Rabbit hepatic arterial anatomy variations: implications on experimental design.

BACKGROUND: The VX2 rabbit model of liver cancer is commonly used to evaluate the efficacy of locoregional anticancer therapy and knowledge of the hepatic arterial anatomy in the rabbit is important for catheter-directed experiments. PURPOSE: To describe the normal anatomy and anatomic variations of the celiac axis and hepatic artery in the rabbit. MATERIAL AND METHODS: Angiograms of 222 rabbits were retrospectively reviewed. The branching pattern of the celiac axis was classified and the diameters of the major branches were measured. Paired t-tests were used to compare the difference between the average sizes of arteries. RESULTS: Variant celiac axis or hepatic artery anatomy was noted in 25.9% of angiograms, with the gastric branches arising from the proper hepatic artery in 23.3% of cases. The celiac axis could be successfully classified into one of five distinct branching patterns in 193 (86.9%) cases. The mean diameters of the right and left hepatic arteries were 0.67 mm (95% CI [0.64, 0.7]) and 1.25 mm (95% CI [1.19, 1.31]), respectively. The mean diameters of the medial and lateral branches of the left hepatic artery were 0.63 mm (95% CI [0.6, 0.67]) and 0.91 mm (95% CI [0.86, 0.96]), respectively. The right hepatic artery was significantly smaller than the left hepatic artery and the lateral branch of the left hepatic artery (all P values <0.0001). CONCLUSION: Arterial variants in the rabbit are not uncommon. The proper hepatic artery often gives origin to gastric artery branches. To facilitate superselective intra-arterial intervention, the left lateral lobe of the liver should be targeted for tumor implantation because of the significant size difference between the right and left hepatic arteries.

Microvascular Injury of the Peribiliary Plexus Associated with Transarterial Chemoembolization: A Study of Surgical Specimens

PURPOSE: We attempted to evaluate the relationship between transarteiral chemoembolization (TACE) and injury to the peribiliary plexus by evaluating the number of microvessels in the peribiliary plexus of surgical specimen. MATERIALS AND METHODS: Surgical specimens were obtained from 78 hepatocellular carcinoma patients and 22 patients with liver metastases. They were divided into 4 groups (Group 1; hepatocellular carcinoma without TACE [n=30], Group 2; hepatocellular carcinoma receiving preoperative TACE once [n=37], Group 3; hepatocellular carcinoma receiving preoperative TACE more than two times [n=11] and Group 4; metastatic carcinoma of the liver patients without a history of liver disease [n=22]). Immunohistochemical staining for factor VIII-related antigen was performed in all the specimens and the number of microvessels in the inner capillary layer and the outer venous layer of the bile duct (> 200 micrometer in diameter) was counted. RESULTS: The mean numbers of microvessels in the inner capillary layer were 4.50, 4.08, 1.64 and 2.05, and those in the outer venous layer were 25.23, 20.00, 18.36 and 12.32 for the 4 groups, respectively. The number of microvessels in group 4 was statistically fewer than that of group 1 (p<0.01). In the hepatocellular carcinoma patients, the number of microvessels was decreased as the number of TACE sessions was increased. CONCLUSION: The number of microvessels in the peribiliary plexus is increased in chronic liver disease patients. It may be from the increased portal pressure and flow stagnation in the sinusoidal and portal venules. TACE can have an effect on microvascular injury of the peibiliary plexus, and this can be a cause of bile duct necrosis and biloma.

Hypervascular Hyperplastic Nodules Appearing in Chronic Alcoholic Liver Disease: Benign Intrahepatic Nodules Mimicking Hepatocellular Carcinoma

