Abstract
Septic arthritis (SA) is a rheumatologic emergency associated with significant morbidity and mortality. Delayed or inadequate treatment of SA can lead to irreversible joint destruction and disability. Current methods of diagnosing SA rely on synovial fluid analysis and culture which are known to be imprecise and time-consuming. We report a novel adaptation of a probe-based real-time PCR assay targeting the 16S rRNA gene for early and accurate diagnosis of bacterial SA. The assay algorithm consists of initial broad-range eubacterial detection, followed by Gram typing and species characterization of the pathogen. The platform demonstrated a high analytical sensitivity with a limit of detection of 101 CFU/ml with a panel of SA-related organisms. Gram typing and pathogen-specific probes correctly identified their respective targets in a mock test panel of 36 common clinically relevant pathogens. One hundred twenty-one clinical synovial fluid samples from patients presenting with suspected acute SA were tested. The sensitivity and specificity of the assay were 95% and 97%, respectively, versus synovial fluid culture results. Gram-typing probes correctly identified 100% of eubacterial positive samples as to gram-positive or gram-negative status, and pathogen-specific probes correctly identified the etiologic agent in 16/20 eubacterial positive samples. The total assay time from sample collection to result is 3 h. We have demonstrated that a real-time broad-based PCR assay has high analytical and clinical performance with an improved time to detection versus culture for SA. This assay may be a useful diagnostic adjunct for clinicians, particularly those practicing in the acute care setting where rapid pathogen detection and identification would assist in disposition and treatment decisions.
Septic arthritis (SA) is a rheumatologic emergency associated with significant morbidity and mortality (6, 9). Delayed or inadequate treatment of SA can lead to irreversible joint destruction with subsequent disability. Accordingly, prompt diagnosis and early initiation of therapy are critical in improving the outcome (7).
The diagnosis of SA in the acute care setting is challenging because of the relatively poor sensitivity and specificity of clinical examination findings and lack of a rapid reliable diagnostic assay. Further, overreliance on conventional laboratory tests for synovial fluid analysis is hindered by the relatively poor performance characteristics of these methods (11, 12, 16). In particular, the sensitivity of Gram staining has been reported in the range of 29% to 50% (3, 4), and the sensitivity of culture may be only 82% (9). Lack of a rapid and accurate diagnostic tool results in acute care clinicians often choosing the conservative approach of hospital admission and empirical broad-spectrum antibiotics for patients with suspected SA. The benefits of this management strategy may be offset, however, by added costs and potential iatrogenic complications associated with unnecessary treatment and hospitalizations, as well as increased rates of antimicrobial resistance. A sensitive, specific diagnostic assay, which allows for rapid definitive diagnosis of SA and directed therapeutic intervention, would thus be invaluable in the acute care setting.
The use of PCR amplification of 16S rRNA gene has been proposed for broad-range detection of eubacteria in synovial fluid (17, 18). However, nearly all broad-based PCR assays reported thus far involve laborious time-consuming postamplification processing (e.g., gel electrophoresis, Southern blotting, or sequencing), making them impractical for routine clinical use (8, 14, 17, 18). For example, to date, the largest recent study using broad-based real-time PCR study for diagnosis of SA demonstrated high sensitivity and specificity but relied on sequencing for definitive pathogen identification (5).
Exploitation of both the highly conserved and hypervariable sequences within the 16S rRNA gene permits design of a platform capable of both eubacterial detection and specific pathogen identification in a single rapid detection platform. We report a novel adaptation of a previously described probe-based real-time PCR assay for early diagnosis and characterization of SA. The assay consists of initial broad-range eubacterial detection targeting the 16S rRNA gene followed by simultaneous parallel PCR analyses, permitting identification of gram-positive and gram-negative type and definitive pathogen characterization of the species. Diagnostic accuracy of our assay was evaluated against conventional culture-based methods using synovial fluid samples from patients presenting with to a tertiary care hospital with suspected acute SA.
MATERIALS AND METHODS
Bacterial species and mock samples.
