Efficacy of dynamic enhancement effects on Gd-EOB-DTPA-enhanced MRI for estimation of liver function assessed by 13 C- Methacetin breath test

Abstract

PURPOSE:

Contrast enhanced magnetic resonance imaging (MRI) is able to assess liver function by characteristic changes of signal intensity (SI).

The aim was to evaluate dynamic contrast-enhanced SI-indices of the abdominal aorta, portal vein and liver.

METHODS:

72 patients underwent Gd-EOB-DTPA-enhanced MRI and a ¹³C-methacetin-based liver breath test (¹³C-MBT) for evaluation of liver function. Region-of-interest measurements in the liver, abdominal aorta and portal vein during native, arterial (AP), late arterial (LAP), portal venous (PVP) and hepatobiliary phase (HBP) were applied to analyze SI-indices in T1-weighted volume-interpolated breath-hold examination (VIBE) sequences with fat-suppression and relative enhancement (RE) analysis was performed.

RESULTS:

The liver (p < 0.001), the portal vein (p < 0.001) and abdominal aorta (p = 0.002) showed significant decrease of REs with decreasing liver function. An increasing trend between logarithmic values of ¹³C-MBT and REs of hepatic parenchyma (HBP; r = 0.662, p < 0.001), portal vein (PVP; r = 0.532, p < 0.001) and abdominal aorta (PVP; r = 0.421, p < 0.001) was observed.

CONCLUSIONS:

RE measurements of the hepatic parenchyma proofed to be a trustable evaluation method for liver function evaluation. In accordance with liver function, changes of REs in the portal vein and abdominal aorta occur.

Keywords

Dynamic contrast enhanced MRI HCC liver metastases hepatic parenchyma portal vein abdominal aorta signal intensity relative enhancement

1 Introduction

Portal hypertension is one of the main implications of liver cirrhosis. It plays a key role in severe complications, such as rupturing of gastroesophageal varices, ascites, the occurrence of hepatorenal syndrome and hepatic encephalopathy. According to Ohm’s law, the blood pressure is dependent on flow and resistance, thus the portal pressure in portal hypertension is increased due to an increase in portal blood flow, an increase in vascular resistance or a combination of both. Hemodynamic changes are critical accessory phenomena of cirrhosis. The architectural distortion of the liver microcirculation caused by fibrosis, scarring and nodule formation as well as active contraction of the vascular smooth muscle cells, myofibroblasts and other contractile elements within or around the hepatic microcirculation have an impact on the intrahepatic vascular resistance. In cirrhotic patients, the development of portal hypertension will increase the risk of death and need for liver transplantation [3].

The liver has a unique dual blood supply which is based on venous blood from the portal vein and arterial blood from the hepatic artery. As there is communication between vessels, including transsinusoidal, transvasal and transplexal routes, vascular compromise will cause perfusion disorders. Established clinical tests to identify portal hypertension are based on different methods as the invasive balloon catheter method [2], analysis of xenobiotic-clearance [8, 15] or ultrasound measurements [4].

In addition to contrast enhanced ultrasound (CEUS)-based evaluation of parenchymal microcirculation [7], contrast enhanced MRI-based methods are able to assess hepatic flow parameters. Here, gadolinium-based contrast agent will be injected and changes in MR imaging will be monitored. Based on mathematical and physical hepatic perfusion models the data acquired during imaging will be fitted by algorithms and representative hepatic flow parameters, e.g. hepatic perfusion index or mean transit time will be assessed [10].

Gadoxetic acid (Gd-EOB-DTPA) is a hepatocyte-specific contrast agent that is capable to display liver function in an MRI-based test by analyzing the signal intensity caused by the interaction of gadolinium chelates with protons [13, 19]. As liver cirrhosis and fibrosis has an impact on the portal pressure, it may be possible to monitor changes using SI-based MR dynamic imaging of the portal vein and abdominal aorta.

The purpose of this study is to investigate the relationship between the liver function estimated by ¹³C-MBT and dynamic MRI-based SI values of the abdominal aorta, the portal vein and the liver parenchyma, respectively.

2 Material and methods

2.1 Patients

In this retrospective study, patients that received Gd-EOB-DTPA-enhanced T1-weighted volume-interpolated breath-hold examination (VIBE) MRI sequences with fat suppression at 3 T during native, arterial (AP), late arterial (LAP), portal venous (PVP) and hepatobiliary phase (HBP), were included. Further additional criteria were contemporary liver function evaluation by a ¹³C-MBT and absence of allergic reactions to liver-specific MRI contrast media, ¹³C-methacetin or specific contraindications to either MRI or Gd-EOB-DTPA administration (Primovist, Bayer Healthcare, Berlin, Germany). Informed consent was obtained from all participating patients and the study was conducted in full accordance with the ethical guidelines of the journal [1].