PURPOSE: Hypervascular hyperplastic nodules in those patients with chronic alcoholic liver disease and who are hepatitis B and C negative have recently been reported on. The purpose of this study was to correlate the radiologic and pathologic findings with the clinical significance of these hypervascular hyperplastic nodules in chronic alcoholic liver disease. MATERIALS AND METHODS: The study included eight hypervascular nodules of seven patients with chronic alcoholic liver disease, and these patients had abused alcohol for more than 20 years. Eight hypervascular nodules were seen on the arterial phase of dynamic CT scans, but the possibility of HCC was excluded pathologically (n=4) or clinically. The radiologic and pathologic findings, and the changes of these nodules on follow up CT scans were retrospectively analyzed. RESULTS: All nodules showed good enhancement on the arterial phase. The tissue equilibrium phase of the dynamic CT scans showed isodensity in seven patients and low density in one patient. Ultrasound scans revealed hypoechoic findings for three nodules, isoechoic findings for two nodules, hyperechoic findings for one nodule, and two nodules were not detected. Angiograms (n=6) showed late incremental tumor staining, and all the nodules were well seen on the sinusoidal phase. CT during hepatic angiography (n=4) showed well stained tumor. CT during arterial portography (n=4) showed no defect in three nodules and nodular defect in one nodule. The MR images (n=3) showed low signal intensity in two nodules and iso-signal intensity in one nodule on T2WI. Five of six cases for which follow up CT scans were performed showed decrease in size and one was disappeared. CONCLUSION: Radiologically, it is often difficult to differentiate the hypervascular hyperplastic nodules seen in the chronic alcoholic liver disease from hepatocellular carcinoma, and histological confirmation is needed for excluded hepatocellular carcinoma. However, late tumor staining during the sinusoidal phase without any blood supply by feeding vessels or any arterioportal shunt on the angiogram, isodensity during the tissue equilibrium phase of dynamic CT and low signal intensity on T2WI may suggest the presence of hypervascular hyperplastic nodule.

Multidetector-row CT Angiography of Hepatic Artery: Comparison with Conventional Angiography

PURPOSE: To determine the usefulness of three-dimensional CT angiography using multidetector-row CT (MDCT) for delineating the arterial anatomy of the liver. MATERIALS AND METHODS: Hepatic arterial three-dimensional CT angiography was performed using MDCT (Lightspeed Qx/I; GE Medical Systems, Milwaukee, Wis., U.S.A.) in 45 patients with HCC undergoing conventional angiography for transcatheter hepatic arterial chemoembolization. The scanning parameters during the early arterial phase were 2.5 mm slice thickness, 7.5 mm rotation of table speed, and a pitch of 3. Images were obtained by one radiologist using maximum intensity projection from axial CT images obtained during the early arterial phase. Two radiologists blinded to the findings of conventional angiography independently evaluated the hepatic arterial anatomy and the quality of the images obtained. RESULTS: Compared with conventional angiography, reader A correctly evaluated the hepatic arterial anatomy depicted at three-dimensional CT angiography. Reader B's evaluation was correct in 40 of 45 patients. Interobserver agreement was good (kappa value, 0.73), and both readers assessed the quality of three-dimensional CT angiography as excellent. CONCLUSION: Three-dimensional CT angiography using MDCT was accurate for delineating the arterial anatomy of the liver, and interobserver agreement was good. The modality may provide, prior to conventional angiography, valuable information regarding a patient's hepatic arterial anatomy.

Atypical Hepatic Angiomyolipoma: CT and MR Findings: A Case Report

Hepatic angiomyolipoma is a rare benign lipomatous tumor of the liver. Radiologic studies usually reveal a fat component, but since this may be minimal, such a component is not always detected. We report a case of atypical hepatic angiomyolipoma which because of the non-visualization of fat at CT and MR imaging, was difficult to differentiate from other hypervascular tumors.

Nodular Hepatocellular Carcinoma: Findings of Single-Level Dynamic CT During Hepatic Arteriography