Thirty-six clinically relevant bacterial organisms and DNA, including the six most common SA-related organisms were obtained from the American Type Culture Collection (ATCC) (Manassas, VA) or the Johns Hopkins Hospital clinical laboratory (Division of Medical Microbiology, Johns Hopkins School of Medicine, Baltimore, MD).
A single isolated colony of each organism was inoculated in tryptic soy broth (TSB) (Becton Dickinson, Sparks, MD) and incubated at 37°C overnight. To determine the limit of detection (LOD), serial dilutions of each of the SA-related organisms (Streptococcus pneumoniae, Staphylococcus aureus, Streptococcus agalactiae, Staphylococcus epidermidis, Escherichia coli, and Neisseria gonorrhoeae) were spiked into culture-negative and DNA-free synovial fluid samples. These mock samples were processed using the procedure (“Extraction of DNA”) described below. LOD was calculated based on CFU per milliliter. DNA was also extracted from our panel of clinically relevant organisms to test the analytical specificity of our gram-positive and gram-negative probe, as well as all pathogen-specific probes, as shown in Table 1.
TABLE 1.
Probe sequences designed in-house
| Target organism | Probe sequencea |
|---|---|
| Gram-positive organisms | 5′-FAM-AGGTGGTGCATGGTTGTCGTCAGC-3′-MGB |
| Staphylococcus aureus | 5′-FAM-CCTTTGACAACTCTAGAGATAGAGCCTTCCC-3′-MGB |
| Staphylococcus epidermidis | 5′-TET-AAAACTCTATCTCTAGAGGGGCTAGAGGATGTCAAG-3′-MGB |
| Streptococcus pneumoniaeb | 5′-TET-TCACCTCTGTCCCGAAGGAAAACTCTATCTCTAGA-3′-MGB |
| Streptococcus agalactiae | 5′-FAM-TGCTCCGAAGAGAAAGCCTATCTCTAGGCC-3′-MGB |
| Gram-negative organisms | 5′-VIC-ACAGGTGCTGCATGGCTGTCGTCAGCT-3′-MGB |
| Escherichia coli | 5′-FAM-ACATTCTCATCTCTGAAAACTTCCGTGGATGTC-3′-MGB |
| Neisseria gonorrhoeae | 5′-FAM-TCTCCGGAGGATTCCGCACATGTCAAAA-3′-MGB |
| Uniprobe | 5′-VIC-CACGAGCTGACGACARCCATGCA-3′-MGB |
Abbreviations: FAM, 6-carboxyfluorescein; MGB, minor groove binder; TET, tetrachlorofluorescein; VIC, Applied Biosystems dye.
The probe may cross-react with the mitis subgroup of the viridans group streptococci (10).
Clinical samples and study location.
Clinical synovial fluid samples were derived from patients who presented with suspected acute SA to one of three clinical sites (the Emergency Department, the Orthopedic Clinic, or the Rheumatology Clinic) of the Johns Hopkins University Hospital and The Johns Hopkins Bayview Medical Center, a large tertiary care hospital, from July 2006 to July 2007. One hundred twenty-one samples were obtained from the microbiology laboratories and were provided for research as “excess” deidentified specimens after the microbiology laboratories had performed standard microbiologic testing including cultivation. The study was approved by The Johns Hopkins Institutional Review Board.
“Excess” samples were processed as follows. (i) The samples were given a random study sample number and taken from the microbiology laboratory to the research laboratory where they were stored at −20°C for later DNA extraction and PCR analysis. (ii) A database which included the microbiology accession number and the random study number was created. (iii) The microbiology database was queried for culture results. (iv) The database was deidentified. (v) Samples were analyzed by PCR. (vi) PCR results were compared with microbiology culture results.
Extraction of DNA.
