal vein, SMA, and SMV in the pancreatic-phase image and in VMI 40 , VMI 55 , and VMI 70 of the PVP, respectively. The copy-and-paste function of the workstation was used to minimize the differences in the size and location of each ROI among the images. Efforts were made to place the ROI in the center of the vascular lumen, and the vessel wall was not included. On each image in which the ROI was drawn in the vascular lumen, the other circular ROI was drawn in the most homogeneous portion of the paravertebral muscle. The average attenuation (AV) and standard deviation (SD) of the Hounsfield unit (HU) were recorded for each ROI.
The conspicuity of PDAC was evaluated using the tumor-to-pancreas CNR and attenuation difference.

Statistical Analyses
The Wilcoxon signed-rank test and paired t-test were used to compare objective parameters and subjective tumor conspicuity. Cohen's simple kappa test was used to evaluate the interreader agreement of subjective parameters. The value of kappa for inter-reader agreement were interpreted as follows; 0-0.20 none, 0.21-0.39 minimal, 0.40-0.59 weak, 0.60-0.79 moderate, 0.80-0.90 strong, 0.91-1.00 almost perfect [1]. Bonferroni correction was performed for multiple comparisons, and a p-value of <.0083 was considered to indicate statistical significance when comparing objective parameters and subjective tumor conspicuity. In comparing tumor size, p<.017 was considered significant. All statistical analyses were performed using the SPSS software (version 25; SPSS).