2.2 Magnetic resonance imaging

For T1-weighted VIBE sequences with fat suppression following conditions were applied: 3.09 ms repetition time, 1.17 and 2.49 ms echo time, 10° flip angle, parallel imaging factor 2 and 64 slices. The reconstructed voxel size was 1.25×1.25×3.0 m³ and the measured voxel size was 1.71×1.25×4.5 mm³. All images had an acquisition time of 14 sec and were acquired during breath-hold before Gd-EOB-DTPA administration (native phase) and during dynamic phases.

The patients received an i.v. bolus injected dose of Gd-EOB-DTPA (0.025 mmol/kg body weight; flow rate, 1 mL/sec) followed by 20 mL 0.9% sodium chloride for contrast-enhanced MRI.

2.3 Image analysis

The sequences were evaluated by region-of-interest (ROI) measurements. Circle shaped ROIs were manually placed at identical locations in every sequence, to obtain representative SI values while avoiding lesions and imaging artifacts. In total 8 ROIs (3 left liver lobe, 3 right liver lobe, 1 portal vein, 1 abdominal aorta) were defined per sequence.

The relative enhancement (RE) of the liver parenchyma (RE_LL), the portal vein (RE_PV) and the abdominal aorta (RE_AA) relative to the native phase was calculated for AP, LAP, PVP and HBP by following formulas:

$\begin{matrix} {RE}_{{LL}_{a}} = \frac{mean ({SI}_{{LL}_{X}}) - mean ({SI}_{{LL}_{native}})}{mean ({SI}_{{LL}_{native}})} \\ {RE}_{{PV}_{a}} = \frac{{SI}_{{PV}_{X}} - {SI}_{{PV}_{native}}}{{SI}_{{PV}_{native}}} \\ {RE}_{{AA}_{a}} = \frac{{SI}_{{AA}_{X}} - {SI}_{{AA}_{native}}}{{SI}_{{AA}_{native}}} \end{matrix}$

a = AP, LAP, PVP, HBP

x = arterial, late arterial, portal venous, hepatobiliary

2.4 ¹³C-methacetin breath test

The ¹³C-MBT was contemporary performed to the MRI scan (24 h) [16, 17]. Before the i.v. injection of ¹³C-methacetin, the ¹³CO₂:¹²CO₂ ratio control was recorded. After injection of 2 mg/kg body weight adapted ¹³C-methacetin and 20 mL 0.9% sodium chloride, a volatile analysis was performed by nondispersive isotope-selective infrared spectroscopy (FANci2-db16, Fischer Analysen Instrumente, Leipzig, Germany). According to the ¹³C-MBT readout values, the patients were classified: normal liver function (¹³C-MBT >315.0 [μg/kg/h]), intermediate liver function (¹³C-MBT: 140.0–315.0 [μg/kg/h]) and severely impaired liver function (¹³C-MBT <140.0 [μg/kg/h]) [17, 18].

2.5 Statistical analysis

The three ¹³C-MBT readout classes were compared as non-parametric independent samples by the Kruskal-Wallis-test. The predictive power of RE values were determined by Spearman-based bivariate correlation analysis, while in all tests the statistical significance level was set to 0.05 (two-sided). All analysis were performed using SPSS software (version 24; IBM, Chicago, IL, USA).

3 Results

3.1 Patient characteristics

In this retrospective study, patients received both a ¹³C-MBT and contrast enhanced dynamic T1-weighted MR imaging of the liver from native to hepatobiliary phase. 72 patients (51 men; 21 women) with a median age of 61.5 years were included (Table 1).

Table 1
Patient characteristics of included cases

Parameter ¹³C-MBT readout

All patients n = 72 >315.0 [μg/kg/h] 140.0–315.0 [μg/kg/h] <140.0 [μg/kg/h]

n = 28 n = 26 n = 18

Age [years] 60.90±9.61 58.89±10.88 64.27±9.63 59.17±5.80

Male 51 (70.8 %) 12 (42.9 %) 23 (88.5 %) 16 (88.9 %)

Female 21 (29.2 %) 16 (57.1 %) 3 (11.5 %) 2 (11.1 %)

