Over the last three years, 53 patients underwent transjugular portosystemic shunting (TIPS). 49 patients were treated successfully (92.5%). Procedure-related morbidity (intention to treat) was seen in 11 patients (20.8%): encephalopathy (n = 5), sepsis (n = 3), right heart failure (n = 2) and progressive liver failure (n = 1). 30-day mortality rate was 13.2% (7/53); five of these patients were in stage Child-Pugh C, one patient in stage B, and one patient had a known coronary heart disease. 30-day rebleeding rate was 6.1% (3/49), but all these patients could be retreated successfully by radiological methods (PTA, embolisation, thrombolysis). Angiographic follow-up (mean six months) of 35 patients detected 30 (85.7%) haemodynamic relevant obstructions (stenosis of stent: n = 4, stenosis of hepatic vein: n = 15, stenosis of stent and hepatic vein: n = 5, occlusion of TIPS-shunt: n = 6). Secondary patency rate following percutaneous reintervention was 91.3%. All rebleedings in the follow-up (n = 7) were treated successfully by TIPS-revision. Five out of 12 patients (41.7%) with refractory ascites were treated successfully by TIPS (complete resolution of ascites after three months: n = 4, significant reduction of ascites: n = 1). We conclude that transjugular portosystemic shunt is an effective way of treating portal hypertension, but there is a need to develop methods to prevent the high incidence of shunt stenosis.