Mortality and complications in patients with portal hypertension who underwent transjugular intrahepatic portosystemic shunt (TIPS) - 12 years experience.

CONTEXT Transjugular intrahepatic portosystemic shunt (TIPS) is the non-surgical treatment option with low level of morbi-mortality and possibility of accomplishment in patients with severe hepatic dysfunction which aims at decompressing the portal system treating or reducing the portal hypertension complications. OBJECTIVE Outline the profile analyze global and early mortality, and the complications presented by cirrhotic patients who underwent TIPS for treatment of digestive hemorrhage by portal hypertension. METHOD Retrospective study based on the data bank of cirrhotic patients' medical reports, who underwent TIPS for digestive hemorrhage by portal hypertension treatment who did not respond to clinical endoscopic treatment, and were assisted from 1998 to 2010 in the Liver Transplant Service at a university hospital. The study was approved by the Committee of Ethics and Research. RESULTS The sample was comprised of 72 (84.7%) patients, being 57 (79.2%) males, average age 47.7 years (age range from 16 to 85 years and SD = 13), 21 (29.2%) patients presented liver disease as cause excessive intake of alcoholic drinks; 21 (29.2%) contamination by hepatitis virus, 16 (22.2%) excessive alcohol intake associated with virus and 14 (19.4%) patients presented other causes. As for initial classification, 14 (20%) had Child-Pugh A, 33 (47.1%) Child-Pugh B and 23 (32.9%) Child-Pugh C. Initial MELD was obtained in 68 patients being 37 (54.4%) higher than 15 points while 31 (45.6%) had up to 15 points. Early death occurred in 19 (26.4%). Global mortality occurred in 41 (60.3%). CONCLUSIONS Mortality is directly related to clinical factors of patients, being Child-Pugh and MELD classifications predictors of mortality, with more impact in patients with Child-Pugh class C and MELD > 15. The complications found were similar to those described in the literature, although the dysfunction by stent stenosis (26.4%) was lower than in the most of the studies and the encephalopathy incidence (58.3%) was higher. Probably, the high incidence of encephalopathy is explained by the low incidence of stenosis.