Virtual monoenergetic images preserve diagnostic assessability in contrast media reduced abdominal spectral detector CT

Abstract

Objectives:

To investigate if low-keV virtual monoenergetic images (VMI_40keV) from abdominal spectral detector CT (SDCT) with reduced intravenous contrast media application (RCM) provide abdominal assessment similar to conventional images with standard contrast media (SCM) dose.

Methods:

78 patients with abdominal SDCT were retrospectively included: 41 patients at risk for adverse reactions who received 44 RCM examinations with 50 ml and 37 patients who underwent 44 SCM examinations with 100 ml of contrast media (CM) and who were matched for effective body diameters. RCM, SCM images and RCM-VMI_40keV were reconstructed. Attenuation and signal-to-noise ratio (SNR) of liver, pancreas, kidneys, lymph nodes, psoas muscle, aorta and portal vein were assessed ROIs-based. Contrast-to-noise ratios (CNR) of lymph nodes vs aorta/portal vein were calculated. Two readers evaluated organ/vessel contrast, lymph node delineation, image noise and overall assessability using 4-point Likert scales.

Results:

RCM were inferior to SCM images in all quantitative/qualitative criteria. RCM-VMI_40keV and SCM images showed similar lymph node and muscle attenuation (p = 0.83,0.17), while for all other ROIs, RCM-VMI_40keV showed higher attenuation (p ≤ 0.05). SNR was comparable between RCM-VMI_40keV and SCM images (p range: 0.23–0.99). CNR of lymph nodes was highest in RCM-VMI_40keV (p ≤ 0.05). RCM-VMI_40keV received equivalent or higher scores than SCM in all criteria except for organ contrast, overall assessability and image noise, where SCM were superior (p ≤ 0.05). However, RCM-VMI_40keV received proper or excellent scores in 88.6/94.2/95.4% of the referring cases.

Conclusions:

VMI_40keV counteract contrast deterioration in CM reduced abdominal SDCT, facilitating diagnostic assessment.

Advances in knowledge:

SDCT-derived VMI_40keV provide adequate depiction of vessels, organs and lymph nodes even at notable CM reduction.

Introduction

Contrast enhancement as achieved by the application of iodinated contrast media (CM) is an essential element of most abdominal CT protocols. It is paramount to ensure sufficient contrast of parenchymal organs, assessability of the vascular system and delineation of abdominal lymph nodes. Moreover, multiphasic contrast-enhanced acquisitions are indispensable to image tumors with characteristic dynamic enhancement patterns, for example, hepatocellular carcinoma (HCC).^1,2 Despite the undisputed value of contrast-enhanced CT, it is known that CM application holds a risk of acute adverse reactions³ and contrast-induced nephropathy.^4–7 In addition, CT contrast agents account for considerable healthcare expenses.^8,9 Hence, reduction of intravenous contrast material has been investigated in numerous previous studies.^10–12

While image contrast in contrast-enhanced CT exams depends on various factors such as body composition, injection parameters and individual circulatory conditions,¹³ a reduced amount of iodine contrast material frequently implies a certain contrast decline. Primarily two methods have been proposed to counteract this effect: lowering tube voltage on the one hand and using dual-energy CT (DECT)-derived low-keV virtual monoenergetic images (VMI) on the other.^14–16 VMIs are calculated as to resemble an appearance which would result from the acquisition with actual monoenergetic X-rays.¹⁷ The iodine contrast increase attained by this method at low-keV levels results from a shift of the energy spectrum towards the physical absorption maximum of iodine, the k-edge at approximately 33 keV. While VMI obtained with prior generation DECT exhibited a higher image noise than conventional CT at low-keV levels,¹⁸ recent DECT scanners provide low-keV VMI with minimal noise levels.^19,20

Various studies investigated the application of low-keV imaging for contrast-reduced abdominal CT examinations. However, most of these studies were either limited to CT angiography^14,21,22 or, if organ and lymph nodes were included, performed on tube-based DECT systems.^23,24 In contrast, assessment of contrast-reduced abdominal CT with low keV VMI derived from spectral detector CT (SDCT), a DECT system based on two detector layers, has only been described in case reports²⁵ and one recent study that was limited to liver assessment.²⁶

SDCT routinely acquires one Compton-weighted and one photoelectron-weighted dataset in temporal and spatial coherence facilitating anticorrelation of noise without additional dose penalty.^27,28 VMI at 40 keV derived from SDCT have been reported to yield the highest contrast increase and best image quality at assessing the abdomen.^26,29,30,30 Based on this evidence, we hypothesized that this particular energy level may be most effective in antagonizing image contrast deterioration in examinations with reduced CM (RCM) dose.³¹ Thus, the purpose of this study was to investigate whether VMI at 40 keV reconstructed from SDCT examinations with 50% CM reduction would yield comparable qualitative and quantitative image parameters as provided by conventional images (CIs

) with standard CM (SCM) application.

