Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jan 1.
Published in final edited form as: J Orthop Res. 2017 Jun 28;36(1):452–458. doi: 10.1002/jor.23611

Viable bacteria persist on antibiotic spacers following two-stage revision for periprosthetic joint infection

Dongzhu Ma #, Robert M Q Shanks §, Charles M Davis III ¢, David W Craft £, Thomas K Wood €, Brian R Hamlin ‡, Kenneth L Urish †,*
PMCID: PMC5694702  NIHMSID: NIHMS878813  PMID: 28543707

Abstract

Treatment in periprosthetic joint infection (PJI) remains challenging. The failure rate of two-stage revision and irrigation and debridement with component retention in PJI suggests that biofilm cells have a high tolerance to antibiotic chemotherapy. Previous work has demonstrated that biofilm cells have high antibiotic tolerance in vitro, but there is little clinical evidence to support these observations. The aim of this study was to determine if retrieved antibiotic spacers from two-stage revision total knee arthroplasty for PJI have evidence of remaining viable bacteria. Antibiotic poly (methyl methacrylate) (PMMA) spacers from two-stage revision total knee arthroplasty for PJI were prospectively collected and analyzed for bacterial 16s rRNA using polymerase chain reaction (PCR), reverse transcription (RT)-PCR, quantitative RT-PCR (qRT-PCR), and single genome analysis (SGA). PCR and RT-PCR identified bacterial species on 53.8% (7/13) of these samples. When initial culture negative cases are excluded, 68% (6/9) samples were identified with bacterial species. A more rigorous qRT-PCR analysis showed a strong positive signal for bacterial contamination in 30.7% (4/13) of cases. These patients did not show any clinical evidence of PJI recurrence after 15 months of follow-up. Because the half-life of bacterial rRNA is approximately a few days, the identification of bacteria rRNA on antibiotic PMMA spacers suggests that viable bacteria were present after conclusion of antibiotic therapy. This study provides evidence for the high tolerance of biofilm cells to antibiotics in vivo and the important role of bacterial persisters in PJI.

Keywords: Periprosthetic Joint Infection, Biofilm Antibiotic Tolerance, Bacterial Persisters, Total Knee Arthroplasty, 16S rRNA, PCR

Introduction

Treatment in periprosthetic joint infection (PJI) remains challenging. The failure rate of two-stage revision and irrigation and debridement with component retention in PJI suggests that biofilm cells have a high tolerance to antibiotic chemotherapy1–12. There are multiple in vitro examples to support this. For example, after exposure to 100 times the minimal inhibitory concentration (MIC) of cefazolin, approximately 10% of a mature methicillin sensitive Staphylococcus aureus biofilm remain on arthroplasty materials13. Also, antibiotic loaded poly methyl-methacrylate (PMMA) bone cement can serve as a substrate for biofilm14; 15. There are many hypothesis to explain these observations including limited diffusion of antibiotics, altered metabolism, and bacterial persisters16; 17. Bacteria persisters are defined as a subpopulation of bacteria that have a phenotypically high tolerance to antibiotics. However, there is little data to support these observations in a clinical setting.

Chronic PJI and treatment with two-stage exchange affords an opportunity to test this observation in vivo. In chronic PJI, a two-stage revision is performed where implants are removed and replaced by an antibiotic PMMA spacer while the infection resolves. After the infection has been controlled, the antibiotic spacer is removed, and final components are then inserted. If biofilm cells have a high tolerance to antibiotics, then viable bacteria could persist on the removed antibiotic PMMA spacer, even in clinical cases where the patient has no recurrence of infection.

We hypothesized that ribosomal nucleic acid (rRNA) could be used to determine the presence of viable bacteria on removed antibiotic spacers after a two-stage exchange. rRNA has an approximate half-life of several days 18; 19. The presence of rRNA six weeks after the completion of antibiotics would suggest a viable biofilm remained on the PMMA spacer in that time period. A prospective clinical study was performed to test this hypothesis. Patients undergoing a two-stage exchange for chronic TKA PJI were enrolled, and we tested for the possible presence of bacterial rRNA using multiple reverse transcriptase polymerase chain reaction (RT-PCR) techniques.

Methods

Study Design

A prospective, multicenter analytic cohort study of patients undergoing explant of antibiotic PMMA spacers for TKA PJI was performed (Level II). Institutional review board approval was obtained. Patients undergoing a two-stage exchange for TKA PJI with an antibiotic cement spacer were enrolled at the time of the removal of the antibiotic cement spacer. Antibiotic spacers were both articulating and static. Inclusion criteria included that antibiotic spacers did not culture any organisms after removal and that there was no evidence of infection recurrence for at least one year following the two-stage exchange. At the time of removal, antibiotic spacers were divided using an osteotome so that one part could be used for microbial culture and one part for rRNA analysis. As a control, a series of antibiotic PMMA spacers were formed in the operating room under sterile conditions, handled in a similar manner as the surgical specimens, and underwent identical handling in sample preparation.

