Abstract
Purpose
-
Growing interest surrounds the role of human gut microbiome in the development of degenerative pathologies such as osteoarthritis (OA), but microbes have recently been detected also in other sites previously considered to be sterile. Evidence emerged suggesting that even native and osteoarthritic knee joints may host several microbial species possibly involved in the osteoarthritic degeneration. This is the first systematic review critically collecting all the available evidence on the existence and composition of knee intra-articular microbiome.
Methods
-
A systematic research on the PubMed, Cochrane and Google Scholar databases was performed. Human clinical studies investigating the presence of intra-articular microbiome in native osteoarthritic knee joints with next-generation sequencing techniques were collected.
Results
-
A total of eight studies were included reporting data on 255 knees. All the included studies reported evidence supporting the existence of an intra-articular microbiome in native knee joints, with detection rates varying from 5.8% to 100%. Bacteria from the Proteobacteria phylum were found to be among the most identified followed by the Actinobacteria, Firmicutes, Fusobacteria, and Bacteroideta phyla. Proteobacteria phylum were also found to be more common in osteoarthritic knees when compared to healthy joints. Furthermore, several pathways correlating those microbes to knee OA progression have been suggested and summarized in this review.
Conclusions
-
Evidence collected in this systematic review suggests that the native knee joint, previously presumed to be a sterile environment, hosts a peculiar intra-articular microbiome with a unique composition. Furthermore, its alteration may have a link with the progression of knee osteoarthritis.
Introduction
Osteoarthritis (OA), most commonly developing in the knee joint, is the main cause of disability in older adults and is characterized by pain, loss of function, and decreased quality of life (QoL). Nowadays, an estimated 240 million individuals worldwide have symptomatic OA, including in 10% of men and 18% of women aged 60 and older (1). Its incidence is rapidly increasing due to an aging and obese population, which causes a huge financial burden globally, expected to be in most countries up to 2.5% of the gross domestic product (2). So far, no effective treatment has shown efficacy in reversing OA progression (3, 4, 5), with joint replacement as the only definitive choice available. This has led to an increasing number of arthroplasty procedures done worldwide despite periods of economic downturn (6) and an average age of patients undergoing joint replacement around the mid-60s, but with increasing numbers of patients younger than 60 years (7). Obviously, this exposes patients to a higher lifetime risk of revision and a not insignificant economic burden, given that in the United States alone the annual cost exceeds US$15 billion (8). Therefore, increasing interest is oriented toward a better understanding of the pathogenesis of OA, with the aim of addressing the trigger factors and implementing preventive actions to delay or slow down its progression. In consequence, as the pathogenesis and treatment of OA have not been completely elucidated, a better knowledge of the natural history of early OA will help identify new targets for intervention (9, 10).
Recently, studies have consistently revealed the fact that both local and systemic inflammations play pivotal roles in OA (11). Among others, gut microbiota is responsible for a series of metabolic, immunological, structural, and neurological functions, such as maintenance of metabolic homeostasis, development and maturation of immune system, resistance to infections, and production of neurotransmitters (12, 13). Dysbiosis, which has emerged as a hidden risk factor inducing the production of pro-inflammatory cytokines and bacterial metabolites, may boost the pathophysiological mechanisms of OA (14). Indeed, a ‘gut–joint’ axis in the development of OA has been hypothesized (15, 16) and preclinical studies on animal models support the idea that migration of pathogenic bacteria from the gut to the synovium in OA patients can contribute to the pathogenesis of this condition (17, 18). Moreover, recent evidence shows that the severity of knee joint damage and synovial inflammation in patients with OA was positively linked to high levels of lipopolysaccharide (LPS), and the production of intestinal bacteria and LPS could induce inflammation and damage in joints through macrophage activation and damage-associated molecular patterns (DAMPs) (19, 20). Thus, the products and metabolites of the gut microbiota appear to influence OA progression by activation of the immune system through different pathways (21).
A great impulse in this quest was given by the introduction of next-generation sequencing (NGS). Indeed, NGS techniques and metagenomics can provide a view of the transcriptome of the host tissue as well as capture all microbial genomes (i.e., bacteria, fungi, and viruses) present in a given sample. This has led to new opportunities to investigate the role of microbiota in sites traditionally assumed to be sterile and has allowed the finding of sequencing data from cerebrospinal fluid (22), amniotic fluid (23), and benign breast tissue (24), suggesting that a native microbial DNA signal could also be present in other sites such as native joints. In fact, there is already evidence indicating the existence of a shoulder joint microbiome (25). Furthermore, prior studies have identified bacteria in hip and knee implants requiring revision for presumed aseptic failure (26), and others, evaluating the utility of NGS for the diagnosis of periprosthetic joint infection (PJI), have identified the presence of microorganisms in 13.9%–71% of patients in the control group without PJI (27, 28, 29). To date, it is still impossible to establish if organisms found with NGS techniques in patients undergoing total knee replacement (TKR) revision for aseptic failure must be attributed to false negatives of standard diagnostic tools such as cultures, or if they are to be seen as initial evidence of the existence of a knee intra-articular microbiome. To further address this issue, NGS evaluations on native OA knee joints are needed, and several studies on this specific topic have been published in the last years. To our knowledge, to date, there is no review summarizing all this evidence; therefore, given the clear value of understanding the potential role of such a microbiome, the aim of the present review is to summarize all the current evidence on the existence of the intra-articular microbiome and its potential correlation with the development of knee OA (KOA).
Materials and methods
The present systematic review was conducted based on the method recommended by the preferred reporting Items for systematic review and meta-analyses (PRISMA) guidelines (2020) (30) and following the PRISMA checklist indications (31). Extensive research on the PubMed, Cochrane, and Google Scholar databases was performed on July 15, 2023, using the following words: (microbiome OR microbiota OR bacteria) AND (knee osteoarthritis OR knee OA). The screening process and analysis were conducted separately by three independent observers (PC, GA, and VL). First, the articles were screened by title and abstract. The following inclusion criteria for relevant articles were used during the screening: (1) clinical trials on humans (2) written in English; and (3) dealing with the sequencing of the knee intra-articular microbiome of patients diagnosed with primary osteoarthritis. Exclusion criteria were: (1) articles written in other languages; (2) animal or in vitro trials; (3) reviews and meta-analyses; (4) trials on joints different from the knee; (5) trials focusing only on revision TKR not including at least an evaluation on native joints; and (6) trials dealing with articular pathologies different from primary osteoarthritis.
