Synopsis
The hepatocyte fraction (HeF) is based on simple pharmacokinetics, and
can quantitatively estimate the fraction of hepatocytes. In this study, the
HeF, liver-spleen contrast ratio and delta T1 value were compared to the
results of 99mTc-GSA scintigraphy using the blood clearance index (HH15) and
receptor index (LHL15). The correlation coefficients of the HH15 were 0.602,
0.544 and 0.773, respectively, and of the LHL15 were 0.612, 0.670 and 0.762,
respectively. The HeF quantification showed the highest correlation with the 99mTc-GSA,
proving it to be useful for a robust evaluation of liver function, compared to
conventional imaging based quantitative methods.Introduction
Most studies using imaging to estimate liver function have reported gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (gadoxetic acid; Gd-EOB-DTPA) as a promising method for the characterization of liver function. Most of these studies used the measurement of the contrast ratio between the liver and the spleen1, 2), T1 and delta T1 value changes of the liver parenchyma3), and calculation of the reduction rate of the T1 value between the pre and post-contrast images4)as markers.
We demonstrated a new method to assess the quantification of the hepatocyte fraction based on simple pharmacokinetics5)using Gd-EOB-DTPA at the last meeting6). Therefore, the purpose of this study was to investigate the correlation between 99mTc-GSA scintigraphy for the evaluation of liver function and the hepatocyte fraction, and conventional image based methods.
Materials and Methods
Forty-eight patients underwent 99mTc-GSA scintigraphy, with measurements
of their blood clearance indexes (HH15) and receptor indexes (LHL15). The patient
criteria were classified into three severity levels, as Normal (n=14), Mild
(n=25) and Moderate/Severe (n=9). These patients were then scanned using a 3T
clinical scanner (Achieva TX, Philips Healthcare) with a MultiTransmit RF
system and a 32 channel phased-array receiver coil. The quantitative
liver-to-spleen contrast ratio (LSR=SIliver/SIspleen) was calculated using the
signal intensities of the liver and spleen in the hepatobiliary phase, from a fat-suppressed
T1-weighted gradient echo sequence with the following scan parameters: T1-TFE,
TE/TR=1.47/3.1 ms, FA=10, 1.32×1.64 mm, 4 mm/-2 mm thickness, 100 slices, shot
interval=5 sec, SENSE factor=1.8 and 12.8 sec scan time with breath holding.
The ΔR1 value of the liver was measured pre and
post-contrast administration, and the latter was done in the hepatobiliary
phase, 17 to 22 minutes after the injection. The T1 map was calculated from the Look-Locker sequence using the following
scan parameters: T1-TFE, TE/TR=1.7/7 ms, 1.37×1.37×8 mm, FA=7, single-slice,
shot interval=5 sec, SENSE factor=2 and scan time=20 sec with breath holding. The
proposed new model of the hepatocyte fraction was calculated from the ΔR1 values
of the liver and spleen (Figure 1).
Finally, the LSR, ΔT1 value and hepatocyte
fraction of each severity level were compared statistically using the Kruskal-Wallis
method, and were compared with the LHL15 and HH15. A p value<0.01 was
considered to be significant.
Results
Overall, the methods were significantly different
among the groups (Kruskal-Wallis method, p value<0.01) (Figure 2), and the p
values of the hepatocyte fractions were p<0.0001 (significantly different).
Graphs were designed to compare the correlation of the HH15 and LHL15 with the LSR,
ΔT1 and hepatocyte fraction (Figure 3). The correlation coefficients of the
HH15 were 0.602, 0.544 and 0.773, respectively, while the correlation
coefficients of the LHL15 were 0.612, 0.670 and 0.762, respectively (Table 1). The
hepatocytefraction had a strong and high correlation for both the HH15 and
LHL15 markers.
Images of the ΔT1 map and hepatocyte fraction map
are shown in Figure 4. The hepatocyte fraction clearly separated the normal and
deteriorated liver functional areas and hepatocellular carcinoma.
Discussion
The values of the LSR and
ΔT1 showed signal
changes in the
hepatocytes, blood and extracellular
extravascular space (EES)
between the pre and post-contrast images. On the other hand, the
hepatocyte
fraction method reflected the intracellular signal change, because the blood
and EES signal were normalized by the measurement of the ΔR1 in the
spleen. Therefore, the
hepatocyte fraction method, as expected, is a more
accurate estimation of the assessment of liver function. Furthermore, the
hepatocyte
fraction was not affected by the
vascularity, showing its utility in the
assessment of the
tumour type, and for follow up examinations providing quantitative information.
Conclusion
The quantification of the
hepatocyte fraction using the
T1 values pre and post-Gd-EOB-DTPA administration has been demonstrated, and
compared with the measurement of the HH15 and LHL15 via
99mTc-GSA
scintigraphy.
The results indicate that the
hepatocyte fraction quantification has a higher
correlation with
99mTc-GSA, and is useful for a robust evaluation of liver
function, when compared to conventional imaging based quantitative methods.
Acknowledgements
No acknowledgement found.References
1)
U. Motosugi et al.
JMRI, 2009; 30:1042-1046.
2)
S. Nakamura et al.
Jpn J Radiol, 2012; 30:25-33
3)
T. Katsube et al.
Investigative Radiology, 2011; 46:277-283.
4)
K. Kamimura et al.
JMRI, 2014; 40:884-889.
5)
O. Dahlqvst et al.
Eur Radiol, 2012; 22:642-653.
6)
T. Okuaki et al.
ISMRM, 2015 proc.