Application of next‐generation sequencing technology in the detection of pathogenic bacteria of the periprosthetic joint infection after arthroplasty

Abstract

To investigate the application value of next‐generation sequencing (NGS) technology in the detection of pathogenic bacteria in the periprosthetic joint infection after arthroplasty. Twenty‐two cases of patients with joint infection after arthroplasty in our hospital from March 2020 to March 2021 were selected, with 11 cases of knee and 11 cases of hip, including 8 cases of male and 14 cases of female, and an average age of 63.55 ± 13.11 years old (range from 28 to 85). Microbiological culture results of synovial fluid and periprosthetic joint tissue and NGS results of periprosthetic joint tissue were collected. The detection rate of NGS and microbiological culture were calculated and statistically analysed by paired χ ² test. Among the 22 patients with joint infection after arthroplasty, the positive rate of NGS was 90.91% (20/22), whereas the positive rate of bacterial culture was 50.00% (11/22). Paired chi‐square test showed a statistically significant difference in the detection rate between the two groups (P = .0029). In the detection of pathogenic microorganism, NGS detected 12 kinds of bacteria, Staphylococcus aureus in 3 patients, Staphylococcus epidermidis in 5 cases, Streptococcus 1 case, Streptococcus dysgalactiae 1 case, Xanthomonas campestris 3 cases, Escherichia coli 2 cases, Bacillus cereus 2 cases, Klebsiella pneumoniae 1 case, Finegoldia magna 1 case, Corynebacterium klopensteriella in 1 case, Brucella 1 case, and Aspergillus flavus 1 case. Bacterial culture detected 6 kinds of bacteria, included 5 cases of Staphylococcus epidermis (including 3 cases of Methicillin‐resistant coagulase‐negative Staphylococcus, (MRSCoN)), 2 cases of Staphylococcus aureus (both Methicillin‐resistant Staphylococcus aureus, (MRSA)), 1 case of Klebsiella pneumoniae, 1 case of Staphylococcus hominis (MRSCoN), 1 case of G⁺ bacillus, and 1 case of Brucella. Compared with bacterial culture, NGS technology has some advantages in the detection efficiency, detection rate, and comprehensiveness, which might be greater diagnostic value in the joint fluid of infection after arthroplasty.

1. INTRODUCTION

Arthroplasty is the most effective treatment for bone and joint diseases, and it is one of the most common operations in the joint field. Among the complications after artificial joint replacement, periprosthetic joint infection (PJI) is the most serious complication, and it is also one of the main reasons for hip and knee revision. ¹ With the aging of our population and the development of society, more and more people have received artificial joint replacement, especially those who received artificial joint replacement in the early years. As a result of the constraints of medical technology and medical environment at that time, more and more people have suffered from prosthesis loosening and artificial joint infection after many years. Once infection occurs after artificial joint replacement, it brings a huge economic burden to patients, their families, and the entire medical industry. Timely diagnosis and identification of pathogenic bacteria are essential to improve the prognosis of patients with artificial joint replacement infection.

Early identification of pathogens can not only target the use of antibiotics, but also play an important role in the selection of treatment options. Currently, preoperative bacterial culture of synovial fluid is the gold standard for the diagnosis of PJI. However, limited by the early stage of the sampling method, culture conditions, culture duration and previous use of antibiotics, the positive rate of bacteria culture generally low. In some patients, the pathogenic bacteria could not be identified after multiple bacterial cultures, resulting in the failure of anti‐infective treatment. ²

High‐throughput sequencing, also known as “Next‐generation” sequencing technology (NGS), is an emerging microbial diagnostic technology. It can detect all nucleic acids present in the specimen at one time, including those from the host and all microorganisms. In recent years, NGS has been applied to the identification and detection of pathogens in infectious diseases, and has a high application value in blood samples. In the field of PJI, NGS applications have gradually been used by everyone. In this study, NGS technology and microbial culture were used to simultaneously detect pathogenic bacteria in joint fluid of patients with joint replacement infection, and to explore the application value of NGS in joint replacement infection.