Secondly, the full text of the selected articles was screened according to the inclusion and exclusion criteria. Thirdly, the reference list of all the retrieved articles was further screened. A flowchart of the systematic review is provided in Fig. 1. Fourthly, discrepancies between the three reviewers were resolved by discussion and consensus, and finally, the results were approved by a senior investigator (TB and EK).
PRISMA 2020 flow diagram for the present systematic review reporting search strategy and outputs.
Citation: EFORT Open Reviews 9, 10; 10.1530/EOR-23-0191
Results
Studies selection and characteristics
Four hundred and thirty-one studies were screened according to the inclusion and exclusion criteria. A total of eight studies were finally included in this systematic review (Table 1). All the included studies were cross-sectional evaluations, while longitudinal studies on microbiome presence and subsequent OA development are totally lacking in the present literature. Seven of the eight included studies were prospective clinical studies, whereas the study from Tsai et al. (32) was a retrospective case-control study in which RNA-sequencing data from synovial tissue of OA patients was aligned to a library of microbial reference genomes to identify microbial reads indicative of different microbial abundance between the two cohorts.
Main features and evidence collected from the included studies on knee intra-articular microbiome.
Authors | Cohort | Main findings | Additional findings |
Goswami et al., 2023 (38) | 113 OA patients undergoing 73 TKR and 40 THR |
|
|
Torchia et al., 2019 (33) | 40 knee OA patients undergoing primary TKR |
|
|
Zhao et al., 2018 (36) | 125 patients with RA and 58 with OA undergoing knee joint surgery |
|
|
Fernández-Rodríguez et al., 2023 (35) | 55 patients/65 knees divided into 5 groups: 15 non-OA, 14 OA, 14 aseptic knee revision, 12 septic knee revision, 10 controlateral non-OA |
|
|
Tsai et al., 2020 (32) | 24 synovial samples: 14 osteoarthritic knees, 10 healthy controls |
|
|
Tarabichi et al., 2018 (27) | 17 primary (8 TKR, 9 THR) and 65 revisions for septic and aseptic failure (39 TKR, 26 THR) arthroplasties |
|
|
Dunn et al., 2020 (34) | 34 hips and 31 knees undergoing TKR/THR, cadaveric samples, mice |
|
|
Témoin et al., 2012 (37) | 36 patients with periodontal disease and knee arthritis (OA or RA) |
|
|
NGS; next-generation sequencing, OA; osteoarthritis, OTUs; operational taxonomic units, PJI; periprosthetic joint infection, RA; rheumatoid arthritis, THR; total hip replacement, TKR; total knee replacement
The sequencing of samples from native osteoarthritic knee joints was conducted in all the included studies: the evaluation was limited to KOA patients (33) or associated with the sequencing of healthy joints (32, 34, 35), arthritic non-OA joints (36, 37), septically and aseptically failed TKR (27, 35, 37) or hip joints (27, 34, 38).
A total of 255 native knee OA joints were sequenced in the included studies: the biggest cohorts were the ones from Goswami et al. (73 patients) (38), Zhao et al. (58 patients with OA) (36) and Torchia et al. (40 patients with OA) (33), while the other cohorts were particularly limited in number (8–31 patients with OA). All the included studies were published in the last 5 years, except the one from Temoin et al. from 2012.
Intra-articular samples generally included synovial fluid and/or synovial tissue, while tissue swabs were performed and evaluated only by Tarabichi et al. and Goswami et al. Interestingly, Temoin et al. considered a cohort composing only patients diagnosed with both periodontal disease and arthritis (25 OA, 11 rheumatoid arthritis) to investigate possible pathogenetic correlations and therefore performed the sequencing of both synovial fluid and periodontal plaque swabs (37). Dunn et al. performed a unique evaluation: eroded and intact sections of 34 hip and 24 knee cartilages were obtained from patients undergoing arthroplasty for end-stage primary OA and were compared to cadaveric OA-free cartilage controls to investigate eventual microbial signature differences in each different tissue (34).
Evidence on OA knee intra-articular microbiome existence
Albeit limited to eight studies for a total of 255 native osteoarthritic knee joints, all the included studies reported evidence supporting the existence of an intra-articular microbiome in native osteoarthritic knee joints.
The proportion of KOA patients in which NGS detected microbes varied consistently from 5.8%, in the study by Temoin et al., to 30% and 35%, in the studies by Torchia and Tarabichi et al, to 96.5%, in the study by Goswami et al. (including both knee and hip OA without a distinct evaluation), to 100%, in the study by Zhao et al. (27, 33, 36, 37, 38). The remaining studies did not report the proportion of samples in which NGS identified microbes (32, 34, 35). In Goswami et al.’s study, NGS led to positive results in 113 out of 117 joint replacements (73 TKR and 40 total hip replacement THR) for a total of 578 different operational taxonomic units (OTUs) (38). Similarly, Tsai et al. identified 299 different OTUs in the 14 KOA samples (versus 84 OTUs found in non-OA patients) (32), and Fernandez-Rodriguez et al. identified 77 OTUs in the 55 included patients (35). Furthermore, they also reported that the highest number of species was found in native OA knees when compared to non-OA knees and aseptic and septic revisions (P < 0.035) (35). Torchia et al. reported an average number of organisms identified by NGS of 4.6 per patient (33) .
As per the other evaluated cohorts, NGS identified microbes in 25% of aseptic revisions in those with negative cultures, while the detection rate in septic revisions was 81.8% in those with negative cultures and 94.1% in those with positive cultures (27). Zhao et al. included patients with both OA (58 patients) and rheumatoid arthritis (125 patients), and NGS identified microbes in 100% of both cohorts (36).
Sequencing of sterile control was conducted by Torchia et al. and no microorganism was found with NGS, thus excluding ambient contamination (33). In the other four studies, controls were collected, but results were not reported in the manuscripts (34, 35, 36, 37).
Characterization of knee OA intra-articular microbiome
The composition and relevant abundance of each microbial species were found to be different in each included study, but some similarities still emerged from the comparison.