Sample Preparation

A portion of the collected antibiotic PMMA spacer was sent for microbiological culture using the institution’s clinical microbiological laboratory following standard clinical protocols. The remaining spacer was then sent for analysis for bacteria rRNA. The sample was submerged in Ambion™ RNAlater® Stabilization Solution (Thermo Fisher Scientific, Carlsbad, CA). Time between specimen collection and contact with RNAlater was limited to no more than two hours. Samples were stored at −80°C.

Bacterial RNA and DNA isolation

PMMA spacer samples were sonicated for 20 minutes (Branson 3510-DTH Ultrasonic Cleaner) and solid material was removed. Samples were then centrifuged at 8,000 × g for 5 minutes at 4°C. Total RNA was isolated via Ambion™ TRIzol® Max Bacterial RNA Isolation kit (Thermo Fisher Scientific, Carlsbad, CA). Invitrogen™ TURBO DNA-free™ kit (Thermo Fisher Scientific, Carlsbad, CA) was used to separate DNA contamination from total RNA. Total DNA was isolated from the interphase and phenol-chloroform layer saved from the phase separation step above according to manufacturer’s recommendations.

Nested PCR and DNA gel extraction

Universal oligonucleotide primers, used in bacteria 16S rRNA gene amplification, are listed in Table 1. Nested PCR is a modification of PCR intended to reduce non-specific binding in products due to the amplification of unexpected primer binding sites 20. Nested PCR reactions were performed using Invitrogen™ AccuPrime™ Pfx DNA polymerase (Thermo Fisher Scientific, Carlsbad, CA). For the first round PCR, primers utilized for the 16S rRNA gene amplification were 16S-rRNA-27-F (5′-AGAGTTTGATYMTGGCTCAG-3′) and 16S-rRNA-1492-R (5′-CGGYTACCTTGTTACGACTT-3′). For the second round PCR, different sets of primers listed in Table 1 were used. Nested PCR reaction mixture was run via 1% agarose gel electrophoresis. DNA bands were extracted via QIAquick® Gel Extraction kit (Qiagen, Hilden, Germany). Oligonucleotide primers were synthesized by the Invitrogen Company (Thermo Fisher Scientific, Carlsbad, CA).

Table 1.

Universal bacterial 16S rRNA primers.

Primer Sequence Reference
16S-rRNA-27-F AGAGTTTGATYMTGGCTCAG Turner et al. 199946
16S-rRNA-337-F ASACTCCTACGGGRSGCWGCART Greisen et al. 199447
16S-rRNA-533-F GTGCCMGCMSCCGCGGTAAT Weisburg et al. 199148
16S-rRNA-785-F GGATTAGATASCCTGGTAGTC Klindworth et al. 201349
16S-rRNA-336-R GTATTACCGCGGCKGCTG Greisen et al. 199447
16S-rRNA-518-R ATTACCGCGGSKGCKGGCAC Ferris et al. 199650
16S-rRNA-536-R GTATTACCGCGGCKGCTG Dahllof et al. 200051
16S-rRNA-805-R GACTACCAGGSTATCTAATCC Tanner et al. 199952
16S-rRNA-926-R CCGTCAATTCMTTTGAGTTT Klindworth et al. 201349
16S-rRNA-1406-R ACGGGCGGTGTGTRCAA Youssef et al. 200953
16S-rRNA-1492-R CGGYTACCTTGTTACGACTT Turner et al. 199946
Open in a new tab

Single Genome Amplification (SGA) of bacterial 16S rRNA gene

Single-cell genomics (SGA)21; 22 was utilized to amplify the 16S rRNA gene from different pathogen strains from spacers with potentially polymicrobial infections. Nested PCR was performed as above by diluting the cDNA or genome DNA in the first round PCR to get less than a 25% positive rate in the second round of PCR. Then, the amplicons were subjected to agarose gel electrophoresis and gel extraction for sequencing. The PCR products were sequenced by the University of Pittsburgh HSCRF Genomics Research Core Sanger Sequencing service. Individual sequence fragments for each amplicon were inspected and edited using the BioEdit (v7.2.5) software, and followed by running BLAST to calculate sequence similarity with NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi) to identify the bacteria with the highest probability of alignment to the isolated sequence from the PMMA spacer.

Colony formation assay

The soft agar colony formation assay is an in vitro cell survival assay based on the ability of a single cell to grow into a colony16; 23. Titanium rods (10 mm) were incubated in TSB growth medium inoculated with 1×106 CFU/ml S. aureus (SH1000) for 48 hours. Rods were then washed once with 1ml PBS and given fresh medium containing 10 μg/mL cefazolin for a 28 day period. At each corresponding time point, samples were sonicated for 20 minutes and centrifuged at 8000 rpm for 5 minutes. The bacterial concentration (CFU/mL) was then determined via CFU assay on TSA II plates (Thermo Fisher Scientific, Carlsbad, CA).

