Proton-pump inhibitors (PPI) are commonly prescribed in an uncritical manner to patients with liver cirrhosis without a clear evidence-based indication. Observational studies suggests that PPI use in cirrhotic patients may be a risk factor for the development of infections, especially spontaneous bacterial peritonitis (SBP). A possible explanation are PPI-associated microbiotic shifts leading to small intestinal bacterial overgrowth with subsequently increased bacterial translocation. Furthermore, PPI therapy in cirrhotic patients may lead to an increased risk for pneumonia and Clostridium difficile-infections. However, the evidence is ambiguous, as other published studies found no evidence for an association of PPI use with an increased risk for SBP or pneumonia. Moreover, an association between episodes of hepatic encephalopathy and PPI use has been reported. Infections and hepatic encephalopathy may often lead to a hospitalization of cirrhotic patients and PPI use at discharge has also been associated to early re-hospitalization. While some studies found an association of PPI and increased mortality in cirrhotic patients, other studies could not observe this association. Thus, some of the current evidence suggests an unfavourable risk profile of PPIs in patients with liver cirrhosis. However, this patient population is considered to be at a high risk of gastrointestinal haemorrhage from peptic ulcers. Importantly, patients with liver cirrhosis have an increased mortality after peptic ulcer bleeding as compared to patients without cirrhosis. Therefore, generous PPI use may also have a yet unproven preventive effect against upper gastrointestinal bleeding. The STOPPIT trial is the first prospective, randomized, controlled, double-blind trial investigating the effect of discontinuation of long-term PPI therapy on hospitalized patients with complicated liver cirrhosis with a pre-existing long-term PPI therapy. Importantly, patients with an evidence-based indication for PPI therapy are excluded from the trial. All study participants (n=476) stop their previous PPI treatment and are then randomized (1:1) to receive either placebo (intervention group) or esomeprazole 20mg/day (control group) for 360 days. The primary hypothesis anticipates a delay of re-hospitalisation and/or death (composite endpoint) in patients who discontinue PPI treatment as compared to patients who continue PPI therapy. Secondary objectives include the assessment of mortality, re-hospitalisation rates, infection rates, rate of acute hepatic decompensation and ACLF, as well as rates of upper and lower gastrointestinal bleeding events in both groups. Impact of prolonged or discontinued PPI therapy on the intestinal microbiota and pharmacoeconomics will be studied as a secondary assessment.