How Much Reduction in Portal Pressure Is Necessary To Prevent Variceal Rebleeding? A Longitudinal Study in 225 Patients With Transjugular Intrahepatic Portosystemic Shunts

• Published in: The American journal of gastroenterology Vol. 96; no. 12; pp. 3379 - 3383
• Main Authors: Rössle, Martin, Siegerstetter, Volker, Olschewski, Manfred, Ochs, Andreas, Berger, Elisabeth, Haag, Klaus
• Format: Journal Article
• Language: English
• Published: New York . 01.12.2001; Wolters Kluwer Health Medical Research, Lippincott Williams & Wilkins
• Subjects: Gastroenterology
• ISSN: 0002-9270; 1572-0241
• DOI: 10.1111/j.1572-0241.2001.05340.x

This longitudinal study determines the risk of rebleeding in relation to the reduction of the portosystemic pressure gradient in patients with a transjugular intrahepatic portosystemic shunt (TIPS) for variceal bleeding. The study included 225 patients in whom a TIPS revision was indicated by the endoscopic finding of varices with a high risk for rebleeding (n = 167) or a recent variceal rebleed (n = 58). The portosystemic pressure gradient was determined before and after TIPS placement and at revision performed after a mean of 10 ± 15 months. The portosystemic pressure gradient at revision approached the index pressure gradient before TIPS implantation (23.1 ± 5.5 mm Hg) by 8.4 ± 31%. Rebleeding was inversely correlated with the reduction in index pressure gradient found at revision. Thus, 80% of rebleedings occurred with pressure gradients close to the index pressure gradient (<25% reduction) or with gradients equal to or greater than the index pressure gradient. In contrast, only one patient (0.4%) and three patients (1.3%) rebled with a pressure gradient of <12 mm Hg or a reduction of the index pressure gradient by >50%, respectively. Kaplan-Meier analysis of rebleeding, which included the 225 patients at risk, showed a probability of rebleeding of 18%, 7%, and 1% for a reduction of the index pressure gradient by 0%, 25–50%, and >50%, respectively. Most rebleedings occurred with pressure gradients similar to the index-pressure gradient measured at first bleeding. Accordingly, a graded reduction by 25–50% sufficiently prevents rebleeding. It can be assumed that, in comparison with the widely used threshold value of 12 mm Hg, a reduction by 25–50% may have a favorable benefit-to-risk ratio with respect to shunt-induced hepatic encephalopathy and liver failure. It should therefore be a goal in the decompressive treatment of portal hypertension and maintained during follow-up of patients with variceal bleeding.