Purpose: Hepatic encephalopathy (HE) is a common complication of transjugular intrahepatic portosystemic shunt (TIPS) procedures. Proton pump inhibitor (PPI) use has recently been implicated as a risk factor for HE, potentially due to alterations in the gastrointestinal tract microbiome or effects on ammonium ion excretion. Although generally supporting an association between PPI use and HE, multiple meta-analyses have reported heterogeneity and publication bias, and only two previous studies have examined the role of PPI use in post-TIPS HE. Therefore further evaluation of the role of PPI use in post-TIPS HE is warranted. The purpose of this study was to determine whether PPI use is associated with an increased risk of post-TIPS HE in an independent patient cohort.
Materials and Methods: This single-institution retrospective study analyzed 86 patients (54 male, mean age 58.2) following TIPS insertion from 1/1/2017 to 12/31/2019. Dates of PPI usage and episodes of new or worsening HE were extracted from medical records. Poisson regression with generalized estimating equations was used to test for an association between PPI use and post-TIPS HE and to test for dose dependence. A Cox proportional hazards model with time dependent covariates was used to analyze post-TIPS survival.
Results: After TIPS creation there were 49 episodes of new or worsening HE among 35 patients on chronic PPI therapy (1.88 episodes per person-year), 44 episodes among 35 patients on intermittent PPI therapy (2.26 per person-year on PPIs and 0.75 off PPIs), and 7 episodes among 16 patients never on PPIs (0.49 per person-year). In univariable analysis PPI use was associated with a 3.65-fold increased rate of new or worsening HE (incidence rate ratio (IRR)=3.65; 95% CI: 1.60-8.35; p=0.002). In multivariable regression after accounting for older age (IRR=1.03; CI: 0.99-1.06; p=0.15) and higher model for end stage liver disease (MELD) score (IRR=1.19; CI: 1.10-1.29; p< 0.001), PPI use was an independent risk factor for post-TIPS HE (IRR=4.70; CI: 1.07-20.6; p=0.04). Further analysis of only those patients who used PPIs showed an increased risk of HE with higher doses (IRR=1.11 per 10 mg omeprazole or equivalent; CI: 1.00-∞; p=0.048). In multivariable survival analysis PPI use did not predict post-TIPS mortality (HR=0.80; CI: 0.36-1.77; p=0.58).
Conclusion: In an independent patient cohort PPI use was associated with increased risk of post-TIPS HE in a dose dependent manner. In addition to being a reproducible risk factor, PPI use may represent one of the few modifiable risk factors associated with post-TIPS HE.