Epidemiology of biliary tract-associated bloodstream infections and adequacy of empiric therapy: an Australian population-based study

Abstract

Abstract Purpose Although the biliary tract is a common source of invasive infections, the epidemiology of cholangitis- and cholecystitis-associated bloodstream infection (BSI) is not well defined. The objective of this study was to determine the incidence, clinical determinants, microbiology of biliary tract-associated BSI, and predicted adequacy of common empiric therapy regimens. Methods All biliary tract-associated BSI in Queensland during 2000–2019 were identified using state-wide data sources. Predicted adequacy of empiric antimicrobial therapy was determined according to microbiological susceptibility data. Results There were 3,698 episodes of biliary tract-associated BSI occurred in 3,433 patients of which 2,147 (58.1%) episodes were due to cholangitis and 1,551 (41.9%) cholecystitis, for age- and sex-standardized incidence rates of 2.7, and 2.0 per 100,000 population, respectively. An increasing incidence of biliary tract-associated BSI was observed over the study that was attributable to an increase in cholangitis cases. There was a significant increased risk for biliary tract-associated BSI observed with advancing age and male sex. Patients with cholangitis were older, more likely to have healthcare associated infection, and have more comorbidities most notably liver disease and malignancies as compared to patients with cholecystitis. The distribution of infecting pathogens was significantly different with polymicrobial aetiologies more commonly observed with cholangitis (18.4% vs. 10.5%; p < 0.001). The combination of ampicillin/gentamicin/metronidazole was predicted to have the overall highest adequacy (96.1%), whereas amoxicillin/clavulanate had the lowest (77.0%). Amoxicillin/clavulanate (75.2% vs. 79.4%, p:0.03) and ceftriaxone/metronidazole (83.4% vs. 89.6%; p < 0.001) showed significantly inferior predicted adequacy for cholangitis as compared to cholecystitis. Conclusions Bloodstream infections related to cholecystitis and cholangitis exhibit different epidemiology, microbiology, and requirements for empiric therapy.