Parenchymal Liver Blood Volume and Dynamic Volume Perfusion CT Measurements of Hepatocellular Carcinoma in Patients Undergoing Transarterial Chemoembolization

Patients and Methods

The Institutional Review Board approved the study (2014-401M-MA-§ 23b MPG). This prospective study was conducted as part of a research initiative comparing flat-panel CBCT (Artis zeego; Siemens Healthcare GmbH, Forchheim, Germany) and a dual-source MDCT (SOMATOM Force, Siemens Healthcare GmbH, Forchheim, Germany).

Study design. Inclusion criteria for the study was the decision for TACE by the local Interdisciplinary Tumor Board defining the patient not eligible for liver transplantation, surgery, or local ablative treatments according to European recommendations (7). Exclusion criteria were any contraindication for TACE. Contraindication for drug-eluting beads (DEB) TACE were a plasma creatinine level above 2.0 mg/ml, an international normalized ratio (INR) above 2.0, a bilirubin level above 2.0 mg/dl, and a more than five-fold increase in transaminase levels.

Initially 25 patients gave their written informed consent to study inclusion. Nine patients had to be excluded from data analysis because no pre/post TACE dVPCT was obtained. Overall 16 patients and 17 lesions were included in the data analysis.

Treatment protocol. TACE was performed using a transfemoral approach. The diagnostic catheter was placed into the common hepatic artery to perform an angiogram of the hepatic vessel anatomy with different angulations to identify the tumor-feeding arteries. A coaxial microcatheter system (Progreat; Terumo, Eschborn, Germany) was advanced superselectively into the tumor-feeding artery for embolization. Chemoembolization was performed with a maximum dose of 75 mg doxorubicin DEB of 30-60 μm in size (Hepasphere™ microsphere; Merit Medical, South Jordan, UT, USA) until observation of stasis. Control hepatic angiography was repeated after embolization to ensure patency of the main hepatic arteries.

Measurement of PLBV. Before and after DEB TACE, a contrast-enhanced CBCT was performed: The process started with a non-enhanced mask run (5 seconds) with breath-hold. After the rotation of the C-arm back to the starting position (5 seconds), contrast medium injection started via the coaxial microcatheter using a flow rate of 3 ml/s and a volume of 36 ml (12 ml contrast medium diluted with 24 ml saline). Contrast-enhanced fill run (5 s) was performed in breath-hold 8 s after the start of contrast medium injection. The overall PLBV acquisition took 23 s. The microcatheter was placed either in the right or left hepatic artery depending on blood flow and collateral arteries to ensure proper flow of contrast medium into the tumor-feeding arteries.

The targeted liver lesions were evaluated on a commercial post-processing workstation (syngo XWP; Siemens Healthcare GmbH). For PLBV measurement, the ratio of the hypervascularized HCC lesion (PLBV_Tumor) and the normal liver parenchyma (PLBV_Liver) was calculated by placing three regions of interest (ROI) within the HCC lesion being treated and one ROI was placed within the normal liver parenchyma in the same slice. This was performed on three different slices resulting in nine PBLV_Tumor ROI and three PLBV_Liver ROI. The same procedure with the same microcatheter position was performed after DEB TACE.

Dynamic volume perfusion CT measurement. Similarly to Wang et al., dVPCT was performed with the MDCT system (8). Scan parameters were as follows: 70 kVp tube voltage (80 kVp if body mass index was above >33 kg/m²), 190 mAs tube current time-product at 70 kVp, or 220 mAs at 80 kVp respectively, 48×1.2 collimation, 4-dimensional spiral mode with variable pitch with a z-axis coverage of 22.4 cm. The total image acquisition time was 71.2 s obtained by an 18 spiral acquisition and an inter-scan delay of 2×3 s, 10×1.5 s, 3×6 s, 2×15, 1×18. The examination was started 10 s after contrast injection: 50 ml iodinated contrast medium (Iomeprol 400; Bracco Imaging S.p.A., Milan, Italy) using venous access best placed in a cubital vein. Flow-rate was 5 ml/s using a power injector (Stellant® D CT Injection System; MEDRAD, Inc., Warrendale, USA) followed by a 50 ml saline chaser with the same flow rate. The patients were asked to hold their breath as long as possible and then to continue with a shallow breathing through the nose.

The dVPCT images were analyzed after reconstruction of the MDCT raw data using a commercial post-processing, multi-modality workstation (syngo.via; Siemens Healthcare GmbH). Three ROI were placed in a single slice within the treated HCC lesion and one ROI was placed in the healthy liver parenchyma in correlation to the PLBV analysis. This was performed within three different slices. The calculated values of arterial liver parchenyma (ALP; ml/100 ml/min), portal venous parenchyma (PVP; ml/100 ml/min), temporal maximum intensity projection [MIP; Hounsfield units (HU)], and hepatic perfusion index (HPI; %) were used for statistical analysis.

Additionally, the arterial phase images were used for dimensional lesion measurement before and after DEB TACE according to mRECIST, for which only the hypervascular part of the HCC lesion was measured (1). Objective response (OR) was defined as stable disease (SD), partial response (PR), and complete response (CR) in contrast to progressive disease (PD).

Statistical analysis. Statistical analysis was performed using JMP 11.0 (SAS Institute, Cary, NC, USA). Data were tested for normal distribution using the Shapiro–Wilk test. Pre- and post-therapy PLBV, ALP, PVP, HPI and dimensional measurements were compared using the Wilcoxon signed-rank test. Additionally, the differences between pre- and post-therapy PLBV, ALP, PVP, HPI and dimensional measurements were analyzed using multivariate analysis: with r²:0.00 to 0.19 being considered very weak correlation, r²:0.20 to 0.39 weak, r²:0.40 to 0.59 as moderate, r²:0.60 to 0.79 as strong, and r²:0.80 to 1.0 as very strong correlation. Two-tailed p-values of less than 0.05 were considered statistically significant.