Bacteria from the Proteobacteria phylum were found to be among the most identified in native osteoarthritic knees. Zhao et al. reported that Proteobacteria was the most identified phylum in their cohort of patients with both OA (55.1% of synovial tissues and 39.1% of synovial fluids) and rheumatoid arthritis (69.0% of synovial tissues and 24.9% of synovial fluids), followed by the BacteroidetaandFirmicutes phyla (36). Dunn et al. identified substantial increases in the proportion of constituent microbial DNA from gram-negative organisms in OA patients compared to non-OA cadaveric controls (P = 0.02) and, among OA specimens, Proteobacteria phylum were particularly enriched. Generally, they reported that microbial alpha-diversity was reduced in human OA vs non-OA controls (P < 0.0001), and in hip samples vs knees (P < 0.0001), suggesting that an increase in OA progression could be linked both to a reduction in microbial diversity and to an increase in the proportion of gram-negative microbes (34).
Torchia et al. found Escherichia coli sp., again belonging to the Proteobacteria phylum, to be the most common microorganism detected in their 40 patients undergoing primary TKR (33) and the same species was among the five most abundant in the 117 joint replacements from the Goswami et al. study, together with Cutibacterium, Staphylococcus, Acinetobacter, and Pseudomonas sp (38). The latter emerged as relevant in Tsai et al.’s retrospective library-based study: among the 43 microbial species that were significantly dysregulated between OA and healthy patients, more than half of these were Pseudomonas sp.(32). On the other hand, Fernandez-Rodriguez et al. reported a high abundance of Proteobacteria phylum in the OA, aseptic revision, and contralateral non-OA groups but further specified that none of the OTUs observed in the study were members of the Pseudomonas sp.(35). This difference could be a result of the microbiome being highly dependent on geography: evidence exists that Pseudomonas sp. has a high prevalence in the microbiome of the Australian population as the one considered in Tsai et al. (39, 40) and may have a lower prevalence in other populations. Furthermore, this difference may be the result of processing different clinical samples (biopsies for Tsai et al. vs synovial fluid for Fernandez-Rodriguez et al.), which could have affected the absolute and relative abundance of different species.
The impact of geography on intra-articular microbiome was evident also in the only multicenter study included in this review, where the hospital of origin of the sample was associated with beta diversity (i.e., bacterial composition), explaining 18.5% of the observed variation among individuals (38).
The four other phyla frequently identified in native osteoarthritic knees were Actinobacteria, Firmicutes, Fusobacteria, and Bacteroideta.
Actinobacteria phylum was reported with high frequency in three studies: Cutibacterium sp. was in all the groups composing the cohort of Fernandez-Rodriguez et al.’s study (35) and the entire Actinobacteria phylum was among the dominant in native joints in the remaining studies (27, 38).
Firmicutes phylumwas generally abundant in both osteoarthritic (34, 36, 38) and non-osteoarthritic (35) knees. Indeed, Cutibacterium, Paracoccus, and Staphylococcus species were reported to be dominant in non-osteoarthritic knees, but the latter was also found to be the most prevalent in septic revisions with positive cultures (35) and culture-negative PJI (27). As already mentioned above, Firmicutes, along with the BacteroidetaandProteobacteria phyla, were the most abundant phyla in all synovial samples of the 183 patients (OA and RA) considered by Zhao et al. (36).
Dunn et al., despite reporting an increase in gram-negative microbes in osteoarthritic samples, identified Firmicutes as a characterizing phylum for knee OA samples while hip OA samples were enriched in Proteobacteria phylum(34).
Fusobacteria phylum was among the three detected with high percentages in native joints by Tarabichi et al. (27) and Fusobacterium nucleatum sp.was found in four out of the five positive arthritic cases by Temoin et al. (37). Interestingly, in their study, identical bacterial clones (F. nucleatum and Serratia proteamaculans spp.) were detected in both the synovial fluid and dental plaque samples of patients diagnosed with both arthritis and periodontitis, suggesting the possibility of infection translocating from the periodontal tissue to the synovium (37).
Bacteroideta phylumwere reported as relevant only in one study: Zhao et al. found Porphyromonas and Bacteroides sp. in 100% (183/183) of the synovial samples they collected. Specifically, they further reported that the synovial tissue of patients with OA was mainly composed of Bacteroides, Megacoccus, Haemophilus, Porphyromonas, and Streptococcus spp., while the synovial fluid was predominantly composed of Bacteroides sp.(36).
Microbiome presence in healthy knees
Only 35 healthy, non-arthritic knee joints were evaluated in the included works, thus limiting the generalizability of the specific findings. Nonetheless, all three studies comparing OA to non-OA patients, which aimed to find different microbial signatures between the two, found Proteobacteria phylum to be more abundant in OA, while Cutibacterium, Staphylococcus, and Paracoccus spp. were dominant in native non-OA knees, comprising nearly 75% of the bacterial abundance (32, 34, 35). Interestingly, Tsai et al. reported that not all 84 OTUs found in non-OA patients were also among the 299 OTUs found in the OA group, with only 36 overlapped (32). As already mentioned above, in their study, 43 microbial species were identified to be significantly dysregulated between OA and non-OA patients, with almost half being Pseudomonas sp.(32).
Intra-articular microbiome and osteoarthritis: is there a correlation?
A possible link between extra and intra-articular microbiome and osteoarthritis was investigated by Temoin et al.: a cohort of 36 patients diagnosed with both periodontal disease and arthritis (25 OA and 11 rheumatoid arthritis) were evaluated by performing the sequencing of both intra-articular synovial fluid and periodontal plaques. The group reported that in five of the 36 synovial fluids (13.9%) bacterial DNA was found, and four of these five were F. nucleatum sp. Narrowing the evaluation to native knee OA patients, the detection rate was brought down to 5.8% (1/17 patients). Interestingly, F. nucleatum sp. was also found in the periodontal plaques of two of those five patients, suggesting the possibility that microbial translocation from the oral cavity to the knee joint had occurred (37). Unfortunately, the study did not include non-osteoarthritic patients; therefore, it is not possible to infer any causative connection between F. nucleatum sp.and OA development.
