Evaluation of the MRSA/SA ELITe MGB Assay for the Detection of Staphylococcus aureus in Bone and Joint Infections

ABSTRACT

Bone and joint infections represent a potentially devastating complication of prosthetic orthopedic joint replacement, thus requiring both rapid and appropriate antibiotic treatment. Staphylococcus aureus is one of the most common pathogens involved in this pathology. Being able to assert its presence is the first step of efficient patient management. This monocenter study evaluated the MRSA/SA ELITe MGB assay for the molecular detection of S. aureus and methicillin-resistant S. aureus (MRSA) in bone and joint biopsy specimens and synovial fluids. This test, together with conventional techniques, including standard cultures and the 16S rRNA amplification assay, was performed on 208 successive perioperative samples collected prospectively for 1 year obtained from 129 patients. Using conventional techniques, we detected a microbial pathogen in 76 samples from 58 patients, 40 of which were identified as S. aureus. The limit of detection (LOD) of the MRSA/SA ELITe MGB assay was experimentally determined for bone and joint biopsy specimens and synovial fluids using negative samples spiked with S. aureus ATCC 43300. The sensitivities of S. aureus detection with the MRSA/SA ELITe MGB assay were 82.5% (33/40 samples) and 97.5% (39/40 samples) using the manufacturer’s LOD and an experimentally determined LOD, respectively. Interestingly, using the osteoarticular specific LOD, 15 additional samples were determined to be positive for S. aureus DNA with the MRSA/SA ELITe MGB assay; in all cases, these samples were obtained from patients considered to be infected with S. aureus according to their clinical and microbiological records. The results were available within 24 h, which could help to expedite therapeutic decisions.

INTRODUCTION

Prosthetic joint replacement is a highly effective intervention that significantly improves patients’ quality of life, providing symptom relief, restoration of joint function, mobility, and independence. However, bone and joint infection (BJI) is a potentially devastating complication of this surgery, with occurrence rates of up to 1% of hip replacements (1) and 1 to 3% of knee replacements (2).

Staphylococcus aureus and non-aureus staphylococci are the main causative pathogens of periprosthetic BJIs, with 29% of cases attributed to each of them (3). While S. aureus mostly causes early infections, non-aureus staphylococci lead to chronic periprosthetic BJIs (4). Among the S. aureus strains responsible for BJIs, methicillin-resistant strains account for 11 to 62% of isolates according to patients’ characteristics and series location, with higher rates of methicillin resistance in North American studies (3, 5–7).

Culture of bone tissue and/or synovial fluid obtained during revision arthroplasty is usually considered the gold standard for diagnosing periprosthetic infection. Thus, two or more intraoperative cultures or the combination of a preoperative aspiration and an intraoperative culture that yield the same organism is considered to be sufficient evidence for BJI (in the case of S. aureus, one of these samples is enough) (8).

However, culture results can take up to 48 h (9), potentially delaying optimal treatment for patients with acute BJI. Direct detection techniques such as real-time (rt) PCR have a rapid turnaround time and have demonstrated higher sensitivity than culture in BJIs (10–13), allowing the rapid initiation of an appropriate antibiotic treatment. A fully automated PCR assay including the extraction step, such as the MRSA/SA ELITe MGB PCR assay on the ELITe InGenius platform, could help to speed up the diagnosis but has not been yet validated for osteoarticular samples.

The aim of our study was to evaluate the fully automated MRSA/SA ELITe MGB PCR assay’s performance in detecting S. aureus and methicillin-resistant S. aureus (MRSA) in perioperative samples obtained during orthopedic surgery compared to culture and 16S rRNA PCR techniques.

MATERIALS AND METHODS

Patients and samples.

Patients undergoing orthopedic surgery were included from June 2019 to July 2020 at the University Hospital of St-Etienne, France. The study was approved by the Institutional Review Board of the University Hospital of St-Etienne (IRBN 1302020/CHUSTE). Samples were collected prospectively during surgery regardless of the clinical status (suspicion of infection or routine surveillance during prosthetic joint replacement surgery). Synovial fluids were collected by the operating surgeon and transferred into a sterile vial (Qualibact 60-mL sterile vial; CEB, France). Tissue specimens were collected separately in the same sterile vials and then transferred into conical tubes (50-mL polypropylene conical Falcon tube; Corning Science, Mexico) containing saline solution in the laboratory. When several samples were collected simultaneously by the operating surgeon, only one synovial fluid and one tissue specimen were included. These samples were selected according to the highest leukocyte determination after direct Gram staining.