Methods and materials

Patient collection

Consultation of the local ethics committee was obtained, and an informed patient consent was waived due to the retrospective character of the study.

At our institution, a RCM protocol is implemented as clinical standard of care for patients at high risk for adverse reactions (e.g., severely reduced kidney function) undergoing SDCT examinations without need for dynamic CM information (i.e., single-phase examinations for morphological assessment of lymphoma size). Application of this protocol is subject to individual clinical decisions and was unrelated to the purpose of this study. Based on the results of previous studies, the amount of contrast medium for the clinical RCM protocol is set to 50% of the respective standard amount for non-DECT examinations.^16,32,33

To identify eligible patients, our RIS database was first screened for patients who underwent abdominal portal-venous phase SDCT with the RCM protocol. Time range for study inclusion was 05/01/2016 to 05/01/2018. 41 patients with 44 RCM abdominal SDCT examinations were found. None of the patients who received RCM abdominal SDCT in the study period was excluded.

Body weight was identified as a possible confounder for comparing RCM and SCM scans as it influences CM distribution within the patient.³¹ Effective body diameter (EBD) has been described as a feasible surrogate parameter for body weight.³⁴

To identify patients who underwent abdominal SDCT with SCM protocol in the same inclusion period, a second database query was carried out for corresponding patients with typical depiction of portal-venous phase contrast and absence of artifacts hampering assessment. Of the patients found, 37 patients who matched the EBDs of the RCM group and who underwent 44 abdominal SDCT examinations were chosen.

Detailed patient characteristics are given in Table 1.

Table 1.

Patient characteristics of reduced (RCM) and standard (SCM) contrast media application cohorts

	Reduced contrast media (RCM) dose cohort	Standard contrast media (SCM) dose cohort	Level of significance
Patient count	41	37
Males/Females	27/14	21/16	p = 0.41
Mean Age	62 ± 13 years	60 ± 17 years	p = 0.76
Effective body diameter	29.2 ± 3.1 cm	29.3 ± 3.1 cm	p = 0.72
Radiation dose (CTDIvol)	10.3 ± 2.3	10.4 ± 2.4	p = 0.79
Contrast volume	50 ml	100 ml
Saline chaser	80 ml	30 ml

Image acquisition and reconstruction

All patients were scanned in supine position on a clinical SDCT system (IQon, Philips, Best, The Netherlands). A consistent scan protocol was used which comprised tube current modulation (DoseRight 3D-DOM, Philips Healthcare), a pitch of 0.671, a rotation time of 0.33 s, a collimation of 64×0.625, a matrix size of 512×512 and a tube voltage of 120 kVp. To receive contrast-enhanced acquisitions, a bolus of iodinated contrast agent (Accupaque 350 mg ml⁻¹, GE Healthcare) was injected via a peripheral vein with a flow rate of 3.5 ml s⁻¹: in the RCM group, a consistent amount of 50 ml CM followed by an 80 ml saline chaser was used. Matched patients from the SCM group received 100 ml CM followed by a 30 ml saline chaser according to the SCM injection protocol used at our institution (< 55 kg: 1 ml/kg; 55–120 kg: 100 ml;> 120 kg: 120 ml, each of which followed by a 30 ml saline chaser). Acquisition of portal-venous phase images was triggered by bolus tracking with a delay of 50 s after a predefined threshold of 150 HU in the descending aorta was reached.