2. MATERIALS AND METHODS

2.1. Study design

We performed a retrospective, observational, cohort study at Honghui Hospital, Xi'an Jiaotong University. According to the inclusion and exclusion criteria, a total of 22 patients who underwent joint revision surgery from March 2020 to March 2021 were selected. There were 11 cases of hip arthroplasty and 11 cases of knee arthroplasty.

2.2. Inclusion criteria

Inclusion criteria: ① patients diagnosed as PJI according to the diagnostic criteria of American Musculoskeletal Infection Society (MSIS). ② NGS test and microbial culture were performed on periarticular tissues of PJI patients. ③ The main evaluation indicators were the detection efficiency of NGS and microbial culture for pathogenic microorganisms; ④ Retrospective study.

2.3. Exclusion criteria

Exclusion criteria: ① Collected synovial fluid <2 mL; ② Culture times less than 2 times; ③ NGS results suggested that the specimens were seriously contaminated during sampling, transport, or processing (more than 2 kinds of background bacteria, and more than 1000 sequences at the same time). ³

2.4. Definitions

Diagnostic criteria of PJI after hip or knee replacement: according to the MSIS diagnostic criteria, it can be diagnosed if it meets one major criteria or more than three minor criteria. The main criteria were: ① The same pathogen was cultured from periprosthetic specimens twice. ② There was a sinus tract communicating with the joint cavity. Minor criteria: ① Elevated serum C‐reactive protein and erythrocyte sedimentation rate. ② The white blood cell count of synovial fluid increased. ③ The percentage of neutrophils in synovial fluid increased. ④ The pathological examination results of periprosthetic tissues were positive. ⑤ Single positive bacterial culture.

2.5. Collection of synovial fluid

All patients underwent joint aspiration after admission. A knee puncture was performed on the lateral side of the patella, and the joint cavity was penetrated medially and inferiorly through the lateral side of the quadriceps tendon. 5–10 mL of synovial fluid was extracted and sent immediately for bacterial culture and NGS. Under the guidance of B‐ultrasound, a hip puncture was inserted into the joint cavity parallel to the femoral neck above the greater trochanter of the lateral thigh, and 5–10 mL of synovial fluid was extracted. The synovial fluid was immediately sent for bacterial culture and NGS.

2.6. Bacterial culture

The extracted synovial fluid samples were sent to the laboratory department of our hospital for routine aerobic and anaerobic bacterial culture for 7 days. (1) Aerobic culture: the specimens were inoculated on solid AGAR plates (containing goat blood) (Wenzhou Kangtai Biological Technology Co., LTD.) and broth medium (Wenzhou Kangtai Biological Technology Co., LTD.), and the plates were cultured in a constant temperature incubator at 37°C. (2) Anaerobic culture: synovial fluid samples were inoculated into anaerobic blood culture bottles (BacT/AlerT FN, Biomerieux, France), which were placed in the Biomerieux BacT/Alert 3D microbial culture system. All synovial fluid samples were identified by Biomrieux Compact‐2 automatic microbial identification drug susceptibility system and Bruker MALDI‐TOF MS identification mass spectrometry.

2.7. Nucleic acid extraction and NGS sequencing

The synovial fluid was broken through the wall, centrifuged, and 600 L of supernatant was collected, and DNA was extracted using a DNA extraction kit (1901, Genskey, Tianjin). The extracted DNA was sonicated to 200‐300 bp fragments, followed by NGS library construction kit (1906, Genskey, Tianjin) for DNA end repair, adapter ligation, and PCR amplification. After amplification, quality control of libraries and insert sizes was performed with the use of a DNA Agilent 2100 bioanalyzer (Agilent Technologies, Santa Clara, United States), and DNA library concentration was measured with the use of fluorescence quantitative PCR. The constructed libraries were pooled and sequenced using Illumina NextSeq 550 sequencer (Illumina, USA) with a sequencing mode of SE75 and a reading volume of no less than 20 M reads. Bioinformatics analysis: The sequences with low quality, low complexity, and length <70 bp were filtered out, and the human reference genome sequence was removed. The obtained high‐quality sequencing data were compared with the microbial genome database to identify microorganisms. The microbial genome database contains 9855 bacterial, 6926 viral, 1582 fungal, 312 parasitic, 184 mycoplasmal, and 177 mycobacterial species.