16S rRNA Degradation assay

Titanium rods were incubated in TSB growth medium inoculated with 1×106 CFU/ml S. aureus (SH1000) for 48 hours. Rods were sonicated for 20 minutes and centrifuged at 8000 rpm for 5 minutes. The pellet was resuspended TE buffer (pH 8.0) and heat shock for 10 minutes at 85°C. After cool to room temperature, 10 mg/ml lysozyme and 10 mg/ml RNase A were added into it and incubated for 30 minutes at 37°C. Then, 100 μg/ml proteinase K/0.2% SDS was added and incubated at 50°C for 10 minutes. Samples were stored at 37°C until each time point. Total RNA extraction was performed as described above on each time point over a twenty-eight-day period. The rRNA quantity was detected by standard curve qRT-PCR. Pure water was used as a negative control, and all samples were analyzed without the addition of reverse transcriptase in the reaction mixture to ensure the complete removal of DNA during RNA isolation.

cDNA synthesis

Reverse transcription of RNA to single-stranded cDNA was performed using Invitrogen™ SuperScript® IV Reverse Transcriptase (Thermo Fisher Scientific, Carlsbad, CA) according to manufacturer’s recommendations. The newly synthesized cDNA was used immediately or frozen at −80°C.

Preparation RNA Standards for Staphylococci 16S rRNA gene

The standard curve method was selected to quantify rRNA copies on the PMMA spacer, as it is a simple and reliable assay that avoids the practical and theoretical problems associated with PCR efficiency assessment24. Possession of a good standard RNA template is a prerequisite for this method. To accomplish this, the 16S rRNA gene fragments were amplified from S. aureus strain (ATCC® 29213™) and S. epidermis strain (ATCC® 12228™). The primers for 16S rRNA gene amplification were Staphylococi-16S-rRNA-standard-out-F (5′-GTACTCGAGATTTATTGGGCGTAAAGCG-3′), Staphylococi-16S-rRNA-standard-out-R (5′-GCTGGATCCGGGACTTAACCCAACATCTC-3′). The fragment was cloned into pBlueScript II SK+ (Agilent Technologies, Santa Clara, CA) and linearized by XbaI as a DNA template for RNA in vitro transcription. Standard RNA template was produced by using Ambion® MEGAscript® kit (Thermo Fisher Scientific, Carlsbad, CA).

Quantitative Real Time-PCR (qRT-PCR)

SYBR green qRT-PCR

Primers used in SYBR green qRT-PCR assay based on previous studies25. 5 μl of the single-stranded cDNA obtained above was analyzed with use of PowerUp™ SYBR Green Master Mix (Thermo Fisher Scientific, Carlsbad, CA). The cycling conditions were 50°C for 10 minutes and 95°C for 5 minutes, followed by 45 cycles of 95°C for ten seconds and 62°C for 30 seconds. For all samples, the threshold cycle number (CT) at which the fluorescence values became logarithmic was determined. The Δ CT value was calculated for each sample as the difference between the sample CT and the control CT.

Standard Curve qRT-PCR

Based on the standard RNA prepared from above, total bacterial 16S rRNA copies were quantified via qPCR by using Applied Biosystems™ TaqMan® Fast Advanced Master Mix (Thermo Fisher Scientific, Carlsbad, CA). The primers and probe were designed based on the Staphylococcus 16S rRNA gene consensus region. In this study, we used Staphylococcus-qPCR-F (5′-TCTGTAACTGACGCTGAKGYK-3′) as the forward primer and Staphylococcus-qPCR-R (5′-CGAATTAAACCACATGCTCCAC-3′) as the reverse primer. The Staphylococcus-16S-rRNA-qPCR-Probe (5′-/56-FAM/CCACGCCGT/ZEN/AAACGATGAGTGCT/3IABkF/-3′) was synthesized by Integrated DNA Technologies Inc. (Coralville, Iowa). Briefly, 5 μl of single-stranded cDNA was obtained via Invitrogen™ SuperScript® III First-Strand Synthesis SuperMix (Thermo Fisher Scientific, Carlsbad, CA) for qRT-PCR and subjected to qPCR cycling as follows: 15 minutes at 95°C followed by 45 cycles of 95°C for 15 seconds and 60°C for 1 minute.

Results

Demographics and Description of Study Participants

A total of 13 patients were identified meeting inclusion and exclusion criteria. Prior to initiating treatment, periprosthetic joint infection was defined by Musculoskeletal Infection Society criteria in 100% of cases. Patients were enrolled after removal of their PMMA spacer. After removal, all antibiotic PMMA spacers did not culture any organisms. The minimum follow up time was 15 months (15–24 months), and no disease recurrence occurred. Preoperative, surgical, and post-operative care was directed by and at the discretion of the patient’s medical team. Prior to removal of the antibiotic spacer, all patients had a minimum of 6 weeks of intravenous antibiotic therapy followed by a minimum of 6 weeks of absence of antibiotics. Selection of antibiotics and dosing was determined by the primary clinical team as a unique, optimized function of the microbe antibiotic sensitivity and patient comorbidities and tolerance. Removal of the antibiotic spacer and placement of final components was determined by the surgical team based on ESR, CRP, and, in some cases, synovial cell count.