PURPOSE: To evaluate the hemodynamics of nodular hepatocellular carcinoma (HCC) using single-level dynamic CT during hepatic arteriography (CTHA) according to the size of the nodule. MATERIALS AND METHODS: Forty 40 patients with nodular HCC confirmed histologically or radiologically underwent single-level dynamic CTHA. Contrast media was injected via the hepatic artery for 20 seconds at a rate of 1.2 ml/sec. Images were obtained prior to this and thereafter at 3-second intervals until 30 seconds after the start of injection, and additional images at 45 and 60 seconds. The CTHA findings were retrospectively reviewed, and cases were assigned to one of three groups, according to the size of the tumor [ 5 cm (group III, n =16)]. We focused on onset time of contrast enhancement (CE), peak CE time and density, and duration of plateau. The patterns of CE were classified as homogeneous, heterogeneous, peripheral, or "nodule in nodule ". RESULTS: The onset time of CE was 3 seconds and mean peak CE time was 18.8 seconds. A plateau was noted in all groups (mean duration, 6.8 seconds). Mean peak CE time and density were 16.5 seconds and 294.1 HU in group I, 18.5 seconds and 324.1 HU in group II, and 20.1 seconds and 114.0 HU in group III. The most frequent pattern of CE (83%) was homogeneous, and this was found in group I. The small HCC group (group I) showed a faster mean peak CE time than the large group (group III) (p < 0.05), and this latter showed a lower peak CE density than the other groups (p < 0.05). In groups II and III, "nodule in nodule "(72%) and heterogeneous patterns (69%) were, respectively, the most frequent findings. CONCLUSION: Single-level dynamic CTHA is useful in evaluating the hemodynamics of HCC.

Pseudolesions around the Gallbladder Fossa: Comparison of Frequency and Radiological Characteristics in Multiphasic CT, CTAP, and CTHA

PURPOSE: The purpose of this study is to compare the frequency with which pseudole-sions around the gallbladder (GB) fossa are revealed by multiphasic CT, by CT during arterial portography (CTAP), and by CT during hepatic arteriography (CTHA) and to determine their radiological characteristics. MATERIALS AND METHODS:Multiphasic CT, CTAP, and CTHA examinations of 81 patients without pathology of the GB and around the GB fossa were evaluated for pseudolesion around the GB fossa. The definition of pseudolesion was as follows: 1) hyperattenuation during the arterial phase and isoattenuation during the delayed phase of multipha-sic CT, or perfusion defect on CTAP and hyperattenuation on CTHA; 2) no Lipiodol tagging on Lipiodol CT; 3) all findings observed adjacent to the gallbladder fossa; and 4) no interval change on follow-up CTAP and CTHA. We compared the frequency of pseudolesions around the GB fossa, as seen on multiphasic CT, CTAP, and CTHA, and determined their size, location, and shape, as revealed by CTHA. RESULTS: The frequency of pseudolesion was 2.5% (2/81) on multiphasic CT, while on CTAP or CTHA, the frequency was 53.1% (43/81), and 58 pseudolesions were identi-fied. Of 58 pseudolesions, 56 were revealed by CTAP and 57 by CTHA. Forty-nine of 58 pseudolesions were larger and all pseudolesions showed more contrast to parenchyma on CTHA than on CTAP. The location of pseudolesions was segment V(32 of 58), IV (25 of 58), and VI (1 of 58), and their size ranged from 5 to 30 (mean, 17.5)mm. Pseudolesions were wedge-shaped (48 of 58), oval (6 of 58), bandlike (3 of 58), or round (1 of 58). CONCLUSION: CTAP and CTHA frequently revealed pseudolesion around the GB fossa. The radiological characteristics of these modalities help differentiate pseudolesions from true tumoral hepatic lesions.

Zebra Pattern in CT during Arterial Portography: Analysis of Associated Factors

PURPOSE: To analyze the factors associated with the zebra pattern in CT during arterial portography(CTAP). MATERIALS AND METHODS: In 275 CTAP procedures, the factors associated with the zebra pattern, such as laminar flow in the portal vein, the presence of liver cirrhosis, the artery selected for CTAP, location of the catheter tip in the superior mesenteric artery(SMA), splenic volume, and the existence of an aberrant right hepatic artery(RHA) emerging from the SMA were analyzed. RESULTS: In 106 of 275 procedures (38.5%), a zebra pattern was apparent. Portal venous laminal flow was seen in 92 % of procedures in the group with this pattern and in 63 % in the group without it. Eighty-three of 235 procedures (35.3 %) in which the SMA was injected and 23 of 40(57.5 %) involving splenic artery injection showed the zebra pattern. In 22 of 35(62.8 %) in which the catheter tip was located in the distal SMA and 61 of 200 ( 30.5 %) in which this was at a proximal site, the zebra pattern was evident. Mean splenic volume was less in the group with the zebra pattern. The effect on the zebra pattern of liver cirrhosis and an aberrant RHA emerging from the SMA was not statistically significant. CONCLUSION: In CTAP, the incidence of the zebra pattern was 38.6%, and was related to laminal flow in the portal vein. The pattern is frequently seen in CTAP involving contrast injection via the splenic artery, distal location of a catheter tip in the SMA, and small splenic volume.