2.8. Statistical analysis

SPSS 23.0 software was used for statistical analysis. The positive detection rates of NGS and synovial fluid bacterial culture in patients with PJI and the positive detection rates of the two methods in hip and knee were calculated and statistically analysed by paired χ ² test, and P < .05 was considered statistically significant.

3. RESULTS

3.1. Results of comparison between one and two detection methods

The results of bacterial culture and NGS sequencing in 22 patients are shown in Table 1. Twenty‐two synovial fluid samples of PJI patients were detected by NGS, of which 20 were positive for pathogenic microorganisms and 2 were negative. The bacterial detection rate of NGS in PJI patients was 90.91%. Pathogenic microorganisms were detected in 11 cases and no pathogenic microorganisms were detected in 11 cases by bacterial culture of synovial fluid. The detection rate of bacterial culture of synovial fluid was 50.00%.

TABLE 1.

Results of NGS and bacterial culture in 22 patients

Gender	Infection site	NGS	Bacterial culture
Male	Hip	Staphylococcus epidermidis	Negative
Male	Hip	Streptococcus	Negative
Female	Hip	Bacillus cereus	Negative
Male	Hip	Finegoldia magna	Negative
Male	Hip	Xanthomonas campestris	Negative
Female	Hip	Corynebacterium klopenstedt	Negative
Female	Hip	Xanthomonas campestris, Staphylococcus epidermidis	Staphylococcus epidermidis (MRSCoN)
Female	Hip	Xanthomonas campestris	Negative
Female	Hip	Escherichia coli	Staphylococcus hominis (MRSCoN)
Male	Hip	Bacillus cereus	Gram‐positive bacillus
Female	Hip	Staphylococcus epidermidis	Staphylococcus epidermidis (MRSCoN)
Male	Knee	Aspergillus flavus	Negative
Male	Knee	Escherichia coli	Negative
Female	Knee	Staphylococcus epidermidis	Staphylococcus epidermidis
Female	Knee	Streptococcus dysgalactiae	Negative
Female	Knee	Staphylococcus aureus	Staphylococcus aureus (MRSA)
Female	Knee	Negative	Negative
Female	Knee	Negative	Staphylococcus epidermidis
Female	Knee	Staphylococcus epidermidis	Staphylococcus epidermidis (MRSCoN)
Female	Knee	Klebsiella pneumoniae	Klebsiella pneumoniae
Female	Knee	Staphylococcus aureus, Bacterium burgeri	Staphylococcus aureus, Bacterium burgeri
Male	Knee	Staphylococcus aureus	Staphylococcus aureus (MRSA)

Comparison of the detection rate of bacteria between NGS and bacterial culture in 22 patients: the final detection rate of bacteria by the two methods was analysed at the species level. Bacillus cereus, Fenegold, Xanthomonas campestris, Corynebacterium Klopenstedt, Escherichia coli, Streptococcus dysgalactiae, and Aspergillus flavus were not detected by bacterial culture. No Staphylococcus hominis was detected by NGS. Bacterial culture detected one patient infected with two kinds of bacteria, and NGS detected two patients infected with two kinds of bacteria (Table 2).

TABLE 2.

