This study evaluated the usefulness of cone-beam computed tomography (CBCT) during ultraselective transcatheter arterial chemoembolization (TACE) for hepatocellular carcinomas (HCC) that could not be demonstrated on angiography. Twenty-eight patients with 33 angiographically occult tumors (mean diameter 1.3 +/- 0.3 cm) were enrolled in the study. The ability of CBCT during arterial portography (CBCTAP), during hepatic arteriography (CBCTHA), and after iodized oil injection (LipCBCT) to detect HCC lesions was retrospectively analyzed. The technical success of TACE was divided into three grades: complete (the embolized area included the entire tumor with at least a 5-mm wide margin), adequate (the embolized area included the entire tumor but without a 5-mm wide margin in parts), and incomplete (the embolized area did not include the entire tumor) according to computed axial tomographic (CAT) images obtained 1 week after TACE. Local tumor progression was also evaluated. CBCTAP, CBCTHA, and LipCBCT detected HCC lesions in 93.9% (31 of 33), 96.7% (29 of 30), and 100% (29 of 29) of patients, respectively. A single branch was embolized in 28 tumors, and 2 branches were embolized in five tumors. Twenty-seven tumors (81.8%) were classed as complete, and 6 (18.2%) were classed as adequate. None of the tumors were classed as incomplete. Twenty-five tumors (75.8%) had not recurred during 12.0 +/- 6.2 months. Eight tumors (24.2%), 5 (18.5%) of 27 complete success and 3 (50%) of 6 adequate success, recurred during 10.1 +/- 6.2 months. CBCT during TACE is useful in detecting and treating small HCC lesions that cannot not be demonstrated on angiography.

Ultraselective TACE induces not only complete tumor necrosis but also peritumoral parenchymal necrosis, similar to that after radiofrequency ablation, when the portal veins are markedly visualized during the TACE procedure.

The purpose of this study was to evaluate the clinical course of main bile duct stricture at the hepatic hilum after transcatheter arterial chemoembolization (TACE) for hepatocellular carcinoma (HCC). Among 446 consecutive patients with HCC treated by TACE, main bile duct stricture developed in 18 (4.0%). All imaging and laboratory data, treatment course, and outcomes were retrospectively analyzed. All patients had 1 to 2 tumors measuring 10 to 100 mm in diameter (mean ± SD 24.5 ± 5.4 mm) near the hepatic hilum fed by the caudate arterial branch (A1) and/or medial segmental artery (A4) of the liver. During the TACE procedure that caused bile duct injury, A1 was embolized in 8, A4 was embolized in 5, and both were embolized in 5 patients. Nine patients (50.0%) had a history of TACE in either A1 or A4. Iodized oil accumulation in the bile duct wall was seen in all patients on computed tomography obtained 1 week later. Bile duct dilatation caused by main bile duct stricture developed in both lobes (n = 9), in the right lobe (n = 3), in the left lobe (n = 4), in segment (S) 2 (n = 1), and in S3 (n = 1). Serum levels of alkaline phosphatase and γ-glutamyltranspeptidase increased in 13 patients. Biloma requiring drainage developed in 2 patients; jaundice developed in 4 patients; and metallic stents were placed in 3 patients. Complications after additional TACE sessions, including biloma (n = 3) and/or jaundice (n = 5), occurred in 7 patients and were treated by additional intervention, including metallic stent placement in 2 patients. After initial TACE of A1 and/or A4, 8 patients (44.4%), including 5 with uncontrollable jaundice or cholangitis, died at 37.9 ± 34.9 months after TACE, and 10 (55.6%) have survived for 38.4 ± 37.9 months. Selective TACE of A1 and/or A4 carries a risk of main bile duct stricture at the hepatic hilum. Biloma and jaundice are serious complications associated with bile duct strictures.

The purpose of this study was to evaluate the detectability of corona enhancement around the hypervascular hepatocellular carcinoma (HCC) by dual-phase cone-beam computed tomography during hepatic arteriography (CBCTHA). Dual-phase CBCTHA was performed for 71 HCC lesions (mean ± SD 1.7 ± 0.9 cm), including seven presenting a nodule-in-nodule appearance and nine hypervascular pseudolesions. The first scan was performed during injection of 30-40 ml half-diluted contrast material at a rate of 1.5-2 ml/s through the hepatic artery. Scanning was initiated 7 s after the beginning of contrast material injection. The second scan was started 30 s after the end of the first scan. Detectability of corona enhancement on second-phase CBCTHA was evaluated. Thickness of corona enhancement was also analyzed as thin (≤2 mm) or thick (>2 mm). Corona enhancement was detected in 63 (88.7%) of 71 tumors (1.8 ± 0.9 cm), but it was not detected in eight tumors (1.0 ± 0.2 cm). Thin corona enhancement was seen in 18 tumors (1.2 ± 0.5 cm), and thick corona enhancement was seen in 45 tumors (2.0 ± 0.9 cm). There was a significant difference in tumor diameter between tumors with and those without corona enhancement (P = 0.0157) and between thin and thick corona enhancement (P = 0.001). In all seven early-stage tumors, corona enhancement was demonstrated around the hypervascular focus within the hypovascular tumor portion. None of the nine pseudolesions showed any corona enhancement. Dual-phase CBCTHA depicted corona enhancement in 88.7% of hypervascular HCC lesions. This technique may improve the diagnostic accuracy of HCC.

