header advert
Bone & Joint Research Logo

Receive monthly Table of Contents alerts from Bone & Joint Research

Comprehensive article alerts can be set up and managed through your account settings

View my account settings

Visit Bone & Joint Research at:

X (Twitter)
Instagram
Article alerts Social media
Current issue
Submit
Open Access

Research

Rapid detection of periprosthetic joint infection using a combination of 16s rDNA polymerase chain reaction and lateral flow immunoassay

A Pilot Study

Download PDF

Abstract

Objectives

The objective of this study was to develop a test for the rapid (within 25 minutes) intraoperative detection of bacteria from synovial fluid to diagnose periprosthetic joint infection (PJI).

Methods

The 16s rDNA test combines a polymerase chain reaction (PCR) for amplification of 16s rDNA with a lateral flow immunoassay in one fully automated system. The synovial fluid of 77 patients undergoing joint aspiration or primary or revision total hip or knee surgery was prospectively collected. The cohort was divided into a proof-of-principle cohort (n = 17) and a validation cohort (n = 60). Using the proof-of-principle cohort, an optimal cut-off for the discrimination between PJI and non-PJI samples was determined. PJI was defined as detection of the same bacterial species in a minimum of two microbiological samples, positive histology, and presence of a sinus tract or intra-articular pus.

Results

The 16s rDNA test proved to be very robust and was able to provide a result in 97% of all samples within 25 minutes. The 16s rDNA test was able to diagnose PJI with a sensitivity of 87.5% and 82%, and a specificity of 100% and 89%, in the proof-of-principle and validation cohorts, respectively. The microbiological culture of synovial fluid achieved a sensitivity of 80% and a specificity of 93% in the validation cohort.

Conclusion

The 16s rDNA test offers reliable intraoperative detection of all bacterial species within 25 minutes with a sensitivity and specificity comparable with those of conventional microbiological culture of synovial fluid for the detection of PJI. The 16s rDNA test performance is independent of possible blood contamination, culture time and bacterial species.

Cite this article: V. Janz, J. Schoon, C. Morgenstern, B. Preininger, S. Reinke, G. Duda, A. Breitbach, C. F. Perka, S. Geissler. Rapid detection of periprosthetic joint infection using a combination of 16s rDNA polymerase chain reaction and lateral flow immunoassay: A Pilot Study. Bone Joint Res 2018;7:12–19. DOI: 10.1302/2046-3758.71.BJR-2017-0103.R2.

Article focus

  • This study presents the initial results of a newly developed test for the rapid (within 25 minutes) intraoperative detection of bacteria from synovial fluid to detect periprosthetic joint infection (PJI). The 16s rDNA test combines a polymerase chain reaction (PCR) for amplification of 16s rDNA with a lateral flow immunoassay in one fully automated system.

Key messages

  • The 16s rDNA test was able to provide a result in 97% of all samples within 25 minutes.

  • The sensitivity and specificity of the 16s rDNA test were comparable with the sensitivity and specificity of conventional synovial fluid culture.

Strengths and limitations

  • This study is the first to develop a 16s rDNA based test within 25 minutes, allowing for a true intraoperative application.

  • Initial results of a pilot study. Future prospective studies for intraoperative application and specialized surgical indications are planned.

Introduction

Before a revision of a total hip arthroplasty (THA) or total knee arthroplasty (TKA), it is mandatory to either confirm or exclude a periprosthetic joint infection (PJI), since the surgical strategies differ significantly between aseptic and septic revisions. In cases of an unclear preoperative diagnosis of PJI, the only other routinely utilized diagnostic option is the histological evaluation of an intraoperative frozen tissue section.1 A previous meta-analysis by Tsaras et al2 has reported convincing diagnostic evidence for the use of frozen tissue sections for the detection of culture-positive PJI in TKA and THA, with a diagnostic odds ratio of 54.7. However, currently there are no commonly utilized diagnostic alternatives for cases of unclear PJI status.

Although a post hoc differentiation between septic and aseptic cases is possible through other non-culture-based diagnostic methods, such as synovial cell count, leucocyte esterase, and α-defensin, none of these have been validated as a diagnostic tool to facilitate intraoperative decision-making.3-8 Additionally, all of these diagnostic methods are dependent on the patient’s immune response towards the presence of bacteria and therefore only allow for an indirect detection of PJI. It was the goal of this study to design a diagnostic test for an intraoperative discrimination between septic and aseptic cases through a direct detection of the causative bacterial species.

To enable an intraoperative diagnosis, a bacterial infection has to be directly detected in order to circumvent the delay due to microbiological culture. A direct detection of bacteria can be realized by multiplex polymerase chain reaction (PCR)-based amplification of 16s rDNA, which encodes highly conservative regions of the 16s ribosomal subunit and is common to all bacterial species. The detection of 16s rDNA, through multiplex PCR, can be performed from different sample materials, such as synovial fluid, periprosthetic tissue samples, and sonicate fluid.9-13 Synovial fluid is the most promising material for such purposes due to the ease of acquisition at the beginning of revision surgery and the wide acceptance within the surgical community of its use to diagnose PJI.13-16

To our knowledge, there are currently only two other studies that were able to achieve a ‘rapid’ molecular diagnosis of PJI, within three hours and 4.5 hours, respectively.12,17,18 While this represents a significant improvement over the standard diagnostic time of culture-based methods, this timeframe is still too great for a true intraoperative application to discriminate between septic and aseptic failures. The aim of this study was to develop a 16s rDNA PCR test system for the detection of PJI that would rapidly (within 25 minutes) facilitate intraoperative discrimination between septic and aseptic prosthetic joint failures.

Patients and Methods

Study design and patient cohort

A total of 77 patients were included in this prospective cohort study, between January 2014 and January 2015, and divided into a proof-of-principle cohort (n = 17, eight cases of PJI) and a validation cohort (n = 60, 23 cases of PJI). The sample size of the validation cohort was determined by power analysis based on the results from the proof-of-principle group (power = 0.80, a = 0.01; two-sided, minimum sample size = 21 per group). All patients provided written informed consent, and the study was approved by the local institutional review board. The proof-of-principle cohort was used to evaluate the test’s functionality and to assess the optimal threshold for discrimination between septic and aseptic patients. The validation cohort was used to evaluate the test’s diagnostic performance.

