The increasing use of TIPS in the treatment of complications of portal hypertension has to be reassessed, taking into account the substantial collective experience gained in many centers around the world. Specific indications need to be demonstrated using well-designed clinical trials, as it is now known that (contrary to the initial expectations), the true rate of post-TIPS encephalopathy is significant and likely to be similar to that observed shortly after shunt surgery. In many patients, the decreased incidence of post-TIPS HE observed during follow-up is related to progressive stenosis of the shunt. This is associated with the recurrence of portal hypertension and, often, with the onset of clinical complications such as ascites or variceal bleeding. Prospective studies are required to define the ideal compromise between portal decompression and maintenance of a certain amount of portal perfusion to the liver and, thus, some portal hypertension, but at a threshold level below that associated with an increased risk of clinical complications. TIPS must be regarded as only one of several therapeutic modalities available to treat portal hypertension. When TIPS induces disabling encephalopathy or accelerated liver failure, closure of the shunt should be performed early, and alternative treatments can be used, including endoscopic sclerotherapy, variceal ligation, large volume paracentesis, and, most importantly, liver transplantation. Hopefully, lessons from the past gleaned from the evaluation of surgical portacaval shunts in the treatment of portal hypertension will help to better define the appropriate use of TIPS in cirrhotic patients to maximize the beneficial effects of reducing the portal pressure while minimizing the deleterious effects of this procedure.