The efficacy of FP 736-04 in experimental liver cirrhosis

PURPOSE: To determine whether the uptake of FP 736-04, a hepatocyte-specific CT contrast agent, is influenced by cirrhotic changes in the liver. MATERIAL AND METHODS: Liver cirrhosis was induced by bile duct ligation (BDL) in Sprague-Dawley rats. Seventy-four animals were divided into three groups comprising: rats with acute BDL; rats with chronic BDL; and normal controls. CT was performed after i.v. infusion of FP 736-04 or saline at a dose of 2 ml/kg b.w., and the mean attenuation in the liver and spleen was measured. The livers from the chronic BDL group were taken for histopathological examination and the extent of the disease was graded according to an arbitrary scale. RESULTS AND CONCLUSION: There was a significant reduction of the native liver attenuation in both chronic and acute BDL groups as compared with the normal controls. FP 736-04 was taken up by the liver parenchyma with a similar degree of enhancement in all three groups.     

X-ray structures of new dipeptide taste ligands

AIMS/BACKGROUND: The mechanism of interaction and the role played by the vesicle lipid composition for the selective association between liposomes and liver cells were studied, at the ultrastructural level, by investigating both in situ and in vitro the interaction between hepatocytes, Kupffer and endothelial liver cells with egg-phosphatidylcholine (eggPC) or eggPC/stearylamine (9:1; mol:mol) reverse-phase evaporation (REV) liposomes. METHODS: Liver cells from rats, isolated by enzymatic perfusion and purified by differential centrifugation, were incubated, in a rotating bath at 37 degrees C, with liposomes (2.5 mM final liposomal lipid concentration). Cell aliquots were withdrawn and processed for electron microscope observation at fixed time intervals. Parallel experiments were carried out by in situ liver perfusion with liposome suspensions. RESULTS AND CONCLUSIONS: Our first conclusions are: 1) lipidic composition affects the rate of liposomes uptake and internalization by hepatocytes; 2) liposome uptake by hepatocytes or Kupffer cells is likely an endocytic process; 3) endothelial cells internalize lipid vesicles as well; 4) liposome uptake was due to a phagocytic activity for all isolated liver cells, while in the in situ observation endothelial cells seem to use another mechanism (fusion); and 5) the rate of internalization is related to the viability of the treated cells. Experimental data seem to indicate that differential behaviour in the internalization of lipid vesicles exists among parenchymal, Kupffer and endothelial liver cells. These differences suggest that clearance of liposomes by these cells involves two mechanisms (i.e., endocytosis or fusion) with different rates of uptake and internalization that facilitate the design of carriers that can deliver drugs preferentially to a specific liver cell type.     

Triphasic spiral computerized tomography of the liver: vascular models of non-cystic focal lesions

PURPOSE: The purpose of this study was to investigate if Triphasic Spiral CT (arterial, portal and equilibrium phases) can improve the characterization of noncystic focal lesions. MATERIAL AND METHODS: Sixty-six patients with suspected focal liver disease underwent Triphasic Spiral CT. After the injection of 120-140 ml contrast material at 3 ml/s the liver was imaged in the arterial (scanning delay: 20-27 s), portal (scanning delay: 45-80 s) and equilibrium (scanning delay: 5-8 min) phases. The enhancement of each lesion was evaluated in each phase and the lesions were grouped by enhancement pattern (11 patterns in all). The reference standards in our 66 patients were surgery (12), biopsy (43), MRI (9), follow-up (9), somatostatin receptor scintigraphy (6). RESULTS: One hundred and twenty-six liver lesions were detected in 66 patients, four of 11 enhancement patterns (hypo/hyper/hyper, hyper/iso/iso, hyper/hyper/iso, hyper/hyper/hyper) were always referrable to benign disease (hemangioma, focal nodular hyperplasia-FNH-adenoma). Four of 11 enhancement patterns (iso/hypo/hypo, iso/iso/hypo, hyper/hypo/hypo, hyper/hyper/hypo) were always referrable to malignant disease (hepatocellular carcinoma-HCC-metastases). The other two patterns (hypo/hypo/hypo, hypo/hypo/hyper) were seen in both benign and malignant diseases. CONCLUSIONS: Triphasic Spiral CT improves the characterization of HCC, FNH, adenoma and hemangioma. The arterial and the equilibrium phases add no information to the yield of the portal venous phase in metastases, except for those from pancreas neuroendocrine tumors in the arterial phase. In our experience, patients with unclassified lesions at US or conventional CT, suspected HCC and metastases from pancreas neuroendocrine tumors should be submitted to Triphasic CT of the liver. This technique however does not appear to be indicated in the study of liver metastases from hypovascular tumors, while it improves the detection of FNH and adenoma.     