Synovial fluid samples were collected preoperatively for the THA and TKA revisions with a preoperative suspicion of PJI, or intraoperatively for THA and TKA revisions without preoperative suspicion of PJI. In addition, synovial aspirations of native joints were performed intra-operatively during primary THA and TKA surgery. These aspirations of native joints functioned exclusively as aseptic negative controls. Within one hour, the samples were transported to our research facility and frozen at -80°C prior to analysis. The proof-of-principle cohort was comprised of two primary TKAs, ten revision THAs or TKAs, and five THA aspirations (Table I). The validation cohort was comprised of seven primary THAs or TKAs, 32 revision THAs or TKAs, and 21 joint aspirations (Table II). Sample harvesting, patient/sample data collection, and documentation were performed in accordance with our institutional guidelines.19 The physical properties of the samples were qualitatively assessed, including variances in synovial viscosity, optical clarity (blood contamination), and sample volumes.

Table I.

Patient and sample characterization for the proof-of-concept cohort (PC)

Sample ID Gender Age (yrs) Preoperative suspicion* Surgical procedure Microbiological culture† PJI‡ 16s rDNA assay
PC_01 Male 19 Aseptic Primary THA/TKA N/A Negative Negative
PC_02 Female 84 Aseptic Primary THA/TKA N/A Negative Negative
PC_03 Male 55 Aseptic Revision THA/TKA Negative Negative Negative
PC_04 Female 52 Aseptic Revision THA/TKA Negative Negative Negative
PC_05 Female 60 Aseptic Revision THA/TKA Negative Negative Negative
PC_06 Female 71 Septic Revision THA/TKA Negative Negative Negative
PC_07 Female 47 Septic Revision THA/TKA Negative Negative Negative
PC_08 Male 87 Unclear Joint aspiration Negative Negative Negative
PC_09 Male 68 Unclear Joint aspiration Negative Negative Negative
PC_10 Male 37 Septic Revision THA/TKA Positive Positive Negative
PC_11 Female 76 Septic Revision THA/TKA Positive Positive Positive
PC_12 Male 60 Septic Revision THA/TKA Negative Positive Positive
PC_13 Female 90 Septic Revision THA/TKA Positive Positive Positive
PC_14 Male 85 Unclear Joint aspiration Positive Positive Positive
PC_15 Male 87 Unclear Joint aspiration Positive Positive Positive
PC_16 Male 74 Unclear Joint aspiration Positive Positive Positive
PC_17 Male 66 Unclear Revision THA/TKA Positive Positive Positive
  1. *

    Based on the preoperative diagnostics

  1. †

    Growth of the same bacterial species in at least two of the following samples: synovial fluid, intraoperative tissue sample, and sonicate fluid cultures (SFC)

  1. ‡

    Final diagnosis of PJI based on intraoperative samples and PJI definition

  1. PJI, periprosthetic joint infection; THA, total hip arthroplasty; TKA, total knee arthroplasty; N/A, not available

Table II.

Patient and sample characterization for the validation cohort (VC)

Sample ID Gender Age (yrs) Preoperative suspicion* Surgical procedure Microbiological culture† PJI‡ 16s rDNA assay
VC_01 Female 65 Aseptic Primary THA/TKA N/A Negative Negative
VC_02 Male 46 Aseptic Primary THA/TKA N/A Negative Negative
VC_03 Female 83 Aseptic Primary THA/TKA N/A Negative Negative
VC_04 Male 48 Aseptic Primary THA/TKA N/A Negative Negative
VC_05 Male 62 Aseptic Primary THA/TKA N/A Negative Negative
VC_06 Male 71 Aseptic Primary THA/TKA N/A Negative Negative
VC_07 Male 81 Aseptic Primary THA/TKA N/A Negative Negative
VC_08 Male 81 Aseptic Revision THA/TKA Negative Negative Positive
VC_09 Male 56 Aseptic Revision THA/TKA Negative Negative Negative
VC_10 Female 78 Aseptic Revision THA/TKA Negative Negative Negative
VC_11 Female 59 Aseptic Revision THA/TKA Negative Negative Negative
VC_12 Female 79 Aseptic Revision THA/TKA Negative Negative Negative
VC_13 Male 54 Aseptic Revision THA/TKA Negative Negative Negative
VC_14 Female 51 Aseptic Revision THA/TKA Negative Negative Negative
VC_15 Male 79 Septic Revision THA/TKA Negative Negative Negative
VC_16 Female 52 Septic Revision THA/TKA Negative Negative Positive
VC_17 Female 46 Septic Revision THA/TKA Negative Negative Negative
VC_18 Female 70 Septic Revision THA/TKA Negative Negative Negative
VC_19 Female 59 Septic Revision THA/TKA Negative Negative Negative
VC_20 Female 70 Septic Revision THA/TKA Negative Negative Negative
VC_21 Male 37 Unclear Joint aspiration Negative Negative Positive
VC_22 Male 62 Unclear Joint aspiration Negative Negative Negative
VC_23 Male 55 Unclear Joint aspiration Negative Negative Negative
VC_24 Male 67 Unclear Joint aspiration Negative Negative Negative
VC_25 Male 77 Unclear Joint aspiration Negative Negative Negative
VC_26 Female 77 Unclear Joint aspiration Positive Negative N/A
VC_27 Male 46 Unclear Joint aspiration Negative Negative Negative
VC_28 Male 60 Unclear Joint aspiration Negative Negative Positive
VC_29 Male 72 Unclear Joint aspiration Negative Negative Negative
VC_30 Male 73 Unclear Joint aspiration Negative Negative Negative
VC_31 Male 68 Unclear Joint aspiration Negative Negative Negative
VC_32 Male 86 Unclear Joint aspiration Negative Negative Negative
VC_33 Female 77 Unclear Joint aspiration Negative Negative Negative
VC_34 Female 71 Unclear Revision THA/TKA Negative Negative Negative
VC_35 Female 70 Unclear Revision THA/TKA Negative Negative Negative
VC_36 Male 85 Unclear Revision THA/TKA Negative Negative Negative
VC_37 Male 74 Unclear Revision THA/TKA Negative Negative Negative
VC_38 Male 61 Septic Joint aspiration Positive Positive Negative
VC_39 Female 89 Septic Revision THA/TKA Positive Positive Positive
VC_40 Female 76 Septic Revision THA/TKA Negative Positive Positive
VC_41 Female 80 Septic Revision THA/TKA Positive Positive Negative
VC_42 Female 78 Septic Revision THA/TKA Positive Positive Positive
VC_43 Female 70 Septic Revision THA/TKA Positive Positive Positive
VC_44 Male 66 Septic Revision THA/TKA Positive Positive Negative
VC_45 Male 60 Septic Revision THA/TKA Positive Positive Positive
VC_46 Male 66 Septic Revision THA/TKA Positive Positive Positive
VC_47 Male 63 Septic Revision THA/TKA Positive Positive Negative
VC_48 Female 75 Septic Revision THA/TKA Positive Positive Positive
VC_49 Male 59 Septic Revision THA/TKA Negative Positive Positive
VC_50 Male 70 Unclear Revision THA/TKA Negative Positive N/A
VC_51 Female 55 Unclear Joint aspiration Positive Positive Positive
VC_52 Female 55 Unclear Joint aspiration Positive Positive Positive
VC_53 Male 66 Unclear Joint aspiration Positive Positive Positive
VC_54 Male 84 Unclear Joint aspiration Positive Positive Positive
VC_55 Male 86 Unclear Joint aspiration Positive Positive Positive
VC_56 Female 64 Unclear Joint aspiration Positive Positive Positive
VC_57 Male 73 Unclear Joint aspiration Positive Positive Positive
VC_58 Female 67 Unclear Revision THA/TKA Negative Positive Positive
VC_59 Female 78 Unclear Revision THA/TKA Positive Positive Positive
VC_60 Male 65 Unclear Revision THA/TKA Positive Positive Positive
  1. *

