Use of Proton Pump Inhibitors and the Risk of Cholangitis

A Nationwide Cohort Study

Yang Won Min; Danbee Kang; Ju-Young Shin; Minwoong Kang; Joo Kyung Park; Kwang Hyuck Lee; Jong Kyun Lee; Kyu Taek Lee; Poong-Lyul Rhee; Jae J. Kim; Eliseo Guallar; Juhee Cho; Hyuk Lee


Aliment Pharmacol Ther. 2019;50(7):760-768. 

In This Article


In this large cohort study using nationally representative data, we found that PPI use was associated with a markedly higher incidence of cholangitis after adjusting for multiple potential confounders. The association was strongest during the period of PPI use and gradually decreased over time after PPI discontinuation. The increased risk of cholangitis in PPI users was consistently observed across pre-defined subgroups of study participants. To our knowledge, the present research is the first cohort study to demonstrate a significant association between PPI use and a risk of cholangitis.

PPIs are a very important class of drugs. They are the most commonly prescribed drugs for the treatment of gastro-oesophageal reflux and peptic ulcer disease, and they are also prescribed to prevent gastric complications from long-term use of aspirin or NSAIDs. PPIs are generally well tolerated and have a low frequency of reported adverse events. PPIs, however, are often used for prolonged periods of time or in inappropriately high dosage, raising safety issues.[26] Several adverse events following PPI use, including pneumonia, enteric infection, gastric neuroendocrine tumour, small intestinal bacterial overgrowth (SIBO), spontaneous bacterial peritonitis, hepatic encephalopathy and gastrointestinal cancer, are likely due to acid inhibition.[27] Furthermore, a prospective cohort study with cirrhotic patients found that the use of PPIs was associated with increased risk of mortality.[28] Of note, a recent review argues that the evidence linking PPI use with many of these adverse events is inadequate, and the association may be due to residual confounding and the overextrapolation of quantitatively small effects.[26]

In spite of potential sources of bias, the evidence supporting an association between gastric acid suppression and various intestinal infections disease is relatively robust in terms of biological plausibility, temporality and specificity.[26] A meta-analysis on the association of PPI use with SIBO identified a strong association in studies in which SIBO was diagnosed by duodenal or jejunal aspirate culture.[5] PPI use was also associated with increased risk of Clostridium difficile, nontyphoidal Salmonella and Campylobacter enteric infections.[29] In addition, small bowel bacterial overgrowth associated with PPI use was significantly associated with an increased incidence of hepatic encephalopathy in patients with cirrhosis.[6] Avoiding non-indicated PPI use and prescribing the lowest possible PPI dose for the shortest possible period are good clinical practices that may thus help reduce the risk of SIBO and other enteric infections in PPI users.[30]

While the association of PPI use with SIBO and with enteric infections is well established, data on the association of PPI use with biliary infection are very limited. Only one case-control study found that PPI use was associated with an 1.23-fold increase in the odds of acute cholecystitis.[17] In addition, a hypothesis-generating study showed that the presence of oropharyngeal flora was more frequent in PPI users, and suggested that PPIs use could increase the number of biliary pathogens and broaden the pathogenic spectrum.[31] In our study, we found a strong association between PPI and the risk of cholangitis, which is consistent with the magnitude of association between PPI use and the risk of bacterial enteric infection observed in a prior meta-analysis.[32] Furthermore, the increased risk in cholangitis associated with PPI use was observed in all subgroups examined and decreased over time after PPI discontinuation.

The exact mechanisms underlying the association of PPI use with an increased risk of cholangitis are unknown. The antimicrobial properties of bile acids contribute to maintaining a sterile biliary tract, but bacteria, most commonly those residing in the intestinal tract, are frequently cultured from the bile of patients with biliary infectious diseases.[33] Studies employing molecular approaches, such as pyrosequencing of bacterial 16S rDNA, have recently evidenced a high similarity between the intestinal and biliary microbiomes in patients with biliary diseases.[34–36] In particular, it has been speculated that there is an association between the microbiota of the proximal small intestine and the bacterial density in the biliary system.[37] In addition, rodent models of small bowel bacterial overgrowth have suggested that intestinal microbiota or bacteria-derived factors affect the bile acid metabolism and contribute to the development of hepatobiliary diseases.[38–40] Accumulating evidence thus suggests that the intestinal microbiota plays an important role in biliary disease development, either through biliary dysbiosis or through direct bile duct injury by products of the gut microbiome. It is thus possible that PPI-induced increases in specific bacterial pathogens in the biliary microbiota may contribute to an increased risk for developing cholangitis.[41]

Our study had multiple strengths, including the use of a large nationwide cohort representative of the general population, a cohort design that is less susceptible to bias compared to case-control studies and the availability of detailed information on various lifestyle parameters associated with biliary disease. Several limitations, however, need to be considered in the interpretation of our findings. First, we did not have information on the reason for PPI prescription, and we could not exclude the possibility that PPIs were indicated for symptoms of cholangitis (reverse causation bias). To reduce the possibility of this type of bias, we excluded the first 7 days of PPI prescription from the definition of exposure periods, and still observed a strong association. In sensitivity analyses, the association was attenuated but still remained significant even when we assumed that PPI exposure could be responsible only for cases of cholangitis occurring 14 days or more after treatment initiation, suggesting that reverse causation cannot fully explain the observed association. Second, our study does not reveal the mechanism of disease development. Third, this is an observational study and we cannot exclude the possibility of residual confounding. However, we adjusted for multiple risk factors and still observed a strong association, adding to the plausibility of the association. Finally, the study was conducted in Korea, and our findings may not generalise to other races/ethnicities or to countries with different comorbidity or health care utilisation patterns.

In conclusion, in this large nationwide analysis, we found a strong association between PPI use and the development of cholangitis. Although PPIs are effective for patients with acid-related disorders, physicians need to consider restricting long-term use of PPIs in individuals at high risk of cholangitis. In addition, additional research is needed to clarify the mechanisms underlying the association between PPIs and cholangitis.