1947
3D hepatocyte fraction index using 3D look locker1Philips Japan, Tokyo, Japan, 2Division of Radiology, Kanazawa University Hospital, Kanazawa, Japan, 3Faculty of Health Sciences, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University, Kanazawa, Japan, 4Philips Healthcare, Tokyo, Japan
Synopsis
The Hepatocyte Fraction index (HeFra), which is based on a simple pharmacokinetic model, can quantitatively estimate the fraction of hepatocytes. It is calculated from R1 change between pre contrast and hepatobiliary phases. 2D look locker (LL) has been used to calculate T1 value in HeFra studies, but covers only a single slice in one breath hold. Clinically 3D scanning is preferable to get a wider coverage of the liver. Therefore, we optimized the imaging parameters of 3D LL. T1 value and HeFra using the optimized 3D LL were the same as those of 2D LL in subjects under T1 1400ms.
INTRODUCTION
Hepatocyte fraction index (HeFra), which is based on a simple pharmacokinetic model1 and can estimate the fraction of hepatocytes, has been described recently2. The HeFra is calculated using the R1 change between pre contrast and hepatobiliary phases. 2D look locker (LL) has been used to calculate T1 (R1) value in HeFra studies2-4 so far, but covers only a single slice in one breath hold (BH). Clinically, 3D scanning is preferable to get a wider coverage of the liver. Therefore, the purpose of this study is to optimize imaging parameters of 3D LL and apply this sequence to the HeFra model.METHODS
This study was approved by the IRB and was performed only after informed consent was obtained. All scans were performed by using an anterior coil and posterior built-in coil on a 3.0T Ingenia (Philips Healthcare, Best, The Netherlands).
Phantom study
A T1 phantom with 8 cylindrical tubes was built (Fig. 1). The tubes were filled with water and different concentrations (0.05-0.8mmol/l) of Gd-DTPA. A phantom with T1 values up to 1400ms was made, considering the T1 value range including normal liver (800ms)5, spleen (1350ms)5 and typical T1 values of diseased liver. For reference values, the T1 phantom was scanned using a 2D LL with the following parameters: TR/TE 6ms/1.55ms, TFE factor 16, shot 2 (multi-shot), FA 7°, phase interval (PI) 96ms, phase number (PN) 40, shot interval (SI) 6000ms, slice 1, scan time 15s. The T1 values ranged from 311ms to 1412ms. Then 3D LL was performed with the following fixed parameters: TR/TE 2.8ms/0.96ms, TFE factor 35, shot 1 (single-shot), FA 7°, PI 100ms, slice 5. We assessed PN and SI directly related to scan time in order to acquire 3D within one BH. First, we scanned 3D LL for PN of 5-15 with fixed value of SI (5000ms), and compared the calculated T1 value using 3D LL with that of 2D LL. For each phantom, a circular region of interest (ROI) was chosen manually and T1 values were determined. Second, we scanned 3D LL with fixed appropriate PN for SI of 3000-5000ms and compared T1 value using 3D LL with that of 2D LL.
Patient study
2D and 3D LL with imaging parameters optimized in the phantom study were performed in two patients (early stage liver cirrhosis and fatty liver). Images were acquired in pre contrast and hepatobiliary phases. HeFra maps based on 2D and 3D LL were calculated from the ΔR1 values of the liver and spleen. For T1 value and HeFra in liver, ROIs were drawn manually in liver on pre-contrast T1 map and were copied to T1 map on hepatobiliary phase. Five ROIs were placed in both lobes of liver parenchyma avoiding focal hepatic lesions, major branches of the portal or hepatic veins. We calculated the percentage error (%) of T1 value and HeFra in two patients.
RESULTS and DISCUSSION
Phantom study
There was good correlation in T1 value between 2D LL and 3D LL for all PN (Fig. 2a) (PN 5, slope 1.26, R2 0.995; PN 7, slope 1.18, R2 0.982; PN 10, slope 1.07, R2 0.99; PN 12, slope 1.02, R2 0.99; PN 15, slope 0.98, R2 0.99). The PNs for which both the slope and R2 are close to 1 are 12 and 15. PN 12 was selected as the optimal value, since for a given shot interval increasing PN leads to longer scan time. There was good correlation in T1 value between 2D and 3D LL (PN of 12) for all SI (Fig. 2b) (SI 3000, slope 0.90, R2 0.990; SI 3500, slope 0.95, R2 0.991; SI 4000, slope 0.98, R2 0.993; SI 4500, slope 1.01, R2 0.994; SI 5000, slope 1.02, R2 0.994). SI 4000 was selected as an optimal value, considering scan time.
Patient study
2D and 3D LL with PN of 12 and SI of 4000 were performed in two patients, and HeFra maps were calculated (Fig. 3). Percentage error between T1 and HeFra using 2D and 3D LL was under 3% in both patients (Table 1).
CONCLUSION
T1 values calculated using 3D LL with PN of 12 and SI of 4000 were almost the same as those obtained using 2D LL, in phantoms with a T1 of less than 1400ms. 3D LL made it possible to acquire multi slice HeFra map in one BH (scan time 18s) in patients.Acknowledgements
No acknowledgement found.References
1. Dahlqvist Leinhard O, Dahlström N, Kihlberg J, et al. Quantifying differences in hepatic uptake of the liver specific contrast agents Gd-EOB-DTPA and Gd-BOPTA: a pilot study. Eur Radiol 2012; 22(3): 642-53.
2. Okuaki T, Morita K, Namimoto T, et al. Assessment of the Hepatocyte Fraction for estimation of liver function. Proceedings of the 24th Annual Meeting of ISMRM, Toronto, 2015 #4301
3. Okuaki T, Morita K, Namimoto T, et al. Comparison of the Hepatocyte Fraction and Conventional Image Based Methods for the Estimation of Liver Function. Proceedings of the 24th Annual Meeting of ISMRM, Singapore, 2016 #165
4. Pan S, Wang XQ, Guo QY. Quantitative assessment of hepatic fibrosis in chronic hepatitis B and C: T1 mapping on Gd-EOB-DTPA-enhanced liver magnetic resonance imaging. World J Gastroenterol 2018; 24(18): 2024-35.
5. de Bazelaire CM1, Duhamel GD, Rofsky NM, et al. MR imaging relaxation times of abdominal and pelvic tissues measured in vivo at 3.0 T: preliminary results. Radiology 2004; 230(3): 652-9.
Figures