Bacterial detection rate of NGS and bacterial culture in 22 patients

Groups	Gram‐positive bacterium					Gram‐negative bacterium
	Staphylococcus epidermidis	Staphylococcus aureus	Streptococcus	Gram‐positive bacillus*	Finegoldia magna	Staphylococcus hominis	Xanthomonas campestris	Escherichia coli	Klebsiella pneumoniae	Bacterium burgeri
NGS	5	3	2	3	1	0	3	1	1	1
Bacterial culture	5	3	0	0	0	1	0	0	1	1
χ 2	0	0	2.1	3.22	1.02	1.02	3.22	1.02	0	0
P	1	1	.15	.07	.31	.31	.07	.31	1	1

3.2. Comparison of the results of the two detection methods in patients with PJI

A total of 22 patients with PJI were simultaneously examined by NGS and synovial fluid bacterial culture. Of all 22 synovial fluid samples, 10 were positive by both assays, 10 were positive by NGS but negative by bacterial culture, 1 was negative by NGS but positive by bacterial culture, and 1 was negative by both assays. The positive rate of NGS in PJI patients was 90.91%, and the positive rate of synovial fluid bacterial culture in PJI patients was 50.00%. Ten cases of NGS and bacterial culture are all positive patients, 9 cases detected the same pathogens (epidermis staphylococcus, Gram‐positive Bacillus, staphylococcus aureus, klebsiella pneumoniae, and brinell coli), 1 case was part of the same more pathogenic bacteria detection (NGS), 1 case of completely different (NGS detection Escherichia Coli, staphylococcus hominis). Paired χ ² test was used for statistical analysis, and the difference between the two was statistically significant (χ ² = 8.84, P = .0029). The bacterial detection rate of NGS in PJI patients was much higher than that of synovial fluid bacterial culture.

3.3. Comparison of the results of the two detection methods in patients with knee PJI

NGS and synovial fluid bacterial culture were performed in 11 patients with knee PJI. Among them, 6 samples were positive for both tests, 3 samples were positive for NGS but negative for bacterial culture, 1 sample was positive for bacterial culture but negative for NGS, and 1 sample was negative for both. In knee PJI patients, the positive detection rate of NGS was 81.82%, while the positive detection rate of bacterial culture was 63.64%. Paired χ ² test was used for statistical analysis, and the difference was not statistically significant (χ ² = 0.9167, P = .3384). However, NGS has a higher bacterial detection rate in knee PJI patients than synovial fluid bacterial culture.

3.4. Comparison of the results of the two detection methods in patients with hip PJI

Of the 11 patients with hip PJI, 4 were positive by both assays, 7 were positive by NGS but negative by synovial fluid bacterial culture, none was negative by NGS but positive by synovial fluid bacterial culture, and none was negative by both assays. In patients with hip PJI, the detection rate of NGS was 100.00%, whereas the detection rate of synovial fluid bacterial culture was 36.36%. The difference between the two groups was statistically significant (χ ² = 10.27, P = .0014). NGS has a higher bacterial detection rate than synovial fluid bacterial culture in patients with hip PJI.

4. DISCUSSIONS

Patients with PJI were included in this study. Once the disease occurred, the disease progressed very rapidly. ⁴ Because of intraoperative wound contamination or poor postoperative wound healing, the patients gradually invaded the deep from the superficial wound, colonised the deep joint, and involved other parts. In our hospital, synovial fluid was extracted for bacterial culture. It usually took 3–5 days to receive the results of aerobic bacteria, and it took about 7 days to receive the results of anaerobic bacteria. This is disadvantageous for early diagnosis as well as for the selection of antimicrobial agents. Therefore, it is essential to discover and explore new detection methods.