The effect of stimulation of the mononuclear phagocyte system on the binding of high- and low-density lipoproteins by the hepatocytes, Kupffer cells and liver endotheliocytes of rats

The role of the mononuclear phagocyte system (MPS) in the regulation of low- (LDL) and high-density lipoprotein (HDL) receptor activity in different rat liver cells was investigated. MPS was activated by intravenous administration of bacterial lipopolysaccharide (LPS) from Serratia marcescens. Liver cells were isolated by in vitro perfusion of the liver with a collagenase solution. Separation of Kupffer and endothelial cells was performed by the method of centrifugal elutriation. In control rats, the hepatocytes bound 1.6 times more 125I-HDL and 2.5 times more 125I-LDL per cell than Kupffer cells. Treatment of rats with LPS resulted in a 4.5-fold decrease in the 125I-HDL binding to Kupffer cells. In contrast, the hepatocytes from LPS-treated rat bound 2 times more 125I-HDL than that from untreated rats. The binding of 125I-LDL and 125I-HDL to endothelial cells and 125I-LDL to hepatocytes were not affected by LPS treatment. These results suggest that the MPS (especially Kupffer cells) plays an important role in the regulation of HDL receptor activity in hepatocytes.     

Different intrahepatic distribution of phosphatidylglycerol and phosphatidylserine liposomes in the rat

Liposomes with diameters of 200 to 400 nm containing phosphatidylserine (PS) or phosphatidylglycerol (PG) were injected intravenously into rats. Two hours after injection, 75% of the injected dose of PS liposomes was found in the liver and only 10% found in the spleen, while 35% of the PG liposomes was found in the liver and as much as 40% was found in the spleen. Cell-isolation experiments revealed the following remarkable difference in the intrahepatic distribution between the two liposome formulations: the PS liposomes distributed in about equal amounts to Kupffer cells and hepatocytes, despite their size (200-400 nm) exceeding that of the endothelial fenestrae (average 150 nm), whereas the PG liposomes were only taken up by the Kupffer cells and not at all by the hepatocytes. Double-label studies, using liposomes in which the lipid-moiety was radio labeled with [3H]cholesteryloleylether ([3H]CE) and the water phase with [14C]sucrose, showed that the liposomes were taken up as intact particles. These observations were confirmed through electron microscopy by determining the in situ localization of liposome-encapsulated colloidal gold particles in thin sections of liver and spleen. The differences in organ distribution are ascribed to differences in opsonization patterns of the two liposomal surfaces. For the difference in intrahepatic distribution, we offer the following two explanations: the exploitation of the blood cell-mediated forced sieving concept and the indication of a PS-specific pharmacological effect on the dimensions of the fenestrations.     

Apoptosis of neutrophils and their elimination by Kupffer cells in rat liver

The fate of neutrophils in the peripheral circulation is poorly understood. In this study, the role of Kupffer cells in eliminating aged and apoptotic neutrophils was investigated. Liver, spleen, lung, and blood samples from Wistar rats were examined by light and electron microscopy, the terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate-biotin nick end labeling (TUNEL) method, and immunohistochemistry after the intravenous injection of OK-432, a streptococcal preparation. Neutrophils were trapped predominantly in the periportal and midzonal regions of hepatic lobules and were in contact with endothelial cells and Kupffer cells, or were surrounded by Kupffer cells. The trapping of neutrophils peaked after 6 hours. Apoptotic neutrophils, with or without buds, were found in the lumen of hepatic sinusoids as early as 6 hours, reached maximal levels after 12 hours, and represented greater than 60% of the total number of neutrophils in the liver. The presence of apoptotic neutrophils was correlated with the degree of neutrophil phagocytosis. Double-staining showed that TUNEL-positive neutrophils were phagocytosed or encircled by ED1- or ED2-positive Kupffer cells. In contrast, apoptosis and phagocytosis of neutrophils were rare in the spleen, lung, and peripheral blood. These results suggested that the appearance of apoptotic neutrophils in the hepatic sinusoids and their rapid clearance by Kupffer cells occurs after the invasion of bacteria (i.e., bacteremia or bacteriotoxemia) or the release of inflammatory mediators into the blood stream. These findings have important implications for the regulation of neutrophil homeostasis, the limitation of inflammation and tissue injury, and provide insight into the physiological removal of circulating, senescent neutrophils.     