Height [cm] 171.92±7.69 169.14±9.31 173.46±6.69 174.00±4.74

Weight [kg] 86.05±16.60 79.16±19.54 90.27±13.33 90.67±12.48

RE_LL_AP 0.16±0.11 0.16±0.13 0.13±0.09 0.19±0.10

RE_LL_LAP 0.59±0.20 0.69±0.17 0.58±0.20 0.45±0.15

RE_LL_PVP 0.61±0.19 0.68±0.17 0.63±0.18 0.49±0.16

RE_LL_HBP 0.85±0.40 1.05±0.29 0.93±0.40 0.43±0.17

RE_PV_AP 2.49±1.78 2.62±0.93 3.21±2.44 1.27±0.86

RE_PV_LAP 2.79±1.43 3.54±1.06 2.67±1.56 1.80±1.09

RE_PV_PVP 2.22±1.07 2.85±0.76 2.05±1.13 1.48±0.85

RE_PV_HBP 0.71±0.41 0.83±0.30 0.74±0.50 0.48±0.35

RE_AA_AP 4.57±2.52 4.77±1.99 5.35±3.10 3.12±1.69

RE_AA_LAP 2.51±1.01 2.80±0.98 2.57±1.06 1.99±0.81

RE_AA_PVP 2.14±0.80 2.41±0.80 2.19±0.76 1.64±0.62

RE_AA_HBP 0.87±0.37 0.83±0.30 0.97±0.50 0.77±0.23

¹³C-MBT [μg/kg/h] 257.99±134.10 397.61±55.52 226.23±52.39 86.67±37.95

Parameter		¹³C-MBT readout
Age [years]	60.90±9.61	58.89±10.88	64.27±9.63	59.17±5.80
Male	51 (70.8 %)	12 (42.9 %)	23 (88.5 %)	16 (88.9 %)
Female	21 (29.2 %)	16 (57.1 %)	3 (11.5 %)	2 (11.1 %)
Height [cm]	171.92±7.69	169.14±9.31	173.46±6.69	174.00±4.74
Weight [kg]	86.05±16.60	79.16±19.54	90.27±13.33	90.67±12.48
RE_LL_AP	0.16±0.11	0.16±0.13	0.13±0.09	0.19±0.10
RE_LL_LAP	0.59±0.20	0.69±0.17	0.58±0.20	0.45±0.15
RE_LL_PVP	0.61±0.19	0.68±0.17	0.63±0.18	0.49±0.16
RE_LL_HBP	0.85±0.40	1.05±0.29	0.93±0.40	0.43±0.17
RE_PV_AP	2.49±1.78	2.62±0.93	3.21±2.44	1.27±0.86
RE_PV_LAP	2.79±1.43	3.54±1.06	2.67±1.56	1.80±1.09
RE_PV_PVP	2.22±1.07	2.85±0.76	2.05±1.13	1.48±0.85
RE_PV_HBP	0.71±0.41	0.83±0.30	0.74±0.50	0.48±0.35
RE_AA_AP	4.57±2.52	4.77±1.99	5.35±3.10	3.12±1.69
RE_AA_LAP	2.51±1.01	2.80±0.98	2.57±1.06	1.99±0.81
RE_AA_PVP	2.14±0.80	2.41±0.80	2.19±0.76	1.64±0.62
RE_AA_HBP	0.87±0.37	0.83±0.30	0.97±0.50	0.77±0.23
¹³C-MBT [μg/kg/h]	257.99±134.10	397.61±55.52	226.23±52.39	86.67±37.95

Arterial phase (AP), late arterial phase (LAP), portal venous phase (PVP) and hepatobiliary phase (HBP). Values indicate the mean±standard deviation. RE_LL: relative enhancement, a function of SI-based indices measured in the liver (LL), RE_PV: relative enhancement, a function of SI-based indices measured in the portal vein (PV), RE_AA: relative enhancement, a function of SI-based indices measured in the abdominal aorta, ¹³C-MBT: ¹³C-methacetin breath test.

The reasons for MR imaging were active monitoring of hepatocellular carcinoma (n = 17), follow-up reasons in the case of secondary malignancies and benign hepatic lesions (n = 13), preinterventional assessment in the case of HCC, known secondary liver malignancies or focal hepatic lesions (n = 11), postinterventional MR imaging (n = 17) and check-up for suspected liver disease or focal hepatic lesions (n = 14). The patients conditions were HCC with cirrhosis (n = 33), HCC without cirrhosis (n = 1), liver cirrhosis without HCC (n = 3), focal nodular hyperplasia (n = 1), carcinoma of the ileum (n = 1), carcinoma of the duodenum (n = 1), cholangiocarcinoma (n = 8), colon cancer (n = 1), hemangioma (n = 4), mamma carcinoma (n = 2), rectal cancer (n = 11), sigma carcinoma (n = 4), thymoma (n = 1) and uveal melanoma (n = 1).

