Renal functional reserve is well maintained in patients with advanced liver cirrhosis and ascites.

BACKGROUND: Glomerular filtration rate (GFR) and effective renal plasma flow (ERPF) in healthy humans can be induced by amino acid stimulation. The rise in GFR from baseline to maximum is referred to as the renal functional reserve (RFR). Recently, we showed that the RFR is preserved in patients with compensated cirrhosis despite impaired renal function. In the present study, we evaluated RFR in decompensated cirrhotics with ascites. METHODS: Steady-state inulin- and para-aminohippurate (PAH) clearances were performed at rest and during amino acid infusion in 22 patients with decompensated liver cirrhosis and ascites. RESULTS: Baseline GFR and ERPF (means +95% confidence intervals) were: GFR 25.2 (21.1-29.2) ml min(-1), ERPF 266.6 (229.7-303.5). Amino acid infusion significantly increased GFR by 67% (38.3-95.8) to 34.6 (29.2-40.0) ml min(-1) (means + (95% confidence intervals), P < 0.001) and ERPF by 29% (11.9-46.3) to 326.3 (274.1-378.5) ml min(-1) (P = 0.002). Renal vascular resistance dropped by 13.4% (3.3-23.5) from 29.4 (24.8-33.9) mmHg ml(-1) min(-1) to 26.4 (22.0-30.7) mmHg ml(-1) min(-1) (P = 0.036). The improved kidney function was accompanied by a decrease in systemic aldosterone levels (P < 0.05). CONCLUSION: In patients with liver cirrhosis and ascites, amino acid infusion improves kidney function. Trials are warranted to test the long-term effects of amino acid infusions in patients with hepatorenal syndrome.

Hemodynamic effects of propranolol and nitrates in cirrhotics with transjugular intrahepatic portosystemic stent-shunt.

BACKGROUND: The combination of tailored TIPS with vasoactive drugs might allow reduction of the rate of subsequent shunt-related sequelae. METHODS: We studied cirrhotic patients 8 weeks (median) after TIPS insertion (8-10 mm) for variceal bleeding. Nitrate (0.1 mg/kg) and propranolol (0.15 mg/kg) alone or combined (same dosages) were infused (I h) sequentially at 1-h intervals (n = 17). Similarly, propranolol was randomly compared to placebo (NaCl, n = 14). We measured mean arterial pressure (MAP, mmHg), heart rate (HR) and portal pressure gradient (PPG: portal minus central venous pressure) prior to and after drugs. RESULTS: Propranolol reduced PPG (mean +/- s, mmHg) significantly (14.8 +/- 3.7 versus 12.1 +/- 3.7; -21% +/- 10%; P < 0.001), while nitrates alone (14.3 +/- 3.4 versus 13.7 +/- 3.4; -11% +/- 3%; P=0.06) or nitrates plus propranolol (12.9 +/- 4 versus 12.4 +/- 4; -7% +/- 8%; P=0.2) induced only minor additive effects on portal pressure. However, nitrate reduced MAP (P < 0.001) and increased HR (P < 0.01), whereas propranolol reduced only HR (P < 0.001) with unchanged MAP, and the combination decreased MAP (P < 0.001). Compared to placebo (no effect), propranolol decreased PPG (14.4 +/- 5.6 versus 11.1 +/- 5.5; -23% +/- 11%; P < 0.001) and HR (P < 0.001). Overall, most patients (92%) responded to propranolol and 54% showed a marked PPG decrease (>20%). CONCLUSIONS: Propranolol significantly reduced portal pressure in cirrhotic patients after TIPS, whereas nitrates induced only minor benefit. TIPS-treated patients might therefore profit from additive propranolol therapy allowing limited shunts to be applied initially and/or to reduce the need for TIPS revisions in the case of shunt-dysfunction during follow-up.

Transjugular intrahepatic portosystemic shunting is not superior to endoscopic variceal band ligation for prevention of variceal rebleeding in cirrhotic patients: a randomized, controlled trial.

