Early use of TIPS in patients with cirrhosis and variceal bleeding

To the Editor: We disagree with several points in the commentary of our study on early use of transjugular intrahepatic portosystemic shunts (TIPS) in patients with cirrhosis and variceal bleeding (1), which was abstracted in ACP Journal Club (2). TIPS should be offered as a first treatment option only to a small proportion of patients with acute variceal hemorrhage—that is, those at high risk for treatment failure. Our study clearly shows that in this high-risk cirrhotic population, mortality is reduced with early use of TIPS. Therefore, to exclude 82% of patients is not a limitation of the study, as suggested by Dr. Korula. The use of vasoactive drugs followed by endoscopic therapy, nonselective β-blockers, and nitrates also cannot be considered a limitation of the study: This is recommended by current guidelines (3).


Early use of TIPS in patients with cirrhosis and variceal bleeding
To the Editor: We disagree with several points in the commentary of our study on early use of transjugular intrahepatic portosystemic shunts (TIPS) in patients with cirrhosis and variceal bleeding (1), which was abstracted in ACP Journal Club (2). TIPS should be offered as a first treatment option only to a small proportion of patients with acute variceal hemorrhage-that is, those at high risk for treatment failure. Our study clearly shows that in this high-risk cirrhotic population, mortality is reduced with early use of TIPS. Therefore, to exclude 82% of patients is not a limitation of the study, as suggested by Dr. Korula. The use of vasoactive drugs followed by endoscopic therapy, nonselective β-blockers, and nitrates also cannot be considered a limitation of the study: This is recommended by current guidelines (3).
Dr. Korula is also concerned that outcomes in the drug therapy plus endoscopic band ligation (DT-EBL) group were worse than the 25% rebleeding and 14% mortality rates reported in a metaanalysis (4) and suggests that this may reflect how patients in the DT-EBL group were managed. We disagree. Our study was in a selected high-risk, poor-prognosis population, unlike the studies included in the meta-analysis. Our study indicates that in highrisk patients, the potential deleterious effects of early TIPS seem to be counterbalanced both by its high efficacy in controlling bleeding and by preventing further deterioration in liver function during the interval from admission to rescue TIPS in patients with failure of standard treatment.

Juan Carlos García-Pagán, MD Jaime Bosch, MD Hepatic Hemodynamic Laboratory, Liver Unit, Hospital Clinic University of Barcelona Barcelona, Spain
In response: In my commentary, I emphasized that 82% of patients with acute variceal bleeding had been excluded to ensure that readers would be clear about the selectiveness of this cohort of patients with advanced cirrhosis. The randomization of the remaining 63 patients reflects a low accrual rate per center (approximately 3 patients per center per year), potentially affecting study quality.
The authors indicate that the current gold standard for prevention of variceal rebleeding is a combination of nonselective βblockers and endoscopic therapy. The meta-analysis by Gonzalez and colleagues (4) supports this, and was referenced in my commentary. However, 39% of patients also received nitrates. There is no consensus that the addition of nitrates to DT-EBL is beneficial. In another study by García-Pagán and colleagues (5), combining nadolol and isosorbide mononitrate with endoscopic ligation had no benefit, since adverse events and hospital readmissions were higher in the DT-EBL group. Similarly, Kumar and colleagues (6) confirmed this lack of benefit. Use of these drugs and dosages in a high-risk cohort is not clearly justified. Portal pressures could also have been measured to identify nonresponders and to withdraw medications that may have contributed to morbidity and death.
Endoscopic variceal ligation is used exclusively in North America. In the study by García-Pagán and colleagues, 25% of patients received sclerotherapy. This is known to be associated with significantly more complications (7), which are expected in advanced cirrhosis, and may have affected results in the DT-EBL group.