Colitis Due to Campylobacter jejuni/coli: Ceftriaxone is Not Effective

Background: Ceftriaxone is often prescribed empirically for patients hospitalized with abdominal pain and fever. If stools culture from such a patient yields Campylobacter jejuni or Campylobacter coli (C. jejuni/coli), clinicians ask about its susceptibility to ceftriaxone, to which it is not routinely tested. We report a case of colitis that raised this question. As we couldn’t find the answer in the published literature, we investigated the sensitivity of strains of C. jejuni/coli to ceftriaxone in our hospital. Methods: We conducted a retrospective study of all strains of C. jejuni and C. coli isolated from specimens of infected patients seen in our tertiary care university-affiliated hospital in Switzerland between March 2009 and December 2010. Campylobacter strains were identified to the species level with matrix-assisted laser desorption ionization time of flight (MALDI-TOF) mass spectrometry and antimicrobial susceptibilities were determined by Etest. Results: Among 108 C. jejuni and 14 C. coli isolates, only 1.6% were fully susceptible to ceftriaxone. Overall, 95.9% of our strains were susceptible to erythromycin but 52.5% of C. jejuni/coli were resistant to ciprofloxacin. Conclusion: Our results suggest that ceftriaxone should be considered ineffective for the treatment of Campylobacter jejuni/coli infections, the major cause of gastro-enteritis in Europe. In the light of increasing resistance to fluoroquinolones, cases of severe Campylobacter colitis or bacteraemia may require a short course of macrolides.


Background
A 25-year-old woman presented to our emergency department with abdominal pain, nausea and fever. She had previously been in good health and did not take any medication. A low-dose computed tomography scan of the abdomen (Figure 1) showed inflammation of the caecum and terminal ileum, making analysis of the appendix difficult. The patient was admitted to the visceral (general) surgery service for surveillance and was started on therapy with intravenous ceftriaxone and metronidazole. Her clinical condition improved, but blood cultures were positive for a Campylobacter coli after 48 hours.
The attending surgeon contacted the infectious diseases specialist to ask if this pathogen was likely to be susceptible to ceftriaxone.
Research of medical literature on human infection with Campylobacter jejuni or coli revealed a striking lack of data on this issue. Because our patient's Campylobacter coli isolate was resistant to both ceftriaxone and ciprofloxacin, she was treated 5 days with oral azithromycin for her bacteremia and discharged from the hospital. Campylobacter species (spp) are the most common cause of bacterial gastroenteritis in industrialized countries [1]. Infections due . There are few published data on the susceptibility of C. jejuni/coli to ceftriaxone. Animal studies have described a high rate of resistance of Campylobacter spp to ceftriaxone [3,4]. A few human studies from Israel and Asia [5][6][7][8] have reported resistance to ceftriaxone, but we found no human cases reported from Europe ( Figure 2). The Hospital of the University of Geneva (HUG) is the largest hospital in Switzerland, with 1900 beds. It serves as a secondary care center for the population of Geneva and as a tertiary care referral center for western Switzerland. To seek further information on the susceptibility of C. jejuni/coli to ceftriaxone in our setting, we performed a retrospective study on all of the isolates of C. jejuni and C. coli isolated in our hospital over almost 2 years.

Methods
During the 22 months, we isolated 122 strains, 108 of C. jejuni and 14 of C. coli on Karmali agar. Campylobacter strains were identified to the species level with MALDI-TOF mass spectrometry. Antimicrobial susceptibility testing was performed on Mueller-Hinton agar (+5% sheep blood) incubated at 37°C in a microaerobic atmosphere, as described in the Clinical and Laboratory Standards Institute (CLSI) guidelines supplement for fastidious bacteria (M45-A2, Second edition, August 2010). E-tests were used to determine the minimal inhibitory concentration (MIC) for ciprofloxacin, ceftriaxone and erythromycin [9][10][11][12]. Erythromycin and ciprofloxacin MIC's were interpreted according to CLSI cut-off values. Neither CLSI nor the European Committee on Antimicrobial Susceptibility Testing (EUCAST) has determined susceptibility breakpoints for ceftriaxone against C. jejuni. We found only two studies that defined the susceptibility breakpoint for ceftriaxone for species of Campylobacter: one [3] considered a MIC ≤ 8 μg/ml as susceptible, while the other [7] used ≤ 2 μg. Thus, we elected to consider strains with a MIC ≤ 2 μg/ml to ceftriaxone as fully susceptible, those with 3-6 μg/ml as intermediate and those with ≥ 8 μg/ml as resistant.

Results and Discussion
Among our C. jejuni/coli isolates, 8/122 (6.6%) were isolated in blood, the rest in stools. Only 2/122 (1.6%) were fully susceptible to ceftriaxone (see supplementary material for MIC values for all antibiotics and all strains). As shown in Fig. 2, more than two-thirds of the isolates had a MIC for ceftriaxone in the resistant range. Nearly all of our C. jejuni/coli strains (117/122; 95.9%) were susceptible to erythromycin, but, as described elsewhere [13][14][15], more than fifty percent (64/122; 52.5%) were resistant to ciprofloxacin.
This small retrospective study confirms the distressingly high resistance rate of human Campylobacter isolates to ciprofloxacin, an agent often used to treat this infection. The rate of resistance has been growing in the last decade [14], leading to restrictions on the use of fluoroquinolones in veterinary medicine [16].

Conclusion
Our study highlights the high level of resistance to ceftriaxone of Campylobacter spp causing human infections in a European setting.
Based on our data, and the few published studies, patients treated empirically with ceftriaxone should benefit from a switch to macrolides when C. jejuni/coli are identified as the cause of the infection, especially if they have an invasive or non-resolving infection.

Declarations Ethics approval and consent to participate
No collection of clinical data for this study. For this type of study, formal consent is not required. Consent for publication Was not searched for as the case description does not contain any picture or specific information that could lead to the identification of the patient.

Availability of data and material
Please refer to "Supplementary Material'' where you'll find the complete Excel table with all the data on which the conclusions of our manuscript rely.

Authors' contributions
SE was the ID specialist implicated in the clinical case. He conceived the study, performed the statistical and clinical analysis and drafted the manuscript. BR and BL participated to the analysis of the data and to the redaction of the manuscript. AR, AC and JS contributed to the acquisition of the data and laboratory analysis, and participated to the redaction of the manuscript. FR was the surgeon in charge of the patient and participated to the redaction of the manuscript. All authors read and approved the final manuscript.