Image reconstruction and preprocessing

For SCM and RCM groups, CIs at 120 kVp as used in clinical routine were reconstructed in axial plane and with a slice thickness and section increment of 2 mm. For the RCM group, VMIs at 40 keV (VMI_40keV) were reconstructed in addition to CIs by linear blending of photoelectric- and compton-weighted datasets, respectively. For reconstruction of both, VMI and CIs, the same spectral image reconstruction algorithm (Spectral, filter B, denoising level 3, Philips Healthcare) was used which operates comparable to the vendors’ hybrid-iterative reconstruction method.²⁰

Automated windowing was used to pre-adjust window width and level of SCM, RCM-CIs and RCM-VMI_40keV to facilitate blinding to image reconstruction and patient subgroup for the qualitative assessment.

Quantitative analysis

To acquire quantitative Hounsfield unit (HU) attenuation values from CIs and VMI_40keV, circular regions of interest (ROIs) were placed by one radiologist with 3 years of experience in oncological imaging in the following regions: liver (right lobe, left lobe), pancreas (head, body), right and left kidney (lower, upper cortex), two physiologically configured retroperitoneal lymph nodes (1 ROIs each), abdominal aorta (suprarenal, infrarenal), portal vein (proximal, distal), psoas muscle and retroperitoneal fat (1 ROIs each). ROIs were drawn in order to cover an area as large as possible without including unrepresentative tissue. ROIs placement and data export of CIs and VMI_40keV were performed in the vendors proprietary image viewer (Intellispace Portal 10, Philips Healthcare). The two ROIs for liver, pancreas, kidneys, lymph nodes, aorta and portal vein were averaged and reported as one corresponding value. Anteroposterior and the transverse abdominal diameters were measured at the level of the first lumbar vertebra to calculate EBD.

Qualitative analysis

For qualitative image analysis, CIs and VMI_40keV obtained from RCM examinations (n = 44 each) as well as CIs reconstructed from SCM examinations (n = 44) were put into a randomized order and subsequently presented to two radiologists with 7 and 2 years of experience who were blinded against patient data, reconstruction (CIs, VMI_40keV) and CM application protocol (RCM, SCM). These radiologists independently reviewed all cases and evaluated the following parameters using 4-point Likert scales: subjective parenchymal contrast of liver, pancreas and spleen, arterial and venous vessel contrast, corticomedullary differentiation of kidneys, delineation of retroperitoneal lymph nodes from adjacent vessels, subjective image noise and overall assessability. Moreover, readers indicated their assumption on whether the scan had been performed with the regular- or reduced contrast protocol in a binary decision. For qualitative assessment, readers were encouraged to freely adjust window settings. Likert scales used for assessing the individual criteria are reported in Table 2.

Table 2.

Criteria of the qualitative image assessment inducing interpretations of each score

Score	Subjective criterion/interpretation
	Parenchymal organ contrast
1	Insufficient organ contrast
2	Limited organ contrast
3	Proper organ contrast
4	Excellent organ contrast
	Corticomedullary differentiation of kidneys
1	Insufficient differentiation
2	Limited differentiation
3	Proper differentiation
4	Excellent differentiation
	Contrast of arterial/venous vessels
1	Insufficient vessel contrast
2	Limited vessel contrast
3	Proper vessel contrast
4	Excellent vessel contrast
	Delineation of retroperitoneal lymph nodes from adjacent vessels
1	Insufficient delineation
2	Limited delineation
3	Proper delineation
4	Excellent delineation
	Subjective image noise
1	High image noise with severe impediment of diagnostic assessment
2	Moderate image with noticeable impediment of diagnostic assessment
3	Little image noise with no or minimal impediment of diagnostic assessment
4	Minimal image noise without impediment of diagnostic assessment
	Overall assessability
1	Poor overall assessability
2	Reduced overall assessability
3	Proper overall assessability
4	Excellent overall assessability

Statistical procedures

As Shapiro Wilk test revealed non-normal data distribution, non-parametric Wilcoxon and Mann-Whitney U test were used to compare quantitative measurements and qualitative Likert scores. Interobserver agreement within the subjective assessment was evaluated using the intraclass correlation coefficient (ICC). EBD was calculated as the square root of the product of the anteroposterior and the transverse abdominal diameter. SNR and CNR were calculated as previously suggested³⁵:

$$SNRx=\frac{{HU}_{ROIx}}{{SD}_{Fat}}$$

$$CNR{}_{lymphnode/vessel}=\frac{{|HU}_{vessel}-{HU}_{lymphnode}|}{\sqrt[2]{{SD}_{vessel}+{SD}_{lymphnode}}}$$

Continuous variables are reported as mean ± standard deviation and Likert scores as median and interquartile range. A p-value below 0.05 is considered to indicate statistical significance.

