Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

The relation between the gut microbiome and osteoarthritis: A systematic review of literature

  • Emanuele Chisari ,

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing

    Affiliations Rothman Orthopaedic Institute at Thomas Jefferson University, Philadelphia, Pennsylvania, United States of America, Department of Medical Microbiology and Infection Prevention, University of Groningen, University Medical Center Groningen, RB, Groningen, Netherlands

  • Marjan Wouthuyzen-Bakker,

    Roles Conceptualization, Investigation, Methodology, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Department of Medical Microbiology and Infection Prevention, University of Groningen, University Medical Center Groningen, RB, Groningen, Netherlands

  • Alex W. Friedrich,

    Roles Conceptualization, Data curation, Supervision, Writing – review & editing

    Affiliation Department of Medical Microbiology and Infection Prevention, University of Groningen, University Medical Center Groningen, RB, Groningen, Netherlands

  • Javad Parvizi

    Roles Conceptualization, Investigation, Methodology, Writing – original draft, Writing – review & editing

    Affiliation Rothman Orthopaedic Institute at Thomas Jefferson University, Philadelphia, Pennsylvania, United States of America



Along with mechanical and genetic factors, emerging evidence suggests that the presence of low-grade inflammation has a role in the pathogenesis of osteoarthritis (OA) and seems to be related to the microbiome composition of the gut.


To provide evidence whether there is clinical or preclinical evidence of gut-joint axis in the pathogenesis and symptoms of OA.


An extensive review of the current literature was performed using three different databases. Human, as well as animal studies, were included. The risk of bias was identified using ROBINS and SYRCLE tools, while the quality of evidence was assessed using GRADE and CAMADARES criteria.


A total of nineteen articles were included. Multiple animal studies demonstrated that both obesity, and high-fat and high-sugar diets resulted in a gut dysbiosis status characterized by increased Firmicutes/Bacteroidetes (F/B) phyla ratio and increased permeability. These changes were associated with increased lipopolysaccharide serum levels, which consequently resulted in synovitis and OA severity. The administration of pre-and probiotics partially reversed this bacterial composition. In addition, in human studies, a decreased amount of gut Bacteroidetes, subsequent increased F/B ratio, have also been observed in OA patients.


Our review confirms preliminary yet sound evidence supporting a gut-joint axis in OA in primarily preclinical models, by showing an association between diet, gut dysbiosis and OA radiological severity and self-reported symptoms. Clinical studies are needed to confirm these findings, and to investigate whether interventions targeting the composition of the microbiome will have a beneficial clinical effect.


Osteoarthritis (OA) is the most common degenerative joint disease [1]. Though it may develop in any joint, it predominantly affects diarthrodial joints (mainly knees, hands, or hips) and, following disease progression, it ultimately leads to joint failure [2]. The understanding of the pathophysiology is still evolving. Among the causes, mechanical factors and genetic factors have been classically shown to play a significant role in the development of OA [3, 4]. Despite the fact that OA traditionally has been considered as “non-inflammatory,” low-grade inflammation seems to play an important role in the initiation and propagation of OA. Emerging evidence suggests that this inflammatory state is triggered by the gastrointestinal microbiome [5].

The gastrointestinal microbiome is defined as the sum of all the genetic material of all microbiota that is present in the gut, their metabolic byproducts, and comprises more than 3 million genera and about 5 thousand bacterial species [6, 7]. While much more has to be explored to better define this biological niche, what is known is that each individual is thought to feature a unique microbiome composition [8] that acts as a fingerprint of each individual subject. Taxonomic studies report that Firmicutes (including Lactobacillus, Bacillus, Clostridium, Enterococcus, and Ruminicoccus genera) and Bacteroidetes are the main phyla, representative of 90% of the total gut microbiome in people with an unspecific diet [9]. Many studies suggest that several intestinal [10] and extra-intestinal diseases [11] are associated with specific bacterial motifs and a modification and disbalance of the microbiome composition (the concept of dysbiosis). Although a causal relationship between dysbiosis and pathophysiology is still under discussion, a direct correlation seems to be plausible, as demonstrated in several chronic conditions and specific pathobionts in a mouse model [12]. Moreover, diseases that have been traditionally considered autoimmune-based are considered to be the results of a significant interaction between innate and adaptive immune response the overall sum of microorganisms, including fungi, virus and their metabolites, in the gut [12].