BACKGROUND: The aim of the present study was to compare the transjugular intrahepatic portosystemic shunt (TIPS) with variceal band ligation (VBL) in the prophylaxis of variceal rebleeding in patients with cirrhosis of the liver. METHODS: Fifty-four cirrhotic patients (21 Child-Pugh class A, 27 class B, 6 class C) were randomized to TIPS (n = 28) or VBL (n = 26) within 2 months after control of esophageal variceal hemorrhage. Statistical analysis was performed on the intention-to-treat principle. RESULTS: Mean follow-up was 2 years. Mortality risk at 1 and 2 years of follow-up was 7.8% +/- 5.3% and 19.9% +/- 8.8% in the TIPS group and 16.5% +/- 7.6% and 16.5% +/- 7.6% in the VBL group, respectively (n.s.); actuarial probability of remaining free from rebleeding was 83.7% +/- 77.4% and 71.4% +/- 10.4% in the TIPS group and 83.9% +/- 7.3% and 78.1% +/- 8.8% in the VBL group at 1 and 2 years, respectively (n.s.). Hepatic encephalopathy within 1 month after randomization was observed in 2 patients in the TIPS group and in 1 in the VBL group. CONCLUSION: TIPS is not superior to VBL in the prevention of variceal rebleeding. Furthermore, similar mortality rates in patients treated with TIPS or VBL negate TIPS as the preferred strategy for prevention of variceal rebleeding.

Total plasma homocysteine and related amino acids in end-stage renal disease (ESRD) patients measured by gas chromatography-mass spectrometry--comparison with the Abbott IMx homocysteine assay and the HPLC method.

Increased concentrations of homocysteine probably contribute to the high cardiovascular morbidity and mortality in hemodialysed end-stage renal disease (ESRD) patients and are determined by a variety of factors such as age, residual renal function, and vitamin status. Fasting plasma concentrations of total homocysteine, methionine, cysteine, and cystathionine were determined by gas chromatography-mass spectrometry (GC-MS) in 131 ESRD patients receiving daily oral folate (160-320 microg) and vitamin B6 (10-20 mg) supplements. Concentrations of homocysteine determined by GC-MS were compared with those measured by high-performance liquid chromatography (HPLC) and an immunofluorescence method (IMx analyzer) using Passing-Bablok regression analysis. Mean plasma concentration of total homocysteine determined by GC-MS (28.7+/-11.9 micromol/l [mean+/-SD]) was significantly lower than that determined by HPLC (34.0+/-14.5 micromol/l; p<0.001) or IMx (32.4+/-13.9 micromol/l; p<0.001). A close correlation existed between GC-MS and HPLC (r=0.931; y=1.203 x+0.279) and GC-MS and IMx (r=0.896; y=1.105 x+0.766). Linear regression analysis showed positive correlations between plasma concentrations of homocysteine and cysteine (r=0.434; p<0.001) and homocysteine and cystathionine (r=0.187; p=0.032). Plasma concentrations of homocysteine correlated negatively with folate (r=-0.281; p=0.001) and vitamin B12 (r=-0.229; p=0.009). GC-MS proved to be a sensitive and reliable method for the determination of total plasma homocysteine and related amino acids. Despite vitamin supplementation, ESRD patients requiring chronic maintenance hemodialysis, have high plasma concentrations of homocyst(e)ine which seems to be metabolized mainly within the transsulfuration pathway, while remethylation to methionine seems to be disturbed.

The effects of prone positioning on intraabdominal pressure and cardiovascular and renal function in patients with acute lung injury.

UNLABELLED: To detect any harmful effects of prone positioning on intraabdominal pressure (IAP) and cardiovascular and renal function, we studied 16 mechanically ventilated patients with acute lung injury randomly in prone and supine positions, without minimizing the restriction of the abdomen. Effective renal blood flow index and glomerular filtration rate index were determined by the paraaminohippurate and inulin clearance techniques. Prone positioning resulted in an increase in IAP from 12 +/- 4 to 14 +/- 5 mm Hg (P < 0.05), PaO(2)/fraction of inspired oxygen from 220 +/- 91 to 267 +/- 82 mm Hg (P < 0.05), cardiac index from 4.1 +/- 1.1 to 4.4 +/- 0.7 L/min (P < 0.05), mean arterial pressure from 77 +/- 10 to 82 +/- 11 mm Hg (P < 0.01), and oxygen delivery index from 600 +/- 156 to 648 +/- 95 mL. min(-)(1). m(-)(2) (P < 0.05). Renal fraction of cardiac output decreased from 19.1% +/- 12.5% to 15.5% +/- 8.8% (P < 0.05), and renal vascular resistance index increased from 11762 +/- 6554 dynes. s. cm(-)(5). m(2) to 15078 +/- 10594 dynes. s. cm(-)(5). m(2) (P < 0.05), whereas effective renal blood flow index, glomerular filtration rate index, filtration fraction, urine volume, fractional sodium excretion, and osmolar and free water clearances remained constant during prone positioning. Prone positioning, when used in patients with acute lung injury, although it is associated with a small increase in IAP, contributes to improved arterial oxygenation and systemic blood flow without affecting renal perfusion and function. Apparently, special support to allow free chest and abdominal movement seems unnecessary when mechanically ventilated, hemodynamically stable patients without abdominal hypertension are proned to improve gas exchange. IMPLICATIONS: Prone positioning is increasingly used to improve gas exchange in patients with acute lung injury. However, during prone positioning an increase in intraabdominal pressure in these critically ill patients may promote dysfunction of other organs. Therefore, we performed a randomized study in mechanically ventilated patients with acute lung injury to investigate the cardiovascular and renal effects of prone positioning.