Results

Study population

The study population comprised 78 patients. 41 patients were included in the RCM group (14 females, 27 males, mean age 62 ± 13 years) and 37 in the SCM group (21 females, 16 males, mean age 60 ± 17 years). Following patient matching between RCM and SCM cohort, radiation exposure (CTDIvol = 10.4±2.4 mGyvs. 10.3 ± 2.3 mGy; p = 0.8) and EBDs (ED = 29.2 ± 3.1 cm vs 29.3 ± 3.1 cm; p = 0.89) were similar.

Quantitative analysis

Highest attenuation was found in RCM-VMI_40keV. It was significantly higher than in SCM images in all ROIs (p ≤ 0.05) except for lymph nodes and psoas muscle, where it was similar (p = 0.83, 0.17, respectively). In conventional RCM images, attenuation was lowest and significantly inferior to both, SCM and RCM-VMI_40keVimages, in all ROIs (p ≤ 0.05).

SNR was comparable between SCM images and RCM-VMI_40keV in all measured ROIs (i.e., liver, pancreas, kidneys, lymph nodes, abdominal aorta, portal vein; p-range: 0.23–0.99). In conventional RCM images, SNR was significantly lower than in SCM images and RCM-VMI_40keV in all ROIs examined (p ≤ 0.05).

CNR of lymph nodes vs abdominal aorta and vs portal vein was highest in RCM-VMI_40keV (23.7 ± 8.9 and 25.4 ± 9.0), lowest in conventional RCM images (8.9 ± 3.2 and 10.2 ± 3.7; p ≤ 0.05) and intermediate in SCM images (15.2 ± 4.9 and 17.5 ± 4.9; p ≤ 0.05). Statistically significant differences regarding CNR_{lymph node/aorta} and CNR_{lymph node/portal vein} were found between all three groups. Figures 1–3 as well as Table 3 give an overview on quantitative results.

Figure 1.

Attenuation within liver, pancreas, kidneys, lymph nodes, aorta and portal vein. Asterisks indicate statistical significance.

Figure 2.

Signal-to-noise ratios (SNR) of liver, pancreas, kidneys, lymph nodes, aorta and portal vein. Asterisks indicate statistical significance.

Figure 3.

Contrast-to-noise ratios (CNR) of lymph node vs abdominal aorta (left) and vs portal vein (right). Asterisks indicate statistical significance.

Table 3.

Mean attenuation, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of all included regions of interest. Asterisks indicate statistically significant differences as compared to the standard contrast media (CM) dose group.

	Regions of interest	Reduced CM dose (Conventional image)	Reduced CM dose (VMI40keV)	Standard CM dose (Conventional image)
Attenuation (HU)	Liver	76.5 ± 14.3*	127.3 ± 29.5*	116.9 ± 26.6
	Pancreas	60.3 ± 10.2*	116.6 ± 27.5*	93.9 ± 18.4
	Kidneys	97.9 ± 15.5*	250.4 ± 52.8*	182.7 ± 35.2
	Psoas	46.7 ± 8.3*	54.4 ± 16.3	60.2 ± 8.6
	Lymph nodes	42.7 ± 18.9*	75.3 ± 32.6	71.9 ± 12.8
	Aorta	94.9 ± 12.9*	226.1 ± 43.5*	161.1 ± 27.4
	Portal vein	95.3 ± 12.2*	238.4 ± 41.4*	172.6 ± 31.2
SNR	Liver	5.2 ± 1.7*	9.1 ± 3.9	10.0 ± 3.0
	Pancreas	4.1 ± 1.3*	8.3 ± 3.5	8.0 ± 2.3
	Kidneys	6.6 ± 1.9*	17.7 ± 6.7	15.8 ± 4.8
	Psoas	3.2 ± 0.9*	3.9 ± 1.8	5.3 ± 1.7
	Lymph nodes	2.9 ± 1.5*	5.3 ± 2.6	6.2 ± 1.8
	Aorta	6.4 ± 1.8*	16.1 ± 6.6	13.8 ± 4.0
	Portal vein	6.4 ± 1.8*	16.9 ± 6.4	14.8 ± 4.2
CNR	Lymph nodes /Aorta	8.9 ± 3.2*	23.7 ± 8.9*	15.2 ± 4.9
	Lymph nodes/ Portal vein	10.2 ± 3.7*	25.4 ± 9.0*	17.5 ± 4.9

