Tel.+49 30 838 62550 Fax.+49 30 838 46029
Background: Campylobacter infection is the most commonly reported bacterial gastroenteritis in Germany, but its epidemiology remains incompletely understood. To investigate the risk factors for Campylobacter infections and the routes of transmission in Germany, we have performed a case control study combined with molecular strain typing (MLST) and source attribution analysis in the framework of the FBI-Zoo network.
Methods: We conducted the study in selected districts of four German federal states. Case patients were recruited through local health authorities (Nov 2011-Feb 2014). Controls were randomly selected from population registries. All participants completed a questionnaire. Campylobacter isolates of a subset of case patients were cultured and further analyzed by MLST of seven housekeeping genes. Using the questionnaires, we conducted univariate logistic regression analyses (SVA), adjusted for age group, sex, and federal state, and multivariate logistic regression analyses (MVA).
Results: In the SVA, travelling abroad was determined to be a statistically significant risk factor (OR 1.11 (95% CI: 1.06-1.17; p-value <0,001; study population: 1814 cases, 3983 controls). For further source-directed analyses, we excluded cases that had travelled abroad. Consumption of chicken meat, preparation of poultry meat in the household, eating out, contact to chickens or ducks/geese, and the use of antacids were identified, among others, as risk factors in SVA. Being a vegetarian, consumption of beef, fresh fruit, vegetables, or herbs were negatively associated with disease. For children <5 years contact with animals (dogs, birds, chickens, ducks/geese), playing in a sand box, and using a pacifier were additional risk factors. Results of MVA will also be presented.
MLST was performed for 614 patient isolates and 543 isolates from animals (338) and food sources (205) that had been collected in the geographical and temporal context of the study. Several new Campylobacter sequence types (STs) were uncovered and currently dominant STs in humans, animals and food were identified.
MLST data combined with detailed source information of current animal and food strains were then combined for a source attribution analysis of the human isolates using Bayesian inference on an asymmetric island model. The main sources attributed to human Campylobacter infection overall, and specifically with C. jejuni, were consumption of chicken and contact with pets. The main source attributed to human C. coli infection was consumption of pork meat.
Conclusions: This first analysis of Campylobacter cases and controls in Germany in combination with molecular typing and source attribution confirmed that chicken meat consumption and its preparation in the household are the most important risk factors for Campylobacter infections in Germany. Further risk factors and sources that were revealed by both approaches were pet contact (in particular for children) and pork meat for C. coli infection. To protect consumers, efforts should be intensified to reduce the Campylobacter load on chicken meat and to inform consumers how they can minimize risk of infection.