The joint environment is traditionally considered sterile. However, preliminary findings suggest that the gut-joint axis exist and is particularly active during the neo-angiogenesis phase of OA [13] when bacteria and bacterial products have a preferential gate from the blood to the cartilage and subchondral bone that usually is less vascularized [1416]. These bacterial elements xmay affect the epigenetic landscape of chondrocytes [17] and prime the innate immune response in the joint via Toll-like receptors signalling [5]. In addition, a recent study [18] suggests that a microbiome may exist inside the OA knee and hip joint as well. The overall ecological interaction among these microbial patterns is still unknown.

Previous studies and methods have underestimated the impact of the microbiome and its products in the pathogenesis of OA [19]. We, therefore, asked whether there is clinical or preclinical evidence of a gut-joint axis in the pathogenesis of OA.


Search strategy

The systematic review was conducted according to the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [20]. A comprehensive search was performed using three electronic medical databases (PubMed, EMBASE, and Cochrane Library) by two independent authors (GA and VI) from January 2020 to April 2020. To achieve maximum sensitivity we combined the search terms “gut microbio* OR gut OR streptococcus OR microbiome” with some terms related to osteoarthritis and inflammation such as “arthritis OR osteoarthritis OR inflammation OR synovial OR synovitis” and typical anatomical landmarks of disease “hip OR knee” as either keywords or Medical Subject Heading (MeSH) terms. The search strategy was repeated during the peer-review process to further check for newly published articles (October 14th, 2021). The reference lists of all included articles, previous reviews on the topic, and top hits from Google Scholar were reviewed to identify further potentially relevant studies, which were assessed using the inclusion and exclusion criteria. To avoid overlap with other ongoing review studies, we searched PROSPERO for any similar reviews.

Selection criteria

The gut microbiome is affected by several factors, including dietary patterns [21], pro- and prebiotics, and drugs (antibiotics, proton-pump inhibitors, non-steroidal anti-inflammatory drugs [22, 23]). Inclusion criteria were (1) both clinical and preclinical studies with any level of evidence (2) published in peer-reviewed journals in English, (3) investigated the role of gut microbiome, with or without dietary interventions, and OA pathogenesis or related-symptoms. We excluded studies in which data was not accessible or missing, or those without an available full-text article. We also excluded duplicates and studies with poor scientific methodology as per risk of bias assessment. Abstracts, case reports, conference presentations, reviews, editorials, and expert opinions were excluded.

Two authors (EC and MW) performed the search and evaluated the articles independently. An experienced researcher in systematic reviews (JP) resolved cases of doubt. 465 articles were retrieved from the three databases after removal of duplicates. First, each of the two investigators read the abstracts of all articles, selected relevant articles according to the inclusion and exclusion criteria, and defined a list of articles for full text reading. Then, the two investigators proceed to full text reading, selected the articles eligible for inclusion and compared the results. An experienced researcher in systematic reviews (JP) resolved cases of doubt. After four weeks, the same studies were reread to ensure the investigators agreed about article selection. There was no disagreement among the investigators. One investigator (EC) extracted data from the full-text articles into an Excel spreadsheet with structured tables to analyze each study descriptively. Doubts and inconsistencies were grouped and resolved. Ultimately, 19 articles were found eligible to be included in the study.

Data extraction and criteria appraisal

Data were extracted from article texts, tables, and figures using the Population, Intervention, Comparison, Outcome framework [24] (PICO) and included the title, year of publication, study design, sample size, study population, patient characteristics, intervention and comparator group (if applicable), outcomes, findings, and conclusions. Two investigators (GA and GG) independently reviewed each article. Discrepancies between the reviewers were resolved by discussion and consensus (Fig 1). After extraction, the data was considered of heterogenous nature both by study design, measure, and method of assessment. Therefore, a descriptive analysis approach was preferred to a metanalysis.

