Improvement of image quality and assessment of respiratory motion for hepatocellular carcinoma with portal vein tumor thrombosis using contrast-enhanced four-dimensional dual-energy computed tomography.

Shingo Ohira,Hayate Washio,Masahiko Koizumi,Teruki Teshima,Naoyuki Kanayama,Toshiki Ikawa,Noriko Kishi,Yoshihiro Ueda,Masayoshi Miyazaki,Takero Hirata,Kentaro Wada,Tsukasa Karino,Dong-Hoon Lee

doi:10.1371/journal.pone.0244079

Shingo Ohira, Hayate Washio + Show 11 more

Open Access

https://doi.org/10.1371/journal.pone.0244079

Copy DOI

Abstract

To assess the objective and subjective image quality, and respiratory motion of hepatocellular carcinoma with portal vein tumor thrombosis (PVTT) using the contrast-enhanced four-dimensional dual-energy computed tomography (CE-4D-DECT). For twelve patients, the virtual monochromatic image (VMI) derived from the CE-4D-DECT with the highest contrast to noise ratio (CNR) was determined as the optimal VMI (O-VMI). To assess the objective and subjective image quality, the CNR and five-point score of the O-VMI were compared to those of the standard VMI at 77 keV (S-VMI). The respiratory motion of the PVTT and diaphragm was measured based on the exhale and inhale phase images. The VMI at 60 keV yielded the highest CNR (4.8 ± 1.4) which was significantly higher (p = 0.02) than that in the S-VMI (3.8 ± 1.2). The overall image quality (4.0 ± 0.6 vs 3.1 ± 0.5) and tumor conspicuity (3.8 ± 0.8 vs 2.8 ± 0.6) of the O-VMI determined by three radiation oncologists was significantly higher (p < 0.01) than that of the S-VMI. The diaphragm motion in the L-R (3.3 ± 2.5 vs 1.2 ± 1.1 mm), A-P (6.7 ± 4.0 vs 1.6 ± 1.3mm) and 3D (8.8 ± 3.5 vs 13.1 ± 4.9 mm) directions were significantly larger (p < 0.05) compared to the tumor motion. The improvement of objective and subjective image quality was achieved in the O-VMI. Because the respiratory motion of the diaphragm was larger than that of the PVTT, we need to be pay attention for localizing target in radiotherapy.

Highlights

Hepatocellular carcinoma (HCC) is the second leading cause of cancer-related deaths in worldwide, and the incidence of portal vein tumor thrombosis (PVTT), which is associated with dismal outcomes, is 30–40% in patients with advanced HCC [1,2,3]
The resultant contrast to noise ratio (CNR) was highest in the virtual monochromatic image (VMI) at 60 keV, and the image was determined as the optimal VMI (O-VMI) in this study (Fig 2C)
The O-VMI demonstrated the superiority of the objective image quality over the standard VMI at keV (S-VMI), showing significantly higher lesion contrast (LC) (134.9 ± 43.6 HU vs 79.1 ± 26.2 HU, p = 0.002) and CNR (4.8 ± 1.4 vs 3.8 ± 1.2, p = 0.002)

Summary

Introduction

Hepatocellular carcinoma (HCC) is the second leading cause of cancer-related deaths in worldwide, and the incidence of portal vein tumor thrombosis (PVTT), which is associated with dismal outcomes (median overall survival of 2.7–4 months), is 30–40% in patients with advanced HCC [1,2,3]. An application of four-dimensional computed tomography (4D-CT) is widely utilized to determine the position and respiratory motion of the target as well as OARs. the difference in CT numbers (Hounsfield unit, HU) between the liver and portal vein is small, and it is difficult to visualize the HCC with PVTT clearly with the conventional 4D-CT. The difference in CT numbers (Hounsfield unit, HU) between the liver and portal vein is small, and it is difficult to visualize the HCC with PVTT clearly with the conventional 4D-CT To overcome this problem, a methodology of contrast-enhanced 4D-CT (CE4D-CT) was presented, and it could allow both enhancement of liver tumor contrast and coverage over the entire breathing cycle [5]. The CE-4D-CT has the potential for improving the accuracy of tumor contouring and localization

Objectives

Methods

Results

Conclusion