Effect of Proton Pump Inhibitors on the Risk and Prognosis of Infections in Patients With Cirrhosis and Ascites

Gitte Dam; Hendrik Vilstrup; Per Kragh Andersen; Lars Bossen; Hugh Watson; Peter Jepsen


Liver International. 2019;39(3):514-521. 

In This Article

Abstract and Introduction


Background & Aims: Many patients with cirrhosis use proton pump inhibitors. We aimed to determine their effects on the risk and prognosis of infections in patients with cirrhosis and ascites.

Methods: We used data from three 1-year trials of satavaptan treatment of ascites (N = 1198) to compare incidence and 90-day mortality of first-time infections between users and nonusers of proton pump inhibitors. With standard and marginal structural Cox models, we adjusted for differences in gender, age, cirrhosis aetiology, Model for End-stage Liver Disease score, serum albumin, lactulose use, severity of ascites, and history of spontaneous bacterial peritonitis or variceal bleeding.

Results: During the follow-up, 446 patients had an infection. At inclusion, 524 patients (44%) used proton pump inhibitors, and 645 (54%) used them at some point during the follow-up. Proton pump inhibitor use increased the rate of infections overall (adjusted hazard ratio = 1.43, 95% CI 1.18–1.74), and it also increased the rate of all specific types of infections except upper respiratory tract infections of presumably viral origin. The estimated cumulative risk of infections was 36.4% for proton pump inhibitor users vs 25.1% for nonusers at 6 months (relative risk = 1.45, 95% CI 1.22–1.73), and 45.2% vs 37.7% at 1 year (relative risk = 1.20, 95% 0.97–1.40). Use of proton pump inhibitors did not affect mortality during the 90 days following infection (adjusted hazard ratio = 0.83, 95% CI 0.53–1.31).

Conclusions: Approximately half of patients with cirrhosis and ascites use proton pump inhibitors. This use increases their risk of bacterial infections, but does not affect their prognosis after an infection occurs.


Patients with cirrhosis are prone to infections, which are often fatal.[1] It is therefore important to identify and eliminate risk factors for infection. Use of proton pump inhibitors (PPIs) is associated with enteric infections in healthy persons,[2–6] because they disrupt the gastric acid barrier, change the gut microbiota[7] and facilitate small intestinal bacterial overgrowth.[8] This is believed to be a causal mechanism behind development of spontaneous bacterial peritonitis, and PPI use has indeed been linked with an increased risk of spontaneous bacterial peritonitis in patients with cirrhosis,[9–15] although some studies found no such association.[16,17] It is clinically important to determine whether it increases the risk of infections in general, due to its widespread use among patients with cirrhosis.[17–21] Moreover, its impact on mortality following an infection needs to be determined. This question has been addressed in a few studies and one meta-analysis,[22–24] but they gave conflicting results and were limited to mortality following spontaneous bacterial peritonitis.

Given the weaknesses and conflicts of the currently available evidence, we decided to extend our previous study of PPI use as a risk factor for spontaneous bacterial peritonitis in patients with cirrhosis.[15] Here, we examine the effects of PPI use on infections in general and on mortality following an infection. We used data from three randomized controlled trials of satavaptan treatment in patients with cirrhosis and ascites.