    Based on the preoperative diagnostics

  1. †

    Growth of the same bacterial species in at least two of the following samples: synovial fluid, intraoperative tissue sample, and sonicate fluid cultures (SFC)

  1. ‡

    Final diagnosis of PJI based on intraoperative samples and PJI definition

  1. PJI, periprosthetic joint infection; THA, total hip arthroplasty; TKA, total knee arthroplasty; N/A, not available

PJI definition and intraoperative sample acquisition

PJI was defined according to the following criteria: intra-articular pus or presence of a sinus tract; histology indicative of infection (type II or III periprosthetic membrane); or positive microbiological culture of the same bacterial species in a minimum of two of the following samples: synovial fluid; intraoperative tissue sample; or sonicate fluid cultures (SFC).20-23 The final diagnosis of PJI was made according to the results of the intraoperative microbiological and histological samples. The final diagnosis of PJI was the benchmark reference, against which the performance of the 16s rDNA test, as well as all calculations for sensitivity and specificity, were referenced.

Synovial fluid sampling was performed in an operating theatre with laminar air flow, utilizing a skin incision, and under fluoroscopic guidance, for all joint aspirations. Intraoperative synovial fluid aspiration was performed under direct visualization of the joint and prior to capsulotomy. Additionally, multiple periprosthetic tissue samples, a histological sampling of the periprosthetic membrane, and SFC were acquired for all cases of revision arthroplasty. The histological evaluation was performed according to the consensus classification of the periprosthetic interface membrane.21 To optimize the microbiological culture methods, both synovial fluid and SFC were incubated in blood culture bottles.24-26 Intraoperative tissue samples were cultured on standard agar plates. To allow for a detection of fastidious bacterial species, all microbiological cultures were incubated for 14 days.27

16s rDNA PCR test system

The 16s rDNA test is based on a targeted PCR and subsequent detection of the PCR products by lateral flow immunoassay. The 16s rDNA test was performed from intraoperatively acquired synovial fluid. The PCR primers target a highly conservative region of the 16s ribosomal subunit that is common to all bacterial species. The complete workflow is illustrated in Figure 1 and requires 25 minutes. Synovial fluid (total volume = 2 µl) was directly combined with the PCR master mix, containing differentially labelled forward (biotin) and reverse (Fluorescein isothiocyanate (FITC)) primers that are specific to a highly conserved 16s rDNA sequence (primers are available on request; Milenia Biotec GmbH, Gießen, Germany). Polymerase chain reaction (30 cycles; 15 minutes) was performed using the Labcycler 48s (SensoQuest GmbH, Göttingen, Germany). In the presence of 16s rDNA, the PCR produced double-labelled (biotin and FITC) DNA products. The PCR mixture was subsequently transferred to the lateral flow immunoassay test unit (Milenia Biotec), where the PCR fragments were captured via their biotin label by specific antibodies in a single-step procedure. The results were displayed as two bands on the test strip. The lower test band indicates the detection of the bacterial 16s rDNA product and the upper band serves as a control, confirming the correct function of the flow assay. The test results were evaluated by spectrometric measurement of the band intensity and quantified by ImageJ software (National Institutes of Health, Bethesda, Maryland; http://imagej.nih.gov).28 The 16s rDNA assay score was calculated as the ratio between the intensity of the test and control bands and expressed in arbitrary units (AU).

General workflow of the 16s rDNA test: a) chronological test principle, with polymerase chain reaction (PCR) followed by subsequent detection of the specific PCR products by lateral flow immunoassay; b) the results are displayed as one test band (T, detection of 16s rDNA); and c) a control band (C), confirming the correct function of assay).

Fig.

General workflow of the 16s rDNA test: a) chronological test principle, with polymerase chain reaction (PCR) followed by subsequent detection of the specific PCR products by lateral flow immunoassay; b) the results are displayed as one test band (T, detection of 16s rDNA); and c) a control band (C), confirming the correct function of assay).

Statistical analysis

All data are given as mean ± sd. The Mann–Whitney U test was used for group comparison and receiver operating characteristic (ROC) analysis was performed to determine the optimal cut-off for discrimination between the two patient groups. Assay sensitivity and specificity were calculated as previously described.29 Sensitivity was defined as true positive (TP) / (TP + false negative (FN)), and specificity was defined as true negative (TN) / (TN + false positive (FP)), All statistical analyses were performed with SPSS software, version 18 (IBM Corp., Armonk, New York), and statistical significance was defined at p < 0.05. The medical patient data and the results from the 16s rDNA PCR test system were evaluated in a double-blinded manner by two authors (SG and VJ).

Results

Test reliability

The 16s rDNA test system provided a diagnostic result within 25 minutes in 97% (75 of 77) of all patients. Two samples could not be evaluated due to massive protein precipitation from the synovial sample during the PCR. Other possible confounding factors for sample evaluation, such as variances in synovial viscosity, blood contamination, small sample volumes, or variances in transport times, did not negatively affect the test reliability.