Synovial fluid samples were thawed at room temperature. Each 500-μl sample aliquot was centrifuged at 3,200 × g for 10 min in an Eppendorf 5415 D centrifuge (Westbury, NY), and the pellet was resuspended in 50 μl of molecular-grade water. In viscous samples which yielded negative internal positive controls (see “Positive, negative, and exogenous internal positive-control preparation” below), these samples were diluted with molecular-grade water (Roche Diagnostics, Basel, Switzerland) in sample:water ratios of 1:10, 1:100, 1:500, and 1:1000 for a final volume of 500 μl before processing. A 10-μl mixture of 1× (0.32 μg/μl) lysozyme (Sigma Aldrich, St. Louis, MO) and 1× (0.5 μg/μl) lysostaphin (Sigma Aldrich) was then added to the sample and incubated at 37°C for 20 min. One-microliter aliquot of 1× proteinase K (MagNA LC kit I; Roche Diagnostics, Indianapolis, IN) was added, and the sample was incubated at 65° C for 10 min. The sample was subjected to a freeze-thaw cycle for 10 min at −80° C and 5 min at 95° C. Samples were then sonicated in a Bransonic T9000 (Shelton, CT) for 10 min before undergoing PCR testing.
Design of primers and probes.
The target site within the 16S rRNA gene (which encompasses the hypervariable V6 region) and design of conserved primers (p891F and p1033R) and probe (Uniprobe) were as previously described (19). Gram-typing and SA-related pathogen-specific probe sequences are shown in Table 1. These probes were designed based on 16S rRNA sequence data obtained from GenBank and aligned with sequences from various clinically relevant bacterial species using the program ClustalW (http://www.ebi.ac.uk/Tools/clustalw2/index.html). The theoretical specificities of all designed primer and probe sequences were further analyzed using NCBI's BLAST (Basic Local Alignment Search Tool) program.
PCR master mix preparation.
Each PCR was performed in a total volume of 50 μl, which comprised 30 μl of PCR master mix and 20 μl of sample input. PCR master mix contained 25 μl of 2× Taqman universal PCR mix (PE Applied Biosystems, Foster City, CA) and 1.5 μl of 67 μM forward primer and reverse primer. The 2× Taqman universal PCR mix and the primers underwent an ultrafiltration step using a Microcon YM-100 centrifugal filter device (Millipore Corporation, Bedford, MA) by centrifugation at 3,200 × g for 10 min to remove potential exogenous background DNA contamination. Following ultrafiltration, an additional 1 μl of 2.5 units of Amplitaq Gold LD (PE Applied Biosystems, Foster City, CA) and 1 μl of 10 μM probe were added to make up the final master mix before sample was added. PCR was then performed using an ABI 7900 HT sequence detection system (PE Applied Biosystems, Foster City, CA). The cycling conditions used were as follows: preincubation at 50° C for 2 min, denaturation at 95° C for 10 min, and 50 repeats at 95° C for 15 s, annealing/extension temperature at 60° C for 60 s.
Positive, negative, and exogenous internal positive-control preparation.
Ultrapure water was used as nontemplate PCR negative control. Culture-negative synovial fluid samples were screened using our universal probe PCR assay. Samples with a threshold cycle (CT) value (see “Post-PCR analysis” below) equal to or higher than nontemplate PCR negative controls were pooled and established for use as a standard negative control. An exogenous internal positive control (IPC) (PE Applied Biosystems, Foster City, CA) was used on all clinical samples according to the manufacturer's instructions in order to rule out sample inhibition to PCR.
PCR assay algorithm.
Clinical synovial fluid samples were tested for the presence of eubacteria using Uniprobe PCR. Positive samples by the Uniprobe PCR were further analyzed with parallel PCRs using both gram-positive and gram-negative probes and our panel of pathogen-specific probes.
Post-PCR analysis.
Amplification data were analyzed by the SDS software (PE Applied Biosystems), which calculates ΔRn using the equation Rn(+) − Rn(−). Rn(+) is the emission intensity of the reports divided by the emission intensity of the quencher at any given time, whereas Rn(−) is the value of Rn(+) prior to amplification. Thus, ΔRn indicates the magnitude of the signal generated. The threshold cycle, or CT, is the cycle at which statistically significant increase in ΔRn is first detected. The CT is inversely proportional to the starting amount of target DNA. Amplification plots were generated by plotting ΔRn versus CT.
