Short and long term effects of bacterial gastrointestinal infection

Sammanfattning: The objectives of this thesis were to increase our understanding of the mortality and complications associated with bacterial gastrointestinal infections, with focus on Salmonella and Campylobacter enteritis. The effect of antibiotics vs. placebo on duration of diarrhea in Campylobacter enteritis was also examined. Persons reported with culture verified bacterial enteritis to the Swedish Institute of Infectious Disease Control 1997 2004 formed the base for three of the studies in this thesis. From the national database, case-based information on age, sex, type of bacteria, date of debut of illness, and country of infection were extracted. Each person in this retrospective cohort was followed until death, an event took place or until study termination. We used the National Tax Board registers to identify deaths in the cohort, and the national Hospital Discharge Register from the National Board of Health and Welfare to observe complications associated with bacterial gastrointestinal infection. Standardized mortality/incidence ratios (SMR/SIR) were used to estimate the relative risk of death or short and long term complications. For persons infected with Campylobacter, the SMR among those infected within Sweden was 2.9 with 95% confidence interval (CI) 1.9 4.0 during the first month after infection, and for those who had acquired Salmonella at home 5.6 (95% CI: 3.4 8.2). No increased SMR within the first 30 days after infection could be found among those infected with Campylobacter or Salmonella abroad: 0.3 (95% CI: 0.04 0.8) and 0.6 (95% CI: 0.2 1.2). We are probably observing a healthy traveler effect , i.e. persons who travel are healthier than the general population. The effect of antibiotics on Campylobacter infection has only been studied in quite small studies. This led us to conduct a quantitative summary analysis of all published randomized controlled trials (RCTs). Eleven RCTs which included a total of 479 study participants were included in a meta-analysis. The summary effect in our random effect model showed a reduction of 1.32 days (95% CI: 0.64 1.99) with symptoms in favor of antibiotics compared to placebo. If one had been infected at home or abroad did not have any interaction effect on the complications observed among patients in our cohorts. We could confirm the associations between EHEC and hemolytic uraemic syndrome, Campylobacter infection and Guillain-Barré syndrome, and Yersinia enteritis and reactive arthritis. We found evidence of an association between Salmonella enteritis and aortic aneurysm, SIR 6.4 (95% CI: 3.1 11.8). Transient bacteremia with non-typhoid Salmonella can probably cause a localized endothelial infection that result in an aneurysm or the enlargement of a pre-existing aneurysm. Salmonella infection was associated with an increased risk for ulcerative colitis, SIR 3.2 (95% CI: 2.2 4.6) and this was also found among those with campylobacterosis, SIR 2.8 (95% CI: 2.0 3.8). No significant increased risk for Crohns disease was shown in the Salmonella cohort, SIR 1.4 (95% CI: 0.8 2.3) or among those reported with Campylobacter enteritis, SIR 1.6 (95% CI: 1.0 2.3). Although similar results have been found elsewhere, more work is needed to refute or confirm our findings.