Test performance

ROC analysis in the proof-of-principle cohort revealed an optimal cut-off level of 0.71 AU, between the test and control bands of the 16s rDNA test strip (Fig. 2). Utilizing this cut-off, the 16s rDNA test system was able to detect seven of eight PJI samples and all of the non-PJI samples correctly, achieving a sensitivity of 87.5% and a specificity of 100% (area under the curve (AUC) = 0.944, p = 0.001) in the proof-of-principle cohort (Fig. 2).

General test performance using the proof-of-principle cohort: a) receiver operating characteristic (ROC) curve for the 16s rDNA test based on the calculated ratio between test and control band to determine the optimal cut-off value to differentiate between periprosthetic joint infection (PJI) and non-PJI samples; and b) box-whisker plot displaying the performance of the 16s rDNA test to differentiate between PJI (n = 8) and non-PJI (n = 9) samples in the proof-of-principle cohort. Dashed lines indicate the optimal cut-off as determined by previous ROC analysis. The Mann–Whitney U test was used to obtain p-values. AUC, area under the curve; CI, confidence interval; AU, assay unit.

Fig.

General test performance using the proof-of-principle cohort: a) receiver operating characteristic (ROC) curve for the 16s rDNA test based on the calculated ratio between test and control band to determine the optimal cut-off value to differentiate between periprosthetic joint infection (PJI) and non-PJI samples; and b) box-whisker plot displaying the performance of the 16s rDNA test to differentiate between PJI (n = 8) and non-PJI (n = 9) samples in the proof-of-principle cohort. Dashed lines indicate the optimal cut-off as determined by previous ROC analysis. The Mann–Whitney U test was used to obtain p-values. AUC, area under the curve; CI, confidence interval; AU, assay unit.

Validation of test performance

Using the predefined cut-off value in the validation cohort, the 16s rDNA test system achieved a sensitivity of 82% (true positive = 18, false negative = 4) and specificity of 89% (true negative = 32, false positive = 4) (AUC = 0.894, p < 0.001). Examination of the ROC curve of the validation cohort confirmed the predetermined cut-off of 0.71 AU as optimal to discriminate between PJI and non-PJI samples (Fig. 3). The performance of the 16s rDNA test was independent of the isolated bacterial species. The complete list of detected bacterial species grouped according to their detection by microbiological culture or 16s rDNA test is displayed in Table III.

Validation of the test performance using the validation cohort: a) receiver operating characteristic (ROC) curve for the 16s rDNA test (ratio between test and control band on the test strip) to differentiate between periprosthetic joint infection (PJI) and non-PJI samples in the validation cohort; and b) box-whisker plot displaying the performance of the 16s rDNA test to differentiate between PJI (n = 23) and non-PJI (n = 37) samples in the validation cohort. Dashed lines indicate the predefined cut-off as determined using the proof-of-principle cohort. The Mann–Whitney U test was used to obtain p-values. AUC, area under the curve; CI, confidence interval; AU, assay unit.

Fig.

Validation of the test performance using the validation cohort: a) receiver operating characteristic (ROC) curve for the 16s rDNA test (ratio between test and control band on the test strip) to differentiate between periprosthetic joint infection (PJI) and non-PJI samples in the validation cohort; and b) box-whisker plot displaying the performance of the 16s rDNA test to differentiate between PJI (n = 23) and non-PJI (n = 37) samples in the validation cohort. Dashed lines indicate the predefined cut-off as determined using the proof-of-principle cohort. The Mann–Whitney U test was used to obtain p-values. AUC, area under the curve; CI, confidence interval; AU, assay unit.

Table III.

Detected bacterial species grouped according to their detection by microbiological culture or 16s rDNA test

Bacterial species Culture positive 16s rDNA test positive
Staphylococcus epidermidis + +
Staphylococcus hominis + +
Staphylococcus caprae + +
Staphylococcus capitis + +
Staphylococcus warneri + +
Staphylococcus aureus + +
Streptococcus agalactiae + +
Propionibacterium acnes + +
Enterococcus faecalis + +
Enterococcus coli + -
Dermabacter hominis + -

Test performance in comparison with conventional microbiological methods

We directly compared the diagnostic performance of the 16s rDNA test with the individual performance of the conventional microbiological diagnostic methods, comprised of synovial fluid and periprosthetic tissue cultures, as well as the histological evaluation of the periprosthetic membrane. The microbiological culture of synovial fluid achieved a lower sensitivity than that of the 16s rDNA test with 80%, and a specificity of 93%. The combination of synovial fluid and tissue sample cultures achieved a sensitivity of 86% and specificity of 86%.

Overall, the correlation between the 16s rDNA test and the microbiological cultures showed a concordance in 75% of all cases, with the 16s rDNA test and the microbiological cultures both being either positive or negative. The correlation between the 16s rDNA test and the histological evaluation of the periprosthetic tissue sample was slightly superior, with a concordance rate of 77%.

Discussion

Despite the longer time period associated with culture-based methods, which precludes an intraoperative application, the detection of PJI by microbiological culture remains the benchmark in PJI diagnostics. To avoid the disadvantages associated with microbiological culture, we developed a test for the rapid detection of bacterial 16s rDNA from synovial fluid (within 25 minutes). To our knowledge, the shortest reported times for the performance of a PCR-based 16s rDNA detection are, in the current literature, three hours and 4.5 hours.12,17,18

A distinct advantage of the 16s rDNA test over other diagnostic methods, such as leucocyte esterase, is the high degree of reliability and resistance to contamination. The detection of leucocyte esterase from synovial fluid is very susceptible to blood contamination, making an evaluation of up to 17% of all samples impossible.2 The high degree of reliability of the 16s rDNA test, with 97% of all samples providing a diagnostic result, and the execution within 25 minutes from only 2 µl of synovial fluid, allow for a true intraoperative application.

Since total joint arthroplasties release wear particles with heterogeneous physicochemical properties, these could theoretically interfere with our 16s rDNA test. To address this issue and to take the heterogeneity of potential patient cohorts into account, patients undergoing arthroplasty revision, as well as primary arthroplasty, were included in our patient collective. The high degree of correlation between the results of our 16s rDNA test and those of the microbiological culture shows that a reliable detection of PJI from synovial fluid is possible even in the presumed presence of wear particles. Although it was not the primary goal of this study to achieve a superior sensitivity over the standard intraoperative microbiological cultures, the 16s rDNA test achieved a slightly higher sensitivity than both the microbiological culture of synovial fluid and periprosthetic tissue cultures.30 The differences in sensitivity and specificity of the 16s rDNA test in the proof-of-principle and validation cohorts could be attributed to the differences in sample size and PJI incidence between the cohorts.