All clinical samples, standardized pooled negative control, and IPCs were performed in triplicate samples. The average and standard deviation for the pooled negative0control replicates from each run were calculated. Due to the potential for day-to-day interrun variability, the cutoff CT value for each run was defined as 3 standard deviations below the negative-control average (1). Any sample with a CT value higher than the cutoff value was considered PCR negative and vice versa.
The accuracy of the Uniprobe PCR was determined by the observed clinical sensitivity and specificity compared to conventional culture results. Ninety-five percent 95% confidence intervals (95% CI) for clinical sensitivity and specificity were estimated by the exact binominal test method.
Discordant analysis.
All samples with discordant findings between PCR and culture results were plated on 5% sheep blood agar plates (Becton Dickinson) to assess for bacterial growth. Any samples with growth on agar were sent to the Johns Hopkins Hospital clinical microbiology laboratory for identification. In addition, amplified PCR products from repeat PCR of the DNA extract from discordant clinical samples were sequenced in-house in the Johns Hopkins University CORE Genetics facility. Sequence match (100%) was performed by accessing the GenBank (www.ncbi.nlm.nih.gov) microbial database via the BLAST program.
RESULTS
LOD and analytical specificity.
The LOD of our Uniprobe PCR was determined by testing mock synovial fluid samples containing serially diluted organisms most commonly found in SA. Our Uniprobe PCR demonstrated a high analytical sensitivity with the LOD on the order of 101 CFU/ml for our panel of common SA-related organisms. We also evaluated the analytic specificity of our Gram-typing probes and our select panel of pathogen-specific probes by testing against DNA extracted from 36 clinically relevant bacterial organisms. All Gram-typing probes and pathogen-specific probes correctly identified their respective target organisms.
Clinical sensitivity and specificity.
One hundred twenty-one clinical synovial fluid samples were collected from patients with suspected SA and tested using our PCR assay. Among the samples collected, 21 were culture positive and 100 were culture negative. As shown in Table 2, 20 of 21 culture-positive samples tested positive by Uniprobe PCR, and 97 of 100 culture-negative samples tested negative. The calculated clinical sensitivity and specificity of Uniprobe PCR are 95.2% (95% CI, 76.2 to 99.9%) and 97.0% (95% CI, 91.5 to 99.4%), respectively. The agreement of Uniprobe PCR with culture was 96.7% (95% CI, 91.8 to 99.1%).
TABLE 2.
Uniprobe PCR results versus culture results with clinical joint fluid samples
| Uniprobe PCR result or parameter | No. of samples with the following culture result:
|
Total no. of samples | |
|---|---|---|---|
| Positive | Negative | ||
| Positive | 20 | 03a | 23 |
| Negative | 1a | 97 | 98 |
| Total no. of samples | 21 | 100 | 121 |
Please refer to Table 3 for further analyses of the discordant results.
Gram-positive and gram-negative probes were tested against our panel of 36 (23 gram-positive and 13 gram-negative) clinically relevant bacterial pathogens and were found to have 100% sensitivity and specificity. Of the 20 Uniprobe PCR-positive samples, our gram-positive and gram-negative probes were in complete concordance with culture results (Table 3); our pathogen-specific probes were concordant with culture results in 16/20 samples (13 isolates were identified as Staphylococcus aureus, 2 isolates were identified as Staphylococcus epidermidis, and 1 isolate was identified as a group B Streptococcus). The 97 samples that tested negative by Uniprobe PCR all had positive IPC results, indicating that there were no inhibitors to PCR present.
TABLE 3.
