Back

Evaluation of commercially available molecular and culture-based assays for rapid detection of methicillin-resistant Staphylococcus aureus

Abstract number: P1558

Sabiiti W., Cortiñas Abrahantes J., Lammens C., Molenberghs G., Aerts M., Goossens H., Malhotra-Kumar S.

Background: The need for rapid methods to accurately detect methicillin-resistant Staphylococcus aureus (MRSA) is widely acknowledged. We assessed 3 commercial assays – 2 molecular, GeneOhm (BD Diagnostics) and GeneXpert (Cepheid) – and 1 culture-based, Baclite (3M) – for their ability to correctly identify MRSA utilising well-characterised isolates, either pure or in mixtures, at varying concentrations.

Methods: Fifty-two isolates (27 MRSA of which 9 were animal strains; 25 non-MRSA of which 8 were MRCoNS, 5 MSSA, and 12 enterococci, Enterobacteriaceae and Acinetobacter spp.), and 21 mixtures of these isolates were tested on the three assays following manufacturer's recommendations. As pure strains, non-MRSA were tested at 1 dilution (105 CFU/ml for GeneOhm and GeneXpert, and 108 CFU/ml for Baclite according to manufacturer's instructions), while serial dilutions (101 to 103 or 104 CFU/ml) of the 27 MRSA strains were tested to determine the limit of detection (LoD) of each assay. Mixtures were made using isolates at varying concentrations according to CLSI guidelines. Moreover, 10 ml of each sample was also simultaneously spiral-plated on blood agar with 6 mg/ml cefoxitin and in case of MRSA positive samples, colony counts were made after overnight incubation. Mean sensitivity and specificity and confidence intervals (CIs) were estimated for each assay by logistic regression model using a penalised likelihood approach.

Results: GeneOhm showed the highest sensitivities for both isolates and mixtures at lower (103 CFU/ml) concentrations (Table).

Of the 27 MRSA tested, 2 strains that could not be detected at 103 CFU/ml were positive at 104 CFU/ml, giving a 100% positivity for GeneOhm at the latter concentration. Mean sensitivity for GeneXpert was geqslant R: gt-or-equal, slanted100% at 104 CFU/ml with narrow CIs indicating high precision of this parameter estimate. Baclite showed the highest mean specificities for both mixtures and isolates. False-positive results with the 2 molecular tests were primarily due to MRCoNS. Minimum LoD for both molecular assays was similar for isolates, while GeneXpert could detect upto 27 CFU/ml of MRSA in mixtures.

Conclusions: A general increase in mean sensitivities of all three assays was observed with increasing MRSA concentrations. GeneOhm and GeneXpert showed comparable performance in terms of sensitivity, specificity and LoD.

Table. Performance of three commercial assays for rapid detection of MRSA

AssayMean sensitivity (95% CI)
at MRSA concentration
Mean specificity (95% CI)Minimum limit of detection
 103 CFU/ml104 CFU/ml  (CFU/ml)
Using isolates as samples
GEneOhm™ MRSA (BD Diagnostics, BE)93.7% (80.2–98.2)ND*77.2% (60.0–88.5)140
GeneXpert™ MRSA (Cepheid, FR)86.8% (75.2–93.5)99.5% (97.9–99.9)82.3% (55.2–94.6)140
Baclite™ MRSA (3M, USA)40.1% (26.1–55.9)93.8% (86.7–97.2)98.2% (90.5–99.7)410
Using mixtures as samples
GEneOhm™ MRSA (BD Diagnostics, BE)97.6% (90.6–99.4)ND**55.6% (31.9–77.1)44
GeneXpert™ MRSA (Cepheid, FR)94.8% (88.3–97.8)99.8% (99.2–100)63.2% (38.7–82.4)27
Baclite™ MRSA (3M, USA)26.9% (14.8–43.8)89.3% (77.7–95.2)95.3% (81.7–98.9)2400
*ND, Not determined. Only samples negative at the lower concentration (n = 2) were tested at this concentration.
**Not determined as all MRSA-positive mixtures were detected at the lower concentration.

Session Details

Date: 16/05/2009
Time: 00:00-00:00
Session name: 19th European Congress of Clinical Microbiology and Infectious Diseases
Subject:
Location: Helsinki, Finland, 16 - 19 May 2009
Presentation type:
Back to top