In addition, the sensitivity of 82% achieved by our synovial fluid 16s rDNA test exceeded the reported sensitivity rates of other 16s rDNA tests which ranged from 64% to 76%.9-11 The achieved correlation rate and sensitivity are independent of the bacterial species, since the utilized primer sequences match to a highly conservative region of bacterial rDNA encoding the 16s ribosomal subunit, which is identical in prokaryotes.31

The 16s rDNA test was able to detect all of the bacterial species isolated by microbiological culture, except in two cases (Table I). Only two isolations of E. coli and Dermabacter hominis were not detected. Both cases represent single positive bacterial isolations in two different patients. The isolation of Dermabacter hominis was only present in the SFC, with all other microbiological cultures remaining negative. The isolation of E. coli represented one of the two 16s rDNA tests which were not analyzable due to massive protein precipitation in the synovial fluid sample.

Our study also has a number of technical limitations. First, specialized equipment, such as a thermocycler, is necessary for an intraoperative application to perform the 16s rDNA test. Therefore, all samples were transported to our research facility for this study and proof of applicability in a true intraoperative scenario is pending. Nevertheless, the test was developed as a fully automated system with a focus on convenience, user friendliness, and rapid detection to allow for an intraoperative application, without further modifications. The translation of the test system into clinical application is the main goal in the continuation of this project. Second, the small patient cohort, to date, should be supplemented by a larger prospective cohort to confirm and validate our current findings. Third, specific indications, such as a prospective comparison between intraoperative frozen tissue sections and the current 16s rDNA test, should be investigated. Finally, our proposed test system has a restricted ability to distinguish between living and dead bacteria, which potentially limits its use to monitor the infection status after antibiotic treatment. Thus, further studies must be performed to validate our findings in synovial fluid from PJI patients after antibiotic treatment. Previous studies have shown that a pre-incubation of biological samples with the membrane-impermeant agent propidium monoazide prevents the amplification of the 16s rDNA from dead cells; such a modification to our 16s rDNA test could be a promising method to further maximize the clinical utility.32,33 Furthermore, our 16s rDNA test rapidly detects the presence of bacteria, but does not detect the specific bacterial strain or, more importantly, a potential antibiotic resistance. Thus, we currently aim to extend our 16s rDNA test towards a multiplex approach, allowing for the simultaneous identification of clinically relevant bacterial strains, as well as specific antibiotic-resistant genes.

In conclusion, the current system can reliably and rapidly detect PJI, enabling an intraoperative application. The direct detection of bacterial 16s rDNA shows encouraging results, and warrants further evaluation in larger patient cohorts. The future addition of the detection of clinically relevant antibiotic resistance will be a focus of further research.

V. Janz; email:
Author Contribution

V. Janz: Designing the study, Acquiring and evaluating the data, Preparing and revising the manuscript.

J. Schoon: Acquiring and evaluating the data.

C. Morgenstern: Acquiring and evaluating the data.

B. Preininger: Designing the study, Acquiring and evaluating the data, Revising the manuscript.

S. Reinke: Acquiring and evaluating the data.

G. Duda: Designing and supervising the study, Revising the manuscript.

A. Breitbach: Designing the study, Evaluating the data, Preparing and revising the manuscript.

C. F. Perka: Designing and supervising the study, Revising the manuscript.

S. Geissler: Designing the study, Acquring and evaluating the data, Preparing and revising the manuscript.

We would like to thank Milenia Biotec GmbH (Gießen, Germany) for the supply of polymerase chain reaction mixes and lateral flow strips. We would also like to thank to the Berlin-Brandenburg Center for Regenerative Therapies (BCRT) Core Unit “Cell Harvesting” for their support with sample processing and molecular experiments. This work was partially supported by grants from the German Federal Ministry of Education and Research (BMBF, 01EC1402B) and the German Research Foundation (DFG, GE2512/2-1) to S. Geissler. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Open access

This is an open-access article distributed under the terms of the Creative Commons Attributions licence (CC-BY-NC), which permits unrestricted use, distribution, and reproduction in any medium, but not for commercial gain, provided the original author and source are credited.

  • Funding Statement

    This work was partially supported by grants from the German Federal Ministry of Education and Research (BMBF, 01EC1402B) and the German Research Foundation (DFG, GE2512/2-1).

  • Conflicts of Interest Statement

    None declared

    • References

    1 Feldman DS , Lonner JH , Desai P , Zuckerman JD . The role of intraoperative frozen sections in revision total joint arthroplasty. J Bone Joint Surg [Am]1995;77-A:1807-1813.PubMed Google Scholar

    2 Tsaras G , Maduka-Ezeh A , Inwards et al. . Utility of intraoperative frozen section histopathology in the diagnosis of periprosthetic joint infection: a systematic review and meta-analysis. J Bone Joint Surg [Am]2012;94-A:1700-1711.CrossrefPubMed Google Scholar

    3 Trampuz A , Hanssen AD , Osmon DR et al. . Synovial fluid leukocyte count and differential for the diagnosis of prosthetic knee infection. Am J Med2004;117:556-562.CrossrefPubMed Google Scholar

    4 Ghanem E , Parvizi J , Burnett RS et al. . Cell count and differential of aspirated fluid in the diagnosis of infection at the site of total knee arthroplasty. J Bone Joint Surg [Am]2008;90-A:1637-43.PubMed Google Scholar

    5 Deirmengian C , Kardos K , Kilmartin P et al. . Diagnosing periprosthetic joint infection: has the era of the biomarker arrived?Clin Orthop Relat Res2014;472:3254-3262.CrossrefPubMed Google Scholar

    6 Deirmengian C , Kardos K , Kilmartin P et al. . The alpha-defensin test for periprosthetic joint infection outperforms the leukocyte esterase test strip. Clin Orthop Relat Res2015;473:198-203.CrossrefPubMed Google Scholar

    7 Wetters NG , Berend KR , Lombardi AV et al. . Leukocyte esterase reagent strips for the rapid diagnosis of periprosthetic joint infection. J Arthroplasty2012;27(Suppl):8-11.CrossrefPubMed Google Scholar