Noncorrelation between thoracic skeletal injuries and acute traumatic aortic tear

PURPOSE: The aim of the study was to investigate whether moderately to well differentiated hepatocellular carcinomas (HCC) possess the same ability to take up a new lipid emulsion contrast medium, FP 736-03, as do hepatocytes. MATERIAL AND METHODS: A rat model of an experimental HCC was used. CT was performed before and after an i.v. bolus injection of 1.0 ml FP 736-03/kg b.w. Attenuation values of normal liver parenchyma and tumour tissue were measured over time. RESULTS: The contrast medium was taken up by the normal liver parenchyma but not by the tumour tissue for which the attenuation remained virtually constant over the observation period. CONCLUSION: The study demonstrated that FP 736-03 was not taken up by the HCC, thus producing a high lesion-to-liver contrast.     

CT with different doses of the hepatocyte-specific contrast medium FP 736-03. Evaluation in a nude-rat model of experimental metastases

PURPOSE: To study the dose-response relationship in FP 736-03 (48 mg I/ml), hepatocyte-specific contrast medium for CT of the liver. MATERIAL AND METHODS: A nude-rat model of experimental hepatic metastases was used. CT of the liver was performed before and after i.v. injection of FP 736-03 at 4 different doses: 0.25, 0.5, 1.0 and 2.0 ml/kg b.w. Attenuation in the normal liver parenchyma and in the metastases was measured and plotted as a function of time. RESULTS: The enhancement of the normal liver parenchyma increased in the dose range studied. No increase was found in the metastases: the attenuation here remained constant during the observation period. The maximum enhancement values at doses of 0.25, 0.5, 1.0 and 2.0 ml/kg b.w. were (mean +/- SD) 13.5 +/- 2.7, 30.1 +/- 4.2, 33.2 +/- 4.5 and 59.7 +/- 13.1 HU respectively.     

A comparison of two MR hepatobiliary gadolinium chelates: Gd-BOPTA and Gd-EOB-DTPA

PURPOSE: Two gadolinium chelates with partial hepatobiliary excretion, Gd-BOPTA and Gd-EOB-DTPA, and one gadolinium chelate with exclusively renal excretion, Gd-HP-DO3A, were compared on MRI at 1.5 T. The time course of enhancement for normal liver, gallbladder, spleen, kidney, and muscle was specifically examined in the rhesus monkey. METHOD: Four animals were evaluated with each agent for a total of 12 MR studies. Breath-hold and non-breath-hold T1 weighted scans were acquired prior to and 1, 2, 3, 4, 5, 15, 30, 45, 60, 75, and 90 min after intravenous contrast medium injection. The same contrast dose, 0.1 mmol/kg, was used for all studies. Images were analyzed by region-of interest measurements. RESULTS: Both hepatobiliary gadolinium chelates achieved sustained enhancement of normal liver parenchyma, superior in magnitude to that following Gd-HP-DO3A injection. On sans 45-90 min following injection, liver enhancement with Gd-BOPTA was superior to that with Gd-EOB-DTPA. This difference was, however, not statistically significant. Liver enhancement decreased more rapidly on delayed scans with Gd-EOB-DTPA than with Gd-Bopta, a result that was statistically significant. Excretion of contrast agent into the gallbladder was noted with both hepatobiliary agents but not with Gd-HP-DO3A. CONCLUSION: Enhancement of normal liver parenchyma peaks at a later time after injection with Gd-BOPTA than with Gd-EOB-DTPA. However, the maximum percent enhancement is comparable when (as in the current evaluation) the two agents are compared at the same dose (0.1 mmol/kg). This finding supports the choice of optimal imaging time post contrast agent administration (for delayed scans) in clinical trials of 20-45 min post injection with Gd-EOB-DTPA and 60-120 min post injection with Gd-BOPTA.     