Opacification of Portal Vein on CT during Hepatic Arteriography Via Normal Vascular Flow

PURPOSE: To evaluate the frequency, distribution, and associated factors of the opacified normal portal veinas seen on CT, during hepatic arteriography(CTHA). MATERIALS AND METHODS: One hundred and eighty-nine patientswho underwent both CTHA and CT during arterioportography (CTAP) during a three-year period were reviewed. Ofthese, 66 patients without anomalous hepatic arterial supply and arterioportal (AP) shunt on celiac angiographywere analyzed. Comparing CTHA with CTAP, we determined whether or not the portal vein(PV) is opacified within thesecond-order branch. The degree of contrast enhancement in the stomach, duodenum, and pancreas was graded asfollows : grade I, all three organs were hypodensely enhanced; grade II, some organ were isodensely enhanced, butothers hypodensely ; grade III, all three organs were isodensely enhanced relative to the CHA. The relationshipbetween opacified portal vein (OPV) and the degree of enhancement of the three organs, amount of contrast media,and Child classification was statistically examined. RESULTS: The PV was opacified in 18 of the 66 patients (27%); This was the main PV in 16, right PV in 13, and left PV in two. Of the single branches, the right post posteriorbranch was most commonly opacified. Among 18 patients with OPV, the degree of three organs (stomach, duodenum, andpancreas) was grade I in two, grade II in two, and grade III in 14 while among 48 patients with nonopacified PV, thefindings were grade I in 27, grade II in seven, and grade III in 14. The relationship between OPV and degree ofenhancement of the three organs was statistically significant (p=0.001). There was however, no statisticallysignificant difference between OPV and Child classification and the amount of contrast media. CONCLUSION: PVopacification during CTHA is not rare and this finding should not therfore be regarded as indicator of apathologic conditions such as AP shunt.

Pseudolesion in Segment IV of the Liver on CT Arterial Portography

PURPOSE: To determine the incidence of pseudolesions seen in the posterior aspect of segment IV of the liveron CT arterial portography(CTAP), and to evaluate the findings of CT hepatic arteriorgraphy(CTHA) and celiacangiography. MATERIALS AND METHODS: We retrospectively analysed 450 cases of patients who had undergone CTHA,CTAP and celiac arteriorgraphy for the diagnosis and treatment of hepatic lesion, and evaluated the incidence andimaging findings of pseudolesions in segment IV of the liver, as seen on CTAP, CTHA and celiac arteriography. RESULT: In 28 of 450 patients(6.2%), pseudolesions of focal perfusion defect were seen in segment IV on CTarterial portography. Pseudolesions were seen on CTAP on 39 different section slices ; these were wedge-shaped in41% of cases(16/39), rectangular in 30.8%(12/39), ovoid-shaped in 23.1%(9/39), and half moon-shaped in 5.1%(2/39); they were from 1 to 3cm(average, 1.5cm) in anteroposterior diameter, from 1 to 3cm(average, 1.9cm) in width, andfrom 1 to 4cm(average, 1.9cm) in craniocaudal diameter. Twenty-seven patients underwent CTHAs ; hyperattenuationwas seen in 13(48.2%), isoattenuation in 12(44.4%), and hypoattenuation in two(7.4%). In 19 of 28 celiacangiograms(67.9%), the right gastric artery was seen to arise from the hepatic artery, and aberrant right gastricveins directly draining into the left lobe of the liver were seen in three(10.7%). CONCLUSION: The incidence ofpseudolesion seen in segment IV of the liver on CTAP was 6.2% (28/450), and in 3 of 28 cases(10.7%), celiacangiography showed aberrant right gastric venous drainage. For pseudolesions, CTHA showed variable attenuation,and this modality is less sensitive than CTAP for the detection of pseudolesion.

