Anne Kvistholm Jensen1,2,3, Jacob Simonsen4, Steen Ethelberg1,2. 1. Department of Infectious Disease Epidemiology, and. 2. Department of Microbiology and Infection Control, Statens Serum Institut. 3. Research Unit of Clinical Microbiology, University of Southern Denmark and Odense University Hospital; and. 4. Department of Epidemiology Research, Statens Serum Institut, Copenhagen, Denmark.
Abstract
BACKGROUND: Recent studies suggest that proton pump inhibitors (PPIs) may increase the risk for listeriosis. We investigated a potential association in cases of nonpregnancy-associated listeriosis using registry data. METHODS: We conducted a population-based, case-control study using Danish health registries. Cases (n = 721) were defined as patients aged ≥45 years notified with listeriosis from July 1994 to December 2012. We selected 34800 controls using risk-set sampling. Controls were individually matched for age, sex, and municipality. Data on use of PPIs and other drugs and hospitalization diagnoses over a 5-year period were extracted from nationwide health registries. A comorbidity index (CMI) was constructed. We calculated the association between use of PPIs and related drugs within 30 days (current use) and other time windows before the index date. Using conditional logistic regression, matched odds ratios (ORs) adjusted for CMI and confounders were estimated. RESULTS: The adjusted OR for current use of PPIs and development of listeriosis was 2.81 (95% confidence interval [CI], 2.14-3.69). PPI usage up to 90 days before the index date remained statistically significant. Subgroup analyses revealed increasing ORs with decreasing age and level of comorbidity and an increased OR for concurrent glucocorticoid treatment (OR, 4.61; 95% CI, 3.01-7.06). No significant association was found for current use of histamine-2-receptor antagonists (adjusted OR, 1.82; 95% CI, 0.89-3.71). CONCLUSIONS: Prescribed PPIs were associated with increased risk of listeriosis. The risk waned with time since last prescription redemption. PPIs may have unwanted side effects in vulnerable populations.
BACKGROUND: Recent studies suggest that proton pump inhibitors (PPIs) may increase the risk for listeriosis. We investigated a potential association in cases of nonpregnancy-associated listeriosis using registry data. METHODS: We conducted a population-based, case-control study using Danish health registries. Cases (n = 721) were defined as patients aged ≥45 years notified with listeriosis from July 1994 to December 2012. We selected 34800 controls using risk-set sampling. Controls were individually matched for age, sex, and municipality. Data on use of PPIs and other drugs and hospitalization diagnoses over a 5-year period were extracted from nationwide health registries. A comorbidity index (CMI) was constructed. We calculated the association between use of PPIs and related drugs within 30 days (current use) and other time windows before the index date. Using conditional logistic regression, matched odds ratios (ORs) adjusted for CMI and confounders were estimated. RESULTS: The adjusted OR for current use of PPIs and development of listeriosis was 2.81 (95% confidence interval [CI], 2.14-3.69). PPI usage up to 90 days before the index date remained statistically significant. Subgroup analyses revealed increasing ORs with decreasing age and level of comorbidity and an increased OR for concurrent glucocorticoid treatment (OR, 4.61; 95% CI, 3.01-7.06). No significant association was found for current use of histamine-2-receptor antagonists (adjusted OR, 1.82; 95% CI, 0.89-3.71). CONCLUSIONS: Prescribed PPIs were associated with increased risk of listeriosis. The risk waned with time since last prescription redemption. PPIs may have unwanted side effects in vulnerable populations.