3.2 Median relative enhancement of the ¹³C-MBT readout classes

A comparison of REs revealed that patients with normal liver function (¹³C-MBT >315.0 [μg/kg/h]) obtained the highest values in hepatic tissue, portal vein and abdominal aorta (Fig. 1). With dynamic MRI sequences, similar increase and decrease patterns, respecticvely, were observed in patients. While the RE of the hepatic tissue steadily increased with dynamic MRI sequences (Fig. 1A), the RE of the portal vein increased until late arterial phase before it started to decrease (Fig. 1B). In the abdominal aorta the highest values of the RE were obtained in the arterial phase (Fig. 1C).

Fig.1

Phase-dependent enhancement of relative enhancement (RE): hepatic parenchyma (RE LL; A), portal vein (RE PV; B) and abdominal aorta (RE AA; C). Cases were stratified in the distinct liver function groups normal liver function (light grey), intermediate liver function (dark grey) and severely impaired liver function (black-white stripes). arterial phase (AP), late arterial phase (LAP), portal venous phase (PVP) and hepatobiliary phase (HBP). Values indicate the mean±standard deviation. RE_LL: relative enhancement, measured in the liver parenchyma, RE_PV: relative enhancement measured in the portal vein (PV), RE_AA: relative enhancement measured in the abdominal aorta (AA), ¹³C-MBT: ¹³C-methacetin breath test, Level of significance: ^*p < 0.05; ^**p < 0.01; ^***p < 0.001.

3.3 Analysis of relative enhancement of tissues

REs of the specific MRI sequences were tested against the ¹³C-MBT readout classes (Table 2). For the hepatic tissue, severely impaired liver function and normal liver function were significantly different (p < 0.002) in the late arterial phase (RE_LL_LAP, p < 0.001) and portal venous phase (RE_LL_PVP; p = 0.003). Additionally, in the hepatobiliary phase (RE_LL_HBP, p < 0.001) severely impaired liver function and intermediate liver function were also significant different (p < 0.001; Fig. 2A). For the portal vein, REs obtained in the arterial phase showed a significant difference between severely impaired and intermediate liver function (p < 0.001) as well as normal liver function (p < 0.001; Fig. 2B). Further significant differences were observed in the late arterial phase (RE_PV_LAP), portal venous (RE_PV_PVP) and hepatobiliary phase (RE_PV_HBP). The abdominal aorta was significant different between severely impaired liver function and normal liver function in the arterial phase (RE_AA_AP; p < 0.05), the late arterial phase (RE_AA_LAP; p < 0.01) and portal venous phase (RE_AA_PVP; p < 0.001; Fig. 2C).

Table 2
Non-parametric test of SI-based indices between distinct liver function groups.

Pairwise comparison

(>315.0) vs. (140.0–315.0) [μg/kg/h] (>315.0) vs. (<140.0) [μg/kg/h] (<140.0) vs. (140.0–315.0) [μg/kg/h]

RE_LL_AP n.s. n.s. n.s.

RE_LL_LAP n.s. p < 0.001 n.s.

RE_LL_PVP n.s. p = 0.002 n.s.

RE_LL_HBP n.s. p < 0.001 p < 0.001

RE_PV_AP n.s. p < 0.001 p < 0.001

RE_PV_LAP p = 0.043 p < 0.001 n.s.

RE_PV_PVP p = 0.027 p < 0.001 n.s.

RE_PV_HBP n.s. p = 0.004 n.s.

RE_AA_AP n.s. p = 0.023 p = 0.030

RE_AA_LAP n.s. p = 0.013 n.s.

RE_AA_PVP n.s. p = 0.002 n.s.

RE_AA_HBP n.s. n.s. n.s.

	Pairwise comparison
RE_LL_AP	n.s.	n.s.	n.s.
RE_LL_LAP	n.s.	p < 0.001	n.s.
RE_LL_PVP	n.s.	p = 0.002	n.s.
RE_LL_HBP	n.s.	p < 0.001	p < 0.001
RE_PV_AP	n.s.	p < 0.001	p < 0.001
RE_PV_LAP	p = 0.043	p < 0.001	n.s.
RE_PV_PVP	p = 0.027	p < 0.001	n.s.
RE_PV_HBP	n.s.	p = 0.004	n.s.
RE_AA_AP	n.s.	p = 0.023	p = 0.030
RE_AA_LAP	n.s.	p = 0.013	n.s.
RE_AA_PVP	n.s.	p = 0.002	n.s.
RE_AA_HBP	n.s.	n.s.	n.s.

REs of hepatic parenchyma, portal vein and abdominal aorta are tested based on ¹³C-MBT readout classification in a non-parametric Kruskal-Wallis test. The post-hoc tests reveal significant differences between the classes. Arterial phase (AP), late arterial phase (LAP), portal venous phase (PVP) and hepatobiliary phase (HBP). Values indicate the mean±standard deviation. RE_LL: relative enhancement, measured in the liver (LL), RE_PV: relative enhancement measured in the portal vein (PV), RE_AA: relative enhancement measured in the abdominal aorta (AA), ¹³C-MBT: ¹³C-methacetin breath test.