Concordant and discordant Uniprobe PCR and type-specific PCR results with clinical synovial fluid samples
| PCR result and sample | Culture resulta | PCR resulta | Organism identification by using the following probeb:
|
Comment(s) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Uni | GP | GN | STAU | STEP | STAG | STPN | ESCO | NEGO | ||||
| Concordant Uniprobe PCR and type-specific PCR results | ||||||||||||
| BAY-051 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BAY-062 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BAY-133 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BAY-160 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0003 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0026 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0039 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0077 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0078 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0084 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0094 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0096 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0115 | STAU | STAU | + | + | − | + | − | − | − | − | − | |
| BTW-J0079 | STEP | STEP | + | + | − | − | + | − | − | − | − | |
| BTW-J0098 | STEP | STEP | + | + | − | − | + | − | − | − | − | |
| BTW-J0102 | STAG | STAG | + | + | − | − | − | + | − | − | − | |
| Discordant Uniprobe PCR and type-specific PCR results | ||||||||||||
| BTW-J0069 | STAU | STEP | + | + | − | − | + | − | − | − | − | Sequenced; no quality data |
| BTW-J0019 | STAU | Negative | − | − | − | − | − | − | − | − | − | Grew in broth after 3 days |
| BAY-157 | Group G streptococcusc | No probe | + | + | − | − | − | − | − | − | − | Sequenced; group G streptococcus |
| BTW-J0030 | Viridans group streptococcid | Streptococcus pneumoniae | + | + | − | − | − | − | + | − | − | Sequenced; Streptococcus pneumoniae |
| BTW-J0031 | Viridans group streptococcid | Streptococcus pneumoniae | + | + | − | − | − | − | + | − | − | Sequenced; Streptococcus pneumoniae |
| BTW-J0049 | Negative | STEP | + | + | − | − | + | − | − | − | − | Sequenced; no quality data |
| BTW-J0086 | Negative | Positive | + | + | + | − | − | − | − | − | − | Sequenced; no quality data |
| BTW-J0101 | Negative | Positive | + | + | + | − | − | − | − | − | − | Sequenced; no quality data |
Abbreviations: STAU, S. aureus; STEP, S. epidermidis; STAG, S. agalactiae.
Abbreviations: Uni, Uniprobe; GP, gram-positive organisms; GN, gram-negative organisms; STAU, S. aureus; STEP, S. epidermidis; STAG, S. agalactiae; STPN, S. pneumoniae; ESCO, E. coli; NEGO, N. gonorrhoeae. Symbols: +, positive; −, negative.
BAY-157 was reported as group B; it was replated, recultured, and identified as group G streptococcus. Sequencing confirmed it as group G streptococcus.
BTW-J0030 and BTW-J0031 were reported as viridans group streptoccus. It was positive for our Streptoccus pneumoniae probe, and sequencing confirmed it as Streptoccus pneumoniae.
Discordant Uniprobe PCR results.
Four samples showed discordant culture and Uniprobe PCR results. Three (BTW-J0049, BTW-J0086, and BTW-J0101) were reported negative by culture but were positive by Uniprobe PCR (Table 3). Repeat culturing of these samples did not show any growth after 3 days of incubation. Sequencing of the PCR product did not yield quality data for organism identification. However, 1/3 samples tested positive with the gram-positive probe, and 2/3 samples were positive with both the gram-positive and gram-negative probes. One sample (BTW-J0019) was reported culture positive for S. epidermidis (grew only after 2 days) but negative by Uniprobe PCR. This sample was reported to grow only in culture broth but not by conventional plating. Repeat culturing of the sample revealed no growth after 3 days of incubation.
Discordant pathogen-specific PCR results.
Four of the 20 Uniprobe-positive, culture-positive samples had discordant results between conventional microbiological methods and our pathogen-specific PCR. One sample (BTW-J0069) was identified as S. aureus by culture but gave a positive test result only with our S. epidermidis probe. Sequencing studies showed no organism-specific quality data. One sample (BAY-157) which was reported as group B streptococcus (S. agalactiae) by culture was group B streptococcus PCR probe negative. Repeat culturing of this sample identified it as a group G streptococcus. Sequencing of the amplicon from this sample also confirmed that is was group G streptococcus. Both the sequencing results and the reculture results were concordant with our PCR probe-negative result but discordant with the initial culture findings (we do not have a group G streptococcus probe). Two samples (BTW-J0030 and BTW-J0031) were positive with Streptococcus pneumoniae probe, but they were reported as viridans group streptococci by the clinical microbiology laboratory.