Qualitative analysis

Between SCM images and RCM-VMI_40keV, Likert scores were equivalent for corticomedullary differentiation of kidneys (both 3 (3-4), p = 0.88), arterial vessel contrast (both 4 (4-4), p = 0.41) and delineation of lymph nodes from adjacent vessels (SCM: 4 (4–4), RCM-VMI_40keV: 4 (3–4), p = 0.11). Venous vessel contrast was rated significantly higher in RCM-VMI_40keV than in SCM images (4 (4–4) vs 4 (3–4), p ≤ 0.05). In contrast, RCM-VMI_40keV received lower average scores than SCM images for parenchymal organ contrast (3 (3–4) vs 4 (3–4), p ≤ 0.05), overall assessability (3 (3–4) vs 4 (3–4), p ≤ 0.05) and subjective image noise (3 (3–4) vs 4 (3–4), p ≤ 0.05). However, for these three criteria, a score of either 3 (“proper”) or 4 (“excellent”) was assigned in 88.6 %, 94.2% and 95.4% of all assessed cases.

For conventional RCM images, Likert scores for all subjective criteria assessed were significantly lower than the corresponding scores assigned in both SCM images and RCM-VMI_40keV (p ≤ 0.05; see also: Table 4).

Table 4.

Median subjective Likert scores for all criteria assessed. Asterisks indicate statistically significant differences as compared to the standard contrast media (CM) dose group

Subjective criteria	Reduced CM dose (Conventional image)	Reduced CM dose (VMI40keV)	Standard CM dose (Conventional image)
Parenchymal organ contrast	2 (2–3)*	3 (3–4)*	4 (3–4)
Corticomedullary differentiation of kidneys	3 (3–3)*	3 (3–4)	3 (3–4)
Arterial vessel contrast	3 (2–3)*	4 (4–4)	4 (4–4)
Venous vessel contrast	2 (2–3)*	4 (4–4)*	4 (3–4)
Delineation of lymph nodes from adjacent vessels	3 (2–3)*	4 (3–4)	4 (4–4)
Subjective image noise	3 (3–3)*	3 (3–4)*	4 (3–4)
Overall assessability	3 (2–3)*	3 (3–4)*	4 (3–4)

Overall interreader agreement implying all subjective criteria for the assessment of SCM images was lowest (0.22) while it was highest for RCM-VMI_40keV (0.44) and intermediate for conventional RCM images (0.35).

In 60.2% of cases from the RCM-VMI_40keV group, blinded reviewers indicated that they assumed the examination had been performed with a regular CM dose, although it had been performed with a RCM dose. Opposingly, reviewers correctly identified RCM examinations without virtual monoenergetic imaging as being performed with RCM protocol in 81.8% of the referring cases. SCM examinations were correctly identified in 86.4% of the cases. Table 3 and Figure 4 show detailed results of the qualitative analysis. Figure 5 shows exemplary images of examinations with reduced and normal CM injection protocol with and without virtual monoenergetic imaging.

Figure 4.

Subjective Likert scores for parenchymal organ contrast, corticomedullary differentiation of kidneys, delineation of lymph nodes from adjacent vessels, arterial and venous vessel contrast, overall assessability and subjective image noise. Asterisks indicate statistical significance between the compared groups.

Figure 5.

Image examples showing improved contrast in RCM-VMI_40keV compared to conventional RCM reconstructions with regard to organ parenchyma (top row), corticomedullary differentiation (second row from the top) and delineation of lymph nodes vs adjacent arterial and venous vessels (second row from the bottom and bottom row, respectively).

Discussion

Iodinated contrast material is used for a plethora of imaging purposes and generally considered to be safe, rarely resulting in acute or delayed adverse reactions including contrast-induced nephropathy; although the question of causality between CM administration and nephropathy is not finally solved,^36–39 the risk is considered highest in patients with a glomerular filtration rate lower than 30 mg/ml.⁷ While CM reduction seems beneficial in terms of patient safety, it inherently leads to a decrease in vessel and organ enhancement.³¹ Previous studies on CT angiography showed that this problem could at least partially be overcome by virtual monoenergetic reconstructions from DECT systems.^14,21,22 In order to objectify this for venous phase abdominal examinations obtained with a detector-based DECT yielding low noise low-keV imaging,²⁰ we reconstructed VMI at 40 keV from abdominal SDCT examinations that were clinically performed with reduced intravenous CM application following increased patient risk for adverse reactions. To enable a comparison with conventional CT examinations, we determined SDCT examinations with default CM application that matched the patient characteristics of the RCM cohort. Between these two patient groups, we compared objective image parameters of relevant abdominal structures and subjective image assessment.