    8 Bingham J , Clarke H , Spangehl M et al. . The alpha defensin-1 biomarker assay can be used to evaluate the potentially infected total joint arthroplasty. Clin Orthop Relat Res2014;472:4006-4009.CrossrefPubMed Google Scholar

    9 Marín M , Garcia-Lechuz JM , Alonso P et al. . Role of universal 16S rRNA gene PCR and sequencing in diagnosis of prosthetic joint infection. J Clin Microbiol2012;50:583-589.CrossrefPubMed Google Scholar

    10 Bémer P , Plouzeau C , Tande D et al. . Evaluation of 16S rRNA gene PCR sensitivity and specificity for diagnosis of prosthetic joint infection: a prospective multicenter cross-sectional study. J Clin Microbiol2014;52:3583-3589.CrossrefPubMed Google Scholar

    11 Gomez E , Cazanave C , Cunningham SA et al. . Prosthetic joint infection diagnosis using broad-range PCR of biofilms dislodged from knee and hip arthroplasty surfaces using sonication. J Clin Microbiol2012;50:3501-3508.CrossrefPubMed Google Scholar

    12 Birmingham P , Helm JM , Manner PA , Tuan RS . Simulated joint infection assessment by rapid detection of live bacteria with real-time reverse transcription polymerase chain reaction. J Bone Joint Surg [Am]2008;90-A:602-608.CrossrefPubMed Google Scholar

    13 Lévy PY , Fenollar F . The role of molecular diagnostics in implant-associated bone and joint infection. Clin Microbiol Infect2012;18:1168-1175.CrossrefPubMed Google Scholar

    14 Font-Vizcarra L , García S , Martínez-Pastor JC , Sierra JM , Soriano A . Blood culture flasks for culturing synovial fluid in prosthetic joint infections. Clin Orthop Relat Res2010;468:2238-2243.CrossrefPubMed Google Scholar

    15 Esteban J , Sorlí L , Alentorn-Geli E , Puig L , Horcajada JP . Conventional and molecular diagnostic strategies for prosthetic joint infections. Expert Rev Mol Diagn2014;14:83-96.CrossrefPubMed Google Scholar

    16 Frommelt L . Aspiration of joint fluid for detection of the pathogen in periprosthetic infection. Orthopade2008;37:1027-1034; quiz 1035-1036. (in German) Google Scholar

    17 Cazanave C , Greenwood-Quaintance KE , Hanssen AD et al. . Rapid molecular microbiologic diagnosis of prosthetic joint infection. J Clin Microbiol2013;51:2280-2287.CrossrefPubMed Google Scholar

    18 Yang S , Ramachandran P , Hardick A et al. . Rapid PCR-based diagnosis of septic arthritis by early Gram-type classification and pathogen identification. J Clin Microbiol2008;46:1386-1390.CrossrefPubMed Google Scholar

    19 Reinke S , Dienelt A , Blankenstein A , Duda GN , Geissler S . Qualifying stem cell sources: how to overcome potential pitfalls in regenerative medicine?J Tissue Eng Regen Med2016;10:3-10.CrossrefPubMed Google Scholar

    20 Krenn V , Morawietz L , Perino G et al. . Revised histopathological consensus classification of joint implant related pathology. Pathol Res Pract2014;210:779-786.CrossrefPubMed Google Scholar

    21 Morawietz L , Classen RA , Schröder JH et al. . Proposal for a histopathological consensus classification of the periprosthetic interface membrane. J Clin Pathol2006;59:591-597.CrossrefPubMed Google Scholar

    22 Zimmerli W , Trampuz A , Ochsner PE . Prosthetic-joint infections. N Engl J Med2004;351:1645-1654. Google Scholar

    23 Osmon DR , Berbari EF , Berendt AR et al. . Diagnosis and management of prosthetic joint infection: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis2013;56:e1-e25.CrossrefPubMed Google Scholar

    24 Janz V , Trampuz A , Perka CF , Wassilew GI . Reduced culture time and improved isolation rate through culture of sonicate fluid in blood culture bottles. Technol Health Care2017;25:635-640.CrossrefPubMed Google Scholar

    25 Portillo ME , Salvadó M , Trampuz A et al. . Improved diagnosis of orthopedic implant-associated infection by inoculation of sonication fluid into blood culture bottles. J Clin Microbiol2015;53:1622-1627.CrossrefPubMed Google Scholar

    26 Shen H , Tang J , Wang Q , Jiang Y , Zhang X . Sonication of explanted prosthesis combined with incubation in BD bactec bottles for pathogen-based diagnosis of prosthetic joint infection. J Clin Microbiol2015;53:777-781.CrossrefPubMed Google Scholar

    27 Schäfer P , Fink B , Sandow D et al. . Prolonged bacterial culture to identify late periprosthetic joint infection: a promising strategy. Clin Infect Dis2008;47:1403-1409.CrossrefPubMed Google Scholar

    28 Rasband WS , Image J . National Institutes of Health, Bethesda, Maryland, USA. 1997-2016. https://imagej.nih.gov/ij/ (date last accessed 25October2017). Google Scholar

    29 Altman DG , Bland JM . Diagnostic tests. 1: sensitivity and specificity. BMJ1994;308:1552.CrossrefPubMed Google Scholar

    30 Meermans G , Haddad FS . Is there a role for tissue biopsy in the diagnosis of periprosthetic infection?Clin Orthop Relat Res2010;468:1410-1417.CrossrefPubMed Google Scholar

    31 Srinivasan R , Karaoz U , Volegova M et al. . Use of 16S rRNA gene for identification of a broad range of clinically relevant bacterial pathogens. PLoS One2015;10:e0117617.CrossrefPubMed Google Scholar

    32 Kobayashi H , Oethinger M , Tuohy MJ , Hall GS , Bauer TW . Distinction between intact and antibiotic-inactivated bacteria by real-time PCR after treatment with propidium monoazide. J Orthop Res2010;28:1245-1251.CrossrefPubMed Google Scholar

    33 Cangelosi GA , Meschke JS . Dead or alive: molecular assessment of microbial viability. Appl Environ Microbiol2014;80:5884-5891. Google Scholar

    Information

    Journal

    Bone & Joint Research

    Volume

    7 No.1 | Pages 12 - 19

    Section

    Research

    Published

    01 January 2018

    DOI

    Expand all

    Center for Musculoskeletal Surgery, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

    viktor.janz@charite.de

    Search for more articles by this author

    Julius Wolff Institute, BCRT & BSRT, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

    Search for more articles by this author

    Center for Musculoskeletal Surgery, Julius Wolff Institute, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

    Search for more articles by this author

    Center for Musculoskeletal Surgery, Julius Wolff Institute, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

    Search for more articles by this author

    Julius Wolff Institute, BCRT & BSRT, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

    Search for more articles by this author

    Julius Wolff Institute, BCRT & BSRT, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

    Search for more articles by this author

    Institut für Pflanzenphysiologie, Martin-Luther-Universität Halle-Wittenberg, Weinbergweg 10, 06120 Halle, Germany.