This study investigated the value of NGS in the diagnosis of PJI after arthroplasty. In 2005, Margulies et al ⁵ performed whole genome sequencing of mycoplasma genitalium and Streptococcus pneumoniae, and NGS technology began to be applied. In 2014, Wilson et al ⁶ used NGS for the diagnosis of infectious diseases. The results show that NGS can be used to diagnose infections caused by clinically uncommon pathogens, and improve the level of clinical diagnosis. In 2018, Tarabichi ⁷ used NGS to diagnose PJI, and the results showed that the positive rate of NGS was 89.3% in patients diagnosed as infection by MSIS criteria, and 25% in patients diagnosed as non‐infection and negative by culture. ⁸ In this study, the bacterial detection rate of NGS in synovial fluid samples of PJI patients was 90.91%, while the detection rate of synovial fluid bacterial culture was only 50.00%. The positive rate of NGS was significantly higher than that of bacterial culture, and it was statistically significant. However, the positive rate of bacterial culture in this study is lower than the 58%–95% reported in the literature, ⁹ , ¹⁰ and there are the following two possible reasons. The first is because of the use of antimicrobial agents prior to sampling. Most of the patients with PJI were treated with broad‐spectrum antibiotics before admission, which reduced the number of bacteria, or the secretion itself contained antibiotics, which affected the propagation and growth of bacteria in the process of culture. Another possible reason is that the time of bacterial culture in our hospital is still not long enough, usually 1 w, not 2 w. ¹¹ Some special microorganisms such as fungi and mycobacteria are difficult to be cultured, resulting in a lower positive rate of bacterial culture than reported in literature.

If there is no timely diagnosis and effective treatment for patients with PJI after arthroplasty, bacteria can form biofilms after 3 weeks, which leads to increased difficulty in the treatment of infection. Therefore, for patients with PJI, it is very important to receive the etiological results as soon as possible. It is necessary to perform rapid, efficient, and comprehensive pathogen detection in patients with PJI after arthroplasty by NGS. The culture time of NGS is generally 2–3 days, which is beneficial for clinicians to select targeted antibiotics according to the bacteriological results as soon as possible. Another advantage of NGS is unbiased sampling, which can not only identify known pathogens, but also discover new microorganisms. ¹² In this study, we found that NGS could detect bacterial species that could not be detected by traditional bacterial culture, and NGS could detect bacterial types that could not be detected by traditional bacterial culture. NGS identified Bacillus cereus, Finegoldia magna, Xanthomonas campestris, Corynebacterium Kroppenstedtii, Streptococcus dysgalactiae and Aspergillus flavus that were not detected by bacterial culture, indicating that NGS was more sensitive than bacterial culture in the detection of the above bacteria. ¹³ The reasons why these pathogens were not detected in bacterial culture may be that Bacillus Cereus is a large gram‐positive bacterium, which is less distributed in synovial fluid and difficult to culture. Finegoldia magna is a Gram‐positive anaerobic coccus that mainly lives in the skin and the mucosa of the gastrointestinal and urogenital tract as a normal commensal, accounting for 5%–12% of all anaerobic infections. ¹⁴ It is considered to be one of the common pathogens of joint infections after prosthetic implantation. However, although Finegoldia magna has been described as a common pathogen of bone infection, its detection rate is low because of its anaerobic characteristics and the complexity of anaerobic culture. However, once infected, it tends to develop drug resistance because of slow growth, often leading to prolonged infection. Corynebacterium Klopenstedt belongs to the bacterial and thermophilic actinomycetes. It is facultative anaerobes. ¹⁵ It grows slowly, faster at 50°C and on carbon yeast fermentation broth AGAR, and is difficult to cultivate by ordinary bacterial culture. Aspergillus flavus belongs to the genus of fungi, which can cause pulmonary, external auditory canal, skin aspergillosis, etc. The culture time is long, usually 1–3 weeks, and the general culture time in our hospital is 1 week, which is difficult to culture, resulting in low positive rate of culture. ¹⁶ There were two patients in this study who were negative for NGS culture, including one who was also negative for bacterial culture. But she can still be diagnosed as PJI according to the MSIS. The reasons for negative NGS culture may be as follows: First, the results of mNGS detection were lost because of the damage of microbial DNA sequence by cross‐contamination in samples or during operation. ¹⁷ Second, because there is a large amount of host human DNA in the sample, the reading of the target sequence is affected by the interference of host information. Third, Grumaz S et al. ¹⁸ reported that antibiotic treatment could reduce the sequence number of pathogenic bacteria in samples during NGS detection, resulting in negative results. Therefore, it is questionable whether 2 weeks of antibiotic cessation before detection is enough. Fourth, insufficient sample can also lead to negative results. Therefore, strict aseptic operation is also required in the application of NGS detection, and sufficient sample should be extracted as far as possible.

