Purpose

To compare the diagnostic accuracy of low-kV MDCT reconstructed with model-based iterative reconstruction algorithm (IMR) to MRI with MRCP sequences in the detection of intraductal papillary mucinous neoplasm (IPMN) and those features useful for their characterization.

Material and methods

We evaluated a total of 90 patients, with no history of pancreatitis, neoplasm and prior surgical procedures in the pancreas, who underwent both low-dose MDCT reconstructed with IMR and MRCP (time interval lower 90 days) for incidental pancreatic cystic lesions. Number of lesions detected in each patient and morphologic features (according to 2018 European Guidelines) of defined target lesion were recorded and analyzed to compare the diagnostic performance of MDCT with IMR algorithm and MRCP.

Results

A higher number of pancreatic cysts was identified on MRI than CT examinations (463 vs 227); however, high intraclass correlation coefficient (ICC) value of 0.76 was found, suggesting a good agreement. Very good agreement (ICC 0.98) was found regarding the diameter of target lesions (21.4 mm CT vs 21.8 mm MRI), location of the target lesion (κ=0.90), detection of MPD dilatation (κ=1.00), MPD communication (κ=0.73) and septa (κ=0.86). Fair agreement with calcifications (κ=0.22) was found.

Conclusion

Low-dose MDCT reconstructed with IMR can be considered equivalent to MRI with MRCP in the evaluation of IPMN, offering detailed morphologic features helpful for their characterization and those correlated with a high risk of malignancy.

Learning objectives

Objectives

To summarize the possible fields of application of Spectral-CT technology to clarify actual advantages and benefits in clinical imaging, derived from recent literature review.

Background

Dual-energy computed tomography (DECT) or spectral-CT is an advanced form of CT that acquires data at different energy levels to enhance material differentiation and tissue characterization. In dual-layer technology the spectral energy separation is dependent on detector composition, the top-layer measures the low-energy photons, while the bottom-layer measures high-energy photons. Complex material decomposition algorithms are then used to generate images for routine clinical interpretation or for more advantage material characterization, including virtual non-contrast (VNC), iodine maps, and effective atomic number, as well as virtual monochromatic image sets (40 to 200keV).

Imaging findings OR Procedure findings

The most common use has been characterization of urinary calculi and gallstones (isoattenuating to bile on conventional CT), detection and correctly definitions of cystic lesions. While the added value in body imaging include: characterization of incidental lesions (very common in body imaging, resulting in substantial cost to healthcare system in terms of follow-up studies), iodine detection and subtraction (possibility to reconstruct a set of monoenergetic images, both at Low and High energy level), assessment of tissue perfusion (using organ iodine distribution), contrast-dose reduction, tissue characterization, and finally the possibility to decrease the number of scan.

Conclusion

Spectral-CT approach has a wide range of advantages, from the increased capability in detecting and characterizing lesions, to the possibility to derive many different maps of visualization, from whom provide a great amount of information to increase confidence in making diagnosis.

Objectives

To describe the difference between primary and secondary hepatic lymphoma (PHL and SHL) and their imaging manifestations.

Background

PHL and SHL represent respectively 0.016% and up to 50% of all non-Hodgkin’s lymphomas, while Hodgkin disease is extremely rare as PHL. Median age of involvement is 50 years in PHL and 40 years in SHL, with a male to female ratio of 2.3:1 and 8:1 respectively.

Chronic liver infection and cirrhosis, immunosuppressive therapy, and autoimmune disease have been associated to the pathogenesis of PHL. The outcome of patients with PHL seems to be better than those with SHL.

Imaging findings OR Procedure findings

Usually lymphoma diagnosis can be simple in case of SHL, mainly due to both simultaneous involvement of nodes and liver, while PHL diagnosis can be challenge as it often mimics HCC, CCC or adenocarcinoma liver metastasis. Liver involvement may have a wide radiological manifestations: solitary lesion, multiple lesions or diffuse infiltration in both cases of PHL and SHL. Focal lesions are typically hypoattenuating on CT and mildly hyperintense on T2-w MRI images, showing restriction on DWI, without significant enhancement in both techniques and typical peripheral enhancement. FDG-PET has proved to be superior to CT in staging lymphoma disease.

Conclusion

Imaging manifestations of hepatic lymphoma, especially PHL, are extremely variable and non-specific, but some radiological features represent useful diagnostic clue in the appropriate clinical setting.

Purpose

To compare radiation dose and image quality of low-dose CT protocol combined with iterative model-based reconstruction algorithm (IMR) with standard-dose CT approach combined with hybrid-iterative reconstruction algorithm (iDose) in the follow-up of abdominal oncologic diseases.