Assay performance time.
The time to detection was 3 h, which included DNA extraction (70 min) and PCR amplification (110 min).
DISCUSSION
Septic arthritis is often clinically indistinguishable from other causes of joint inflammation. We demonstrated the potential utility of a real-time broad-based PCR assay which could be a useful adjunct for rapid, accurate diagnosis of SA in acute clinical settings. In comparison with other published assays involving broad-based PCR for SA (3, 6, 9, 12, 13), notable features of our platform include the following: (i) high sensitivity and specificity, (ii) capacity for early pathogen characterization (i.e., Gram stain specificity and species identification), and (iii) rapidity coupled with simple sample processing and identification.
Although PCR amplification has high theoretical sensitivity, clinical reliability of the assay requires sufficient quantities of DNA for amplification. Incorporation of a combination of chaotropic, thermal, and enzymatic inductions of cell lysis in the sample processing protocol allowed us to achieve high detection sensitivity via effective extraction of microbial DNA, even from the difficult-to-lyse cell walls of gram-positive organisms. The LOD of our Uniprobe PCR was comparable if not better than those of other reported PCR assays for synovial fluid samples (13, 8). Notably, only one PCR-negative, culture-positive sample was found in our study, and the culture may have resulted from a laboratory contamination event, since bacterial growth from this sample was detected only in culture broth, not by conventional plating.
We observed the presence of either PCR inhibitors or excess DNA, as evidenced by negative internal positive controls from some highly viscous joint fluid samples, which required predilution before testing positive by PCR. Similar findings were also reported by Rosey et al. (14).
The high specificity of our Uniprobe PCR assay was likely attributable to the methods we employed for minimizing exogenous eubacterial DNA contamination, which can be pronounced in broad-based 16S rRNA eubacterial PCR assays. We employed a previously reported decontamination measure using size-based ultrafiltration to reduce contaminating eubacterial DNA from component PCR reagents (19). Finally, samples were given a positive score by Uniprobe PCR only if their CT values were at least 3 standard deviations below the CT values of the nontemplate controls, minimizing the likelihood of false-positive results (1). Nevertheless, three Uniprobe PCR-positive but culture-negative cases were observed. Although we cannot exclude the possibility of false-positive PCR results, the relatively high bacterial load detected based on low CT values in these three samples suggested that these may have been truly positive samples. Potential explanations for the discrepant findings include superior sensitivity of PCR over culture methods or administration of antibiotics prior to sample collection. Although it was not possible to definitively explain the reason why sequencing of these samples did not yield quality data, a likely explanation may be the presence of multiple species in the prepared sample (either as contaminants or etiologic organisms), rendering sequencing problematic. Use of a third “resolver” test (e.g., PCR targeting an alternative pathogen-specific gene) may help adjudicate these discrepant cases in future analyses.
One of the design objectives of the assay is to obtain the most microbiological information rapidly to allow for early directed antimicrobial selection in the acute care setting. We recognize that there is a myriad of potential etiologic agents, including previously unrecognized ones, and using a probe-based approach for species identification does limit our detection scope to a finite number of anticipated pathogens. Therefore, in addition to our panel of specific probes to identify the “most common” SA-related pathogens, we have designed Gram-type-specific probes as an additional measure to characterize the Uniprobe-detected pathogens. Our Gram-type-specific probes demonstrated 100% specificity in both the test panel of organisms and all of the culture-positive clinical samples. Moreover, BLAST search against the microbial database from GenBank under the most stringent criteria confirmed 100% Gram specificity (data not shown). Our panel of six pathogen-specific probes was selected to detect the majority (∼80%) of etiologic agents responsible for SA (3, 7, 15).