No abstract

There are usually multiple caudate arteries arising from the right, left, and middle hepatic arteries, and they are frequently connected to each other. Therefore, hepatocellular carcinoma (HCC) in the caudate lobe is frequently fed by multiple branches arising from different origins. HCC located in the Spiegel lobe is usually fed by the caudate arteries derived from the right and/or left hepatic artery. HCC in the paracaval portion is mainly fed by the caudate artery derived from the right hepatic artery; with low frequency, it is fed by the caudate artery derived from the left hepatic artery. HCC in the caudate process is usually fed by the caudate artery derived from the right hepatic artery. Because of the complexity and overlap of vascular territories, the tumor-feeding branch of a recurrent HCC lesion in the caudate lobe frequently changes on follow-up arteriograms. In addition, several extrahepatic collateral vessels supply the recurrent tumor. To perform effective transcatheter arterial chemoembolization (TACE) for HCC in the caudate lobe, radiologists should have sufficient knowledge of vascular anatomy supplying HCC in the caudate lobe.

Abstract:Purpose: The present study aimed to distinguish between glioblastomas and primary central nervous system lymphomas (PCNSLs) using 1 H-Magnetic Resonance Spectroscopy (MRS), especially glutamate (Glu) / creatine (Cr) and Glu/Glu + glutamine (Gln) ratios.Materials and methods: A total of 46 patients (31 cases diagnosed with glioblastoma, 15with PCNSL) were examined by in vivo single-voxel proton 1 H-MRS with a 3-T MR imaging system. Differences in absolute concentration of Cr, choline/Cr, lipid (1.3ppm)/Cr, Glu+Gln/Cr, Glu /Cr, and Glu/Glu+Gln ratios among groups were evaluated with Mann-Whitney U test.Results: PCNSLs (3.408 ± 1.194 [standard deviation]) showed significantly higher Glu/Cr ratios as compared to glioblastomas (2.220 ± 0.942; P= 0.003) (Glu/Cr cutoff ratio of 2.509showed a sensitivity of 88% (7/8) and a specificity of 92% (22/24)), while glioblastomas (0.539 ± 0.098) showed significantly lower Glu/Glu+Gln ratios as compared to PCNSLs (0.728 ± 0.147; P<0.001) (Glu/Glu+Gln cutoff ratio of 0.558 showed a sensitivity of 69%(18/26) and a specificity of 100% (13/13)). And PCNSLs (1.101 ± 0.387) showed significantly higher Cho/Cr ratios as compared to glioblastomas (0.850 ± 0.465; P= 0.026). Conclusion:Glu/Cr, Glu/Glu+Gln, and Cho/Cr ratios may be useful in distinguishing (Figure 1d, 1e). The independent quantification of Glu and Gln can be considerably improved by moving from 1.5T to 3T.With regard to Glu+Gln, it has been reported that Glu+Gln/Cr ratios from contrast-enhancing regions do not differ significantly between glioblastomas and PCNSLs [2]. No differences inGlu or Gln alone in in vivo 1 H-MRS between them have been reported, to the extent that we could determine.On the other hand, it has been reported that in high-grade gliomas (World HealthOrganization grades III and IV), microdialysates in the tumor periphery consistently show significantly higher extracellular Glu relative to microdialysates in non-tumoral regions, unlike PCNSL [5]. And, high extracellular Glu has been noted to stimulate Gln synthetase and promote the synthesis of Gln from Glu in cultured astrocytes [6].Based on these reports, we hypothesized that in glioblastomas due to high extracellular Glu the synthesis of Gln from Glu would be promoted [6], and so Glu/Glu+Gln ratios might be lower than those in PCNSLs. The purpose of this study was to analyze the differences in the 1 H-MRS findings among glioblastomas and PCNSLs with special reference to Glu and Gln expression. Materials and Methods:Patients:This retrospective analysis of the data was approved by the institutional review board of our university. All patients gave informed consent prior to inclusion in this study. All patients Clinical parameters in the analyzed groups are summarized in Table 1. All tumors were confirmed histopathologically after preoperative diagnosis with magnetic resonance imaging (MRI) and 1 H-MRS, and were not subjected to biopsy, surgical resection, chemotherapy or radiation therapy before the preoperative diagnosis with MRI and 1 H-MRS. We...

The inferior phrenic artery (IPA) is the most common extrahepatic collateral vessel to hepatocellular carcinoma (HCC); however, there are many anatomical variations in its origin and branches. In addition, the IPA is frequently reconstructed through several pathways, mainly through the retroperitoneal network, because of the occlusion of its orifice due to atherosclerosis or previous catheter manipulation. Infrequently, selective catheterization into the IPA is impossible even using a microcatheter, particularly in the IPA that originates from the proximal or distal portion of the celiac trunk or from the aorta with an acute angle. In this article, we describe anatomical variations of the IPA and catheterization techniques, such as a catheter with a large side hole and a catheter with a cleft, to facilitate catheterization into the IPA that is difficult using a conventional coaxial technique. Radiologists should have sufficient knowledge of such variations and catheterization techniques to perform transcatheter arterial chemoembolization for HCCs through the IPA effectively and safely.