Quality Control Strains Used in Susceptibility Testing of Campylobacter spp.  

Antimicrobial resistance among Campylobacter isolates is an increasing problem all over the world. In particular, resistance to fluoroquinolones increased dramatically during the 1990s (2, 5). At present, susceptibility testing of Campylobacter is important for both clinical and epidemiological reasons. Due to the special growth conditions of Campylobacter spp., however, the quality control limits given for nonfastidious organisms in an aerobic atmosphere may not be adequate for these pathogens. Indeed, the lack of a generally accepted protocol for susceptibility testing (including resistance breakpoints) makes the comparison of different publications on Campylobacter jejuni antimicrobial resistance difficult. We describe here the MICs for two ATCC strains determined in the atmosphere recommended for cultivation of Campylobacter. These strains are the two most commonly used controls in C. jejuni MIC determinations, but only rarely have any data on the test results been published (4).

The MICs for Staphylococcus aureus ATCC 29213, Escherichia coli ATCC 25922, and C. jejuni RH 3583 (a local control strain, originally isolated in Edinburgh, United Kingdom; C. jejuni 143483) were determined by the agar plate dilution method. Mueller-Hinton II agar (BBL, Becton Dickinson and Company, Cockeysville, Md.) supplemented with 5% sheep blood was used as the culture medium. The plates were incubated at 35°C for 48 h in a microaerobic atmosphere (CampyPak; BBL). The antimicrobials evaluated were ciprofloxacin, nalidixic acid, erythromycin, azithromycin, clindamycin, ampicillin, amoxicillin-clavulanic acid, cefotaxime, imipenem, gentamicin, tetracycline, and chloramphenicol. The measurements were repeated 10 to 14 times, and four different lots of medium were used.

The MIC results for the ATCC strains were compared to the quality control range recommended by the National Committee for Clinical Laboratory Standards (NCCLS) for nonfastidious organisms, using Mueller-Hinton broth without blood or other supplements (3). Most of the values fell within control ranges (Table 1). However, the MICs of azithromycin for S. aureus ATCC 29213 were 2 to 5 dilutions higher than those recommended by the NCCLS. It is known that an increased level of carbon dioxide does decrease the effect of basic antimicrobials such as macrolides (1). This will certainly happen in the microaerobic atmosphere required for the growth of Campylobacter. The MICs of ciprofloxacin for E. coli ATCC 25922 were also systematically higher than the NCCLS quality control range. A similar but minor trend was observed for both of these ATCC strains with several additional antimicrobial agents. The MIC test results for C. jejuni RH 3583 fell within the range of 2 to 3 dilutions for all antimicrobials tested. No quality control limits are established for Campylobacter spp., which made the reliable interpretation of these MIC results difficult.

TABLE 1.

Agar plate dilution quality control results for S. aureus ATCC 29213, E. coli ATCC 25922, and C. jejuni RH 3583 under study growth conditions^a

Antimicrobial agent	S. aureus ATCC 29213			E. coli ATCC 25922			C. jejuni RH 3583 MIC range (μg/ml)
	MIC range (μg/ml)	NCCLS QC range (μg/ml)b	% of MIC results within QC range	MIC range (μg/ml)	NCCLS QC range (μg/ml)b	% of MIC results within QC range
Ciprofloxacin	0.125-0.5	0.12-0.5	100	0.03-0.06	0.004-0.015	0	0.25-0.5
Nalidixic acid	16-64	—c	NAd	2-4	1-4	100	4-8
Erythromycin	0.5-2	0.25-1	92	>128	—	NA	1-2
Azithromycin	8-16	0.5-2	0	32-64	—	NA	0.25-0.5
Clindamycin	0.25-0.5	0.06-0.25	17	128->128	—	NA	0.5-1
Ampicillin	0.25-4	0.5-2	92	4-8	2-8	100	2-8
Amoxicillin-clavulanic acid	0.25-0.5	0.12-0.5	100	2-8	2-8	100	1-2
Cefotaxime	0.5-1	1-4	83	0.03-1	0.03-0.12	79	2-8
Imipenem	<0.03-0.06	0.015-0.06	100	0.125-0.5	0.06-0.25	82	0.06-0.25
Gentamicin	1-2	0.12-1	42	0.5-2	0.25-1	79	0.5-1
Tetracycline	0.25-1	0.12-1	100	1-2	0.5-2	100	<0.125-0.25
Chloramphenicol	8-16	2-8	17	4-16	2-8	71	<2-4

^aEach measurement was repeated 10 to 14 times.

^bThe quality control (QC) range recommended by the NCCLS for nonfastidious organisms using Mueller-Hinton broth without blood or other supplements.

^c—, the value has not been given by the NCCLS.

^dNA, not applicable.

These data indicate that the quality control limits recommended by the NCCLS for nonfastidious organisms may not be adequate for Campylobacter susceptibility testing. We believe that a set of Campylobacter strains might work better for quality control purposes. The global surveillance of the increasing resistance problem among Campylobacter spp. would be considerably fostered by generally accepted recommendations for susceptibility testing of these pathogens. Such recommendations are urgently needed.