NGS also has shortcomings, the main one being that microbial nucleic acids in samples can be influenced by human host nucleic acids. The vast majority of sequences (usually >99%) are derived from human hosts, which limits the sensitivity of this method, ¹⁹ and the detection of samples, reagents, and microorganisms in the laboratory environment will all affect the accuracy of the results. ²⁰ , ²¹ In addition, it is also a problem to be solved that only using NGS to detect pathogenic bacteria cannot do drug sensitivity test at the same time. Drug sensitivity analysis should be performed on the basis of positive bacterial culture, which can guide clinical medication more effectively. NGS can only identify bacteria, but cannot determine whether the detected bacteria are drug‐resistant bacteria. To ensure the success rate of joint replacement surgery, clinicians often choose high‐level antibiotics, which will lead to the generation of drug‐resistant bacteria because of the overuse of antibiotics. Therefore, although bacterial culture takes a long time and the positive rate is slightly low, it is still the gold standard for the diagnosis of clinical bacterial infectious diseases. ²² In this study, we found a case with NGS results and bacterial culture results that were completely different. NGS showed Escherichia coli, and bacterial culture showed staphylococcus hominis. The possible reason for NGS is that the specimen of NGS is the preoperative synovial fluid, which does not exclude the possibility of contamination, and the bacterial culture is the tissue cut during the operation. The tissues from four different parts were taken during the operation, of which two bacterial cultures were Staphylococcus hominis (MRSCoN), and the other two bacterial cultures were negative. After vancomycin anti‐infection treatment, the patient's hemameba, CRP, and ESR were decreased, and the infection was significantly improved. The “gold standard” for clinical microbiological diagnosis is bacterial culture, ²³ and NGS can be used as an auxiliary means of reference, especially for synovial fluid samples with low culture‐positive rate.

This study also has some shortcomings, such as small sample size will lead to large bias, and large sample research can be carried out in the future. This study shows that NGS technology can quickly obtain etiological data in the diagnosis of PJI after arthroplasty, and the detection rate is relatively high. It can detect not only the bacteria that are consistent with bacterial culture, but also the bacteria that are not detected by bacterial culture. Therefore, NGS technology has a good application value in the diagnosis of PJI after arthroplasty.

5. CONCLUSIONS

Compared with traditional bacterial culture, NGS can not only improve the detection rate of pathogenic bacteria in PJI patients, detect the pathogens that are consistent with bacterial culture, but also detect a variety of other potential pathogenic bacteria. The short detection time is more conducive to the early diagnosis and timely treatment of PJI, which has certain application value. However, there was only 22 cases in this study, so further multi‐centre randomised controlled prospective studies with larger samples are needed to improve the accuracy and reliability of the conclusions by combining the curative effects of PJI patients.

AUTHOR CONTRIBUTIONS

JH conceived and designed the research. YC and KJ wrote the manuscript. LXZ and FY performed the research. All authors read and approved the final manuscript

FUNDING INFORMATION

This work was supported by Natural Science Foundation of Shaanxi Province, China (Grant No. 2020JQ‐962).

CONFLICT OF INTEREST

The authors declare that they have no conflicts of interest to declare.

ETHICS STATEMENT

This is a retrospective study approved by the Honghui Hospital, Xi'an Jiaotong University Ethics Committee (202201005) and conducted according to the principles of the Declaration of Helsinki. The requirement for informed consent was waived by the Honghui Hospital, Xi'an Jiaotong University Ethics Committee in view of the retrospective nature of the study.

Chang Y, Jiang K, Zhang L, Yang F, Huang J. Application of next‐generation sequencing technology in the detection of pathogenic bacteria of the periprosthetic joint infection after arthroplasty. Int Wound J. 2023;20(6):2121‐2128. doi: 10.1111/iwj.14087

DATA AVAILABILITY STATEMENT

The datasets used during the current study are available from the corresponding author upon reasonable request.