Association between recent use of proton pump inhibitors and nontyphoid salmonellosis: a nested case-control study.

BACKGROUND The association between proton pump inhibitors (PPIs) and nontyphoid salmonellosis (NTS) continues to be debated. The current study was designed to determine the association between use of oral PPIs and the diagnosis of NTS. METHODS The Taiwan National Health Insurance Research Database from 2000 to 2010 was searched for cases of NTS, defined by the International Classification of Disease, Ninth revision, Clinical Modification. A nested case-control study in hospitalized population was conducted using 4 controls for each case patient (14 736 case patients and 58 944 controls), matched for age, month and year of entry, Charlson comorbidity index score, and well-known predisposing factors for NTS, including autoimmune diseases, acquired immunodeficiency syndrome, diabetes, cirrhosis, transplantation, gastrointestinal operations or diseases, and malignancies. RESULTS Persons with NTS had a higher rate of using oral PPIs within the prior year (adjusted odds ratio [OR], 2.09; 95% confidence interval [CI], 1.95-2.24; P < .001). The association was greatest for current PPI use (adjusted OR, 5.39; 95% CI, 4.79-6.06; P < .001). Although use of H2-receptor antagonists (adjusted OR, 1.84; 95% CI, 1.71-1.98), antibiotics (5.21; 4.81-5.64), steroids (3.18; 2.99-3.39), and nonsteroidal anti-inflammatory drugs (2.37; 2.26-2.48) within the 30 days were also associated with NTS, the linkage between PPI use and NTS remained significant in the subgroup without these medications. CONCLUSIONS The use of oral PPIs was associated with the occurrence of NTS. The risk waned with time after discontinuation.