Microbiological culture.

Tissue specimens were ground (Ika Ultra Turrax T18 digital dispersing device) prior to inoculation on the different media. A 10-μL volume of each specimen was streaked onto agar plates using sterile loops. Blood agar plates (COS; bioMérieux, Marcy-l’Etoile, France) and chocolate agar plates (PVX; bioMérieux) were incubated at 35°C for 48 h under aerobic conditions and under 5% CO₂, respectively. A 50-μL volume of each specimen was also inoculated into a culture broth (Schaedler broth, 0.02%; bioMérieux) and incubated at 35°C for up to 14 days. All of the microorganisms were identified using matrix-assisted laser desorption ionization–time of flight mass spectrometry (Microflex Smart LS; Bruker, Brenem, Germany). Antimicrobial susceptibility tests (ASTs) were done using a Vitek 2 instrument (AST P631 card; bioMérieux) and interpreted according to the CA-SFM/EUCAST guidelines v1.0. 2019 (14).

16S rRNA PCR assay.

DNA extracts from the MRSA/SA ELITe MGB PCR assay were used to perform universal 16S rRNA PCR as previously described (15). PCR products were analyzed by capillary electrophoresis (DNA 7500 kit, 5067-1507; Agilent, France). Samples yielding a single band on electrophoresis were then subjected to Sanger sequencing analysis (DTCS quick start kit; Beckman Coulter). Sequences were blast searched using the Bioinformatics Bacteria Identification (BIBI) database (16). Sequence similarities greater than 96 and 98%, respectively, were used to define the identification at the genus and species levels. PCR results were considered valid if the β-globin gene, which was used as a control, was correctly amplified in each sample.

MRSA/SA ELITe MGB PCR assay.

The MRSA/SA ELITe MGB fully automated real-time PCR assay was performed according to the manufacturer’s instructions (MRSA/SA ELITe MGB; Elitech, Turin, Italia) using the ad hoc instrument (ELITe Ingenius; Elitech). The tissue samples were first ground as described for culture. A 200-μL volume of sample was pretreated with 50 μL of proteinase K (Eurobio Scientific, Les Ulis, France) for at last 2 h, and then a 200-μL volume transferred into the sample tube of MRSA/SA ELITe MGB PCR assay (Elitech). This commercially available assay detects sequences within the ldh1 gene of S. aureus for the species identification and the mecA and mecC genes for methicillin resistance, as well as a sequence of an exogenous external control called CPE. Results were considered invalid if no target was amplified and the CPE was not detected or detected below the manufacturer’s recommended cycle threshold (C_T) of 37.

LOD determination.

The limit of detection (LOD) was determined by serial 5-fold dilutions of the MRSA strain ATCC 43300 spiked in synovial fluids and fluidified bone and joint biopsy specimens, with a final bacterial concentration ranging from 5 × 10³ CFU/mL to 1.6 CFU/mL. The concentration of the spiked specimen was quantified on blood agar plate using an automated plater (easySpiral Dilute; Interscience, St-Nom-la-Bretèche, France) and a colony counter (Scan 1200; Interscience) (17). Then, 200-μL portions of each 5-fold dilution were tested in 10 replicates on either synovial fluid or fluidified tissue specimen using the MRSA/SA MGB PCR assay. The LOD relative to the value corresponding to a probability of 0.95 of detecting the evaluated strain (18) was determined by probit regression analysis (19).

Interpretation of results.

Samples were considered positive for S. aureus if at least one microbiological culture or 16S rRNA PCR assay identified this bacterium. Samples determined to be positive with the MRSA/SA ELITe MGB PCR assay and negative by both culture and the 16S rRNA PCR assay were further analyzed by a specific S. aureus assay targeting the femA gene (20), and the patient’s medical records were reviewed by a clinical microbiologist and an infectious diseases specialist. The femA PCR was performed on the DNA extracted by using the InGenius instrument, as previously described (20). According to recent IDSA proposals (8), BJI was diagnosed when at least one of the following criteria was evident: (i) the presence of a sinus tract that communicates with the prosthesis is definitive evidence of BJI; (ii) the presence of acute inflammation, as seen on histopathologic examination of periprosthetic tissue at the time of surgical debridement or prosthesis removal, as defined by the attending pathologist, is highly suggestive evidence of BJI; (iii) the presence of purulence without another known etiology surrounding the prosthesis is definitive evidence of BJI; and (iv) two or more intraoperative cultures or the combination of a preoperative aspiration and an intraoperative culture that yield the same organism (indistinguishable based on common laboratory tests, including genus and species identification or identical antibiogram) may be considered definitive evidence of BJI. The growth of a virulent microorganism (e.g., S. aureus) in a single specimen from a tissue biopsy or synovial fluid may also represent BJI (8).