Bacterial 16S rRNA detection demonstrates considerable bacterial contamination of spacers

To determine if bacterial rRNA could be detected on antibiotic PMMA spacers, we used PCR and RT-PCR with universal 16s primers. Bacterial 16s rRNA gene or bacterial rRNA could be detected in 7 samples of the 13 samples (53.8%; Table 2). We then attempted to semi-quantify bacteria 16s rRNA levels on PMMA spacers using qRT-PCR. A separate set of primers were used in these experiments, optimized for qRT-PCR (Table 3). The mean threshold cycle number (ΔCT, the difference between the sample CT and the control CT) using universal primers was 27.63±4.27 (minimum: 15.3; maximum: 31.6). The ΔCT for sample1, 3, 10 and 11 were high, indicating a considerable amount of 16S rRNA existed on these spacer samples. Control PMMA spacers were included in all of these experiments. Standard PCR controls were used including treating samples with DNase to remove residual DNA contamination and withholding reverse transcriptase from samples as a negative control, and negligible contamination was observed.

Table 2.

The bacteria strains detected by SGA in the spacer samples

Sample ID Initial Culture species Strain-1 Strain-2 Strain-3 Match with initial culture species(Y/N)
1 Staphylococcus aureus Staphylococcus aureus Acidovorax Acinetobacter Y
2 Staphylococcus aureus Staphylococcus aureus Acinetobacter Bacillus Y
3 Culture Negative Uncultured bacterium Y
4 Culture Negative Undetected Y
5 Staphylococcus epidermidis Streptococcus Pseudomonas N
6 Staphylococcus epidermidis Undetected N
7 Cornyebacterium Undetected N
8 Culture Negative Undetected Y
9 Culture Negative Streptococcus Saccharomyces Cerevisiae N
10 Staphylococcus aureus Staphylococcus aureus Streptococcus Y
11 Staphylococcus aureus Staphylococcus aureus Y
12 Staphylococcus Undetected N
13 Staphylococcus epidermidis Undetected N
Open in a new tab

Table 3.

qRT-PCR results of different sets of primers (ΔCT) and absolute numbers

Sample ID qPCR-set1 (ΔCT) qPCR-set2 (ΔCT) qPCR-set3 (ΔCT) qPCR-Universal (ΔCT) 16S rRNA copies STD curve control
1 28.6 Undetermined Undetermined 25.8 15017 Undetermined
2 31.7 Undetermined Undetermined 29.3 0 Undetermined
3 27.6 28.4 25.1 15.3 168 Undetermined
4 28.9 26.1 29.9 31.6 174 Undetermined
5 36.9 Undetermined Undetermined 29.8 5 Undetermined
6 31.2 Undetermined Undetermined 29.4 270 Undetermined
7 33.4 Undetermined Undetermined 29.7 50 Undetermined
8 36.6 Undetermined Undetermined 29.6 86 Undetermined
9 28.8 Undetermined Undetermined 30.4 2366 Undetermined
10 19.6 Undetermined Undetermined 24.4 14567 Undetermined
11 31.9 Undetermined Undetermined 25.3 5558 Undetermined
12 32.5 Undetermined Undetermined 29.7 148 Undetermined
13 29.6 Undetermined Undetermined 28.9 157 Undetermined
Open in a new tab

In this study, a qRT-PCR standard-curve method was designed to quantify bacteria 16s rRNA copies from spacer samples (Figure 1). High 16S rRNA copies were observed in 30.8% of samples (4/13) suggesting these samples may contain viable bacteria. Control PMMA spacers were included in all of these experiments.

Figure 1.

Figure 1

Open in a new tab

16S rRNA copies in clinical spacer samples. Based on the standard RNA we prepared, a qRT-PCR standard curve was determined for bacterial 16S rRNA quantification. From the qRT-PCR standard curve, the quantity of 16S rRNA copies were calculated. 15,014, 2,366 14,567 and 5,558 16S rRNA copies were detected from samples 1, 9, 10 and 11, respectively.

Single Genome Analysis Identified Bacteria on the Spacer Identical to the Initial Cultured Organism

Single genome analysis was used to identify the species of bacteria present due to the qRT-PCR signal (Table 2). Using qRT-PCR 16s rRNA universal primers (Table 1), a bacterial species could be identified in 53.8% of spacers (7/13). In PJI, 7% to 34% of cases are reported to be culture negative26–28. When cases that were culture negative at the initiation of treatment were excluded, SGA identified the same species as initial cultures prior to initiating treatment in 40% of cases (4/10). Of those detected, many samples yielded mono- and polymicrobial bacterial communities including skin commensals and pathogens 71.4% (5/7).

16S rRNA Degradation

An in vitro assay was used to quantify rRNA degradation to test the assumption that rRNA would not be present for more than one month after an initial heat shock treatment. Viable bacteria capable of reproduction on blood agar from the biofilm were quantified. The qRT-PCR standard curve was used to quantify remaining copies of rRNA. At day 28, the bacterial 16S rRNA became undetectable (Figure 2).

Figure 2.

