
Methods: CDC’s National Antimicrobial Resistance Monitoring System (NARMS) receives Campylobacter isolates from 10 sites participating in the Foodborne Diseases Active Surveillance Network (FoodNet). We linked NARMS isolate data to FoodNet patient data for outcome and recent international travel. We categorized isolates with resistant or intermediate MICs defined by the Clinical and Laboratory Standards Institute as resistant to erythromycin (ERY-R) and ciprofloxacin (CIP-R). We used Fisher’s exact test to compare characteristics of patients with blood and stool isolates.
Results: During 2005–2014, we tested 12,200 Campylobacter blood and stool isolates; 115 (0.9%) were from blood, of which 68% were C. jejuni, 14% C. fetus, 10% C. coli, 6% C. upsaliensis, and 2% other species. Bacteremia was associated with non-jejuni/coli species, age ≥65 years, hospitalization, and death (all p<0.01). Travel was less likely among patients with blood isolates (p<0.01). Among blood isolates, 2% were ERY-R and 21% were CIP-R; this did not significantly differ from resistance among stool isolates (2% ERY-R, 23% CIP-R). Comparing CIP-R to susceptible isolates from blood, the proportions of patients age ≥65 years, hospitalized, or reporting travel did not significantly differ.
Conclusion: Blood isolation of Campylobacter was rare; however, it was more likely in elderly patients and for non-jejuni/coli species. Bloodstream infections were associated with adverse outcome and nearly one quarter of bloodstream infections were ciprofloxacin-resistant. Clinicians should consider the possibility of resistance when considering fluoroquinolones for treatment of invasive Campylobacter infection.

J. Reynolds,
None
J. Huang, None
F. Medalla, None