Statistical analysis.

The statistical analyses were performed with Medcalc statistical software version 19.5.3 (MedCalc Software, Ltd., Ostend, Belgium). P values below 0.05 were considered statistically significant.

RESULTS

Patients and samples.

A total of 240 osteoarticular samples from 145 patients were included from June 2019 to July 2020 (Fig. 1). Twenty-one samples (8.8%) were excluded due to invalid MRSA/SA ELITe MGB PCR results, including four samples determined to be positive for S. aureus by culture or 16S rRNA PCR, and 11 samples (4.6%) were excluded due to invalid 16S rRNA PCR results. In these 32 cases, not enough material was available to allow retesting (Fig. 1). A total of 208 samples, including 71 synovial fluid specimens and 137 tissue specimens, originating from 129 patients were analyzed. The median age of patients was 68 years, and the male/female ratio was 1.30. Perioperative samples were collected the from hip (n = 81; 38.9%), knee (n = 60; 28.8%), upper leg (n = 15; 7.2%), lower leg (n = 15; 7.2%), ankle (n = 10; 4.8%), shoulder (n = 8; 3.8%), foot (n = 8; 3.8%), wrist (n = 5; 2.4%), elbow (n = 3; 1.4%), and upper arm (n = 3; 1.4%). Among the 208 samples, 166 (79.8%) were collected from patients with implanted osteoarticular prosthesis, and 95 (45.7%) were obtained from patients who had received antimicrobial agents up to 14 days prior to the surgery.

FIG 1.

Processing of samples and endpoint assessment used for the evaluation of the MRSA/SA ELITe MGB PCR assay evaluation for the detection of S. aureus in osteo-articular samples. Asterisks: *, including 13 samples with 16S rRNA PCR negative, 3 samples without 16S rRNA PCR result due to insufficient quantity, and 1 sample with 16S rRNA PCR not interpretable; **, including 13 samples with 16S rRNA PCR negative, and 2 samples with 16S rRNA PCR not interpretable. IQ, insufficient quantity; NI, not interpretable; neg., negative; pos., positive.

Of the 208 samples included, cultures and/or 16S rRNA PCR assay were able to diagnose an infection in 76 (36.5%) samples, including 74 (97.4%) monomicrobial and two polymicrobial (2.6%) infections. The microorganisms isolated from these 76 samples consisted of 40 S. aureus strains (52.6%), including 5 MRSA strains (12.5%), 18 non-aureus staphylococci strains (23.7%; including 9 methicillin-resistant isolates [50.0%]), and 18 strains belonging to other species (23.7%) (see Table S1). Among the 40 S. aureus-positive samples, 17 were detected by cultures only, 12 were detected by 16S rRNA PCR assay only, and 11 were detected by both techniques. Overall, cultures and 16S rRNA PCR were found to detect the presence of S. aureus in 70% (28/40) and 57.5% (23/40) of the samples, respectively (Table 1).

TABLE 1.

Comparison of culture, 16S rRNA PCR, and MRSA/SA ELITe MGB techniques for the detection of S. aureus in osteoarticular samples

Technique	Sample statusa (n)
	S. aureus positive	S. aureus negative	Se (%)b	Sp (%)b
Culture
S. aureus positive	28	0	70	100
S. aureus negative	12	168
16S rRNA PCR
S. aureus positive	23	0	57.5	100c
S. aureus negative	17	165
MRSA/SA ELITe MGB
S. aureus positive	39	15	97.5	91.1
S. aureus negative	1	153

^aThe sample status was considered positive for S. aureus if the bacterial culture or 16S rRNA PCR result was positive; otherwise, the sample status was considered negative for S. aureus. The femA PCR was not used to assess samples status.

^bSe, sensitivity; Sp, specificity.

^cThree samples were not tested due to insufficient quantity.