Figure 2

Open in a new tab

Bacterial 16s rRNA degradation. Titanium rods were incubated in TSB growth medium inoculated with 1×106 CFU/ml Staphylococcus aureus (SH1000) for 48 hours (biofilm). The rods were sonicated for 20 minutes and centrifuged at 8000 rpm for 5 minutes to collect the bacteria. The bacteria were lysed by lysozyme solution with RNase A added, followed by proteinase K/SDS treatment. Samples were stored at 37°C for up to 28 days. At each time point, total RNA extraction was performed for subsequent qRT-PCR analysis. At day 28, the bacterial 16S rRNA became undetectable.

Discussion

In this study, we found evidence suggesting viable bacteria remain on antibiotic PMMA spacers after long-term antibiotic treatment in PJI. Bacteria with a phenotypic high tolerance to antibiotics are one of the properties of bacterial persisters, and bacteria persister cells are believed to be a source of chronic or recurrent infections. There are multiple in vitro examples demonstrating biofilm cells have a high tolerance to antibiotics as compared to planktonic bacteria29–31, but there is a lack of direct clinical evidence supporting this observation. In this clinical study, qRT-PCR was used to determine if viable bacteria remained on PMMA spacers retrieved after antibiotic treatment for TKA PJI. During a two-stage revision for PJI, an antibiotic PMMA spacer is removed after there is no additional evidence of remaining infection for placement of final arthroplasty implants. High levels of bacterial 16S rRNA present on these PMMA spacers suggest viable bacteria remain on these PMMA spacers. The presence of viable bacteria on PMMA spacers after an extended course of intravenous antibiotics with no evidence of continuing clinical infection provides strong clinical evidence of biofilm antibiotic tolerance in PJI.

A series of molecular techniques demonstrated the continued presence of bacterial rRNA on the PMMA spacers. Quantitative PCR has been used for bacterial identification in many conditions, including sepsis and meningitis32–36. Using RT-PCR and semi-quantitative RT-PCR, we found a signal was present on all samples. The strong signal present on some samples provides evidence of the presence of bacteria rRNA on the PMMA spacer; however, these methods are binary with the potential for a high false positive rate, especially by using bacterial universal primers. Absolute quantification by the standard-curve method is frequently used in environmental microbiology 37. We designed a standard-curve to directly quantify copies of bacteria rRNA. Of these samples 30.8% (4/13) had high 16S rRNA copies suggestive of remaining viable bacteria (Figure 1). These results corroborated our PCR and semi-qRT-PCR results. Negative signal could be the result of three possibilities including a lack of remaining viable bacteria, a primer with insufficient specificity for the low quantities of biofilm, or the signal may have been compromised during preparation from a series of freeze thaw cycles. We modeled our standard curve primer based on the reference genome for S. aureus, but with a low biofilm mass, there could be a divergence of results between different bacteria strains based on primer specificity.

Single genome analysis provided a second piece of evidence supporting viable bacteria on retrieved PMMA spacers. In patients that were initially culture positive, we were able to isolate rRNA that corresponded with the initial cultured bacteria species prior to the initiation of treatment on 44% (4/9) of the PMMA antibiotic spacers. In the remaining 5 cases where SGA did not agree with initial culture, there were 4 samples where SGA was unable to detect a species. This suggests that either viable bacteria were not present, rRNA on the sample had been lost to repeated freeze-thaw cycles, or we did not have a specific enough PCR primer to amplify a low signal for that specific bacterial strain. In the one sample where SGA detected a species that did not agree with initial culture results, the species had a relatively similar antibiotic susceptibility pattern. The disagreement in speciation could be the result of either a contaminant on sample processing or detecting a possible polymicrobial infection where the initial cultured organism was not detected. In clinical studies, multispecies biofilm may contribute to wound chronicity 38, intracellular infection in otitis media 39 and diabetic chronic wounds 40. Our data support that the multispecies microbial biofilm may contribute to chronic PJI.

An important limitation of this study is the possible high false positive rate of 16s qRT-PCR. To limit this possibility, we tested the accuracy of the assumption that rRNA would have limited stability making it an appropriate marker for bacteria viability. Antibiotics had been discontinued for a minimum of 6 weeks prior to removal of the PMMA spacer. In our RNA degradation assay, at day 28, the 16S rRNA became undetectable (Figure 2). In other studies, rRNA has been used to detect viable bacteria with stability less than one week 41 and a weak to non-existent signal at 3 weeks 18. Degradation of RNA plays a central role in RNA metabolism. Messenger RNA (mRNA) decay and stable RNA degradation generally have been considered distinct and separate from RNA maturation19. rRNA is more stable than mRNA and believed to have a stability on the order of days 42. rRNA is generally not degraded during exponential growth, but extensive degradation of these molecules occurs under other physiologic conditions 43. Our in vitro assay agrees with this previous literature, however possess the same limitations in correlating an in vitro assay with in vivo observations. If rRNA degradation occurs on an antibiotic spacer in the treatment of PJI at a similar rate as measured with our in vitro assay, then the identified bacterial 16s rRNA could have come from only viable bacteria.