    Search for more articles by this author

    Center for Musculoskeletal Surgery, Julius Wolff Institute, BCRT & BSRT, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

    Search for more articles by this author

    Julius Wolff Institute, BCRT & BSRT, Charité – Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany

    Search for more articles by this author
    Periprosthetic joint infection
    Rapid detection
    16s rDNA
    Polymerase chain reaction
    Lateral flow
    Immunoassay
    Submit Letter to the Editor

    Share

    Share article on social media
    Facebook
    X (Twitter)
    LinkedIn
    Email

    Figures

    General workflow of the 16s rDNA test: a) chronological test principle, with polymerase chain reaction (PCR) followed by subsequent detection of the specific PCR products by lateral flow immunoassay; b) the results are displayed as one test band (T, detection of 16s rDNA); and c) a control band (C), confirming the correct function of assay).

    Fig.

    General workflow of the 16s rDNA test: a) chronological test principle, with polymerase chain reaction (PCR) followed by subsequent detection of the specific PCR products by lateral flow immunoassay; b) the results are displayed as one test band (T, detection of 16s rDNA); and c) a control band (C), confirming the correct function of assay).
    Go to Article
    General test performance using the proof-of-principle cohort: a) receiver operating characteristic (ROC) curve for the 16s rDNA test based on the calculated ratio between test and control band to determine the optimal cut-off value to differentiate between periprosthetic joint infection (PJI) and non-PJI samples; and b) box-whisker plot displaying the performance of the 16s rDNA test to differentiate between PJI (n = 8) and non-PJI (n = 9) samples in the proof-of-principle cohort. Dashed lines indicate the optimal cut-off as determined by previous ROC analysis. The Mann–Whitney U test was used to obtain p-values. AUC, area under the curve; CI, confidence interval; AU, assay unit.

    Fig.

    General test performance using the proof-of-principle cohort: a) receiver operating characteristic (ROC) curve for the 16s rDNA test based on the calculated ratio between test and control band to determine the optimal cut-off value to differentiate between periprosthetic joint infection (PJI) and non-PJI samples; and b) box-whisker plot displaying the performance of the 16s rDNA test to differentiate between PJI (n = 8) and non-PJI (n = 9) samples in the proof-of-principle cohort. Dashed lines indicate the optimal cut-off as determined by previous ROC analysis. The Mann–Whitney U test was used to obtain p-values. AUC, area under the curve; CI, confidence interval; AU, assay unit.
    Go to Article
    Validation of the test performance using the validation cohort: a) receiver operating characteristic (ROC) curve for the 16s rDNA test (ratio between test and control band on the test strip) to differentiate between periprosthetic joint infection (PJI) and non-PJI samples in the validation cohort; and b) box-whisker plot displaying the performance of the 16s rDNA test to differentiate between PJI (n = 23) and non-PJI (n = 37) samples in the validation cohort. Dashed lines indicate the predefined cut-off as determined using the proof-of-principle cohort. The Mann–Whitney U test was used to obtain p-values. AUC, area under the curve; CI, confidence interval; AU, assay unit.

    Fig.

    Validation of the test performance using the validation cohort: a) receiver operating characteristic (ROC) curve for the 16s rDNA test (ratio between test and control band on the test strip) to differentiate between periprosthetic joint infection (PJI) and non-PJI samples in the validation cohort; and b) box-whisker plot displaying the performance of the 16s rDNA test to differentiate between PJI (n = 23) and non-PJI (n = 37) samples in the validation cohort. Dashed lines indicate the predefined cut-off as determined using the proof-of-principle cohort. The Mann–Whitney U test was used to obtain p-values. AUC, area under the curve; CI, confidence interval; AU, assay unit.
    Go to Article

    Metrics

    Downloaded 1511 times

    References

    1 Find in article

    Feldman DS , Lonner JH , Desai P , Zuckerman JD . The role of intraoperative frozen sections in revision total joint arthroplasty. J Bone Joint Surg [Am]1995;77-A:1807-1813.

    PubMed Google Scholar
    2 Find in article

    Tsaras G , Maduka-Ezeh A , Inwards et al. . Utility of intraoperative frozen section histopathology in the diagnosis of periprosthetic joint infection: a systematic review and meta-analysis. J Bone Joint Surg [Am]2012;94-A:1700-1711.

    Crossref PubMed Google Scholar
    3 Find in article

    Trampuz A , Hanssen AD , Osmon DR et al. . Synovial fluid leukocyte count and differential for the diagnosis of prosthetic knee infection. Am J Med2004;117:556-562.

    Crossref PubMed Google Scholar
    4 Find in article

    Ghanem E , Parvizi J , Burnett RS et al. . Cell count and differential of aspirated fluid in the diagnosis of infection at the site of total knee arthroplasty. J Bone Joint Surg [Am]2008;90-A:1637-43.

    PubMed Google Scholar
    5 Find in article

    Deirmengian C , Kardos K , Kilmartin P et al. . Diagnosing periprosthetic joint infection: has the era of the biomarker arrived?Clin Orthop Relat Res2014;472:3254-3262.

    Crossref PubMed Google Scholar
    6 Find in article

    Deirmengian C , Kardos K , Kilmartin P et al. . The alpha-defensin test for periprosthetic joint infection outperforms the leukocyte esterase test strip. Clin Orthop Relat Res2015;473:198-203.

    Crossref PubMed Google Scholar
    7 Find in article

    Wetters NG , Berend KR , Lombardi AV et al. . Leukocyte esterase reagent strips for the rapid diagnosis of periprosthetic joint infection. J Arthroplasty2012;27(Suppl):8-11.

    Crossref PubMed Google Scholar
    8 Find in article

    Bingham J , Clarke H , Spangehl M et al. . The alpha defensin-1 biomarker assay can be used to evaluate the potentially infected total joint arthroplasty. Clin Orthop Relat Res2014;472:4006-4009.