LOD of the MRSA/SA ELITe MGB PCR assay.

The manufacturer recommends using a 35 C_T as a cutoff point for the detection of the ldh1 target, and a 37 C_T for the detection of the mecA/mecC target, based on data collected from nasal swabs or blood samples. Analysis of the data set using the manufacturer cutoff points revealed that only 33 of the 40 positive S. aureus samples were detected by the MRSA/SA ELITe MGB PCR assay, which resulted in a sensitivity of 82.5% for the detection of S. aureus. However, the MRSA/SA ELITe MGB PCR assay amplified the ldh1 target for 6 of the 7 false-negative results, with a C_T that was slightly above 35. Therefore, a new cutoff point was determined using the ATCC 43300 MRSA strain, yielding a LOD of 161 CFU/mL (P = 0.0001) for S. aureus and 133 CFU/mL (P = 0.0001) for the mec target (Fig. 2; see also Fig. S1 in the supplemental material). These new LODs correspond to C_T values of 37.5 for ldh1 and 38.0 for mecA/mecC, respectively. All of the data presented here were analyzed using those cutoff values.

FIG 2.

Experimentally determined LOD of MRSA/SA ELITe MGB PCR assay. The LOD of S. aureus (ldh1 target) (A and C) and the LOD of methicillin resistance (mec target) (B and D) were determined using serial 5-fold dilutions of the methicillin-resistant S. aureus strain ATCC 43300 spiked in synovial fluids and fluidified bone and joint biopsy specimens.

Performance of the MRSA/SA ELITe MGB PCR assay for the detection of S. aureus in osteoarticular samples.

The results of MRSA/SA ELITe MGB PCR assay for the detection of S. aureus are presented in Table 1. The MRSA/SA ELITe MGB PCR assay was able to detect 39 of the 40 samples positive for S. aureus in culture and/or in 16S PCR, which gives a sensitivity of 97.5% (95% confidence interval [CI] = 86.8 to 99.9%) and a specificity of 91.1% (95% CI = 85.7 to 95.0%). For the sample missed by the MRSA/SA ELITe MGB PCR assay, the growth of S. aureus could only be detected after a broth enrichment culture of 14 days. In our lab, due to the overnight treatment of the samples, the results of the MRSA/SA ELITe MGB PCR assay were obtained within 24 h, whereas those obtained using species identification by culture were obtained within 24 to 72 h and 1 day later for the antimicrobial susceptibility test. The MRSA/SA ELITe MGB PCR assay was able to detect S. aureus 1 day earlier in approximately half of the cases (data not shown).

Furthermore, 15 samples originating from 10 patients were determined to be positive for S. aureus using the MRSA/SA ELITe PCR assay (median [range] C_T value of 33.0 [30.8 to 37.5]), while both reference tests (culture and 16S rRNA PCR assay) remained negative. These discrepant results were further analyzed by using the result of a specific PCR targeting the femA S. aureus-specific gene and by reviewing the patients’ medical records (Table 2; see also Table S2). Overall, these data revealed that each of the 15 samples was isolated from a patient exhibiting a documented S. aureus infection, with either a positive femA PCR test or previous microbiological and/or clinical evidence of S. aureus infection (Table 2).

TABLE 2.

Patients and samples determined to be positive for S. aureus using the MRSA/SA ELITe MGB PCR assay and negative by culture and 16S rRNA PCR

Patient	Sample	CT of MRSA/SA ELITe MGB assaya	femA PCR resultb	Detection of S. aureus by culture or 16S rRNA PCRc			Statusd
				Prior SA+ sample (days from detection)	SA+ samples during the surgery	Ulterior SA+ sample (days to detection)	Clinical	Antibiotic
1	1	31.1	+	+ (16)	0/5	–	+	+
	2	33.4	+
2	1	36.0	+	–	3/6	–	+	–
3	1	34.1	–	+ (15)	0/7	–	+	+
4	1	33.1	+	+ (22)	0/4	–	+	+
5	1	37.5	IQ	+ (20)	0/7	–	+	+
	2	31.1	IQ
6	1	32.7	IQ	+ (126)	0/3	–	+	+
7	1	33.9	IQ	+ (42)	0/5	+ (60)	–	+
8	1	30.9	+	+ (15)	0/5	–	–	–
	2	35.5	–
9	1	30.8	+	+ (18)	0/3	+ (15)	+	+
	2	32.3	+
	3	32.7	IQ
10	1	33.9	+	+ (26)	0/2	–	+	+

^aCycle threshold (C_T) value for S. aureus detection were determined. C_T values of <37.8 are considered positive.