Biofilms play an important role in medical device infections based on their inherent resistance to antibiotic chemotherapy 44. The continued presence of bacterial rRNA on PMMA spacers suggests that viable bacteria remain after explant of implants and extedned systemic antibiotic therapy had been discontinued in PJI. This study extends in vitro observations that biofilms are highly tolerant to antibiotics 45. This suggests that subclinical biofilm cells remains after treatment and provides evidence for the importance of the host immune system at either removing or controlling any remaining biofilm. In other words, our data suggests that a two-stage revision for PJI does not completely remove the infection, but is a critical step in helping the host control the infection. In a two-stage revision for PJI, approximately 15% of patients are unable to sustain a reimplantation of components, and a complete two-stage revision has a treatment failure rate between 10–20% (PMID: 26378265). In these patients, our findings suggest at least a piece of the mechanism behind PJI treatment failure is the inability of antibiotics to completely remove the biofilm and provides evidence for considering the utility of long-term antibiotic suppression to control the infection in select patients at high risk for treatment failure.

Acknowledgments

The study was supported in part by funding from the National Institutes of Health (NIH) through the grants K08 AR071494 (KLU), KL2 TR000146 (KLU), KL2 TR001856 (KLU), and 5 UL1 TR0000005 (KLU, RMS) from the National Center for Advancing Translational Sciences (NCATS). Additional funding support was provided in part by the Woodward Family Endowment in Biomedical Engineering (CMD, TKW, KLU) and the Pellegrini Endowment to the Department of Orthopaedics and Rehabilitation at the Pennsylvania State University (KLU).

Footnotes

Author Contributions Statement: All authors contributed to study design, data analysis, and manuscript preparation. DM, RMS, TKW, KLU contributed to sample analysis. CMD, DWC, BRH, and KLU contributed to sample collection and preparation. All authors read and approved a final draft of the manuscript.