VMI of CM-reduced examinations (RCM-VMI_40keV) showed similar or even higher attenuation and SNR as well as a significantly higher CNR of lymph nodes vs adjacent vessels. RCM-VMI_40keV received comparable ratings as SCM examinations except for parenchymal organ contrast, image noise and overall assessability; however, for these criteria, excellent or good scores were assigned in 88.6 %, 94.2% and 95.4 % of the referring cases, respectively, indicating fully preserved assessability. Only 38.8% of the RCM-VMI_40keV examinations were correctly identified as CM-reduced examinations, while this rate was 82.8% for the same examinations without virtual monoenergetic imaging.

Low-keV imaging in CM-reduced abdominal DECT has been previously investigated. The majority of referring studies were performed on tube-base DECT scanners.^{14,21,22,24,40} Lv et al reported an optimal assessment of RCM examinations in VMI at 50 keV.⁴⁰ The underlying reason was that VMI at 40 keV, while providing an even more pronounced attenuation increase than higher energy levels, showed a substantial increase of image noise compared to VMI at 50 keV and CIs. Based on this premise, Noda et al did not include VMI at 40 keV in their assessment using a rapid kV switching system.²⁴ While more recent iterative reconstruction methods help to reduce noise in low-keV VMI,¹⁹ it can be considered a technical advantage of SDCT to acquire projection data in temporal and spatial coherence allowing for further noise reduction by anticorrelation of image noises within the two datasets.²⁸ Concordantly, SNR observed in our study was comparable between RCM-VMI_40keV and SCM images while RCM-VMI_40keV yielded significantly higher CNR of lymph nodes vs adjacent vessels. Contextualizing these results with the low CM dose applied in our study equaling 219 mgI/kg (vs 300 mgI/kg used by Lv et al and 400 mgI/kg by Noda et al) illustrates the potential of SDCT-derived VMI at 40 keV for reestablishing vessel and organ enhancement even at very low CM doses.

Importantly, our study did not investigate lesion assessment, to which the results we found should not be generalized. For some abdominal lesions such as complex renal or liver lesions, depiction of heterogeneity of enhancement is an important factor, which might be hampered in VMI at 40 keV due to possible contrast blooming. Notably, Nagayama et al reported that conspicuity and image quality in assessing HCC at multiphasic CM-reduced SDCT examinations was rated highest in VMI at 40 keV which is a first indication that these reconstructions might be suitable for lesion assessment.²⁶ However, larger scale studies including a greater variety of lesions should be encouraged before assessment of such lesions should be performed based on VMI from CM-reduced examinations. Our study has further limitations that need to be addressed. First, it is a retrospective study based on patients who received a contrast-reduced abdominal CT because they were at increased risk for contrast-induced side effects and had appropriate indications (e.g., lymphadenopathy yes/no). For this reason, there was no direct control group as in a prospective study. Instead, we had to assemble examinations used for comparison from other SDCT examinations with SCM application protocol. Second, our results only apply for the detector-based DECT system we investigated. Third, we refrained from including different keV levels in our analysis but focused on the lowest level available as numerous recent studies revealed VMI at 40 keV from SDCT to yield the most pronounced contrast increase and image quality improvements.^29,30 Last, influence of RCM administration on renal function was not evaluated; although this is an interesting and current subject of ongoing scientific discussions, large cohorts with precise assessment of various known confounders are pivotal to provide valuable contribution to the existing body of evidence; this was beyond the scope of this study.

Conclusion

To conclude, this study shows that VMI at 40 keV provided by SDCT preserve qualitative assessability and quantitative image parameters in CM-reduced abdominal CT. Although accurate organ lesion assessment may require higher amounts of CM and further investigation, clinical application of this technique seems suitable when no dedicated assessment of lesion enhancement patterns is requested.