Material and methods

Ninety-four patients with known abdominal oncological diseases who underwent both low dose CT performed on 256-row scanner (100 kV, automated mAs modulation, model- based reconstruction algorithm) and standard-dose CT performed on 256-row scanner (120 kV, automated mAs modulation, iDose algorithm), were enrolled. Density values in liver and spleen and signal-to-noise ratio (SNR), along with image noise, dose parameters and diagnostic quality with 4-point scale, were evaluated.

Results

Noise of images expressed as SD values, measured in liver and spleen, was significantly lower in IMR images (liver 11,42 vs 14,57, p<0,001) whereas SNR was statistically higher (liver 10,68 vs 7,94, p<0,001) compared to iDose reconstruction. Volumetric-Computed-Tomographic-Dose-Index (CTDIvol) and Dose-Length-Product (DLP) were significantly lower in IMR gorup compared to iDose (DLP 559,83 vs 959,77 mGy*cm, p<0,001), with overall dose reduction of 45,7%. 4-point scale qualitative analysis did not reveal any significant differences in terms of diagnostic quality (p=0,04).

Conclusion

Automatic tube-current modulation combined with IMR algorithm and low kV setting allows dose reduction of 45,7% in whole body CT imaging without loss of diagnostic quality, thus representing a useful diagnostic approach in reducing dose exposure in oncologic patients who undergo several follow-up studies.

Purpose

To investigate whether portal vein thrombosis (PVT) influences perfusion computed tomography (pCT) analysis in patients with advanced hepatocellular carcinoma (HCC) treated with sorafenib.

Material and methods

106 pCT study consecutively performed in 38 patients with advanced HCC were analysed. pCT was based on the acquisition of 16 dynamic slices/scan per 40 scans, performed on a 256-slice MDCT scanner, after iv bolus injection of 50 ml iodinated contrast agent (350 mgI/ml) at a flow rate of 5 ml/s. The following pCT parameters were calculated for both normal liver parenchyma and HCC: hepatic perfusion (HP,ml/s/100g), arterial perfusion (AP,ml/s), portal perfusion (PP,ml/s) and hepatic perfusion index (HPI,%). Continuous variables are expressed as mean and standard deviation.

Results

No statistically significant differences were found in HP, AP, PP and HPI values of HCC lesions between patients with and without PVT. HP value of normal liver parenchyma was significantly higher in patients with PVT in comparison to those without (14.6 ± 5.4 vs 18.1 ± 11.3, p=0.002) while HPI was significantly lower (24.1 ± 28.1 vs 13.2 ± 13.2, p=0.002). No differences were found in AP and PP values.

Conclusion

The appearance of PVT lead to parenchyma perfusion changes due to the reduction of portal venous flood and the increase of arterial flow. However the hemodynamic changes do not modify the perfusion of HCC due to the development of unpaired arteries not associated with portal vein branches. This allows to performed pCT analysis also in patients with advanced HCC and PVT.

Learning objectives

Purpose

To compare the diagnostic accuracy of low-kV MDCT reconstructed with model-based iterative reconstruction algorithm (IMR) to MRI with MRCP sequences in the detection of intraductal papillary mucinous neoplasm (IPMN) and those features useful for their characterization.

Material and methods

We evaluated a total of 90 patients, with no history of pancreatitis, neoplasm and prior surgical procedures in the pancreas, who underwent both low-dose MDCT reconstructed with IMR and MRCP (time interval lower 90 days) for incidental pancreatic cystic lesions. Number of lesions detected in each patient and morphologic features (according to 2018 European Guidelines) of defined target lesion were recorded and analyzed to compare the diagnostic performance of MDCT with IMR algorithm and MRCP.

Results

A higher number of pancreatic cysts was identified on MRI than CT examinations (463 vs 227); however, high intraclass correlation coefficient (ICC) value of 0.76 was found, suggesting a good agreement. Very good agreement (ICC 0.98) was found regarding the diameter of target lesions (21.4 mm CT vs 21.8 mm MRI), location of the target lesion (κ=0.90), detection of MPD dilatation (κ=1.00), MPD communication (κ=0.73) and septa (κ=0.86). Fair agreement with calcifications (κ=0.22) was found.

Conclusion

Low-dose MDCT reconstructed with IMR can be considered equivalent to MRI with MRCP in the evaluation of IPMN, offering detailed morphologic features helpful for their characterization and those correlated with a high risk of malignancy.