The Safety of TIPS in High MELD Patients: Re-examining Established Practices: 1023

Introduction: Transjugular intrahepatic portosystemic shunt (TIPS) plays an important role in the management of complications of portal hypertension. Historically, studies have shown that there is a significantly increased mortality risk when performing the procedure on patients with a MELD score >18. This study aims to re-examine the accuracy of using a MELD > 18 as a cut-off for post TIPS survival and hepatic dysfunction. We hypothesized that a less conservative cut-off would not result in increased mortality at 1 month and 3 months following TIPS.Figure: Kaplan-Meier survival curves based on MELD scores show decreased survival for those with MELD score > 21 (log rank p < 0.001).Methods: We conducted a retrospective analysis of 290 patients who underwent TIPS using polytetrafluoroethylene (PTFE) covered stents at a single academic center from 2006 to 2016. Elective and emergent TIPS were included. For each patient, MELD at the time of TIPS was noted. Survival data at 1 month and 3 months was obtained. Receiver operating curve (ROC) analysis was used to determine optimal cut-off for MELD that influenced survival. Kaplan-Meier analysis was performed to complement the findings in our ROC analysis. Results: Median age at the time of TIPS was 56 with 60% males and 84% Caucasian. The pre-TIPS MELD scores ranged from 7 to 40 (median 14), with 22% of patients having a score MELD > 18 at the time of TIPS. Death rates at 1 month and 3 months post TIPS were 11%, and 14% respectively. ROC curves using the Youden's Index demonstrated an optimal cut-off of 21 with respect to post TIPS short-term survival. Kaplan-Meier analysis comparing survival between those with MELD > 21 and those with MELD ≤ 21, as well as when further stratified into groups by MELD, demonstrated a significantly poorer survival in those with MELD > 21 (p < 0.001) using the log rank test (see figure). Conclusion: These observations from a large single center cohort support using a MELD cut-off of 21 when predicting short-term mortality from TIPS. In our current era of a rising average MELD score on the liver transplant waitlist, these findings suggest the need to reevaluate established practices that deem a MELD >18 as predictive of increased mortality following TIPS.