Prolongation of discordant heart xenograft survival after liver hemoperfusion: role of nonparenchymal liver cells

In the discordant guinea pig (GP)-to-rat combination, heart xenografts are hyperacutely rejected. The aim of the present study was to demonstrate that a donor-species-specific extracorporeal liver hemoperfusion can prolong survival of discordant heart xenografts and to specify the role of non-parenchymal cells. GP hearts were grafted into Brown Norway rats (group 1) In group 2, heart xenografting was carried out immediately after a 15-min GP hemoperfusion. In group 3, Kupffer cells of the GP liver were blockaged by intravenous injection of dextran sulfate (4 mg/100 g) 30 min before hemoperfusion. In group 4, Kupffer cells of the liver were activated by intravenous injection of muramyl dipeptide (MDP; 500 micrograms/250 g) 24 h before hemoperfusion. Lymphocytotoxic antibodies were detected according to a complement-dependent antibody assay. A donor-specific liver hemoperfusion can delay hyperacute rejection of heart xenografts (67.6 +/- 47.1 min in group 2 versus 8.0 +/- 2.4 min in group 1; p < 0.01) and reduce the level of lymphocytotoxic antibodies. Deposits of immunoglobulins and complement were significant on the hemoperfused liver and moderate on the transplanted heart. In group 3, after blockade of Kupffer cells with dextran, heart xenograft survival was less prolonged (31.8 +/- 8.2 min) and the decrease in antibody levels was not significant and associated with moderate deposits of immunoglobulins and complement on the hemoperfused liver and significant deposits on heart xenografts. In group 4, after stimulation of Kupffer cells by MDP, a significant decrease in antibody levels was present, and significant deposits were observed. These results show that donor-specific liver hemoperfusion can prolong the survival of discordant heart xenografts and support the hypothesis that nonparenchymal liver cells play a major role by absorption of preformed antibodies and complement.     

Influence of a portacaval shunt on the distribution of 14C-chol-PC-DCP-liposomes and liposome entrapped 3H-methotrexate in the organs of rats

The individual roles of hepatic parenchymal cells and non-parenchymal cells in the uptake of intravenously injected liposomes by the liver have not yet been clearly determined. Experimentally, it is suggested that a portacaval shunt reduces the phagocytic capacity of the liver RES. In our experiments a portacaval shunt does not influence the hepatic uptake of 14C-Chol-PC-DCP-liposomes and the liposome entrapped 3H-Methotrexate in rats. In normal rats and in rats with a portacaval shunt, most of the 14C- and 3H-radioactivity is found in the liver and the ratio between the amounts of free and liposome entrapped radioactivity 3 to 15 hours after i.v. injection is approximately 1:10. Therefore, it is concluded that the liver RES does not play an integral role in the enrichment of liposomes in the liver.     

Role of Kupffer cells in arresting circulating tumor cells and controlling metastatic growth in the liver

Metastasis to the liver is a common event in clinical oncology. Blood-borne tumor cells (TCs) arriving to the liver sinusoids run into a special vascular bed. The lining of liver sinusoids is shared by Kupffer cells (KCs) and endothelial cells. KCs, liver-fixed macrophages, are responsible for detection and removal of "non-self" particles. To investigate their role in arresting blood-borne TCs and controlling tumor growth, we injected a syngeneic colon carcinoma cell line into a mesenteric vein of two groups of rats; one group was without Kupffer cells and the other normal controls. We removed the liver of these animals at different time intervals and performed immunohistochemical analysis with monoclonal antibodies (MoAbs) against our tumor cell line, three macrophage subpopulations, natural killer cells, and B and T lymphocytes. Additionally, we showed in vitro spontaneous cytotoxicity of KCs against our tumor cell line. Results suggest that KCs play a relevant role in arresting circulating TCs at the liver sinusoid, although it is limited to a small number of malignant cells. They also seem to play a major role in clearing neoplastic cells from the liver parenchyma, in controlling tumor growth in the very early stages of metastatic development, and in modulating the host immune response to cancer cells.     