Double Uncture fo Single Common Femoral Artery for CT Hepatic Arteriography and CT Arterial Portography

PURPOSE: To evaluate the safety and efficacy of double puncture of a single common femoral artery for CThepatic arteriography(CTHA) and CT arterial portography(CTAP) in patients with hepatocellular carcinoma. MATERIAL AND METHODS: Between October and December 1995, 35 patients with hepatocellular carcinoma underwent doublepuncture of a single common femoral artery for combined CTHA and CTAP. Preangiographic laboratory data were asfollows : platelet count from 28,000 to 250,000/mm3 (average, 124,500/mm3) ; prothrombin time from 45.8% to100%(average, 85.3%). In the inguinal area, a pair of 21G puncture needles were used unilaterally for the firsttwo femoral punctures, at a distance of 5-7mm ; a 0.018" guidewire for the insertion of a 4-F sheath in a coaxialmicropuncure introducer set ; and a 0.035" guidewire and 4 F check-flo sheath for the insertion of 4-F catheters.After being moved to the CT room, patients then underwent spiral CTHA and CTAP for further detection ofhepatocellular carcinoma nodules. Transarterial chemoembolization followed, and the punctured site was thencompressed by the usual finger compression method. After initial compression and one day later, we observed thepuncture site for complications such as hematoma formation, thromboembolization or arteriovenous fistula ; ect. RESULTS: Except for the formation of two mild hematomas, no remarkable severe complications were noted. Fifteenpatients who had previously undergone combined spiral CTHA and CTAP using the bilateral femoral puncture method(with a 5-F check-flo sheath) felt more comfortable than when an earlier method was used. CONCLUSION: For combinedCTHA and CTAP, double puncture of a single common femoral artery is safer and more comfortable than the bilateralfemoral puncture method.

Normal Anatomical Variation of the Intrahepatic Portal Vein in Korean

PURPOSE: To determine the prevalence of variations in the intrahepatic portions of portal veins, as visualized by arterial portograms, and to examine the surgical implications of these findings in Koreans. MATERIALS AND METHODS: Five hundred and nineteen arterial portograms of 519 patients with no evidence of vascular tumor invasion or distortion were retrospectively reviewed. In all patients, the main, right, right anterior, right posterior and left portal vein were visualized. RESULTS: Variations in intrahepatic portal anatomy were found in 102 patients (19.6%) and involved an immediate trifurcation of the main portal vein in 82 (15.8%) a right posterior segment from the main portal vein in ten (1.9%), a right anterior segment from the left portal vein in nine (1.7%) and an absent horizontal portion from the left portal vein in one (0.2%). CONCLUSION: On arterial portogram, variations in the intrahepatic portions of the portal veins are frequently seen. The recognition of such variations is important in the preoperative evaluation of patients with hepatic tumors, since the variations may have implications for tumor resection, for the localization of tumor thrombi and in interventional radiologic procedures involving the portal vein.

Recurrent Pyogenic Cholangitis: Angiographic Findings and Its Significance

PURPOSE: To analyse the angiographic findings of recurrent pyogenic cholangitis. MATERIALS AND METHODS: Hepatic arteriography and portography were performed preoperatively in 34 patients with intrahepatic stones and recurrent cholangitis. Twenty five of these underwent partial resection of the liver and in nine, the biliary tract was drained. Hepatic arteriogram and portogram findings correlated with liver atrophy and were analysed retrospectively by two radiologists; angiographic and pathologic findings also correlated. RESULTS: In the arterial phase, abnormal stained areas, were seen in 17 of 34 cases, periarterial staining in 14, mass-like staining in two and a mixed pattern in one. The hepatic artery showed spastic change in 11 of 34 cases and tortuous change in 17. Arteriovenous shunting was not seen. In the portal phase, abnormal findings of the portal veins were noted in 16 cases; decreased size and nonvisualization were seen in eleven patients, and decreased size only, in five. Fifteen cases showed liver atrophy; in 13 of these, portal vein abnormalities were also present. CONCLUSION: In recurrent pyogenic cholangitis, angiographic findings may be normal or findings of abnormal periarterial staining, mass-like staining, spastic and tortuous change of the hepatic artery, and abnormal portal vein can be present. The differential diagnosis of hepatitis, hepatic mass and cirrhosis should be considered.