^bIQ, insufficient quantity.

^cSA⁺, S. aureus positive.

^dThe clinical status was considered positive if the patient was determined to be septic by the referent infectious disease specialists at the time of surgery. The antibiotic status was considered positive if the patient was receiving treatment against staphylococci up to 14 days prior or at the time of the surgery.

Performance of the MRSA/SA ELITe MGB PCR assay for the detection of methicillin resistance in S. aureus.

By comparison to the AST results of the S. aureus strains, the sensitivity and specificity were 100% (95% CI = 47.8 to 100%) and 97.5% (95% CI = 94.3 to 99.2%), respectively. All five samples with MRSA isolated in the culture were correctly detected by the MRSA/SA ELITe MGB PCR assay. In addition, the latter assay was also able to detect MRSA in four samples that were detected neither by culture nor by 16S rRNA PCR assay. These four discrepant samples were collected from two patients. The first patient had been previously diagnosed with a chronic BJI due to a MRSA strain, with a positive sample 19 days before, and was receiving antistaphylococcal treatment at the time of surgery. The second patient was determined to be infected by a methicillin-susceptible S. aureus strain (detected in culture) at the time of the study, but clinical failure of the specific treatment initiated led to a second surgery 4 months later, during which samples revealed the presence of a methicillin-resistant non-aureus staphylococcal isolate (S. epidermidis). In addition, 13 samples from 9 patients were determined to be positive for mecA/mecC only; 7 of these 13 samples were positive in culture with Staphylococcus non-aureus isolates, with concordant results for methicillin resistance in 6 of them (see Table S1 in the supplemental material).

DISCUSSION

No consensus exists about the best choice for techniques to be used in the diagnosis of osteoarticular infection (number of culture media, duration of culture, and best indication for molecular methods). A classical approach would be to use a combination of culture media, specific S. aureus PCR, and broad-range 16S rRNA gene PCR, with several limitations to these techniques, such as turnaround time (for culture) and lower sensitivity and specificity, as well as failure to detect antimicrobial resistance (for 16S rRNA PCR) (21–23). Specific S. aureus PCRs have been found highly reliable in detecting S. aureus and MRSA, with higher sensitivity compared to culture and 16S rRNA PCR in all kinds of matrices, including osteoarticular samples (11–13, 24, 25).

In this context, we evaluated the MRSA/SA ELITe MGB PCR assay for the detection of S. aureus and MRSA directly in perioperative osteoarticular samples. A total of 240 samples were initially collected; however, 32 had to be excluded from the study: 21 samples (8.8%) were too viscous and yielded invalid MRSA/SA ELITe MGB PCR assay results. These issues arose in the earlier phase of the study, and the pretreatment period was extended (from a 2-h proteinase K [PK] treatment initially to an overnight period), which reduced the number of invalid results. Eleven (4.6%) more samples were excluded due to invalid 16S rRNA PCR results and insufficient quantity to allow retesting. The presence of PCR inhibitors in synovial fluids, such as hem or glycoproteins, has been noted in most of the studies analyzing those viscous specimen (26–28). It has been suggested to use pelleted fluids instead of crude specimen to minimize inhibitors (29). Unfortunately, the latter protocol could not be applied to the invalid samples of our study due to the low amounts of specimen.

Among the 208 samples included, 76 (35.6%) were considered microbiologically positive, including 40 positive for S. aureus using culture and/or 16S rRNA PCR assay. The initial analysis of the data set showed that, using C_T recommended by the manufacturer, only 33 positive-S. aureus samples would have been detected by the MRSA/SA ELITe MGB PCR assay, which would have resulted in a sensitivity of 82.5% for the detection of S. aureus. Therefore, these initial C_T values, which were derived from testing blood and nasal swab samples, appeared to not be adapted for testing bone and joint specimens, in which bacterial loads are expected to be lower. The determination of the new C_T values of 37.5 for ldh1 and 38.0 for mecA/mecC proved to be more fitting and showed more accurate results. Indeed, 39 (97.5%) of the 40 S. aureus-positive samples were now being detected by the MRSA/SA ELITe MGB PCR assay. The false-negative sample came from a patient who was not under antibiotic treatment at the time of surgery. Regarding this specific sample, initial culture and 16S rRNA PCR remained negative, and a S. aureus strain was isolated only after 14 days from the enrichment broth, suggesting a very low bacterial inoculum.