References

  • 1.Wilson MG, Kelley K, Thornhill TS. Infection as a complication of total knee-replacement arthroplasty. Risk factors and treatment in sixty-seven cases. J Bone Joint Surg Am. 1990;72:878–883. [PubMed] [Google Scholar]
  • 2.Brandt CM, Sistrunk WW, Duffy MC, et al. Staphylococcus aureus prosthetic joint infection treated with debridement and prosthesis retention. Clin Infect Dis. 1997;24:914–919. doi: 10.1093/clinids/24.5.914. [DOI] [PubMed] [Google Scholar]
  • 3.Deirmengian C, Greenbaum J, Stern J, et al. Open debridement of acute gram-positive infections after total knee arthroplasty. Clin Orthop Relat Res. 2003:129–134. doi: 10.1097/01.blo.0000092996.90435.35. [DOI] [PubMed] [Google Scholar]
  • 4.Hartman MB, Fehring TK, Jordan L, et al. Periprosthetic knee sepsis. The role of irrigation and debridement. Clin Orthop Relat Res. 1991:113–118. [PubMed] [Google Scholar]
  • 5.Marculescu CE, Berbari EF, Hanssen AD, et al. Outcome of prosthetic joint infections treated with debridement and retention of components. Clin Infect Dis. 2006;42:471–478. doi: 10.1086/499234. [DOI] [PubMed] [Google Scholar]
  • 6.Mont MA, Waldman B, Banerjee C, et al. Multiple irrigation, debridement, and retention of components in infected total knee arthroplasty. J Arthroplasty. 1997;12:426–433. doi: 10.1016/s0883-5403(97)90199-6. [DOI] [PubMed] [Google Scholar]
  • 7.Rand JA. Alternatives to reimplantation for salvage of the total knee arthroplasty complicated by infection. J Bone Joint Surg Am. 1993;75:282–289. doi: 10.2106/00004623-199302000-00018. [DOI] [PubMed] [Google Scholar]
  • 8.Rasul AT, Jr, Tsukayama D, Gustilo RB. Effect of time of onset and depth of infection on the outcome of total knee arthroplasty infections. Clin Orthop Relat Res. 1991:98–104. [PubMed] [Google Scholar]
  • 9.Schoifet SD, Morrey BF. Treatment of infection after total knee arthroplasty by debridement with retention of the components. J Bone Joint Surg Am. 1990;72:1383–1390. [PubMed] [Google Scholar]
  • 10.Bradbury T, Fehring TK, Taunton M, et al. The fate of acute methicillin-resistant Staphylococcus aureus periprosthetic knee infections treated by open debridement and retention of components. J Arthroplasty. 2009;24:101–104. doi: 10.1016/j.arth.2009.04.028. [DOI] [PubMed] [Google Scholar]
  • 11.Wilde AH, Ruth JT. Two-stage reimplantation in infected total knee arthroplasty. Clin Orthop Relat Res. 1988:23–35. [PubMed] [Google Scholar]
  • 12.Ocguder A, Firat A, Tecimel O, et al. Two-stage total infected knee arthroplasty treatment with articulating cement spacer. Arch Orthop Trauma Surg. 2010;130:719–725. doi: 10.1007/s00402-010-1054-y. [DOI] [PubMed] [Google Scholar]
  • 13.Urish KL, DeMuth PW, Kwan BW, et al. Antibiotic-tolerant Staphylococcus aureus Biofilm Persists on Arthroplasty Materials. Clin Orthop Relat Res. 2016;474:1649–1656. doi: 10.1007/s11999-016-4720-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Swearingen MC, Granger JF, Sullivan A, et al. Elution of antibiotics from poly(methyl methacrylate) bone cement after extended implantation does not necessarily clear the infection despite susceptibility of the clinical isolates. Pathog Dis. 2016;74:ftv103. doi: 10.1093/femspd/ftv103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Stoodley P, Nistico L, Johnson S, et al. Direct demonstration of viable Staphylococcus aureus biofilms in an infected total joint arthroplasty. A case report. J Bone Joint Surg Am. 2008;90:1751–1758. doi: 10.2106/JBJS.G.00838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Wood TK, Knabel SJ, Kwan BW. Bacterial persister cell formation and dormancy. Appl Environ Microbiol. 2013;79:7116–7121. doi: 10.1128/AEM.02636-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Wang X, Lord DM, Cheng HY, et al. A new type V toxin-antitoxin system where mRNA for toxin GhoT is cleaved by antitoxin GhoS. Nat Chem Biol. 2012;8:855–861. doi: 10.1038/nchembio.1062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Cenciarini C, Courtois S, Raoult D, et al. Influence of long time storage in mineral water on RNA stability of Pseudomonas aeruginosa and Escherichia coli after heat inactivation. PLoS One. 2008;3:e3443. doi: 10.1371/journal.pone.0003443. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Deutscher MP. Degradation of RNA in bacteria: comparison of mRNA and stable RNA. Nucleic Acids Res. 2006;34:659–666. doi: 10.1093/nar/gkj472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Yourno J. A method for nested PCR with single closed reaction tubes. PCR Methods Appl. 1992;2:60–65. doi: 10.1101/gr.2.1.60. [DOI] [PubMed] [Google Scholar]
  • 21.Clarridge JE., 3rd Impact of 16S rRNA gene sequence analysis for identification of bacteria on clinical microbiology and infectious diseases. Clin Microbiol Rev. 2004;17:840–862. doi: 10.1128/CMR.17.4.840-862.2004. table of contents. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.McLean JS, Lasken RS. Single cell genomics of bacterial pathogens: outlook for infectious disease research. Genome Med. 2014;6:108. doi: 10.1186/s13073-014-0108-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Borowicz S, Van Scoyk M, Avasarala S, et al. The soft agar colony formation assay. J Vis Exp. 2014:e51998. doi: 10.3791/51998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Larionov A, Krause A, Miller W. A standard curve based method for relative real time PCR data processing. BMC Bioinformatics. 2005;6:62. doi: 10.1186/1471-2105-6-62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Bergin PF, Doppelt JD, Hamilton WG, et al. Detection of periprosthetic infections with use of ribosomal RNA-based polymerase chain reaction. J Bone Joint Surg Am. 2010;92:654–663. doi: 10.2106/JBJS.I.00400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Scorzolini L, Lichtner M, Iannetta M, et al. Sonication technique improves microbiological diagnosis in patients treated with antibiotics before surgery for prosthetic joint infections. New Microbiol. 2014;37:321–328. [PubMed] [Google Scholar]
  • 27.Choi HR, Bedair H. Mortality following revision total knee arthroplasty: a matched cohort study of septic versus aseptic revisions. J Arthroplasty. 2014;29:1216–1218. doi: 10.1016/j.arth.2013.11.026. [DOI] [PubMed] [Google Scholar]