On the other hand, this possible causative connection was investigated in the cohort of Tsai et al. in which Gene Set Enrichment Analysis and the Weighted Gene Co-expression Network Analysis were conducted to correlate the expression of genes in immune signaling pathways and signatures involved in osteoarthritis development to microbes that were found to be dysregulated between OA and non-OA patients. Among those dysregulated species, nine were found to be significantly correlated to immune signatures and OA pathways, with five of those being part of Pseudomonas sp. The authors concluded that those microbes might play a role in increased inflammation-induced extracellular matrix remodeling and could be involved in decreased cell signaling pathways crucial for joint and immune function. More specifically, those microbes were associated with the downregulation of both neutrophil degranulation and Wnt and Rho GTPase signaling pathways, with all factors being linked to increased catabolic activity. Furthermore, those microbial species were found to have a positive correlation with activated mast cell infiltration and a negative correlation with CD8+ T cells. Although high CD8+ T cells seem to be linked to OA progression, the authors remarked on the importance of the CD4+/CD8+ ratio that has been reported to be higher in OA synovial tissue. The reported suppression of CD8+ cells by high microbial abundance could therefore lead to a skewed CD4+/CD8+ ratio, thus promoting knee OA progression (32).
Dunn et al. performed a functional analysis and identified several bacterial functional pathways predicted to be altered in association with OA: an increase in LPS production, phosphatidylinositol signaling, and nitrogen metabolism and decreases in sphingolipid metabolism were associated with both hip and knee OA (34).
The impact of previous procedures on intra-articular microbiome presence and diversity was investigated in two studies: Torchia et al. stated that all five patients who underwent prior knee surgical procedures did not have any organisms identified by NGS at the time of TKA (33) while Goswami et al. found prior corticosteroid injection within 6 months before arthroplasty to be associated with an elevated abundance of several lineages, with Corynebacterium kroppenstedtii sp. being the most common species (60% of previously injected patients) (38). Nevertheless, it is worth mentioning that those presented are exclusively cross-sectional studies, while longitudinal long-term studies on healthy patients subsequentially developing knee OA are lacking in the present literature. Studies with this specific design could help better define the real causative relationship between intra-articular microorganisms and knee OA development.
NGS technique in knee intra-articular microbiome detection
The included studies adopted either the Illumina (4) or the IonTorrent (2) platform to perform NGS, while Goswami et al. (38) used both sequencing platforms because their study spanned a period during which their diagnostic laboratory switched from IonTorrent to the Illumina platform. Unfortunately, there are no studies differentiating the impact of these two specific platforms on the detection of intra-articular microbiome, thus limiting the possibility of comparing the detection rates in the included studies. Nonetheless, NGS generally appears as a precious and effective tool in detecting knee intra-articular microbiome. Tarabichi et al. found NGS to have a sensitivity of 89.3% and a specificity of 73% in detecting any bacteria in revision arthroplasties, which makes NGS more sensitive but less specific when compared to cultures (60.7%, 97.3% respectively) (27). In the same study, NGS and cultures had a concordance of 88.2%, while Fernandez-Rodriguez et al. reported that NGS confirmed the culture results in all the patients with PJI (35). Furthermore, NGS was powerful in detecting microorganisms in patients with PJI: Tarabichi et al. reported an 89.3% detection rate in the cases considered to be infected, 94.1% in those with positive cultures, and 81.8% in those with negative cultures diagnosed as septic using the Musculoskeletal Infection Society (MSIS) criteria (27).
A remarkable variability in the included studies was also found in the type of samples that were used to detect intra-articular microbiome, including synovial fluid, collected either percutaneously or intra-operatively, synovial tissue, and swabs. The results on their specific efficacy are contrasting; therefore, until now, it is still not possible to determine the best sample to detect intra-articular microbiome. Goswami et al. found swabs to outperform both synovial fluid and synovial tissue in terms of profiling capacities (38) while Tarabichi et al. had positive samples only from synovial tissue, whereas swabs and fluids were all negative at NGS (27).
Discussion
The main finding of the present systematic review is the existence of a native intra-articular knee microbiome composed of different microbial phyla that can be identified with NGS procedures. Nevertheless, it is worth mentioning that, through this systematic review of the present literature, only eight studies for a total of 255 native osteoarthritic knee joints were available, thus remarking the paucity of the evidence on this specific topic and limiting the generalizability of these findings.
In recent times, interest in identifying correlations between alterations in the human microbiota and the development of osteoarthritis has grown enormously. This was recently investigated in a study systematically reviewing evidence on the gut–joint cross-talk (41): the authors, despite reporting a lack of high-quality studies on the topic, concluded that the available evidence suggests that gut microbiome dysbiosis seems to aggravate OA by enhancing the immune response, thus increasing the inflammatory state of the articular environment. Specifically, gut microbiome dysbiosis seems to induce higher levels of systemic LPS and subsequent elevation of other proinflammatory factors, such as TNF-α, IL-1b, IL-6, and c-reactive protein (CRP), thus supposedly enhancing OA progression (41). From a clinical point of view, studies have identified a clinical correlation between pain scores in OA patients, and the bacterial composition of stool samples (42, 43) and small clinical trials suggest that modification in the gut microbiome could be linked to symptomatic improvements in patients diagnosed with knee OA (44, 45).
This ‘indirect’ gut–joint cross--talk, mediated by systemic inflammatory factors, may not be the only pathway linking microbial species to the development of osteoarthritis. Indeed, with the technological advance of tools such as NGA, evidence of microbes was found in sites previously presumed to be sterile: meconium, amniotic fluid (23), cerebrospinal fluid (22), and breast tissue (24) were all found to host microbial species. Furthermore, a study from Qiu et al. reported the presence of a low-abundance microbiome in the rotator cuff and, potentially, in other shoulder tissues of patients undergoing primary arthroplasty surgery (25).
As for the knee joint, NGS has been proven to be a precious tool in the identification of microbes in PJIs, detecting species that traditional cultures may fail to isolate (28, 29). Furthermore, reports are emerging on NGS isolating microbes in joint replacements undergoing revisions for presumedly aseptic failure: Carr et al. collected 248 samples from 41 patients undergoing 21 TKR and 20 THR revisions for clinically aseptic failures (4 of 41 later found to have positive intra-operative cultures) and found a higher and differentially abundant microbial diversity in patient samples compared to that of open-air controls (46). There is still a lack of agreement on whether this must be considered an initial evidence of a true microbiome present in native knee joints (already prior to primary TKR surgery) or a clinically uninfected joint implants’ colonization. Initial evidence of a native knee intra-articular microbiome has been recently published, but a comprehensive review of these evidences is not available in the literature. For this reason, and to further clarify this topic, we performed a systematic review of the literature including all the available studies investigating the presence and composition of the native osteoarthritic knee microbiome and excluding studies that included only joint revision surgeries and articular pathologies different from primary OA.