    Crossref PubMed Google Scholar
    9 Find in article

    Marín M , Garcia-Lechuz JM , Alonso P et al. . Role of universal 16S rRNA gene PCR and sequencing in diagnosis of prosthetic joint infection. J Clin Microbiol2012;50:583-589.

    Crossref PubMed Google Scholar
    10 Find in article

    Bémer P , Plouzeau C , Tande D et al. . Evaluation of 16S rRNA gene PCR sensitivity and specificity for diagnosis of prosthetic joint infection: a prospective multicenter cross-sectional study. J Clin Microbiol2014;52:3583-3589.

    Crossref PubMed Google Scholar
    11 Find in article

    Gomez E , Cazanave C , Cunningham SA et al. . Prosthetic joint infection diagnosis using broad-range PCR of biofilms dislodged from knee and hip arthroplasty surfaces using sonication. J Clin Microbiol2012;50:3501-3508.

    Crossref PubMed Google Scholar
    12 Find in article

    Birmingham P , Helm JM , Manner PA , Tuan RS . Simulated joint infection assessment by rapid detection of live bacteria with real-time reverse transcription polymerase chain reaction. J Bone Joint Surg [Am]2008;90-A:602-608.

    Crossref PubMed Google Scholar
    13 Find in article

    Lévy PY , Fenollar F . The role of molecular diagnostics in implant-associated bone and joint infection. Clin Microbiol Infect2012;18:1168-1175.

    Crossref PubMed Google Scholar
    14 Find in article

    Font-Vizcarra L , García S , Martínez-Pastor JC , Sierra JM , Soriano A . Blood culture flasks for culturing synovial fluid in prosthetic joint infections. Clin Orthop Relat Res2010;468:2238-2243.

    Crossref PubMed Google Scholar
    15 Find in article

    Esteban J , Sorlí L , Alentorn-Geli E , Puig L , Horcajada JP . Conventional and molecular diagnostic strategies for prosthetic joint infections. Expert Rev Mol Diagn2014;14:83-96.

    Crossref PubMed Google Scholar
    16 Find in article

    Frommelt L . Aspiration of joint fluid for detection of the pathogen in periprosthetic infection. Orthopade2008;37:1027-1034; quiz 1035-1036. (in German)

    Google Scholar
    17 Find in article

    Cazanave C , Greenwood-Quaintance KE , Hanssen AD et al. . Rapid molecular microbiologic diagnosis of prosthetic joint infection. J Clin Microbiol2013;51:2280-2287.

    Crossref PubMed Google Scholar
    18 Find in article

    Yang S , Ramachandran P , Hardick A et al. . Rapid PCR-based diagnosis of septic arthritis by early Gram-type classification and pathogen identification. J Clin Microbiol2008;46:1386-1390.

    Crossref PubMed Google Scholar
    19 Find in article

    Reinke S , Dienelt A , Blankenstein A , Duda GN , Geissler S . Qualifying stem cell sources: how to overcome potential pitfalls in regenerative medicine?J Tissue Eng Regen Med2016;10:3-10.

    Crossref PubMed Google Scholar
    20 Find in article

    Krenn V , Morawietz L , Perino G et al. . Revised histopathological consensus classification of joint implant related pathology. Pathol Res Pract2014;210:779-786.

    Crossref PubMed Google Scholar
    21 Find in article

    Morawietz L , Classen RA , Schröder JH et al. . Proposal for a histopathological consensus classification of the periprosthetic interface membrane. J Clin Pathol2006;59:591-597.

    Crossref PubMed Google Scholar
    22 Find in article

    Zimmerli W , Trampuz A , Ochsner PE . Prosthetic-joint infections. N Engl J Med2004;351:1645-1654.

    Google Scholar
    23 Find in article

    Osmon DR , Berbari EF , Berendt AR et al. . Diagnosis and management of prosthetic joint infection: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis2013;56:e1-e25.

    Crossref PubMed Google Scholar
    24 Find in article

    Janz V , Trampuz A , Perka CF , Wassilew GI . Reduced culture time and improved isolation rate through culture of sonicate fluid in blood culture bottles. Technol Health Care2017;25:635-640.

    Crossref PubMed Google Scholar
    25 Find in article

    Portillo ME , Salvadó M , Trampuz A et al. . Improved diagnosis of orthopedic implant-associated infection by inoculation of sonication fluid into blood culture bottles. J Clin Microbiol2015;53:1622-1627.

    Crossref PubMed Google Scholar
    26 Find in article

    Shen H , Tang J , Wang Q , Jiang Y , Zhang X . Sonication of explanted prosthesis combined with incubation in BD bactec bottles for pathogen-based diagnosis of prosthetic joint infection. J Clin Microbiol2015;53:777-781.

    Crossref PubMed Google Scholar
    27 Find in article

    Schäfer P , Fink B , Sandow D et al. . Prolonged bacterial culture to identify late periprosthetic joint infection: a promising strategy. Clin Infect Dis2008;47:1403-1409.

    Crossref PubMed Google Scholar
    28 Find in article

    Rasband WS , Image J . National Institutes of Health, Bethesda, Maryland, USA. 1997-2016. (date last accessed 25October2017).

    https://imagej.nih.gov/ij/ Google Scholar
    29 Find in article

    Altman DG , Bland JM . Diagnostic tests. 1: sensitivity and specificity. BMJ1994;308:1552.

    Crossref PubMed Google Scholar
    30 Find in article

    Meermans G , Haddad FS . Is there a role for tissue biopsy in the diagnosis of periprosthetic infection?Clin Orthop Relat Res2010;468:1410-1417.

    Crossref PubMed Google Scholar
    31 Find in article

    Srinivasan R , Karaoz U , Volegova M et al. . Use of 16S rRNA gene for identification of a broad range of clinically relevant bacterial pathogens. PLoS One2015;10:e0117617.

    Crossref PubMed Google Scholar
    32 Find in article

    Kobayashi H , Oethinger M , Tuohy MJ , Hall GS , Bauer TW . Distinction between intact and antibiotic-inactivated bacteria by real-time PCR after treatment with propidium monoazide. J Orthop Res2010;28:1245-1251.

    Crossref PubMed Google Scholar
    33 Find in article

    Cangelosi GA , Meschke JS . Dead or alive: molecular assessment of microbial viability. Appl Environ Microbiol2014;80:5884-5891.

    Google Scholar