Infarction and laceration of liver parenchyma caused by wedged CO(2) venography before tips insertion.

We describe the fatal outcome of an elective TIPS procedure performed in a 43-year-old man with alcoholic cirrhosis. Wedged hepatic venography with CO(2) was the reason for infarction and laceration of liver parenchyma resulting in a subcapsular hematoma and subsequent intra-abdominal bleeding. This is the first report of this complication after the use of CO(2) in a cirrhotic patient.

Comparison of portal vein velocity and the hepatic venous pressure gradient in assessing the acute portal hemodynamic response to propranolol in patients with cirrhosis.

OBJECTIVE: The aim of this prospective study was to compare noninvasive Doppler sonography and invasive measurement of the hepatic venous pressure gradient (HVPG) to determine the acute portal hemodynamic response to propranolol in patients with liver cirrhosis. METHODS: In a blinded study design, portal vein velocity (PVV) and HVPG were simultaneously assessed in 11 cirrhotic patients for 4 h after oral ingestion of 40 mg propranolol. RESULTS: Both HVPG (17.2% +/- 4.3%, p < 0.0001) and PVV (15.6% +/- 2.1%, p < 0.0002) showed a highly significant reduction during the study period versus baseline. Based on HVPG measurements, four patients (36%) were classified as nonresponders. These patients had a significantly lower PVV reduction compared to the responders (responders: 18.8% +/- 2.0% vs nonresponders: 10.0% +/- 2.1%, p < 0.05). Nonresponders were identified by Doppler sonography with a sensitivity of 1.0, specificity of 0.86, and positive predictive value of 0.9 when a threshold of 20% PVV reduction 120 min after drug intake was applied. CONCLUSIONS: Doppler sonography is a useful tool for assessment of the acute portal hemodynamic effect of propranolol. To distinguish portal hemodynamic nonresponders from responders to propranolol, PVV measurements should be carried out 2 h after drug administration, and PVV reduction should be not <20% in propranolol responders.

Is the increase in serum cystathionine levels in patients with liver cirrhosis a consequence of impaired homocysteine transsulfuration at the level of gamma-cystathionase?

BACKGROUND: It has been suggested that the major metabolic block in the methionine catabolic pathway in cirrhotics exists at the level of the enzyme S-adenosylmethionine synthetase because in previous studies using conventional amino-acid analyzers, no intermediates of transmethylation/transsulfuration were found to accumulate in plasma downstream of S-adenosylmethionine synthesis. We therefore measured serum concentration intermediates of methionine transmethylation/transsulfuration using an improved gas chromatography/mass spectrometry technique. METHODS: Serum concentrations of methionine, homocysteine, cystathionine, N,N-dimethylglycine, N-methylglycine, methylmalonic acid, 2-methylcitric acid and alpha-aminobutyric acid were determined by gas chromatography/mass spectrometry in 108 consecutive patients with liver cirrhosis at Child stages A (mild cirrhosis, n = 27) and B/C (severe cirrhosis, n = 81), 18 outpatients with non-cirrhotic liver disease, and 55 healthy individuals. RESULTS: Serum levels of methionine, N,N-dimethylglycine, N-methylglycine, cystathionine, and homocysteine were significantly higher in patients at Child stages B/C compared with those of healthy controls (P < 0.01), and they were also significantly higher than in patients with non-cirrhotic liver disease (P < 0.01 and P < 0.05 for homocysteine, respectively). They also correlated with the Child-Pugh score (P < 0.01). Homocysteine, cystathionine, N,N-dimethylglycine, N-methylglycine, methylmalonic acid, and 2-methylcitric acid correlated with serum creatinine. The mean cystathionine concentration was significantly higher in patients with creatinine > or = 1.4 mg/dl than in patients with normal creatinine values (P < 0.01). However, the differences between cirrhotics and healthy controls were still significant after correcting for creatinine. CONCLUSIONS: Our data provides indirect evidence for two hitherto unrecognized alterations of methionine metabolism in cirrhotics, i.e. impairment of the transsulfuration of homocysteine at the level of cystathionine degradation and a shift in remethylation of homocysteine towards the betaine-homocysteine-methyltransferase reaction.