One of the inherent limitations of exploiting 16S rRNA gene in designing probes for species identification is that even within our target region, which contains the highest degree of sequence variability (V6) (2), sequences between phylogenetically related species may still be identical. This is the case for S. pneumoniae and some viridans subgroup streptococci, including S. mitis, S. sanguis, and S. oralis (10). For definitive species identification in S. pneumoniae probe-positive cases, specific confirmatory molecular assays which query other genetic loci or culturing would need to be performed (20). Nonetheless, from a clinical practice standpoint, management decision in the acute care setting would not be affected in either case. We observed in at least one case (BAY-157) superior specificity of our genotyping approach over traditional phenotypic methods for species identification.
Minimizing the number of processing steps for sample preparation together with use of real-time PCR chemistry methods provided an assay which is rapid and relatively simple compared to traditional culture or PCR methods. The complete process (from specimen collection to target detection) can be achieved in approximately 3 h with use of the most up-to-date high-speed thermocyclers. This compares with 1 to 2 days minimum for routine culture (longer for fastidious organisms) The two-step PCR algorithm (first Uniprobe PCR is performed; if the Uniprobe PCR is positive, it is followed by Gram-specific and appropriate panels of pathogen-specific PCRs) offers potential cost savings, as it reduces unnecessary PCR testing associated with Uniprobe-negative cases or gram-positive or -negative positive results. The use of molecular beacons for direct enzyme-independent amplicon hybridization, as well as melting curve analysis of the amplicons, is under active investigation in our laboratory as a time- and cost-saving improvement of this assay.
In conclusion, our findings provide proof of concept that a real-time multiprobe eubacterial PCR assay can diagnose SA with speed and accuracy. The clinical applicability of our assay algorithm as a “molecular triage tool” in the Emergency Department could extend beyond SA detection. Prompt recognition and characterization of systemic bacterial infections in otherwise “sterile body fluids” in acutely ill febrile patients would be invaluable in routine clinical care. We envision that our PCR assay may potentially obviate the use of conventional cultures if negative Uniprobe PCR results can reliably predict the absence of SA; but in Uniprobe-positive cases, cultures should still be performed for confirmation and antimicrobial susceptibility testing. Further large-scale prospective studies are required for clinical validation of this approach, in addition to evaluation of the potential utility in acute care practice in regards to feasibility, cost-effectiveness, and turnaround time.
Acknowledgments
This work is supported by the Mid-Atlantic Regional Centre of Excellence grant NIH AI-02-031 from NIAID.
Footnotes
Published ahead of print on 27 February 2008.
REFERENCES
- 1.Bobo, L., B. Munoz, R. Viscidi, T. Quinn, H. Mkocha, and S. West. 1991. Diagnosis of Chlamydia trachomatis eye infection in Tanzania by polymerase chain reaction/enzyme immunoassay. Lancet 338847-850. [DOI] [PubMed] [Google Scholar]
- 2.Chakravorty, S., D. Helb, M. Burday, N. Connell, and D. Alland. 2007. A detailed analysis of the 16S rRNA gene segments for the diagnosis of pathogenic bacteria. J. Microbiol. Methods 69330-339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cuchacovich, R., S. Quinet, and A. M. Santos. 2003. Applications of polymerase chain reaction in rheumatology. Rheum. Dis. Clin. N. Am. 291-20. [DOI] [PubMed] [Google Scholar]
- 4.Faraj, A. A., O. D. Omonbude, and P. Godwin. 2002. Gram staining in the diagnosis of acute septic arthritis. Acta Orthop. Belg. 68388-391. [PubMed] [Google Scholar]
