Phenotypic versus genotypic identification of bacteria in the clinical microbiology laboratory: the possible impact on patient care

Abstract number: 1733_88

Adler A., Temper V., Moses A.E., Nahor O., Zimchoni O., Block C.

Objective: Phenotypic bacterial identification methods have numerous strengths but often fail because the phenotype may be variable and subject to biases of interpretation. Sequencing of 16S rRNA and other genes is a more accurate and objective method of identification of microorganisms. We report the study of the causes for phenotypic misidentification in four cases of severe bacterial infections and the possible impact of these errors on patients' care in these cases.

Methods: Phenotypic identification was performed using conventional manual methods and commercial identification kits. Genotypic identification was performed by comparing an 800 bp amplicon of the 16S rRNA gene to the GeneBank database. The impact of the phenotypic misidentification was assessed by reviewing the patients' files and questioning the Infectious Diseases specialists that were involved in these cases.

Results: The cases summaries, the phenotypic methods and results, the genotypic results and the possible causes for the phenotypic misidentification are presented in the table. The initial misidentification had no apparent effect on the patients' management and outcome in the first two cases. In the third case, it might have led to missed diagnosis of infectious endocarditis, to inappropriate duration of antibacterial therapy and eventually to valvular replacement surgery. In the forth case, the misidentification of Brucella melitensis led to a suboptimal choice of antibacterial therapy that might have led to persistent infection. Also, it led to inadequate safety measures in the laboratory that resulted in laboratory-associated brucellosis in a technician.

Conclusion: Phenotypic misidentification is multifactorial. It may be caused by an unusual phenotype or by erroneous selection or interpretation of commercial identification systems. Since the consequence of misidentification might be crucial, we believe that genotypic identification of microorganisms should be considered in any case of significant infection, when the results obtained by phenotypic methods are ambiguous or a rare organism is identified.

Case summary and Source of isolatesPhenotypic identificationPhenotypic MethodsGenotypic identificationPossible causes for phenotypic misidentification
Pus from a brain abscess in a two-year old girl with congenital cyanotic heart defectGemella speciesManual methodsStreptococcus intermediusHuman error
Blood culture (2 sets) from a 68-year old female patient with metastatic cancerListeria grayiiManual methods and API CORYNE (bioMérieux)Lactobacillus plantarumPseudocatalase activity resembling catalase led to misidentification by commercial kit.
Blood and valvular tissue from a 59-year old male patient with infective endocarditisPasteurella haemolytica (isolated from blood culture; valvular tissue culture – no growth)Manual methods, API 20E (bioMérieux) and RapID NF plus (Remel)Agregatibacter (Haemophilus) aphrophilus (blood culture isolate and valvular tissue)Erroneous characterisation of the isolate as ``glucose non-fermenters'' that led to selection of inappropiate identification systems.
Blood culture from a 24-year old female patient with fever and persistent bacteraemiaActinobacillus ureaeVITEK 1 (the isolate was transferred from another hospital)Brucella melitensisInappropriate reliance on automated indentification system.

Session Details

Date: 31/03/2007
Time: 00:00-00:00
Session name: European Society of Clinical Microbiology and Infectious Diseases
Location: ICC, Munich, Germany
Presentation type:
Back to top