Interestingly, there were also 15 positive MRSA/SA ELITe MGB PCR results (positive for S. aureus but negative for mec) for which the reference tests remained negative. All C_T values of the MRSA/SA ELITe MGB PCR assay were unambiguously positive except for one sample. These 15 discrepant results were obtained from 10 different patients. The analysis of the medical records of these patients revealed that nine of them were suffering from chronic S. aureus BJIs, and the last one was suffering from acute BJI, with S. aureus being detected in three of six perioperative samples. Eight of the patients were receiving antistaphylococcal treatment, which can explain the negativity of cultures. Furthermore, traditional tissue culture might not be able to recover bacteria captured in biofilms, as well as bacteria lysed during specimen transport, because of the release of locally delivered antibiotics at the time of prosthesis removal (30, 31). It could also be argued that the persistence of dead bacterial DNA in the sample could explain these discrepancies, since it has been demonstrated that bacterial DNA can still be detected up to 43 days after successful treatment (32). These results underline the complexity of diagnosing BJIs with conventional techniques. In our study, the MRSA/SA ELITe MGB PCR assay outperformed culture and 16S rRNA PCR, which argues for the use of specific real-time PCRs that help to provide more accurate results in detecting S. aureus DNA in synovial fluids and bone and joint specimens (33, 34).

The antimicrobial resistance of the causative microorganism can be part of the decision process regarding the different types of strategies considered (early versus delayed reimplantation, cemented versus noncemented prosthesis, and antibiotic-containing cements for the fixation of the prosthesis) (8). Concerning methicillin resistance, all MRSA isolates identified by culture were also correctly identified by the MRSA/SA ELITe MGB PCR assay. Moreover, the MRSA/SA assay yielded four more MRSA-positive results concerning two different patients. The first one was suffering from a chronic infection due to a MRSA isolate that had been previously diagnosed and treated. The second patient was later found suffering from chronic BJI caused by both S. aureus and a methicillin-resistant S. epidermidis, which could explain the presence of the mecA/mecC gene amplification signal (35, 36). Discrepant values (ΔC_T > 2) between the S. aureus and mecA/mecC targets has been found to suggest a mix of methicillin-susceptible S. aureus and methicillin-resistant coagulase-negative staphylococci (37), but this was not the case in our study. Furthermore, the detection of a mec gene without S. aureus DNA amplification should draw attention to the possible presence of non-aureus staphylococci, which could then require further investigations using other microbiological techniques, and if confirmed, lead to a reevaluation of the patient’s antibiotic treatment. Since this study was neither designed to include enough of these strains, nor provide a confirmation method for these microorganisms, we suggest the MRSA/SA ELITe MGB PCR assay to be evaluated in detecting methicillin resistance in non-aureus staphylococci. Likewise, even if all MRSA-infected samples of our cohort were correctly detected by the MRSA/SA MGB PCR assay, a study including more patients infected by MRSA would be valuable. A limit of the present study is the fact that the femA-specific S. aureus PCR assay was performed only for specimens determined to be positive exclusively by the MRSA/SA ELITe MGB PCR assay, which may slightly overestimate the sensitivity of the latter. Finally, even with an overnight treatment of the samples, the PCR assay was shown to provide an earlier detection of pathogen in most cases. In addition, it allowed us to detect 15 more staphylococcal infections in culture/16S rRNA-negative patients with other stigmata of infection by S. aureus.

To conclude, this study showed that the MRSA/SA ELITe MGB PCR assay was accurate for detecting both S. aureus and methicillin-resistant S. aureus DNA in perioperative bone and joint samples. As with other fully automated specific S. aureus PCRs, the rapid turnaround time and high sensitivity and specificity could allow faster treatment initiation or adaptation in patients found to be positive by this test and thus could dramatically improve the diagnosis and clinical management of patients suffering from BJI (11, 25, 34). Moreover, the MRSA/SA ELITe MGB PCR assay allows the possibility to perform complementary molecular tests such as the 16S rRNA PCR assay or specific PCRs on the remaining DNA extract, as exemplified by the detection of Kingella kingae DNA in pediatric BJIs.