A Case-Controlled Study of the Safety and Efficacy of Transjugular Intrahepatic Portosystemic Shunts After Liver Transplantation

The role of transjugular intrahepatic portosystemic shunt (TIPS) insertion in managing the complications of portal hypertension is well established, but its utility in patients who have previously undergone liver transplantation is not well documented. Twenty-two orthotopic liver transplantation (OLT) patients and 44 nontransplant patients (matched controls) who underwent TIPS were analyzed. In the OLT patients, the TIPS procedure was performed at a median of 44.8 months (range = 0.3-143 months) after transplantation. Eight (36.4%) had variceal bleeding, and 14 (63.6%) had refractory ascites. The underlying liver disease was cholestatic in 10 (45.4%) and viral in 4 (18.2%). The mean pre-TIPS Model for End-Stage Liver Disease (MELD) score was 13.4 ± 5.1. There were no significant differences in age, sex, indication, etiology, or MELD score with respect to the control group. The mean initial portal pressure gradients (PPGs) were similar in the 2 groups (21.0 versus 22.4 mm Hg for the OLT patients and controls, respectively), but the final PPG was lower in the control group (9.9 versus 6.9 mm Hg, P < 0.05). The rates of both technical success and clinical success were higher in the control group versus the OLT group [95.5% versus 68.2% (P < 0.05) and 93.2% versus 77.2% (P < 0.05), respectively]. The rates of complications and post-TIPS encephalopathy were similar in the 2 groups, and there was a trend toward increased rates of shunt insufficiency in the OLT group. The mortality rate of the patients with a pre-TIPS MELD score > 15 was significantly higher in the OLT group [hazard ratio (HR) = 4.32, 95% confidence interval (CI) = 1.45-12.88, P < 0.05], but the mortality rates of the patients with a pre-TIPS MELD score < 15 were similar in the 2 groups. In the OLT group, the predictors of increased mortality were the pre-TIPS MELD score (HR = 1.161, 95% CI = 1.036-1.305, P < 0.05) and pre-TIPS MELD scores > 15 (HR = 5.846, 95% CI = 1.754-19.485, P < 0.05). In conclusion, TIPS insertion is feasible in transplant recipients, although its efficacy is lower in these patients versus control patients. Outcomes are poor for OLT recipients with a pre-TIPS MELD score > 15.