Effects of methyl palmitate on cytokine release, liver injury and survival in mice with sepsis

The effects of methyl palmitate (MP), a known inhibitor of Kupffer cells, were studied in a model of polymicrobial sepsis induced in CD-1 mice by cecal ligation and puncture (CLP). The inhibition of Kupffer cells by pretreatment with MP was shown by the reduced phagocytosis, the production of tumor necrosis factor (TNF) and interleukin-6 (IL-6) after lipopolysaccharide (LPS) challenge. The reduced activation of Kupffer cells resulted in lower levels of inflammatory products after CLP. TNF and IL-6 were significantly reduced in serum 2 h and 24 h respectively after CLP, interleukin-1 beta (IL-1 beta) was reduced in liver 4 h after CLP, nitric oxide (NO) and serum amyloid A (SAA) were significantly reduced 8 and 24 h respectively after CLP. Liver toxicity was significantly reduced in MP-treated mice and survival was significantly prolonged at all intervals, reaching 45% after six to ten days compared with 3% in control mice. These findings suggest that Kupffer cells play an important role in liver damage and survival in sepsis.     

Regression of metastatic liver tumors in rats treated with angiogenesis inhibitor TNP-470: occurrence of apoptosis and necrosis

To clarify the mechanism of the reduction of metastatic liver tumors in rats treated with angiogenesis inhibitor TNP-470, the death of tumor cells was examined pathologically and ultrastructurally. Liver metastases were developed by intravenous injection of AH-130 cells. TNP-470 was given subcutaneously after tumor cell injection. Alterations in the size and number of metastatic tumors were examined at various time points, in association with the analysis of cell death pattern. The metastatic nodules were divided into 4 groups according to the morphological patterns of cell death; no cell death, scattered apoptosis, central necrosis, and diffuse necrosis. The number and size of the metastatic tumors at 2 weeks in untreated rats were larger than those in treated rats. The number of tumors in untreated rats decreased, but the tumor size increased. All rats treated with TNP-470 were alive and free from tumors after 4 weeks, whereas all the untreated rats died of liver metastases. The percentages of the tumors with necrosis in untreated rats (61.2% at 2 weeks and 100% at 4 weeks) were significantly higher than that (31.8% at 2 weeks) in treated rats (P < 0.01). The percentage of the tumors containing apoptotic cells in treated rats was significantly higher than that in untreated rats (54.5% vs. 30.6%; P < 0.05). The growth of metastatic tumors without treatment might be faster than the growth of vessels in untreated tumors, resulting in central necrosis due to ischemia. On the other hand, the reduction of metastatic liver tumors treated with TNP-470 might be caused by inhibition of angiogenesis, providing a weak ischemic stimulus which triggers apoptosis, rather than by a direct cytotoxic effect on tumor cells, because previous in vivo experiments demonstrated that TNP-470 affected endothelial cells but not tumor cells.     

Small structural ensembles for a 17-nucleotide mimic of the tRNA T psi C-loop via fitting dipolar relaxation rates with the quadratic programming algorithm

The kinetic behaviour of a naproxen human serum albumin conjugate (Nap23-HSA) was investigated in rats and in isolated perfused rat livers (IPRL), as compared to its active metabolite naproxen-lysine (Nap-lysine) and free naproxen. Through covalently linking the anti-inflammatory drug naproxen to HSA, this drug can be selectively delivered to non parenchymal cells of the liver. Liver endothelial and Kupffer cells play an important role in the pathogenesis of inflammatory liver diseases. Targeting naproxen to these cells might increase its efficacy and reduce the side effects. The altered kinetic properties of Nap23-HSA, after i.v. injection of 22 mg x kg(-1), as compared to an equimolar amount of the uncoupled drug, were demonstrated in vivo by a decrease in the steady state volume of distribution (41 +/- 5 vs. 134 +/- 19 ml x kg(-1)), a decrease in its clearance (0.48 +/- 0.05 vs. 0.63 +/- 0.1 ml x min(-1) x kg(-1)), a shorter plasma half life (60 +/- 11 vs. 152 +/- 44 min) and a sustained biliary excretion. Liver targeting of Nap23-HSA was clearly demonstrated: drug content of the liver 180 min after injection was about 30 times higher for Nap23-HSA as compared to naproxen itself. The IPRL experiments showed that the Vmax of hepatic removal of the conjugate was 40 microg x min(-1) x g liver(-1). With doses below receptor saturation a rapid removal of the conjugate (t1/2 = 6 min) from the perfusion medium was found. In conclusion, this study demonstrates the saturable uptake of Nap23-HSA and its lysosomal degradation in both in vivo and IPRL experiments. Covalently linked naproxen is released as Nap-lysine. This active metabolite accumulates in Kupffer and endothelial cells in which it reaches therapeutic concentrations. Release from these cells leads to rapid uptake by hepatocytes and carrier mediated excretion into bile. Levels of Nap-lysine in bile and plasma reflect the slowest step in its generation: the proteolytic release in endothelial and Kupffer cells.     