- 5.Fenollar, F., R. Veronique, S. Andreas, D. Michel, and D. Raoult. 2006. Analysis of 525 samples to determine the usefulness of PCR amplification and sequencing of the 16S rRNA gene for diagnosis of bone and joint infections. J. Clin. Microbiol. 441018-1028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gupta, M. N., R. D. Sturrock, and M. Field. 2003. Prospective comparative study of patients with culture proven and high suspicion of adult onset septic arthritis. Ann. Rheum. Dis. 62327-331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ho, G., Jr., and E. Y. Su. 1982. Therapy for septic arthritis. JAMA 247797-800. [PubMed] [Google Scholar]
- 8.Jalava, J., M. Skurnik, A. Toivanen, P. Toivanen, and E. Eerola. 2001. Bacterial PCR in the diagnosis of joint infection. Ann. Rheum. Dis. 60287-289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kaandorp, C. J., P. Krijnen, J. H. Moens, J. D. Habbema, and D. van Schaardenburg. 1997. The outcome of bacterial arthritis: a prospective community-based study. Arthritis Rheum. 40884-892. [DOI] [PubMed] [Google Scholar]
- 10.Kawamura, Y., X. G. Hou, F. Sultana, H. Miura, and T. Ezaki. 1995. Determination of 16S rRNA sequences of Streptococcus mitis and Streptococcus gordonii and phylogenetic relationships among members of the genus Streptococcus. Int. J. Syst. Bacteriol. 45406-408. [DOI] [PubMed] [Google Scholar]
- 11.Krey, P. R., and D. A. Bailen. 1979. Synovial fluid leukocytosis: a study of extremes. Am. J. Med. 67436-442. [DOI] [PubMed] [Google Scholar]
- 12.Margaretten, M., J. Kohlwes, D. Moore, and S. Bent. 2007. Does this patient have septic arthritis? JAMA 2971478-1488. [DOI] [PubMed] [Google Scholar]
- 13.Mariani, B. D., D. S. Martin, M. J. Levine, R. E. Booth, and R. S. Tuan. 1996. PCR detection of bacterial infection in total knee arthroplasty. Clin. Orthop. 33111-22. [DOI] [PubMed] [Google Scholar]
- 14.Rosey, A.-L., E. Abachin, G. Quesnes, C. Cadilhac, Z. Pejin, C. Glorion, P. Berche, and A. Ferroni. 2007. Development of a broad-range 16S rDNA real-time PCR for the diagnosis of septic arthritis in children. J. Microbiol. Methods 6888-93. [DOI] [PubMed] [Google Scholar]
- 15.Ross, J. J. 2005. Septic arthritis. Infect. Dis. Clin. N. Am. 19799-817. [DOI] [PubMed] [Google Scholar]
- 16.Soderquist, B., I. Jones, H. Fredlund, and T. Vikerfors. 1998. Bacterial or crystal-associated arthritis? Discriminating ability of serum inflammatory markers. Scand. J. Infect. Dis. 30591-596. [DOI] [PubMed] [Google Scholar]
- 17.van der Heijden, I. M., B. Wilbrink, A. E. M. Vije, L. M. Schouls, F. D. C. Breedveld, and P. P. Tak. 1999. Detection of bacterial DNA in serial synovial samples obtained during antibiotic treatment from patients with septic arthritis. Arthritis Rheum. 422198-2203. [DOI] [PubMed] [Google Scholar]
- 18.Wilbrink, B., I. M. van der Heijden, L. M. Schouls, J. D. A. van Embden, J. M. W. Hazes, F. C. Breedveld, and P. P. Tak. 1998. Detection of bacterial DNA in joint samples from patients with undifferentiated arthritis and reactive arthritis, using polymerase chain reaction with universal 16S ribosomal RNA primers. Arthritis Rheum. 41535-543. [DOI] [PubMed] [Google Scholar]
- 19.Yang, S., S. Lin, G. D. Kelen, T. C. Quinn, J. D. Dick, C. G. Gaydos, and R. E. Rothman. 2002. Quantitative multiprobe PCR assay for simultaneous detection and identification to species level of bacterial pathogens. J. Clin. Microbiol. 403449-3454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yang, S., S. Lin, A. Khalil, C. Gaydos, E. Nuemberger, G. Juan, J. Hardick, J. G. Bartlett, P. G. Auwaerter, and R. E. Rothman. 2005. Quantitative PCR assay using sputum samples for rapid diagnosis of pneumococcal pneumonia in adult emergency department patients. J. Clin. Microbiol. 433221-3226. [DOI] [PMC free article] [PubMed] [Google Scholar]