Research Article: Cost-Effectiveness of Blood Agar for Isolation of Mycobacteria

Date Published: November 21, 2007

Publisher: Public Library of Science

Author(s): Michel Drancourt, Didier Raoult, Pamela Small

Abstract: BackgroundMycobacterium species are grown using specific media that increase laboratory cost, thus hampering their diffusion in resource-limited countries. Preliminary data suggested that versatile blood agar may be also used for mycobacterial culture.MethodologyWe examined the growth of 41 different Mycobacterium species on 5% blood agar. Over a 24-month period we analysed isolation of mycobacteria after parallel inoculation of clinical specimens into both a reference automated system (BACTEC 9000 MB broth) and 5% blood agar slant tubes, after NaOH decontamination, and compared the cost of performing 1,000 analyses using these two techniques.ConclusionsMycobacterium reference species cultured on blood agar, with the exception of Mycobacterium ulcerans. Inoculation of 1,634 specimens yielded 95 Mycobacterium isolates. Blood agar performed significantly more efficiently than BACTEC 9000 MB broth (94 vs 88 isolates, P = 0.03). Decontamination of Candida albicans in 5 specimens by addition of amphotericin B in blood agar yielded one more M. tuberculosis isolate that could not be isolated in BACTEC broth. Uneven distribution of time to culture positivity for M. tuberculosis had a median (range) of 19±5 days using blood agar and 26±6 days using BACTEC 9000 MB broth. Cost for 1,000 analyses in France was estimated to be of 1,913 euros using the blood agar method and 8,990 euros using the BACTEC 9000 MB method. Blood agar should be regarded as a first-line medium for culturing Mycobacterium species. It saves time, is cost-effective, is more sensitive than, and at least as rapid as the automated method. This is of particular importance for resource-limited countries in which the prevalence of tuberculosis is high.

Partial Text: Specific media, such as egg-based media (e.g. Lowenstein Jensen medium), agar-based media (e.g. Middlebrook media) and liquid media (e.g. Middlebrook and BACTEC broths) are recommended for culturing Mycobacterium species [1]. Such requirements pose logistic and economic problems, especially in resource-limited areas where bacteriological culture facilities are few and the prevalence of mycobacterial infections, notably tuberculosis, is high. The cost of automatic detection using reference automated mycobacterial culture systems is high. The rationale for using specific media for culturing mycobacteria was to ensure the growth of Mycobacterium species without supporting the growth of contaminants. However, effective decontamination procedures have been developed for non-sterile samples [1]. We therefore questioned the utility and cost-effectiveness of an alternative medium, blood-agar, for the isolation and growth of Mycobacterium species as it is relatively commonly available and inexpensive. Moreover, it may save time and expense by avoiding duplicate inoculation for sterile specimens when other microorganisms are suspected. Sporadic papers have reported the isolation of Mycobacterium tuberculosis on standard blood-agar incubated for 3 weeks or more [2],[3] and proof-of-principle studies have been published [4],[5]. However, it has not been evaluated whether blood agar is useful for the isolation of opportunistic non-tuberculous Mycobacterium species, what the contamination frequency is for non sterile samples, and what the cost-effectiveness of such a diagnostic approach is.

This study extends the findings of previous anecdotal [2], [3]–[9] and proof-of-principle studies [4], [5]–[13] and demonstrates that blood-agar in slant tubes outperforms the reference automated method for isolation of mycobacteria from clinical specimens. Most Mycobacterium species encountered in the clinical microbiology laboratory readily grew onto 5% sheep-blood agar, except for M. ulcerans, a fastidious organism which is rarely isolated from diseased tissues in patients with Buruli ulcer [14]. The characterisation of adverse factors preventing this species culture on blood agar was beyond the scope of this study. These data extend previous information to a number of Mycobacterium species routinely encountered in microbiological laboratory practice, regardless of whether the species were slow or rapid growers. Previous study indeed demonstrated that blood-agar was at least equivalent to egg-based medium for the isolation of M. tuberculosis from respiratory and lymph node specimens [5]. In present study, the sensitivity of blood agar for culturing Mycobacterium isolates from clinical specimens was 98.9% (1/95 isolate was cultured in the BACTEC bottles only). This M. tuberculosis isolate had been recovered from a diseased lymph node and there was no obvious reason for the lack of growth on blood agar. No clinical material was left in order to reproduce the parallel inoculation. Seven M. tuberculosis isolates made on blood-agar medium failed to grow in BACTEC bottles, thus giving a significantly lower sensitivity (92.6%) of the BACTEC medium compared to blood agar for the isolation of mycobacteria. M. marinum grew in blood agar incubated at 30°C and not in BACTEC incubated at 37°C. This observation agrees with the reported optimal growth temperature of this species and that some M. marinum isolates grow better on blood than on Middlebrook agar [15]. Blood agar slants offer the possibility of incubation at different temperatures whereas all automatic systems, including BACTEC, do not allow for such modulation. We suggest that the M. tuberculosis complex isolates which did not grow in BACTEC bottles, failed to grow as the inocula were too small to promote growth in the 40-ml volume of broth but large enough to promote minute growth on solid medium. The contamination frequency of 1.5% on blood agar for respiratory tract specimens was low and warrants further evaluation in other laboratories. We used amphotericin B to decontaminate samples inoculated on blood agar, whilst amphotericin B cannot be used in BACTEC system as it interferes with the detection system. Accordingly, one respiratory tract M. tuberculosis isolate was made on blood agar after amphotericin B decontamination.



Leave a Reply

Your email address will not be published.