Risk of bias assessment

Risk of bias assessment of non-randomized clinical studies was performed according to the ROBINS risk of bias tool [25] for non-randomized study and RoB 2.0 [26] for randomized studies. This assessment used “low,” “moderate,” and “high” as judgment keys. “Low” indicated a low risk of bias, “moderate” indicated a moderate risk of bias, and “high” indicated a high risk of bias. The assessment was performed by two authors (EC and AWF) independently. The inter-rater agreement was 90%. Any discrepancy was solved by consensus. The risk of bias assessment for preclinical studies was instead performed using the SYRCLE’s tool [27]. A score ranging from 0 (lowest) to 10 (highest) was used to classify potential study bias. Two investigators evaluated the studies independently (EC and AWF) with a 95% inter-rater agreement. 12 studies were retrieved after full-text reading and were assessed for the risk of bias. One clinical study was excluded because of a serious risk of bias. Three clinical studies were assessed to have a moderate risk of bias, and one article was excluded because of a serious risk of bias. One randomized clinical study was assessed to have “some concerns” according to the RoB tool. The preclinical animal studies were assessed as having a low to moderate risk of bias (mean SYRCLE score 3.9, range 3–6). (Supplementary Material) (AUTHOR NOTE: weblearn does not allow for supplementary material submission, happy to share by email).

Study quality assessment

The quality of clinical studies was assessed using the GRADE method [28]. Each study was classified as “low”, “moderate” or “high” according to the quality of evidence. With the exception of one study that featured a low quality of evidence (already excluded after the risk of bias assessment), all the studies ranked “high”.

The Collaborative Approach to Meta-Analysis and Review of Animal Data from Experimental Studies (CAMADARES) checklist [29, 30] was used to assess the quality of the preclinical studies (n = 10) (Supplementary Material) (AUTHOR NOTE:weblearn does not allow for supplementary material submission, happy to share by email). The assessment was performed independently by two authors (EC, AWF) with an inter-rater agreement of 94%. Each study was assessed and scored on a scale from 0 (lowest) to 10 (highest) points. The overall quality was moderate among the included studies (mean CAMADARES score 4.45, range 3–6).


Study characteristics

19 studies were included in the final analysis, of which 10 were on animal models [3138], 8 were human clinical or ex vivo studies [5, 18, 19, 3847], and one was based on analysis of publicly available databases [40].

Concerning the human clinical trials, 3 were non-randomized observational studies [5, 18, 19], and 3 were randomized clinical trials [4143]. The main features of the studies included are reported in Table 1. Data was retrieved from 1,548 individuals, of which 58.2% were women. The population was aged between 50 and 65 years. The models investigated in the preclinical studies were adult mice [33, 36, 4850], and adult rats [31, 34, 35] (Table 1).

Most studies analyzed the relationship between OA and the variations in microbial populations of the gastrointestinal tract or in the local articular cartilage (specifically knee and hip samples) by 16S ribosomal RNA gene sequencing, internal transcribed spacer (ITS) amplification and/or shotgun metagenomics [18]. Overall, all studies showed different level of evidence of the role of the gut-joint axis in the pathogenesis and progression of OA.

One study analyzed, in a model of sex steroid deficiency, the role of specific microbial species and strains (administrated by oral gavage of probiotic supplementation) in the maintenance of intestinal barrier integrity. Metabolic alterations of the microbiome and their correlation to gut permeability were investigated by measuring the serum concentration of lipopolysaccharides (LPS), a component of the outer membrane of Gram-negative bacteria [18, 19, 31, 32, 35, 36]. Diet and obesity were analyzed concomitantly [31, 32, 35] in relation to gut microbiome in mice.

Some studies analyzed the role of other dietary aspects in relation to OA severity: physical exercise [31, 49], the administration of nutritional supplementation [34] and/or prebiotics in rodents [3134] and humans [41] (Table 1).

Effects of diet regimens and the gut-joint axis

High-fat diet (HFD) and steroid-deficiency induced obesity are common models of gut epithelium disruption, gut dysbiosis, and low-grade systemic inflammation [31, 35].

When mice were on an HFD, OA-like and inflammatory changes were observed during histological analysis. However, when pre-and probiotics were administered contextually to the food, the inflammatory effects on the gut-joint axis were partially reversed and prevented [31, 35]. Another study, reported similar results with the administration of oligofructose in obese mice with increased gut permeability [32]. However, Collins et al. [38] reported how these changes can happen even independently of HF diet, but based on direct relationship between the microbiome and the OA phenotype. A study tried to replicate what was previously found by investigating the gut microbiome and LPS level in a case-control study of OA vs non-OA human patients [39]. While alpha e beta diversity did not differ among the two groups, the LPS level were strongly associated with OA severity (104.9 ± 45.8 EU/mL vs 61.3 ± 33.9 EU/mL; p<0.0001) highlighting the role of gut permeability in OA.

