Potential conflict of interest: Nothing to report.
Early transjugular intrahepatic portosystemic shunt in patients with cirrhosis and variceal bleeding†
Article first published online: 29 OCT 2010
Copyright © 2010 American Association for the Study of Liver Diseases
Volume 52, Issue 5, pages 1847–1850, November 2010
How to Cite
Rössle, M., Grandt, D. (2010), Early transjugular intrahepatic portosystemic shunt in patients with cirrhosis and variceal bleeding. Hepatology, 52: 1847–1850. doi: 10.1002/hep.23982
- Issue published online: 29 OCT 2010
- Article first published online: 29 OCT 2010
García-Pagán JC, Caca K, Bureau C, Laleman W, Appenrodt B, Luca A, et al. Early use of TIPS in patients with cirrhosis and variceal bleeding. N Engl J Med 2010;362:2370-2379. (Reprinted with permission.)
Background: Patients with cirrhosis in Child–Pugh class C or those in class B who have persistent bleeding at endoscopy are at high risk for treatment failure and a poor prognosis, even if they have undergone rescue treatment with a transjugular intrahepatic portosystemic shunt (TIPS). This study evaluated the earlier use of TIPS in such patients. Methods: We randomly assigned, within 24 hours after admission, a total of 63 patients with cirrhosis and acute variceal bleeding who had been treated with vasoactive drugs plus endoscopic therapy to treatment with a polytetrafluoroethylene-covered stent within 72 hours after randomization (early-TIPS group, 32 patients) or continuation of vasoactive-drug therapy, followed after 3 to 5 days by treatment with propranolol or nadolol and long-term endoscopic band ligation (EBL), with insertion of a TIPS if needed as rescue therapy (pharmacotherapy–EBL group, 31 patients). Results: During a median follow-up of 16 months, rebleeding or failure to control bleeding occurred in 14 patients in the pharmacotherapy–EBL group as compared with 1 patient in the early-TIPS group (P=0.001). The 1-year actuarial probability of remaining free of this composite end point was 50% in the pharmacotherapy–EBL group versus 97% in the early-TIPS group (P<0.001). Sixteen patients died (12 in the pharmacotherapy–EBL group and 4 in the early-TIPS group, P=0.01). The 1-year actuarial survival was 61% in the pharmacotherapy–EBL group versus 86% in the early-TIPS group (P<0.001). Seven patients in the pharmacotherapy–EBL group received TIPS as rescue therapy, but four died. The number of days in the intensive care unit and the percentage of time in the hospital during follow-up were significantly higher in the pharmacotherapy–EBL group than in the early-TIPS group. No significant differences were observed between the two treatment groups with respect to serious adverse events. Conclusions:In these patients with cirrhosis who were hospitalized for acute variceal bleeding and at high risk for treatment failure, the early use of TIPS was associated with significant reductions in treatment failure and in mortality. (Current Controlled Trials number, ISRCTN58150114.)
This study by García-Pagán et al.1 is the first randomized study comparing the use of early transjugular intrahepatic portosystemic shunt (TIPS) treatment with the current standard treatment in patients with liver cirrhosis and acute esophageal variceal bleeding. Only patients with an advanced risk of bleeding-related mortality (Child-Pugh class C and B patients with active bleeding on endoscopy)2, 3 were included. The study showed that the early use of TIPS (within 3 days of admission) reduced the 6-week mortality rate to 3% (33% with medical treatment) and the 1-year mortality rate to 14% (39% with medical treatment). When TIPS was used as a rescue treatment after the failure of medical treatment, the mortality rate was high (four of seven patients in the study by García-Pagán et al.), and this was comparable to previous results.4 Other (expected) beneficial effects of early TIPS placement included reduced rates of ascites, hepatorenal syndrome, and spontaneous bacterial peritonitis and significantly fewer days in the intensive care unit and in the hospital (P < 0.014).
This study might influence the current treatment strategy for variceal bleeding in patients with cirrhosis and lead to the stratification of these patients into groups with a high or low risk of bleeding-related mortality. As outlined in Fig. 1, patients with a high rate of bleeding-related mortality [Child-Pugh class C patients (score < 13) and Child-Pugh class B patients with active bleeding on endoscopy] may receive early TIPS treatment. They may then be followed with duplex sonography to confirm shunt patency. In contrast, as stated by the researchers, early TIPS should not be used for Child-Pugh class A patients because they have low rates of medical treatment failure and mortality. Such patients may be treated according to current recommendations with a step-up strategy using β-blocking agents, endoscopic band ligation, and rescue TIPS.5
The results of this study may influence not only the treatment of acute bleeding but also primary and secondary prophylaxis. The availability of a highly effective treatment with a very low rate of bleeding-related mortality (3%) even in high-risk patients might call into question the need for primary prophylaxis for variceal bleeding. Thus, the need for (and adverse effects of) regular endoscopic procedures and years of drug therapy could be avoided, and this would probably improve patients' quality of life. In this context, the knowledge that primary prophylaxis delayed neither the occurrence of varices nor the first occurrence of variceal bleeding is important.6 Furthermore, in patients who receive early TIPS for their first variceal bleeding, the role of secondary prophylaxis in the prevention of rebleeding will be limited. In these patients, drugs and endoscopic treatments might be primarily applied as temporary measures to stop bleeding until TIPS implantation is performed.
