Use of a new and simple microarray assay (Prove-it Sepsis) for rapid identification of bacteria involved in bone and joint infections directly from clinical samples

Abstract number: P1053

Allais E., Boisset S., Freydiere A-M., Benito Y., Ferry T., Lustig S., Vandenesch F., Laurent F.

Objectives: The microbiological diagnosis of bone and joint infections (BJIs) currently relies on standard cultures, which are time consuming and are subject to false negative results in case of fastidious organisms and previous antibiotic therapy. Prove-itTM Sepsis (Mobidiag^®) is a new commercial PCR and microarray assay that rapidly and reliably identifies more than 50 bacterial species present in positive blood cultures in less than 3 hours. The aim of our study was to evaluate the performance of this assay for the detection and identification of bacteria directly from BJI clinical samples compared to conventional culture and universal 16S rDNA PCR.

Methods: We selected 59 clinical samples (33 articular fluids and 26 biopsies), for which a microbiological diagnosis and identification had been established previously either by culture (C+) or PCR only (C-/PCR+). All these positive samples were chosen within the spectrum of bacteria included in the microarrays. Ten negative clinical samples (C-/PCR-) were also included. DNA was extracted with the automated MagNa Pure System (Roche) and the Prove-itTM assay was performed according to the manufacturer's protocol initially dedicated to blood culture. A human b-globin PCR was used to control DNA extraction and absence of inhibitors.

Results:

•C-/PCR- (n = 10): all samples were negative by using Prove-ItTM assay,

•C+ (n = 10): 7 samples were positive with concordant identification, whereas 3 remained negative (S. epidermidis, S. aureus, and mixed culture S. aureus+P. acnes),

•C-/PCR+ (n = 49):

–34 samples (69.4%) were positive with concordant identification with the following bacteria: S. aureus (n = 10), S. epidermidis (n = 10), S. pyogenes (n = 3), S. agalactiae (n = 3), S. dysgalactiae (n = 3), K. pneumoniae (n = 3), S. marcescens (n = 1) P. aeruginosa (n = 1),

–1 sample was positive indicating a mixture of 4 species, of which only one was identified by universal 16S PCR,

–14 samples were negative which corresponded to the following diagnoses by 16S rDNA PCR: 4 belonging to Staphylococcus sp, 7 to Streptococcus sp, 2 Enterobacteriaceae and 1 anaerobe.

Conclusion: These promising preliminary results indicate that this new microarray method allows a rapid and reliable detection of bacteria involved in BJIs. Providing additional probes relevant in BJIs (such as Kingella kingae, Propionibacterium acnes) being included in the array, it could replace conventional 16S rDNA PCR for clinical samples from patient suspect of BJIs.