The first evidence found on our review is the paucity of studies on the specific topic: we managed to collect only eight studies with a total of 255 native OA knee joints evaluated using sequencing techniques. This paucity raises some relevant and legitimate doubts about the generalization of the presented findings but should serve as a boost in the conduction of new, possibly longitudinal, studies. Indeed, all the included studies were cross-sectional studies that, collecting data at a single time point, can be useful in identifying initial correlations but are ineffective in ascertaining causality effects. On the other hand, longitudinal studies, which collect data at multiple time points, could better clarify the causative role of specific microbiomes found at baseline in the progressive development of knee OA. Nonetheless, all the included studies found evidence of the presence of a native intra-articular knee microbiome, with percentages of positive samples ranging from 5.8% to 100% (36, 37). When considering the largest cohorts, NGS led to positive results in 96.5% (113/117) of combined hip and knee native joints (38), 100% of the 58 osteoarthritic knee joints (36), and 30% of the 40 patients undergoing TKR for primary knee OA (33).
As for the composition of the identified microbiome, five microbial phyla were found to be dominant in native osteoarthritic knees: Proteobacteria, Actinobacteria, Firmicutes, Fusobacteria, and Bacteroideta. The Proteobacteria phylumwas the most identified in the majority of the studies (33, 34, 35, 36) and an increase in gram-negative constituents in the microbial DNA of OA joints was generally reported. Among Proteobacteria, E. coli sp. (33) and Pseudomonas sp. (32) were frequently identified. Three studies included both osteoarthritic and non-osteoarthritic knee joints and tried to compare the identified microbiomes to evaluate if specific microbes could be present exclusively in OA patients, thus suggesting a possible pathogenetic link (32, 34, 35). It is worth mentioning that the total amount of native, healthy, non-OA joints in the included studies was limited to 25 joints in addition to 10 cadaveric samples. Nonetheless, all three studies found Proteobacteria phylumto be more abundant in OA, while Cutibacterium, Staphylococcus, and Paracoccus spp.were dominant in native non-OA knees, comprising nearly 75% of the bacterial abundance in the study by Fernandez-Rodriguez et al. (32, 34, 35). An interesting study from Hammad et al., which was not included in the present review since it focused on patients diagnosed with rheumatoid arthritis, similarly found that the microbiome of nine healthy controls was predominated by Proteobacteria (83.5%) and Firmicutes (16.1%) and, to a much lesser extent, the Actinobacteria (0.2%) and Bacteroideta (0.1%) phyla (47). Those evidences agree with the present literature reporting Proteobacteria phylum as a major constituent of the gut microbiome (48, 49) and its expansion as a possible actor in organ dysfunction (50).
On the other hand, Firmicutes phylum, also known to be constituents of the normal gut microbiome (51), was generally abundant in both osteoarthritic (34, 36, 38) and non-osteoarthritic (35) knees. Dunn et al. further found the Firmicutes phylumto bea characterizing phylum for knee OA samples while hip OA samples were enriched in Proteobacteria phylumand reported that microbial alpha-diversity was generally reduced in human OA vs non-OA controls (P < 0.0001) (34) and this reflects the reduction in microbiome diversity seen in other rheumatic diseases (52, 53, 54).
Three of the eight included studies tried to further evaluate how the intra-articular microbiome could impact OA induction and progression. Temoin et al. found identical bacterial clones (F. nucleatum and S. proteamaculans sp.) in both the synovial fluid and dental plaque of patients diagnosed with both periodontitis and knee osteoarthritis, suggesting a translocation of the bacteria from the oral cavity to the knee joint (37). Tsai and colleagues found that 43 microbial species were significantly dysregulated between OA and healthy patients and further determined specific functional patterns related to those dysregulated species (32): the two main pathways that were found to be targeted were decrease in cell signaling pathways crucial for joint and immune function and an increase in inflammation-induced extracellular matrix remodeling and catabolism, supporting the preclinical literature that reports that microbes seem to enhance the joint catabolism in OA pathogenesis (55). Lastly, Dunn et al. performed a functional analysis in both non-OA and OA samples. In the latter, found several bacterial functional pathways previously reported to be altered in similar conditions, such as the increase in LPS production, phosphatidylinositol signaling and nitrogen metabolism, and decreases in sphingolipid metabolism (34).
It is worth mentioning that this research has some intrinsic limitations. First, we chose to include only studies evaluating patients with native knee osteoarthritis in which intra-articular samples had been sequenced to detect microbes. This led to the exclusion of several studies including exclusively TKR revision or other rheumatic diseases that have nonetheless been mentioned above in the text. This further led to a restricted number of included studies (eight) and knee joints (255), thus limiting the generalization of the presented results. The generalizability of these results is additionally limited by the relevant methodological variation found in the included studies in terms of designs, cohorts, sample collection, and sequencing techniques. Secondly, it is worth mentioning that NGS detection does not prove the presence of live bacteria or causation of disease, and therefore clinical studies with this specific aim are required. Furthermore, no longitudinal, long-term studies on healthy patients were found in the present literature. This prevents us from clearly identifying the real cause–effect relationship between the intra-articular microbiome and subsequent knee OA development. Lastly, contamination cannot be definitively ruled out in the majority of the included studies, given that only one reported the NGS evaluation of sterile controls (33).
Future research will have to deal with these limitations. Specifically, further evidence is needed from bigger cohorts of arthritic and non-arthritic joints and mainly from longitudinal studies on healthy patients at risk of knee OA in order to better define the impact of different microbial species in the development of knee osteoarthritis. Subsequently, future research should then focus on interventional strategies targeted at microbiome modifications with the aim of possibly altering the osteoarthritic degeneration.
