Development and validation of an automated 16S rDNA assay for the detection of bacterial DNA in clinical samples

Abstract number: P1989

de Boer R.F., Schuurman T., Kooistra-Smid A.M.D.

Objectives: Since 2002 the department of Research and Development of the Laboratory for Infectious Diseases started to analyse clinical specimens on request for the presence of bacterial DNA. The assay consists of a manual DNA extraction method, followed by conventional broad-range PCR. PCR positive products are subsequently sequenced to identify the organism that was present in the clinical specimen (ref. Schuurman et al., JCM. 2004, 42;2 p734). However, the assay is very laborious and time consuming. To decrease hands on time (HOT) the extraction and detection procedure were automated. This study describes the validation of an automated 16S rDNA assay (A) for the detection of bacterial DNA in clinical samples. Method A was compared with our conventional detection method (C).

Methods: Prior to DNA extraction, samples were treated with mechanical disruption for both methods. DNA extraction was performed with the manual Boom method (C) or with the automated easyMAG specific A protocol (bioMérieux) (A). Conventional (C) and real-time (A) PCR were performed with universal primers (probe), which amplified/detected the whole 16S rRNA gene. All positive PCR products were subsequently sequenced. The analytical sensitivity of method A was assessed by dilution series of different model organisms (n = 3), spiked in phosphate buffered saline. Also, both methods were compared in a validation performed on clinical samples (n= 89).

Results: Analytical sensitivity of both methods were comparable, with limits of detection of approximately 400 – 40 CFU/PCR, 1450 – 145 CFU/PCR, and 800 – 80 CFU/PCR for Gram-negative (E. coli) and Gram-positive organisms (S. aureus and S. pneumoniae) respectively. Thirty eight positive and 46 negative samples were in complete concordance for both methods (94.4%). Method A detected 4 additional clinical significant positive samples (S. intermedius, S. maltophilia, S. pneumoniae and a mixed sequence). Method C yielded 1 additional positive result (S. aureus), which also was regarded clinical significant. PCR inhibition was detected in 10% of all samples for method A in comparison with 49% for method C. Also, HOT for obtaining a PCR result decreased (~50%) when method A was applied.

Conclusions: With regards to sensitivity both methods were comparable in detecting bacterial DNA from clinical samples. However, when method A was applied HOT was reduced considerably, PCR results were easier to interpret, and inhibition rates decreased dramatically.

Session Details

Date: 19/04/2008
Time: 00:00-00:00
Session name: 18th European Congress of Clinical Microbiology and Infectious Diseases
Location: Barcelona, Spain
Presentation type:
Back to top