Fig.2

Boxplot analysis of SI-based indices separated by ¹³C-MBT readout categories. RE values obtained during the hepatobiliary phase of the liver (RE_LL_HBP; A) show significant differences between severely impaired liver function (median: 0.437) and intermediate (median: 0.901) as well as normal liver function (median: 1.069). RE values obtained in the portal vein during the late arterial phase (RE_PV_AP; B) are also significant different between severely impaired liver function (median: 1.036) and intermediate (median: 2.633) as well as normal liver function (median: 2.482). In contrast, RE values obtained in the abdominal aorta during the portal venous phase (RE_AA_PVP; C) show only between severely impaired (median: 1.537) and normal liver function (median: 2.336) a significant differentiation. ¹³C-MBT >315.0 [μg/kg/h]: normal liver function, ¹³C-MBT: 140.0–315.0 [μg/kg/h]: intermediate liver function, ¹³C-MBT <140.0 [μg/kg/h]: severely impaired liver function, ¹³C-MBT: liver function-dependent ¹³C-methacetin metabolism measured in a breath test, Level of significance: ^*p < 0.05; ^**p < 0.01; ^***p < 0.001.

3.4 Correlation analysis of tissue and sequence specific RE to ¹³C-MBT readout values

In a bivariate Spearman correlation analysis the REs of the tissues at specific time points were tested against logarithmic values of ¹³C-MBT readouts (Table 3). According to the Spearman correlation analysis, REs of the hepatic tissue had the highest correlation with logarithmic ¹³C-MBT readout values in the hepatobiliary phase (r = 0.648; p < 0.001; Fig. 3A). For portal vein, best correlations were reveived the portal venous phase, (r = 0.528; p < 0.001; Fig. 3B) and for abdominal aorta in the arterial phase (r = 0.405; p < 0.001; Fig. 3C).

Table 3
Linear spearman correlation analysis of RE-values to logarithmic values of ¹³C-MBT

r p-value

log(¹³C-MBT) RE_LL_AP –0.153 n.s.

RE_LL_LAP 0.528 <0.001

RE_LL_PVP 0.471 <0.001

RE_LL_HBP 0.648 <0.001

RE_PV_AP 0.347 0.003

RE_PV_LAP 0.521 <0.001

RE_PV_PVP 0.528 <0.001

RE_PV_HBP 0.329 0.005

RE_AA_AP 0.277 0.019

RE_AA_LAP 0.325 0.005

RE_AA_PVP 0.405 <0.001

RE_AA_HBP 0.035 n.s.

Arterial phase (AP), late arterial phase (LAP), portal venous phase (PVP) and hepatobiliary phase (HBP). Values indicate the mean±standard deviation. RE_LL: relative enhancement, measured in the liver (LL), RE_PV: relative enhancement measured in the portal vein (PV), RE_AA: relative enhancement measured in the abdominal aorta (AA), ¹³C-MBT: ¹³C-methacetin breath test, r: correlation coefficient, n.s.: not significant.

Fig.3

Correlation analysis of SI-based indices to logarithmic values of ¹³C-MBT readout in scatterplots. The RE values of the hepatic tissue show the strongest correlation to logarithmic values of ¹³C-MBT readout (r = 0.648; A) compared to RE values of the portal vein (r = 0.528; B) and abdominal aorta (r = 0.405; C). RE_LL_HBP: RE values obtained during the hepatobiliary phase of the liver, RE_PV_PVP: RE values obtained during the portal venous phase of the portal vein, RE_AA_PVP: RE values obtained during the portal venous phase of the abdominal aorta, ¹³C-MBT: liver function-dependent ¹³C-methacetin metabolism measured in a breath test.