Conclusion
Although limited to eight studies, initial evidence of intraarticular microbiome was found in both healthy and OA native knees. Further studies will be needed to further support this evidence and to assess the possible correlation between such presence and the onset and evolution of OA disease. We believe that those presented could be precious references inspiring researchers to perform new studies on this relevant topic.
ICMJE conflict of interest statement
All the authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding statement
This work was partially supported by ''Ricerca Corrente'' funding from Italian Ministry of Health to IRCCS Humanitas Research Hospital as part of the Keystone project, grant number GR-2019-12371158.
Author contribution statement
TB conceived the study, developed the research protocol, and coordinated the study. PC, GA, and VL performed the literature search, screened the manuscripts, and performed data abstraction. TB, PC, GA, and VL prepared the first manuscript draft. EK, MR, and MM finally contributed to final edits and critical revisions prior to submission. All the authors gave their approval to the final version of the present manuscript.
References
- 1↑
Abramoff B, & Caldera FE. Osteoarthritis: pathology, diagnosis, and treatment options. Medical Clinics of North America 2020 104 293–311. (https://doi.org/10.1016/j.mcna.2019.10.007)
- 2↑
Puig-Junoy J, & Ruiz Zamora A. Socio-economic costs of osteoarthritis: a systematic review of cost-of-illness studies. Seminars in Arthritis and Rheumatism 2015 44 531–541. (https://doi.org/10.1016/j.semarthrit.2014.10.012)
- 3↑
Zhu Z, Li J, Ruan G, Wang G, Huang C, & Ding C. Investigational drugs for the treatment of osteoarthritis, an update on recent developments. Expert Opinion on Investigational Drugs 2018 27 881–900. (https://doi.org/10.1080/13543784.2018.1539075)
- 4↑
Di Matteo B, Anzillotti G, Gallese A, Vitale U, Gaggia GMMC, Ronzoni FL, Marcacci M, & Kon E. Placenta-derived products demonstrate good safety profile and overall satisfactory outcomes for treating knee osteoarthritis: a systematic review of clinical evidence. Arthroscopy 2023 39 1892–1904. (https://doi.org/10.1016/j.arthro.2023.03.033)
- 5↑
Anzillotti G, Conte P, Di Matteo B, Bertolino EM, Marcacci M, & Kon E. Injection of biologic agents for treating severe knee osteoarthritis: is there a chance for a good outcome? A systematic review of clinical evidence. European Review for Medical and Pharmacological Sciences 2022 26 5447–5459. (https://doi.org/10.26355/eurrev_202208_29413)
- 6↑
Kurtz SM, Ong KL, Lau E, & Bozic KJ. Impact of the economic downturn on total joint replacement demand in the United States: updated projections to 2021. Journal of Bone and Joint Surgery. American Volume 2014 96 624–630. (https://doi.org/10.2106/JBJS.M.00285)
- 7↑
Pabinger C, Lothaller H, & Geissler A. Utilization rates of knee-arthroplasty in OECD countries. Osteoarthritis and Cartilage 2015 23 1664–1673. (https://doi.org/10.1016/j.joca.2015.05.008)
- 8↑
Kunkel ST, Sabatino MJ, Kang R, Jevsevar DS, & Moschetti WE. The cost-effectiveness of total hip arthroplasty in patients 80 years of age and older. Journal of Arthroplasty 2018 33 1359–1367. (https://doi.org/10.1016/j.arth.2017.11.063)
- 9↑
Madry H, Luyten FP, & Facchini A. Biological aspects of early osteoarthritis. Knee Surgery, Sports Traumatology, Arthroscopy: Official Journal of the ESSKA 2012 20 407–422. (https://doi.org/10.1007/s00167-011-1705-8)
- 10↑
Conte P, Anzillotti G, Di Matteo B, Gallese A, Vitale U, Marcacci M, & Kon E. Orthobiologic injections for treating degenerative meniscus lesions: a matter of facts? Ten years of clinical experience in a systematic review. Journal of Cartilage & Joint Preservation 2023 3 100104. (https://doi.org/10.1016/j.jcjp.2023.100104)
- 11↑
Zhu Z, Jin X, Wang B, Wluka A, Antony B, Laslett LL, Winzenberg T, Cicuttini F, Jones G, & Ding C. Cross-sectional and longitudinal associations between serum levels of high-sensitivity C-reactive protein, knee bone marrow lesions, and knee pain in patients with knee osteoarthritis. Arthritis Care and Research 2016 68 1471–1477. (https://doi.org/10.1002/acr.22834)
- 12↑
Adak A, & Khan MR. An insight into gut microbiota and its functionalities. Cellular and Molecular Life Sciences 2019 76 473–493. (https://doi.org/10.1007/s00018-018-2943-4)
- 13↑
Arumugam M, Raes J, Pelletier E, Le Paslier D, Yamada T, Mende DR, Fernandes GR, Tap J, Bruls T, Batto JM, et al.Enterotypes of the human gut microbiome. Nature 2011 473 174–180. (https://doi.org/10.1038/nature09944)
- 14↑
Biver E, Berenbaum F, Valdes AM, Araujo de Carvalho I, Bindels LB, Brandi ML, Calder PC, Castronovo V, Cavalier E, Cherubini A, et al.Gut microbiota and osteoarthritis management: an expert consensus of the European society for clinical and economic aspects of osteoporosis, osteoarthritis and musculoskeletal diseases (ESCEO). Ageing Research Reviews 2019 55 100946. (https://doi.org/10.1016/j.arr.2019.100946)
- 15↑
Hao X, Shang X, Liu J, Chi R, Zhang J, & Xu T. The gut microbiota in osteoarthritis: where do we stand and what can we do? Arthritis Research and Therapy 2021 23 42. (https://doi.org/10.1186/s13075-021-02427-9)
- 16↑
Guido G, Ausenda G, Iascone V, & Chisari E. Gut permeability and osteoarthritis, towards a mechanistic understanding of the pathogenesis: a systematic review. Annals of Medicine 2021 53 2380–2390. (https://doi.org/10.1080/07853890.2021.2014557)
- 17↑
Collins KH, Paul HA, Reimer RA, Seerattan RA, Hart DA, & Herzog W. Relationship between inflammation, the gut microbiota, and metabolic osteoarthritis development: studies in a rat model. Osteoarthritis and Cartilage 2015 23 1989–1998. (https://doi.org/10.1016/j.joca.2015.03.014)