Liver perfusion scintigraphy prior to and after transjugular intrahepatic portosystemic shunts (TIPS) in patients with portal hypertension.

PURPOSE: This investigation was performed to compare the hemodynamic results of the transjugular intrahepatic portosystemic shunt, a new interventional treatment for portal hypertension, with those observed after the established surgical shunt interventions. METHODS: We examined 22 patients with portal hypertension due to liver cirrhosis before and after elective TIPS by liver perfusion scintigraphy. The relative portal perfusion was determined before and after the shunt procedure. Additionally, we measured the portal pressure gradient (PPG: portal-central venous pressure, mmHg). RESULTS: Prior to TIPS, the relative portal perfusion was significantly reduced to 22 +/- 9.1%. After the intervention we calculated values of 23.1 +/- 10.7% in the TIPS-group (p = 0.67; not significant). In spite of unchanged portal perfusion, the portal pressure was significantly (p < 0.001) reduced from 25.6 +/- 5.3 to 14.8 +/- 4 mm Hg. CONCLUSION: These results suggest that the reduction of portal hypertension by TIPS is effective. The portal perfusion is maintained by TIPS suggesting that liver perfusion is preserved to a higher degree.

Long term outcome after transjugular intrahepatic portosystemic stent-shunt in non-transplant cirrhotics with hepatorenal syndrome: a phase II study.

BACKGROUND: Recent small studies on hepatorenal syndrome (HRS) indicate some clinical benefit after transjugular intrahepatic portosystemic stent-shunt (TIPS) but sufficient long term data are lacking. AIM: We studied prospectively feasibility, safety, and long term survival after TIPS in 41 non-transplantable cirrhotics with HRS (phase II study). PATIENTS AND METHODS: HRS was diagnosed using current criteria (severe (type I) HRS, n=21; moderate (type II) HRS, n=20). Thirty one patients (14 type I, 17 type II) received TIPS (8-10 mm) while advanced liver failure excluded shunting in 10. During follow up (median 24 months) we analysed renal function and survival (Kaplan-Meier). RESULTS: TIPS markedly reduced the portal pressure gradient (21 (5) to 13 (4) mm Hg (mean (SD)); p<0.001) with one procedure related death (3.2%). Renal function deteriorated without TIPS but improved (p<0.001) within two weeks after TIPS (creatinine clearance 18 (15) to 48 (42) ml/min; sodium excretion 9 (16) to 77 (78) mmol/24 hours) and stabilised thereafter. Following TIPS, three, six, 12, and 18 month survival rates were 81%, 71%, 48%, and 35%, respectively. As only 10% of non-shunted patients survived three months, total survival rates were 63%, 56%, 39%, and 29%, respectively. Multivariate Cox regression analysis revealed bilirubin (p<0.001) and HRS type (p<0.05) as independent survival predictors after TIPS. CONCLUSIONS: TIPS provides long term renal function and probably survival benefits in the majority of non-transplantable cirrhotics with HRS. These data warrant controlled trials evaluating TIPS in the management of HRS.

Nitrite and nitrate levels in patients with cirrhosis of the liver: influence of kidney function and fasting state.