  • 28.Berbari EF, Marculescu C, Sia I, et al. Culture-negative prosthetic joint infection. Clin Infect Dis. 2007;45:1113–1119. doi: 10.1086/522184. [DOI] [PubMed] [Google Scholar]
  • 29.Bernier SP, Lebeaux D, DeFrancesco AS, et al. Starvation, together with the SOS response, mediates high biofilm-specific tolerance to the fluoroquinolone ofloxacin. PLoS Genet. 2013;9:e1003144. doi: 10.1371/journal.pgen.1003144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Gupta K, Marques CN, Petrova OE, et al. Antimicrobial tolerance of Pseudomonas aeruginosa biofilms is activated during an early developmental stage and requires the two-component hybrid SagS. J Bacteriol. 2013;195:4975–4987. doi: 10.1128/JB.00732-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Stewart PS. Antimicrobial Tolerance in Biofilms. Microbiol Spectr. 2015:3. doi: 10.1128/microbiolspec.MB-0010-2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Biagi E, Vitali B, Pugliese C, et al. Quantitative variations in the vaginal bacterial population associated with asymptomatic infections: a real-time polymerase chain reaction study. Eur J Clin Microbiol Infect Dis. 2009;28:281–285. doi: 10.1007/s10096-008-0617-0. [DOI] [PubMed] [Google Scholar]
  • 33.Boutaga K, Savelkoul PH, Winkel EG, et al. Comparison of subgingival bacterial sampling with oral lavage for detection and quantification of periodontal pathogens by real-time polymerase chain reaction. J Periodontol. 2007;78:79–86. doi: 10.1902/jop.2007.060078. [DOI] [PubMed] [Google Scholar]
  • 34.Rosey AL, Abachin E, Quesnes G, et al. Development of a broad-range 16S rDNA real-time PCR for the diagnosis of septic arthritis in children. J Microbiol Methods. 2007;68:88–93. doi: 10.1016/j.mimet.2006.06.010. [DOI] [PubMed] [Google Scholar]
  • 35.Smith-Vaughan H, Byun R, Nadkarni M, et al. Measuring nasal bacterial load and its association with otitis media. BMC Ear Nose Throat Disord. 2006;6:10. doi: 10.1186/1472-6815-6-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.VanGuilder HD, Vrana KE, Freeman WM. Twenty-five years of quantitative PCR for gene expression analysis. Biotechniques. 2008;44:619–626. doi: 10.2144/000112776. [DOI] [PubMed] [Google Scholar]
  • 37.Brankatschk R, Bodenhausen N, Zeyer J, et al. Simple absolute quantification method correcting for quantitative PCR efficiency variations for microbial community samples. Appl Environ Microbiol. 2012;78:4481–4489. doi: 10.1128/AEM.07878-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Ganesh K, Sinha M, Mathew-Steiner SS, et al. Chronic Wound Biofilm Model. Adv Wound Care (New Rochelle) 2015;4:382–388. doi: 10.1089/wound.2014.0587. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Thornton RB, Rigby PJ, Wiertsema SP, et al. Multi-species bacterial biofilm and intracellular infection in otitis media. BMC Pediatr. 2011;11:94. doi: 10.1186/1471-2431-11-94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Dowd SE, Wolcott RD, Sun Y, et al. Polymicrobial nature of chronic diabetic foot ulcer biofilm infections determined using bacterial tag encoded FLX amplicon pyrosequencing (bTEFAP) PLoS One. 2008;3:e3326. doi: 10.1371/journal.pone.0003326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Cenciarini-Borde C, Courtois S, La Scola B. Nucleic acids as viability markers for bacteria detection using molecular tools. Future Microbiol. 2009;4:45–64. doi: 10.2217/17460913.4.1.45. [DOI] [PubMed] [Google Scholar]
  • 42.Rauhut R, Klug G. mRNA degradation in bacteria. FEMS Microbiol Rev. 1999;23:353–370. doi: 10.1111/j.1574-6976.1999.tb00404.x. [DOI] [PubMed] [Google Scholar]
  • 43.Deutscher MP. Degradation of stable RNA in bacteria. J Biol Chem. 2003;278:45041–45044. doi: 10.1074/jbc.R300031200. [DOI] [PubMed] [Google Scholar]
  • 44.Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: a common cause of persistent infections. Science. 1999;284:1318–1322. doi: 10.1126/science.284.5418.1318. [DOI] [PubMed] [Google Scholar]
  • 45.Jefferson KK, Goldmann DA, Pier GB. Use of confocal microscopy to analyze the rate of vancomycin penetration through Staphylococcus aureus biofilms. Antimicrob Agents Chemother. 2005;49:2467–2473. doi: 10.1128/AAC.49.6.2467-2473.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Turner S, Pryer KM, Miao VP, et al. Investigating deep phylogenetic relationships among cyanobacteria and plastids by small subunit rRNA sequence analysis. J Eukaryot Microbiol. 1999;46:327–338. doi: 10.1111/j.1550-7408.1999.tb04612.x. [DOI] [PubMed] [Google Scholar]
  • 47.Greisen K, Loeffelholz M, Purohit A, et al. PCR primers and probes for the 16S rRNA gene of most species of pathogenic bacteria, including bacteria found in cerebrospinal fluid. J Clin Microbiol. 1994;32:335–351. doi: 10.1128/jcm.32.2.335-351.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Weisburg WG, Barns SM, Pelletier DA, et al. 16S ribosomal DNA amplification for phylogenetic study. J Bacteriol. 1991;173:697–703. doi: 10.1128/jb.173.2.697-703.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Klindworth A, Pruesse E, Schweer T, et al. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2013;41:e1. doi: 10.1093/nar/gks808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Ferris MJ, Muyzer G, Ward DM. Denaturing gradient gel electrophoresis profiles of 16S rRNA-defined populations inhabiting a hot spring microbial mat community. Appl Environ Microbiol. 1996;62:340–346. doi: 10.1128/aem.62.2.340-346.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Dahllof I, Baillie H, Kjelleberg S. rpoB-based microbial community analysis avoids limitations inherent in 16S rRNA gene intraspecies heterogeneity. Appl Environ Microbiol. 2000;66:3376–3380. doi: 10.1128/aem.66.8.3376-3380.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Tanner MA, Shoskes D, Shahed A, et al. Prevalence of corynebacterial 16S rRNA sequences in patients with bacterial and “nonbacterial” prostatitis. J Clin Microbiol. 1999;37:1863–1870. doi: 10.1128/jcm.37.6.1863-1870.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Youssef N, Sheik CS, Krumholz LR, et al. Comparison of species richness estimates obtained using nearly complete fragments and simulated pyrosequencing-generated fragments in 16S rRNA gene-based environmental surveys. Appl Environ Microbiol. 2009;75:5227–5236. doi: 10.1128/AEM.00592-09. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES