Secondary prophylaxis of oesophageal variceal haemorrhage: a role for transjugular intrahepatic portosystemic shunt? Phillip Harrison a and John Karani b • Cirrhotic patients who survive an episode of acute variceal haemorrhage are at high risk of recurrent bleeding. • Transjugular intrahepatic portosystemic shunt (TIPS) is more effective than medical (drug and endoscopic) therapies in preventing rebleeding in patients with cirrhosis. • Randomized studies comparing TIPS to medical therapies have shown that TIPS has no effect on survival when used as secondary prophylaxis in patients with cirrhosis following index variceal haemorrhage. • Medical therapies cause less encephalopathy and more frequent improvement in Child–Pugh class at a lower cost than TIPS. • Transjugular intrahepatic portosystemic shunt should not be used as first-line therapy to prevent variceal rebleeding in patients with cirrhosis, but should be limited to rescue patients with uncontrolled or recurrent variceal bleeding after failed medical therapy. Cirrhotic patients who survive an episode of acute variceal haemorrhage are at high risk of recurrent bleeding. Many treatments have been found to be effective at preventing rebleeding, including drug therapy, endoscopic treatments and transjugular intrahepatic portosystemic stunt (TIPS). In this issue of European Journal of Gastroenterology & Hepatology, Jalan et al . compared three historical cohorts of patients with cirrhosis after index variceal bleed and found a lower rebleeding rate in patients receiving TIPS (16.2%) compared to either band ligation (39.3%) or endoscopic variceal sclerotherapy (74.6%). Despite the efficacy of TIPS in preventing variceal rebleeding, there was no significant difference in survival between the three cohorts. However, sub-group analysis found a lower mortality in patients with Child–Pugh class C cirrhosis receiving TIPS. Unfortunately, this promising observation in patients with advanced cirrhosis is not supported by the results of meta-analysis of randomized studies. Therefore, we do not recommend TIPS as first-line therapy to prevent variceal rebleeding in patients with cirrhosis. Eur J Gastroenterol Hepatol 14:591–593 & 2002 Lippincott Williams & Wilkins European Journal of Gastroenterology & Hepatology 2002, 14:591–593 Keywords: sclerotherapy, band ligation, pharmacotherapy, portal hypertension, meta-analysis a The Institute of Liver Studies and b Department of Radiology, King’s College Hospital, Denmark Hill, London SE5 9PJ, UK Correspondence to Dr Phillip Harrison PhD MD FRCP, Senior Lecturer, Institute of Liver Studies, King’s College Hospital, Denmark Hill, London, SE5 9PJ, UK Tel: +44 20 7346 3169; fax: +44 20 7346 3167; e-mail: phillip.harrison@kcl.ac.uk Received 5 April 2002 Cirrhotic patients who survive an episode of acute variceal haemorrhage are at high risk of recurrent bleeding. Many treatments have been found to be effective in preventing rebleeding, including drug ther- apy, endoscopic treatments and portosystemic shunts. Most of these treatments, except endoscopic therapy, act by decreasing portal pressure. Since its introduction over 10 years ago, transjugular intrahepatic portosyste- mic shunt (TIPS) has become very popular in the management of portal hypertension. TIPS is relatively easy to perform and effectively reduces the portal pressure gradient below the threshold associated with variceal haemorrhage. In this issue of European Journal of Gastroenterology & Hepatology , Jalan and colleagues from Edinburgh, UK present their experience with the secondary prophylaxis of oesophageal variceal haemor- rhage [1]. They retrospectively compared three cohorts of patients with cirrhosis who, following their index variceal bleed, were treated with endoscopic variceal sclerotherapy (1986–1991), band ligation (1991–1995) or TIPS (1993–1996). Owing to local management policy, these cohorts were relatively discretely sepa- rated over time, although there was a 2-year overlap period in which patients received either band ligation or T IPS. Jalan et al. observed a lower rebleeding rate in patients receiving TIPS (16.2%) compared to either of the endoscopic therapies, and confirmed the efficacy of band ligation (39.3% rebleeding rate) compared to endoscopic variceal sclerotherapy (74.6% rebleeding rate). This finding is in agreement with that of a meta- Leading article 591 0954–691X & 2002 Lippincott Williams & Wilkins