Analysis of Anatomic Variation of the Hepatic Artery by Spiral CT Hepatic Arteriography, Lipiodol CT, and Angiography

PURPOSE: To analyse the branching of the hepatic artery and respective arterial blood supply of liver segments, based on the Counaud classification. MATERIALS AND METHODS: The analysis was based on 176 patients who had undergone celiac arteriography, superior mesenteric arteriography, spiral CT hepatic arteriography(CTHA) andlipiodol CT for the detection of hepatic lesions. We compared CTHA findings with those of angiography and Lipiodol CT. RESULTS: There were five groups of hepatic artery variations and 23 of segmental supply pattern. The mostcommon branching pattern was that all branches arose from the common hepatic artery(131 cases ; 74%). The most common segmental supply pattern, seen in 38% of cases, is that the right hepatic artery supplied segments V-VIII, and the left hepatic artery supplied segments II-IV. Segment IV supply was mainly from the left hepatic artery, with 34% being provided by the right hepatic artery ; the supply to segment IV from the hepatic artery also satisfied 76% of the supply needs of segment 1. CONCLUSION: These results could be helpful for the planning of treatment such as transarterial chemoembolization or subsegmentectomy.

Anatomical variation of the Main and Right Portal Vein on Indirect Portogram: Correlated with CT and Hepatic Angiogram

PURPOSE: To describe variations of the main and right portal veins as visualized by indirect portograms and to examine the surgical implications of these findings. MATERIALS AND METHODS: A retrospective review was conducted of 632 indirect portograms of 632 patients in whom third-order branches were visualized. All patientsalso underwent dynamic CT and AP and oblique hepatic angiography. RESULTS: Variations of the main portal veinwere found in 165 patients(26.1%) and involved an immediate trifurcation of the main portal vein in 102 patients(16.1%), early division of the right posterior portal vein in 53(8.4%), simultaneous division into fourbranches in three (0.5%), five branches in two (0.3%), a replaced P4 originating from the right portal vein in two(0.3%), and an accessory P6 originating from the main portal vein, a replaced P6 originating from the main portalvein, or a replaced left lateral portal vein originating from the right portal vein in one (0.2%), respectively.Of 468 patients who had a common right portal vein, absence of the right posterior portal vein was seen in 26 patients(5.6%) and an accessory subsegmental branch originating from the right portal vein in 17 patients(3.6%). CONCLUSIONS: Recognition of such variations of the portal vein assists in the localization and appreciation ofthe hepatic segmental anatomy, in preoperative evaluations for hepatic resections, and in radiologic interventional procedures through the portal vein.

Small Arterioportal Shunt: A Pseudolesion Mimicking Hepatocellular Carcinoma in Angiography

PURPOSE: To document the findings of a small arterioportal(AP) shunt on hepatic angiography through the various diagnostic modalities, and to determine whether this shunt is related to hepatocelluar carcinoma. MATERIALS AND METHODS: During the hepatic angiography in 223 patients, a small AP shunt and subsequent focalparenchymal staining more than 1cm in diameter were found in 28 patients. We compared this angiographic abnormality using MR imaging(n=10), CTAP(n=12), iodized-oil CT(n=23), intraoperative ultrasonography(n=5) and follow-up hepatic angiography(n=11), as well as with conventional ultrasonography and CT scan. RESULTS: Arterioportal shunts noted on angiographic study matched with perfusion defects noted in CTAP(10/12) and also with the area of arterial enhancement seen on dynamic CT scan(3/4). In no case was a lesion found on MR and no Lipiodol uptake was seen on CT. There was no evidence of tumor growth around the AP shunts on follow-up angiographies, andno tumor was found during surgery. CONCLUSION: A small AP shunt was not related to the presence of a tumor. Ifthe hemodynamic changes resulfing from a small AP shunt are understood, confusion can be avoided in the interpretation of vascular imaging including CTAP and dynamic CT.