According to this study, early TIPS placement might be beneficial only in a minority of patients with variceal bleeding. Thus, only 63 of 359 patients (17.5%) with acute variceal bleeding were randomly allocated to the treatment groups: 18 refused to participate; 112 had Child-Pugh class A or B cirrhosis without active bleeding on endoscopy; and 166 were excluded for various reasons, such as isolated gastric variceal bleeding, Child-Pugh scores greater than 13 points, previous failure to respond to treatment with drugs and endoscopic band ligation, age greater than 75 years, portal vein thrombosis, hepatocellular carcinoma, and renal failure. However, in everyday practice, many of the patients excluded from this randomized study might be considered good candidates for early TIPS treatment. In particular, patients with gastric variceal bleeding, patients with renal failure, and patients who have failed to respond to previous medical treatment might benefit from the early use of TIPS. Patients older than 75 years might also be regarded as good candidates for early TIPS placement because they have poor tolerance for rebleeding. In addition, the general exclusion of patients with hepatocellular carcinoma from early TIPS treatment might not be justified. TIPS could have a place as a palliative treatment in patients with an adequate prognosis and an increased risk of rebleeding.
The largest group excluded from the study was the group of patients with Child-Pugh class A or B disease without active bleeding on endoscopy (31%). Because of the 97% survival rate at 6 weeks in patients with Child-Pugh class B or C disease, we might suggest that the survival of patients with Child-Pugh class A or B disease who received early TIPS placement would be close to 100%, which could hardly be improved by any other treatment. In addition, rebleeding after TIPS placement would be a rare occurrence in such patients, and thus secondary prevention could be avoided. This may be particularly true for Child-Pugh class A and B patients with a high baseline portosystemic pressure gradient (>20 mm Hg), which is an additional risk factor for rebleeding and bleeding-related mortality.7, 8 Finally, the increase in the rate of hepatic encephalopathy (HE) by the shunt, the strongest argument against the TIPS treatment, was not confirmed by the study of García-Pagán et al.1 If this finding were extended to patients with Child-Pugh class A or B disease, they might also be regarded as candidates for early TIPS treatment.
Because of the great influence of this study on the treatment strategy for variceal bleeding, specific attention should be paid to those results differing from previous studies or experiences. The results of the medical group are largely as expected.9, 10 In contrast, some results of the early TIPS group are unexpected. The fact that the authors used bleeding and not survival as the primary endpoint reveals that they expected a small difference in survival requiring an impossibly high sample size. Indeed, the patients had advanced disease and a mean bilirubin concentration of 3.7± 4.8 mg/dL at the baseline. Thus, approximately half of the patients had a bilirubin concentration greater than 3 mg/dL, which predicts reduced survival after TIPS.4, 11, 12 When such patients are treated electively, they have 6-week and 1-year survival rates of only 85% and 75%, respectively.13 Survival rates were, however, comparable between Child-Pugh class A and B patients treated electively (95% and 85%) and the early TIPS group (97% and 86%). It can be speculated that bleeding might cause an acute but transient deterioration that upgrades a patient's Child-Pugh score, which does not reliably reflect the baseline liver function.
In contrast to the study under discussion, randomized studies of secondary prophylaxis did not find a survival benefit for TIPS patients.4, 9 This may be due to the fact that these studies excluded acute bleeders and thus selected survivors with a lower risk of bleeding-related deaths. In addition, the previous studies used uncovered stents with a high rate of shunt insufficiency, which led to a higher rate of recurrent bleeding. As for HE, the results of the study by García-Pagán et al.1 and the studies of secondary prophylaxis are also different. Although TIPS increased the incidence of HE in patients treated for secondary prophylaxis,4, 9 this was not observed in the study by García-Pagán et al. (8 patients with early TIPS and 12 patients with medical treatment). The lower rate of HE, as expected, may be due to the fact that in the study by García-Pagán, stents were initially dilated to 8 mm, and a further dilatation to 10 mm was performed only if the gradient did not decrease below the threshold of 12 mm Hg. Despite this, the mean pressure gradient after TIPS of 6.2 ± 3 mm Hg was lower than that needed and provided the chance for a further reduction in the incidence of HE with even smaller shunts. This good result may also confirm the assumption that patients' Child-Pugh scores did not correctly reflect the baseline liver function because of the influence of acute bleeding.
In conclusion, this study by García-Pagán et al.1 suggests that in Child-Pugh class C (score < 13) and B patients with active bleeding on endoscopy, early TIPS may be used as a first-line treatment. Because of the excellent survival and long-term efficiency of early TIPS, the need for prophylactic treatment may be reconsidered. In patients without these characteristics, the current step-up strategy may be continued. Future studies including Child-Pugh class A and B patients are needed to confirm the study results and the treatment concept.