- 18↑
Ulici V, Kelley KL, Azcarate-Peril MA, Cleveland RJ, Sartor RB, Schwartz TA, & Loeser RF. Osteoarthritis induced by destabilization of the medial meniscus is reduced in germ-free mice. Osteoarthritis and Cartilage 2018 26 1098–1109. (https://doi.org/10.1016/j.joca.2018.05.016)
- 19↑
Huang ZY, Stabler T, Pei FX, & Kraus VB. Both systemic and local lipopolysaccharide (LPS) burden are associated with knee OA severity and inflammation. Osteoarthritis and Cartilage 2016 24 1769–1775. (https://doi.org/10.1016/j.joca.2016.05.008)
- 20↑
Huang Z, & Kraus VB. Does lipopolysaccharide-mediated inflammation have a role in OA? Nature Reviews. Rheumatology 2016 12 123–129. (https://doi.org/10.1038/nrrheum.2015.158)
- 21↑
Sun C, Zhou X, Guo T, & Meng J. The immune role of the intestinal microbiome in knee osteoarthritis: a review of the possible mechanisms and therapies. Frontiers in Immunology 2023 14 1168818. (https://doi.org/10.3389/fimmu.2023.1168818)
- 22↑
Perlejewski K, Bukowska-Ośko I, Nakamura S, Motooka D, Stokowy T, Płoski R, Rydzanicz M, Zakrzewska-Pniewska B, Podlecka-Piętowska A, Nojszewska M, et al.Metagenomic analysis of cerebrospinal fluid from patients with multiple sclerosis. Advances in Experimental Medicine and Biology 2016 935 89–98. (https://doi.org/10.1007/5584_2016_25)
- 23↑
Stinson LF, Boyce MC, Payne MS, & Keelan JA. The not-so-sterile womb: evidence that the human fetus is exposed to bacteria prior to birth. Frontiers in Microbiology 2019 10 1124. (https://doi.org/10.3389/fmicb.2019.01124)
- 24↑
Hieken TJ, Chen J, Hoskin TL, Walther-Antonio M, Johnson S, Ramaker S, Xiao J, Radisky DC, Knutson KL, Kalari KR, et al.The microbiome of aseptically collected human breast tissue in benign and malignant disease. Scientific Reports 2016 6 30751. (https://doi.org/10.1038/srep30751)
- 25↑
Qiu B, Al K, Pena-Diaz AM, Athwal GS, Drosdowech D, Faber KJ, Burton JP, & O’Gorman DB. Cutibacterium acnes and the shoulder microbiome. Journal of Shoulder and Elbow Surgery 2018 27 1734–1739. (https://doi.org/10.1016/j.jse.2018.04.019)
- 26↑
Bereza PL, Ekiel A, Auguściak-Duma A, Aptekorz M, Wilk I, Kusz DJ, Wojciechowski P, & Martirosian G. Identification of silent prosthetic joint infection: preliminary report of a prospective controlled study. International Orthopaedics 2013 37 2037–2043. (https://doi.org/10.1007/s00264-013-1955-9)
- 27↑
Tarabichi M, Shohat N, Goswami K, Alvand A, Silibovsky R, Belden K, & Parvizi J. Diagnosis of periprosthetic joint infection: the potential of next-generation sequencing. Journal of Bone and Joint Surgery. American Volume 2018 100 147–154. (https://doi.org/10.2106/JBJS.17.00434)
- 28↑
Tarabichi M, Shohat N, Goswami K, & Parvizi J. Can next generation sequencing play a role in detecting pathogens in synovial fluid? Bone and Joint Journal 2018 100–B 127–133. (https://doi.org/10.1302/0301-620X.100B2.BJJ-2017-0531.R2)
- 29↑
Street TL, Sanderson ND, Atkins BL, Brent AJ, Cole K, Foster D, McNally MA, Oakley S, Peto L, Taylor A, et al.Molecular diagnosis of orthopedic-device-related infection directly from sonication fluid by metagenomic sequencing. Journal of Clinical Microbiology 2017 55 2334–2347. (https://doi.org/10.1128/JCM.00462-17)
- 30↑
Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, et al.The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021 372 n71. (https://doi.org/10.1136/bmj.n71)
- 31↑
Moher D, Liberati A, Tetzlaff J, Altman DG & PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement. PLOS Medicine 2009 6 e1000097. (https://doi.org/10.1371/journal.pmed.1000097)
- 32↑
Tsai JC, Casteneda G, Lee A, Dereschuk K, Li WT, Chakladar J, Lombardi AF, Ongkeko WM, & Chang EY. Identification and characterization of the intra-articular microbiome in the osteoarthritic knee. International Journal of Molecular Sciences 2020 21 8618. (https://doi.org/10.3390/ijms21228618)
- 33↑
Torchia MT, Amakiri I, Werth P, & Moschetti W. Characterization of native knee microorganisms using next-generation sequencing in patients undergoing primary total knee arthroplasty. Knee 2020 27 1113–1119. (https://doi.org/10.1016/j.knee.2019.12.013)
- 34↑
Dunn CM, Velasco C, Rivas A, Andrews M, Garman C, Jacob PB, & Jeffries MA. Identification of cartilage microbial DNA signatures and associations with knee and hip osteoarthritis. Arthritis and Rheumatology 2020 72 1111–1122. (https://doi.org/10.1002/art.41210)
- 35↑
Fernández-Rodríguez D, Baker CM, Tarabichi S, Johnson EE, Ciccotti MG, & Parvizi J. Mark Coventry Award: human knee has a distinct microbiome: implications for periprosthetic joint infection. Journal of Arthroplasty 2023 38 S2–S6. (https://doi.org/10.1016/j.arth.2023.03.084)
- 36↑
Zhao Y, Chen B, Li S, Yang L, Zhu D, Wang Y, Wang H, Wang T, Shi B, Gai Z, et al.Detection and characterization of bacterial nucleic acids in culture-negative synovial tissue and fluid samples from rheumatoid arthritis or osteoarthritis patients. Scientific Reports 2018 8 14305. (https://doi.org/10.1038/s41598-018-32675-w)
- 37↑
Témoin S, Chakaki A, Askari A, El-Halaby A, Fitzgerald S, Marcus RE, Han YW, & Bissada NF. Identification of oral bacterial DNA in synovial fluid of patients with arthritis with native and failed prosthetic joints. Journal of Clinical Rheumatology 2012 18 117–121. (https://doi.org/10.1097/rhu.0b013e3182500c95)
- 38↑
Goswami K, Clarkson S, Tipton C, Phillips CD, Dennis DA, Klatt BA, O’Malley M, Smith EL, Gililland J, Pelt CE, et al.The microbiome of osteoarthritic hip and knee joints: a prospective multicenter investigation. Journal of Bone and Joint Surgery. American Volume 2023. (https://doi.org/10.2106/jbjs.22.00594).
