Optimal detection of Campylobacter spp in stools

In view of the importance of Campylobacter spp and related organisms in human disease, and the awareness of the under-reporting of these organisms from stools, we read with interest the recent reports of McClurg and associates¹ and of Kulkarni et al.² These two studies independently compared the recovery of campylobacter and related species from human stools using non-selective filtration, selective plating, and polymerase chain reaction (PCR) detection.

Since 1977, the Red Cross Children’s Hospital in Cape Town, South Africa, has been isolating Campylobacter spp. In 1990, for cost containment reasons, the use of antibiotic containing selective media was discontinued, and the “Cape Town” protocol introduced. This protocol^3,4 combines both membrane filtration on to antibiotic free blood agar plates and incubation in an H₂ enriched microaerobic atmosphere (Oxoid BR 38 (Oxoid, Basingstoke, UK) or BBL 70304 without catalyst (BBL, Kansas City, USA)). Although other workers have advocated filtration for isolation, the Cape Town protocol is the first to combine filtration with a hydrogen enhanced microaerophilic growth atmosphere.

The use of antibiotic free plates allows the growth of antibiotic sensitive campylobacter strains, and the increased H₂ in the incubation atmosphere permits the growth of species with an essential requirement for H₂ such as C concisus and C rectus. Some strains of C jejuni subspecies jejuni, C jejuni subspecies doylei, and C upsaliensis grow poorly, or not at all, under conventional microaerobic conditions, but will flourish in an H₂ enhanced microaerobic atmosphere.^3,5 Incubation in an H₂ enhanced microaerobic atmosphere increased stool cultures positive for campylobacter by 78% compared with stools incubated under conventional microaerobic conditions.³ McClung and colleagues¹ and Kulkarniand colleagues² described their growth conditions as microaerobic¹ or microaerophilic,² without stating the hydrogen concentration. These laboratories experienced overgrowth of the membrane filters by commensal faecal flora,² and used membrane filters with a pore size of 0.45 μm.² We have found filters with a pore size of 0.60 μm to be optimal for the Cape Town protocol (unpublished results, 1990). These factors may have contributed to the isolation of fewer campylobacter strains.^1,2

We agree that PCR diagnosis of campylobacter in stools is not practical for diagnostic laboratories,² especially those in developing countries. We also agree that it may not be cost effective to use several selective media and/or filtration for efficient enteropathogenic campylobacter isolation.¹ For the past 12 years we have efficiently isolated and biochemically speciated hundreds of strains each year of 17 species or subspecies of campylobacter, arcobacter, and helicobacter from the diarrhoetic stools and blood cultures of our patients, without selective media, but with the use of the Cape Town protocol.^3–5 When tested against a variety of selective media, the Cape Town protocol was consistently superior for the isolation of campylobacteraceae.^3–5 Differences in the prevalence of campylobacter may well exist between Cape Town and the UK, and the application of the Cape Town protocol in future UK studies, as has been suggested,² should improve the campylobacter isolation rate and, hopefully, answer this question.