4 Discussion

Increased portal hypertension is associated with severe clinical complications of liver cirrhosis like variceal bleeding, ascites or portal hypertensive gastropathy. According to anatomical location, the aetiology of increased portal resistance can be categorized into pre-hepatic, intra-hepatic and post-hepatic causes [2]. While cirrhotic portal hypertension is associated with an increased hepatic venous pressure gradient, in non-cirrhotic patients suffering from portal hypertension, the hepatic venous pressure gradient is normal or only mildly elevated but lower than the portal venous pressure [9]. Architectural changes influence the sinusoidal blood flow and cause an increase of hepatic resistance. This results in an increased splanchnic vasodilation and increased blood flow into the portal venous system and initial changes regarding hemodynamic and microcirculation in liver parenchyma might be detected [6]. Similar to other hepatic disorders, the initial presence of portal hypertension occurs almost unremarkable. Nevertheless, the assessment of the hepatic venous pressure gradient can be a strong indication for portal hypertension. Accurate measurements of the hepatic venous pressure gradient rely on invasive techniques. Therefore a balloon catheter needs to be introduced into the hepatic vein to allow a direct measurement [2]. Another option to access hepatic perfusion parameters is by analyzing the flow limited clearance of xenobiotics like indocyanine green or sorbitol [8, 15]. But, as in cases of advanced liver disease the perfusion as well as intracellular enzymatic and transport processes are altered, so the application of xenobiotics for hepatic perfusion assessment may not represent the perfusion status accurate enough. Another option is to assess hepatic perfusion parameters by a hepatocyte model for pharmacokinetic analysis which is based on dynamic liver MR imaging over time [10].

The radiological imaging of Gadoxetic-acid in MRI-based functional imaging is an established tool for assessing liver function and displaying hepatic lesions. One convincing benefit of the contrast agent is the resistance to enzymatic metabolism. In pioneer-like studies, the use of Gadolinum-based dynamic MRI for the noninvasive assessment of hepatic perfusion parameters was evaluated [11]. They showed that MRI-based compartmental analysis of signal intensity versus time curves can be applied as alternative method to measure hemodynamic parameters, as hepatic flow parameters correlated with the severity of cirrhosis and portal hypertension.

Liver cirrhosis is associated with a loss of functioning hepatocytes, bridging of portal spaces or nodular regeneration of the liver parenchyma, which hinders hepatocyte contrast agent uptake and causes decreased SI-based values [12, 14]. The estimation of SI-based indices provides insights into regional hepatocyte-specific function, efficiency, functionality, condition and perfusion.

In order to reduce the number of different examinations and to reduce ionizing radiation dose, researchers evaluate established tests in order to extract further information about patients’ health status. Different studies proofed that hepatic flow parameters achieved by contrast agent-based MR images are able to display liver function. We hypothesized that T1-weighted contrast-enhanced SI-based MR imaging of the portal vein and the abdominal aorta will be able to display liver function. Therefore we analyzed time-dependent RE values and compared them to liver function, represented by ¹³C-MBT readout values.

Our results demonstrate that SI-based MR-imaging of the portal vein and abdominal aorta show a characteristic pattern which is determined by liver function and time-dependent image acquisition. Similar to Nilsson et al. [13], we observed that patients with liver disease achieve maximum RE after a shorter time than patients with better liver function. Our data permit a crude assumption on hepatic function already in the portal venous phase, based on MR imaging of liver parenchyma, as patients with severely impaired liver function show now further enhancement. In accordance to the finding that hepatic perfusion index is dependent on fibrosis severity made by Leporq et al. [10], we observed that dynamic SI-indices of the liver, portal vein and abdominal aorta are decreased with increasing progression of liver damage. In contrast to our measurements, Leporq et al. used a dual-input three-compartment hepatocyte model for pharmacokinetic analysis. We in contrast expected to see an effect on SI-values without applying a perfusion model, as SI-indices depend indirect on contrast agent concentration, which is influenced by aortic flow and pressure. Our results revealed a significant correlation between SI-based indices of hepatic tissue, portal vein (r = 0.528; p < 0.001) and abdominal aorta (r = 0.405; p < 0.001) and ¹³C-MBT readout values, which correlates to Child-Pugh classes (r = –0.707; p < 0.001) [5].

A recent published study by Wildner et al. [20] demonstrated a significant dependence of angiogenesis parameters assessed by dynamic contrast-enhanced ultrasound (DCEUS) to specific malignant liver lesions. Their data showed that benign lesions (focal-nodular-hyperplasia) and malignant focal liver lesions (hepatocellular carcinoma, cholangiocellular carcinoma, pancreatic adenocarcinoma, breast cancer, colorectal cancer and melanoma) experience specific absolute signal intensity values and characteristic washout. While the focus of our study was set on MRI-based signal intensity values of portal vein and abdominal aorta, an additional analysis based on specific hepatic lesion and DCEUS assessed angiogenesis parameters would improve the informative value.

Our study has several limitations. In order to decrease the influence of nonhomogeneous distributed parenchymal changes, analysis were done with the mean of six distinct ROI measurements in the case of hepatic parenchyma. Another limitation is the absence of healthy patients without hepatic diseases. Also, an analysis based on specific hepatic diseases and hepatic venous pressure gradient would be beneficial, as they might be crucial parameters for portal hypertension which now are not taken into account. An analysis of relaxometry values might add further information, which may improve the evaluation of the abdominal aorta and portal vein.