- 39↑
Dinsdale NK, Castaño-Rodríguez N, Quinlivan JA, & Mendz GL. Comparison of the Genital microbiomes of pregnant aboriginal and non-aboriginal women. Frontiers in Cellular and Infection Microbiology 2020 10 523764. (https://doi.org/10.3389/fcimb.2020.523764)
- 40↑
Ozkan J, Coroneo M, Willcox M, Wemheuer B, & Thomas T. Identification and visualization of a distinct microbiome in ocular surface conjunctival tissue. Investigative Ophthalmology and Visual Science 2018 59 4268–4276. (https://doi.org/10.1167/iovs.18-24651)
- 41↑
Liu S, Li G, Xu H, Wang Q, Wei Y, Yang Q, Xiong A, Yu F, Weng J, & Zeng H. ‘Cross-talk’ between gut microbiome dysbiosis and osteoarthritis progression: a systematic review. Frontiers in Immunology 2023 14 1150572. (https://doi.org/10.3389/fimmu.2023.1150572)
- 42↑
Boer CG, Radjabzadeh D, Medina-Gomez C, Garmaeva S, Schiphof D, Arp P, Koet T, Kurilshikov A, Fu J, Ikram MA, et al.Intestinal microbiome composition and its relation to joint pain and inflammation. Nature Communications 2019 10 4881. (https://doi.org/10.1038/s41467-019-12873-4)
- 43↑
Sánchez Romero EA, Meléndez Oliva E, Alonso Pérez JL, Martín Pérez S, Turroni S, Marchese L, & Villafañe JH. Relationship between the gut microbiome and osteoarthritis pain: review of the literature. Nutrients 2021 13 716. (https://doi.org/10.3390/nu13030716)
- 44↑
Lei M, Guo C, Wang D, Zhang C, & Hua L. The effect of probiotic Lactobacillus casei Shirota on knee osteoarthritis: a randomised double-blind, placebo-controlled clinical trial. Beneficial Microbes 2017 8 697–703. (https://doi.org/10.3920/BM2016.0207)
- 45↑
Coulson S, Butt H, Vecchio P, Gramotnev H, & Vitetta L. Green-lipped mussel extract (Perna canaliculus) and glucosamine sulphate in patients with knee osteoarthritis: therapeutic efficacy and effects on gastrointestinal microbiota profiles. Inflammopharmacology 2013 21 79–90. (https://doi.org/10.1007/s10787-012-0146-4)
- 46↑
Carr C, Wilcox H, Burton JP, Menon S, Al KF, O’Gorman D, Lanting BA, Vasarhelyi EM, Neufeld M, & Teeter MG. Deciphering the low abundance microbiota of presumed aseptic hip and knee implants. PLoS One 2021 16 e0257471. (https://doi.org/10.1371/journal.pone.0257471)
- 47↑
Hammad DBM, Liyanapathirana V, & Tonge DP. Molecular characterisation of the synovial fluid microbiome in rheumatoid arthritis patients and healthy control subjects. PLoS One 2019 14 e0225110. (https://doi.org/10.1371/journal.pone.0225110)
- 48↑
Shin NR, Whon TW, & Bae JW. Proteobacteria: microbial signature of dysbiosis in gut microbiota. Trends in Biotechnology 2015 33 496–503. (https://doi.org/10.1016/j.tibtech.2015.06.011)
- 49↑
Rizzatti G, Lopetuso LR, Gibiino G, Binda C, & Gasbarrini A. Proteobacteria: a common factor in human diseases. BioMed Research International 2017 2017 9351507. (https://doi.org/10.1155/2017/9351507)
- 50↑
Li J, Zhao F, Wang Y, Chen J, Tao J, Tian G, Wu S, Liu W, Cui Q, Geng B, et al.Gut microbiota dysbiosis contributes to the development of hypertension. Microbiome 2017 5 14. (https://doi.org/10.1186/s40168-016-0222-x)
- 51↑
Jandhyala SM, Talukdar R, Subramanyam C, Vuyyuru H, Sasikala M, & Nageshwar Reddy D. Role of the normal gut microbiota. World Journal of Gastroenterology 2015 21 8787–8803. (https://doi.org/10.3748/wjg.v21.i29.8787)
- 52↑
Scher JU, Joshua V, Artacho A, Abdollahi-Roodsaz S, Öckinger J, Kullberg S, Sköld M, Eklund A, Grunewald J, Clemente JC, et al.The lung microbiota in early rheumatoid arthritis and autoimmunity. Microbiome 2016 4 60. (https://doi.org/10.1186/s40168-016-0206-x)
- 53↑
Xu YY, Tan X, He YT, Zhou YY, He XH, & Huang RY. Role of gut microbiome in ankylosing spondylitis: an analysis of studies in literature. Discovery Medicine 2016 22 361–370.
- 54↑
Manasson J, Shen N, Garcia Ferrer HR, Ubeda C, Iraheta I, Heguy A, Von Feldt JM, Espinoza LR, Garcia Kutzbach A, Segal LN, et al.Gut microbiota perturbations in reactive arthritis and postinfectious spondyloarthritis. Arthritis and Rheumatology 2018 70 242–254. (https://doi.org/10.1002/art.40359)
- 55↑
Wang Q, Lepus CM, Raghu H, Reber LL, Tsai MM, Wong HH, von Kaeppler E, Lingampalli N, Bloom MS, Hu N, et al.IgE-mediated mast cell activation promotes inflammation and cartilage destruction in osteoarthritis. eLife 2019 8 e39905. (https://doi.org/10.7554/eLife.39905)