BACKGROUND: Increased serum nitrite/nitrate (NOx) levels, the stable metabolites of nitric oxide (NO), have been reported in patients with cirrhosis. NOx levels, however, are influenced not only by endogenous NO synthesis but also by urinary NOx excretion and dietary intake. We attempted to elucidate factors that influence NOx levels independently of endogenous NO production and to determine the conditions under which NOx levels reflect endogenous production in patients with cirrhosis and healthy controls. METHODS: NOx serum concentrations and urinary NOx excretion were determined by means of the Griess reaction in relation to Child-Pugh score, kidney function, fasting state, and after exposure to tap water. RESULTS: Multifactor regression analysis showed inulin clearance (P = 0.0074) and Child-Pugh score (P = 0.0001) to be independent factors predicting NOx levels in patients with cirrhosis. NOx serum levels correlated negatively with the inulin clearance (P < 0.0001), which deteriorated with progressive loss of liver function. NOx levels decreased by about 30% within a 24-h fasting period. After 24 h fasting urinary NOx excretion was not significantly increased in patients with advanced cirrhosis. CONCLUSION: NOx serum levels, taken as a surrogate for endogenous NO formation, have to be viewed with caution in patients with cirrhosis because they often have impaired kidney function. However, in steady-state conditions after an adequate fasting period NOx levels might be good prognostic markers in patients with cirrhosis since they reflect two possible sequelae of liver insufficiency-namely, increased NO formation and impaired kidney function.

Effects of different octreotide dosages on splanchnic hemodynamics and glucagon in healthy volunteers.

AIMS: This study evaluated the dependence of portal and mesenteric blood flow and plasma glucagon levels on octreotide dosage and its mode of application. METHODS: Two groups of 10 individuals each received octreotide either subcutaneously (placebo, 100 and 200 microgram) or intravenously (100- microgram bolus i.v., 25 and 100 microgram/h) in a double-blind, random order. Using Doppler ultrasound, we examined portal and mesenteric blood flow and measured plasma glucagon levels at regular intervals within a 4-hour period under fasting conditions. RESULTS: Contrary to placebo, octreotide caused a decrease in portal blood flow (PVF) and in superior mesenteric artery blood flow (SMAF) together with an increase in the mesenteric pulsatility index (PI). The same total dose of 100 microgram octreotide caused a similar PVF response, averaged over 4 h, given either subcutaneously (-28.0 +/- 4.8%), intravenously (-29.4 +/- 4.3%) or as a continuous infusion (-29.3 +/- 4.6%). As concerns intravenous infusions, 100 microgram/h was more effective than 25 microgram/h (-37.8 +/- 6.2 vs. -29.3 +/- 4.6%). The PVF reduction remained constant during intravenous infusion, whereas glucagon levels decreased progressively over the entire observation time. CONCLUSIONS: The decrease in PVF is dependent on the octreotide dose. However, this is not constantly paralleled by a decrease in plasma glucagon concentration.

[Diagnosis and therapy of Budd-Chiari syndrome]. / Diagnostik und Therapie des Budd-Chiari-Syndroms.

PURPOSE: Budd-Chiari syndrome is a fairly uncommon disease in Europe. This often leads to its late diagnosis. The syndrome is characterised by portal hypertension and splanchnic congestion due to obstruction of hepatic venous outflow. This paper describes the treatment of three patients with Budd-Chiari syndrome by interventional therapeutic techniques and discusses alternative treatment modalities. PATIENTS AND METHODS: The first patient presented with veno-occlusive disease and was treated by the placement of a transjugular intrahepatic portosystemic stent-shunt. The second patient showed an occlusion of the major hepatic veins. After percutaneous recanalisation, a stent was placed in the right hepatic vein which remained patient. The third patient had a membranous obstruction of the right hepatic vein which was treated by percutaneous balloon dilatation. RESULTS: In all patients the clinical symptoms resolved completely after treatment and no complications were encountered. CONCLUSIONS: The authors conclude that interventional therapeutic techniques offer a wide variety of possibilities for the treatment of patients with Budd-Chiari syndrome and are safe, effective and relatively inexpensive. However, further studies are required to assess the long-term results and survival rates of these patients.

Endoscopic manometry of esophageal varices: evaluation of a balloon technique compared with direct portal pressure measurement.