In conclusion, evaluation of dynamic SI-based indices of liver parenchyma, portal vein and abdominal aorta can give a strong hint on liver function with best results for liver parenchyma evaluation.

References

Ethical guidelines for publication in Clinical Hemorheology and Microcirculation: Update 2016. Clin Hemorheol Microcirc, 2016;63(1):1–2. doi:10.3233/CH-162058.

Bloom

, Kemp

, Lubel

. Portal hypertension: Pathophysiology, diagnosis and management. Intern Med J. 2015;45(1):16–26. doi:10.1111/imj.12590.

Bosch

, Garcia-Pagan

. Complications of cirrhosis. I. Portal hypertension. J Hepatol. 2000;32(1 Suppl):141–56.

Clevert

, D’Anastasi

, Jung

. Contrast-enhanced ultrasound and microcirculation: Efficiency through dynamics– current developments. Clin Hemorheol Microcirc. 2013;53(1-2):171–86. doi:10.3233/CH-2012-1584.

Geisel

, Lüdemann

, Hamm

, Denecke

. Imaging-based liver function tests– past, present and future. RoFo. 2015;187(10):863–871.

Haimerl

, Jung

, Beyer

, Pregler

, Dollinger

, Sieron

, et al. Chronic liver disease: Correlation of CEUS-based microperfusion and indocyanine green clearance. Clin Hemorheol Microcirc. 2015;61(2):195–204. doi:10.3233/CH-151990.

Haimerl

, Brunn

, Poelsterl

, Beyer

, Wiesinger

, Stroszczynski

, et al. Quantitative evaluation of real-time maximum liver capacity (LiMAx) and time intensity curve (TIC) analysis in CEUS-based microperfusion. Clin Hemorheol Microcirc. 2017;67(3-4):373–82. doi:10.3233/CH-179217.

Keiding

, Engsted

, Ott

. Sorbitol as a test substance for measurement of liver plasma flow in humans. Hepatology. 1998;28(1):50–6. doi:10.1002/he510280109.

Khanna

, Sarin

. Non-cirrhotic portal hypertension – Diagnosis and management. Journal of Hepatology. 2014;60(2):421–41. doi:10.1016/j.jhe2013.08.013.

10.

Leporq

, Daire

, Pastor

, Deltenre

, Sempoux

, Schmidt

, et al. Quantification of hepatic perfusion and hepatocyte function with dynamic gadoxetic acid-enhanced MR imaging in patients with chronic liver disease. Clin Sci. 2018. doi:10.1042/cs20171131.

11.

Materne

, Smith

, Peeters

, Dehoux

, Keyeux

, Horsmans

, et al. Assessment of hepatic perfusion parameters with dynamic MRI. Magn Reson Med. 2002;47(1):135–42.

12.

Miles

, Hayball

, Dixon

. Functional images of hepatic perfusion obtained with dynamic CT. Radiology. 1993;188(2):405–11. doi:10.1148/radiology.188.2.8327686.

13.

Nilsson

, Blomqvist

, Douglas

, Nordell

, Jonas

. Assessment of liver function in primary biliary cirrhosis using Gd-EOB-DTPA-enhanced liver MRI. HPB (Oxford). 2010;12(8):567–76. doi:10.1111/j.1477-2574.2010.00223.x.

14.

Popper

. Pathologic aspects of cirrhosis. A review. Am J Pathol. 1977;87(1):228–64.

15.

Reichen

. Assessment of hepatic function with xenobiotics. Semin Liver Dis. 1995;15(3):189–201. doi:10.1055/s-2007-1007277.

16.

Stockmann

, Lock

, Riecke

, Heyne

, Martus

, Fricke

, et al. Prediction of postoperative outcome after hepatectomy with a new bedside test for maximal liver function capacity. Ann Surg. 2009;250(1):119–25. doi:10.1097/SLA.0b013e3181ad85b5.

17.

Stockmann

, Lock

, Malinowski

, Niehues

, Seehofer

, Neuhaus

. The LiMAx test: A new liver function test for predicting postoperative outcome in liver surgery. HPB (Oxford). 2010;12(2):139–46. doi:10.1111/j.1477-2574.2009.00151.x.

18.

Stockmann

, Lock

, Malinowski

, Seehofer

, Puhl

, Pratschke

, et al. How to define initial poor graft function after liver transplantation? - a new functional definition by the LiMAx test. Transpl Int. 2010;23(10):1023–32. doi:10.1111/j.1432-2277.2010.01089.x.

19.

Verloh

, Haimerl

, Rennert

, Muller-Wille

, Niessen

, Kirchner

, et al. Impact of liver cirrhosis on liver enhancement at Gd-EOB-DTPA enhanced MRI at 3 Tesla. Eur J Radiol. 2018;82(10):1710–5. doi:10.1016/j.ejrad.2013.05.033.