Proliferating cell nuclear antigen, plasma fibronectin, and liver regeneration rate after seventy percent hepatectomy in normal and cirrhotic rats

BACKGROUND: The difference in liver regeneration rate in relation to proliferating cell nuclear antigen (PCNA) and plasma fibronectin level in the cirrhotic and control liver after 70% hepatectomy were examined in the rat. METHODS: Liver cirrhosis was induced by intraperitoneal injection of thioacetamide for 12 weeks; rats without thioacetamide administration served as controls. On the day before and days 1, 2, 3, and 7 after 70% hepatectomy, PCNA labeling index of hepatocyte, plasma fibronectin level, and percentage of the initial liver weight were determined. RESULTS: Liver regeneration rate as expressed by percent of initial liver weight was impaired in the cirrhotic liver, and significantly lower regeneration rate was observed on days 3 and 7 after hepatectomy in the cirrhotic rats as compared with controls. PCNA labeling index was higher in the cirrhotic liver before hepatectomy. Little change of PNA labeling index was observed in the cirrhotic liver, although the index increased significantly in the control liver; significantly higher values were observed on days 1, 2, and 3, with the maximal value on day 2 after hepatectomy. The plasma fibronectin level was significantly lower in the cirrhotic rat before and after hepatectomy. CONCLUSIONS: The results show that liver regeneration is impaired in the cirrhotic liver associated with little activation of PCNA and with lower plasma fibronectin level.     

Inactivation of Kupffer cells prevents early alcohol-induced liver injury

It is well recognized that consumption of alcohol leads to liver disease in a dose-dependent manner; however, the exact mechanisms remain unclear. Hypoxia subsequent to a hypermetabolic state may be involved; therefore, when it was observed recently that inactivation of Kupffer cells prevented stimulation of hepatic oxygen uptake by alcohol, the idea that Kupffer cells participate in early events that ultimately lead to alcohol-induced liver disease became a real possibility. The purpose of this study was to test that hypothesis. Male Wistar rats were exposed to ethanol continuously by means of intragastric feeding for up to 4 weeks using the model developed by Tsukamoto and French. In this model, ethanol causes fatty liver, necrosis and inflammation--changes characteristic of alcohol-induced liver disease in human beings. Kupffer cells were inactivated by twice weekly treatment with gadolinium chloride (GdCl3), a selective Kupffer cell toxicant. AST levels were elevated to 192 +/- 13 and 244 +/- 56 IU/L in rats exposed to ethanol for 2 and 4 wk, respectively (control value, 88 +/- 7). This injury was prevented almost completely by GdCl3 treatment. Fatty changes, inflammation and necrosis were also all reduced dramatically by GdCl3 treatment. The average hepatic pathological score of rats treated with ethanol for 4 wk was 4.3 +/- 0.6, which was reduced significantly in ethanol- and GdCl3-treated rats to 1.8 +/- 0.5 (p < 0.05). Rates of ethanol elimination were elevated 2- to 3-fold in rats exposed to ethanol for 2 to 4 wk. This elevation was blocked by GdCl3 treatment.(ABSTRACT TRUNCATED AT 250 WORDS)     

Lipiodol avidity of neuroendocrine liver metastases

AIMS: Lipiodol has been shown to concentrate in most hepatocellular carcinomas as well as in some liver metastases, including those of neuroendocrine origin. Our aim was to determine the proportion of neuroendocrine liver metastases that take up lipiodol and to identify tumour characteristics that predict avidity. METHODS: Avidity was assessed in 12 patients with neuroendocrine liver metastases by performing an abdominal CT scan immediately after selective hepatic arterial injection of 5 ml of unlabelled lipiodol and this was correlated with number and size of lesions as well as angiographic and plain CT scan features. RESULTS: In seven patients the tumours displayed lipiodol avidity (four solitary, three multiple); five patients had non-avid lesions (all multiple). A large dominant liver tumour was the only predictor of avidity (mean diameter of largest lesion 9 cm vs. 3 cm for patients with non-avid tumours: P=0.01). Avidity was not related to vascularity or CT density of lesions. CONCLUSIONS: Although this is a small study, it would appear that approximately 50% of neuroendocrine liver metastases selectively concentrate lipiodol, which could have implications for targeted cancer therapy.