BACKGROUND/AIMS: Recently, a non-invasive endoscopic balloon technique for esophageal manometry was published. In the present study, we assess its methodological aspects together with the relationship to portal pressure. METHODS: In 20 patients with liver cirrhosis who had received an intrahepatic portosystemic stent-shunt (TIPS), we evaluated portal and variceal pressure before and after balloon occlusion of TIPS (random order). Portal pressure was measured continuously via a portal venous catheter, and variceal pressure was determined at the same time independently by two endoscopists using two balloon techniques (inflation until varix collapses; deflation until varix reappears). RESULTS: Overall, mean (+/-SD) portal pressure (28.5+/-7 mmHg) was significantly higher (p<0.001) than mean variceal pressure (24.4+/-6 mmHg). Balloon manometry-determined variceal pressure values were 10+/-15% higher with the inflation technique (26.2+/-7 mmHg) than with the balloon deflation technique (22.6+/-6 mmHg, p<0.001). Portal pressure and variceal pressure correlated significantly (p<0.001; balloon inflation: r=0.61, balloon deflation: r=0.66, mean values of inflation and deflation: r=0.68). Short-term TIPS occlusion led to mean increases of 52% and 35% in portal pressure and variceal pressure, respectively. The manometry results of both endoscopists correlated well with either balloon technique (r> or =0.93; p<0.001) and we saw no adverse effects. CONCLUSIONS: Variceal balloon manometry provides non-invasive variceal pressure data which correlate to portal pressure assessed prior to and after short-term TIPS occlusion. However, probably due to variance in collateral anatomy, variceal pressure does not exactly predict portal pressure and its acute changes in the individual patient. The averaged variceal pressure of the inflation and deflation balloon technique provides the best relation to portal pressure combined with a good interobserver reliability and warrants further clinical evaluation.

[TIPSS: technical and clinical results after 4 years]. / TIPSS: Technische und klinische Ergebnisse nach 4 Jahren.

PURPOSE: Retrospective analysis of the technical and clinical results after transjugular portosystemic stent shunt (TIPSS) procedure. METHOD: Between 1992 and 1996 we tried to establish a TIPSS in 90 patients. The indications were: recurrent variceal haemorrhage (n = 74), refractory ascites (n = 12), hepatorenal syndrome (HRS) (n = 4). Due to advanced liver cirrhosis 16 patients suffered of severe renal dysfunction (HRS). 57 patients had ascites. RESULTS: TIPSS implantation was technically successful in 96.7% (1992-1994: 5.1%, 1995-1996: 100%) of the patients. Complications occurred in 14.9% (1992-1994: 25.6%, 1995-1996: 6.3%). TIPSS-associated mortality was 2.3% (1992-1994: 5.1%, 1995-1996: 0%). 76.1% of the patients required reinterventions. 85.3% of reinterventions were necessary in the first year after TIPSS placement. The survival rate without reintervention was 28%, 21% and 9% for Child A, B, and C patients, respectively. Recurrent variceal haemorrhage occurred in 12.7%. De novo hepatic encephalopathy developed in 13.8%. Ascites improved in 79.2% and renal function in 75% of the patients. CONCLUSION: TIPSS is an effective method to treat recurrent variceal haemorrhage, refractory ascites and HRS. Complication and mortality rate depend on the investigator's experience and on the technique used.

Acute Budd-Chiari syndrome: treatment with transjugular intrahepatic portosystemic shunt.

The case of a 28-year-old man with acute Budd-Chiari syndrome due to veno-occlusive disease is reported. Transjugular intrahepatic portosystemic shunt (TIPS) was performed after upper gastrointestinal endoscopy, duplex sonographic and abdominal computed tomographic examination, inferior cavogram with hepatic venous catheterization, and transvenous biopsy. A 10-mm parenchymal tract was created. The patient did well after the procedure; ascites resolved and liver function improved markedly. The shunt has remained patent up to now for 6 months.

[Carbon dioxide (CO2) as contrast medium for the new installation and follow-up of a TIPS]. / Kohlendioxid (CO2) als Kontrastmittel zur Neuanlage und Kontrolle von TIPS.

PURPOSE: To determine the efficacy of CO2 as negative contrast medium compared with iodinated contrast medium in creation and control of TIPS. MATERIAL AND METHODS: CO2 was used during TIPS procedures in 33 patients. In 21 patients a wedged hepatic venography was obtained for planning the shunt tract. Additional TIPS-control DSA was performed in 42 cases by direct portal venography to verify the TIPS function and patency. In all cases CO2 gas was used in addition to iodinated contrast medium. RESULTS: CO2 produced excellent wedged hepatic venographies in all patients. Visualisation of the portal veins and collaterals was superior to iodinated contrast medium. The TIPS-control DSA performed with CO2 were comparable to those performed with iodinated contrast medium. Complications were not observed in our study. CONCLUSION: CO2 is an effective contrast medium for TIPS procedures. In particular the visualisation of portal veins performed by CO2-wedged hepatic venography is superior to iodinated contrast medium.