20.

Wildner

, Schellhaas

, Strack

, Goertz

, Pfeifer

, Fiessler

, et al. Differentiation of malignant liver tumors by software-based perfusion quantification with dynamic contrast-enhanced ultrasound (DCEUS). Clin Hemorheol Microcirc. 2018. doi:10.3233/ch-180378.

Parameter		¹³C-MBT readout
	All patients n = 72	>315.0 [μg/kg/h]	140.0–315.0 [μg/kg/h]	<140.0 [μg/kg/h]
		n = 28	n = 26	n = 18
Age [years]	60.90±9.61	58.89±10.88	64.27±9.63	59.17±5.80
Male	51 (70.8 %)	12 (42.9 %)	23 (88.5 %)	16 (88.9 %)
Female	21 (29.2 %)	16 (57.1 %)	3 (11.5 %)	2 (11.1 %)
Height [cm]	171.92±7.69	169.14±9.31	173.46±6.69	174.00±4.74
Weight [kg]	86.05±16.60	79.16±19.54	90.27±13.33	90.67±12.48
RE_LL_AP	0.16±0.11	0.16±0.13	0.13±0.09	0.19±0.10
RE_LL_LAP	0.59±0.20	0.69±0.17	0.58±0.20	0.45±0.15
RE_LL_PVP	0.61±0.19	0.68±0.17	0.63±0.18	0.49±0.16
RE_LL_HBP	0.85±0.40	1.05±0.29	0.93±0.40	0.43±0.17
RE_PV_AP	2.49±1.78	2.62±0.93	3.21±2.44	1.27±0.86
RE_PV_LAP	2.79±1.43	3.54±1.06	2.67±1.56	1.80±1.09
RE_PV_PVP	2.22±1.07	2.85±0.76	2.05±1.13	1.48±0.85
RE_PV_HBP	0.71±0.41	0.83±0.30	0.74±0.50	0.48±0.35
RE_AA_AP	4.57±2.52	4.77±1.99	5.35±3.10	3.12±1.69
RE_AA_LAP	2.51±1.01	2.80±0.98	2.57±1.06	1.99±0.81
RE_AA_PVP	2.14±0.80	2.41±0.80	2.19±0.76	1.64±0.62
RE_AA_HBP	0.87±0.37	0.83±0.30	0.97±0.50	0.77±0.23
¹³C-MBT [μg/kg/h]	257.99±134.10	397.61±55.52	226.23±52.39	86.67±37.95

	Pairwise comparison
	(>315.0) vs. (140.0–315.0) [μg/kg/h]	(>315.0) vs. (<140.0) [μg/kg/h]	(<140.0) vs. (140.0–315.0) [μg/kg/h]
RE_LL_AP	n.s.	n.s.	n.s.
RE_LL_LAP	n.s.	p < 0.001	n.s.
RE_LL_PVP	n.s.	p = 0.002	n.s.
RE_LL_HBP	n.s.	p < 0.001	p < 0.001
RE_PV_AP	n.s.	p < 0.001	p < 0.001
RE_PV_LAP	p = 0.043	p < 0.001	n.s.
RE_PV_PVP	p = 0.027	p < 0.001	n.s.
RE_PV_HBP	n.s.	p = 0.004	n.s.
RE_AA_AP	n.s.	p = 0.023	p = 0.030
RE_AA_LAP	n.s.	p = 0.013	n.s.
RE_AA_PVP	n.s.	p = 0.002	n.s.
RE_AA_HBP	n.s.	n.s.	n.s.

		r	p-value
log(¹³C-MBT)	RE_LL_AP	–0.153	n.s.
	RE_LL_LAP	0.528	<0.001
	RE_LL_PVP	0.471	<0.001
	RE_LL_HBP	0.648	<0.001
	RE_PV_AP	0.347	0.003
	RE_PV_LAP	0.521	<0.001
	RE_PV_PVP	0.528	<0.001
	RE_PV_HBP	0.329	0.005
	RE_AA_AP	0.277	0.019
	RE_AA_LAP	0.325	0.005
	RE_AA_PVP	0.405	<0.001
	RE_AA_HBP	0.035	n.s.

Efficacy of dynamic enhancement effects on Gd-EOB-DTPA-enhanced MRI for estimation of liver function assessed by 13 C- Methacetin breath test

Abstract

PURPOSE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1 Introduction

2 Material and methods

2.1 Patients

2.2 Magnetic resonance imaging

2.3 Image analysis

2.4 13C-methacetin breath test

2.5 Statistical analysis

3 Results

3.1 Patient characteristics

References

2.4 ¹³C-methacetin breath test