Comparison of two different laboratory systems for the identification of bacteria isolates

Project No. BGIA 2055


completed 06/2006


Bacteria species are identified in the classical manner by the assessment of morphological criteria of colonies and single cells, of cytological characteristics, physiological criteria, oxygen reaction, and certain characteristics of metabolism. Miniaturized test systems were developed for the detection of various pathogenic bacteria obtained from different manufacturers: these systems examine the specific metabolic productivity for the group of microorganisms concerned. Before now, only a test system applying visual identification has been employed in the microbiological laboratory of the BG Institute for Occupational Safety and Health (BGIA). Bacteria samples from work environments are mixed populations which may include both harmless, ubiquitous environmental organisms, and pathogenic bacteria of medical interest. The system in use yields sufficiently reliable results only for the pathogenic bacteria as a part of the whole mixed population. For this reason, a second test system was obtained. This new system was originally developed for the characterization of mixed populations in environmental samples. Its principle of analysis is also based upon a test of the microorganisms' metabolic properties and subsequent comparison of reaction data. In contrast to the earlier used system, however, the microtitration plates used in this case employ photometric rather than visual detection. Owing to its basic concept for the detection of environmental organisms, the new system promises to deliver a wider area of application than that yielded by the old one: the latter is designed specifically for the detection of bacteria of medical interest. The objective of this comparison of test systems was the selection of one of the two test systems, which is then to be employed routinely in future for the identification of bacteria.


The objective was to reach conclusions regarding the comparability of results obtained using the two systems. For this purpose, studies were conducted on a total of five test strains obtained from multiple strain collections (DSMZ, the German national collection of microorganisms and cell cultures, Braunschweig; American Type and Culture Collection (ATCC), Manassas, Virginia, USA) and on five own isolates from environmental samples. The selected test strains were primarily bacteria commonly found in liquid environmental samples and potentially relevant as possible pathogens when present in high concentrations in working areas. One bacteria species frequently isolated from atmospheric samples was also studied. The media employed for preincubation of the pure cultures and the duration and temperature of incubation were varied in a number of test arrays (comparison of the manufacturer's specifications for the two test systems with the standard laboratory procedure for measurement of the total colony count of bacteria from atmospheric and material samples).


The preculture of the bacteria on various media had no bearing upon the identification results with either system. Differences in temperature at which the test preparations were incubated also had no influence upon the final result. The results were however influenced by the duration of incubation: the largest number of plausible identifications was attained following incubation of the test preparations for 24 hours. This duration corresponds to the requirements of the API system, but was also employed for the studies involving the BIOLOG-System, for which an incubation time of 20 hours should be sufficient according to the manufacturer's information. Four of the five test strains from strain collections were identified either decisively or with plausible probability (= 99.8%) by means of the API system. Only three of the five test strains were identified reliably by means of the BIOLOG system. BIOLOG was more successful than API in one case in identifying unknown isolates from environmental samples.
Altogether, the BIOLOG system offered no advantages compared to the API system employed to date. The BGIA's standard internal laboratory procedure for identifying bacteria isolates from atmospheric or material samples from working environments will not therefore be modified for the time being. The limitations of the API system with regard to isolates from environmental samples are known and will continue to be tolerated. The BGIA will first conduct a further comparison with the identification system of a third manufacturer. The intention here is to improve and validate the quality of results for identification of bacteria isolates from workplace samples.

Further informations:

Last Update:

30 Jan 2007


Financed by:
  • Hauptverband der gewerblichen Berufsgenossenschaften (HVBG)
Research institution(s):
  • Berufsgenossenschaftliches Institut für Arbeitsschutz - BGIA

-cross sectoral-

Type of hazard:

biological agents


biological agents

Description, key words:

identification, bacteria species, systems for identification, biochemical test