Few studies investigated the use of glucosamine sulfate in similar models, and while a non-blinded randomized clinical trial [41] found that both Glucosamine sulfate and Green-lipped mussel used as prebiotic, were effective in improving joint function and pain in knee OA patients. However, a preclinical trial showed how glucosamine was effective only in co-administration with Lactobacillus supplementation [34], which was shown to prevent the changes even when administered alone.

Microbial composition changes associated with osteoarthritis

When obesity was induced by administering a HFD in mice models, the most common finding was an increased Firmicutes/Bacteroidetes (F/B) phyla ratio in the gut compared to chow-fed mice [31, 32, 35] (Table 1) (F/B = 5–10 [35]. This finding was found to be strongly correlated with OA severity according to the modified Mankin Criteria, using a histologic/histochemical grading system.

Quantitatively, the F/B ratio increase has shown to be mostly dependent of a decrease in Bacteroidetes phylum population (especially Bacteroides and Prevotella genera), whereas the total Firmicutes abundance mostly remained the same. However, the Firmicutes composition was different in OA patients, mainly showing a decrease in Lactobacillus spp. and an increase in the Clostridiales order [35]. The largest OA human population investigating the gut microbiome, showed that the relative most abundant species among Lactobacillales, was Streptococcus spp., which have shown to represent 20% of Firmicutes in OA patients and this was positively correlated with higher WOMAC OA pain scores and lower functionality [5]. These results were validated by replication in an independent cohort and by meta-analysis of the final results. Finally, they presented evidence that this association was driven by local inflammation (synovitis and joint effusion) in the joint assessed through a subgroup MRI analysis in 373 patients. Interestingly, a recent study looking a publicly available sequencing data, used mendelian randomization approach to identify microbial taxa associated to OA in humans [40]. Their finding were in agreement with what previously reported.


The results of our systematic review demonstrate that a gut-joint axis exist in preclinical and clinical models. We included 15 studies of level 3 and 4 evidence, and one randomized study. Since these designs cannot prove causality, future large longitudinal controlled studies are encouraged to strengthen this concept. To our knowledge, this is the first review exploring the gut-joint axis in an evidence-based systematic fashion (Fig 2).

Fig 2. The model of gut-joint axis.

FOS: fructooligosaccharides; LPS: lipopolysaccharide; OA: osteoarthritis. In the picture are described the three main actors of the gut-joint axis in OA development. Environmental factors act as either stressor or protectants of gut integrity and microbiota health. If this balance is disrupted, bacterial metabolites and byproducts such as LPS and chemokines pass the gut barrier into the blood flow causing distant damage to diarthrodial joints such as the knee. In the long-term, similar damage together with immune priming can initiative OA changes and affect pain and functionality.

Gut permeability as a foundational element of the gut-joint axis

Gut microbiome changes such as bacterial abundance, activity or diversity, are observed in animal models of obesity, and can be either acute or chronic. Indeed, obesity is associated with specific microbiome patterns that ultimately lead to impairment of the gut mucosa and microbiome translocation [51]. All of these changes in the gut are commonly described as gut dysbiosis. Interestingly, together with physical exercise, weight loss is strongly associated with OA control [52]. Both interventions are associated with reduced pain and disease progression [53, 54]. In addition, along with functional food use [33, 41], these interventions have been shown to be effective in reverting the acute microbiome compositional changes that are associated with OA, as depicted in this systematic review.

When gut dysbiosis occurs as a result of an underlying chronic disease (e.g., inflammatory bowel disease), chronic antibiotic treatment, or lifestyle modification (e.g., obesity and metabolic disease), gut permeability is significantly affected [5560]. It has been demonstrated that a HFD depletes eosinophils in the gut and is associated with an increase in gut permeability [61], reduced tight-junctions molecules, mucus, and antimicrobial peptides production [62]. Moreover, microorganisms are spatially redistributed in the intestine, mainly occupying intervillous/cryptal spaces [62]. If such changes progressively occur, innate immune receptors in the gut become activated by microbial products (such as LPS) and stimulate the production of pro-inflammatory mediators (IL-1beta; IL-17, IL-18) both locally and systemically once similar byproducts reach the systemic circulation [39]. Thus, both inflammatory mediators and bacterial toxins (e.g., LPS) are subsequently translocated into the circulation [57] with priming action on the immune system that has been associated with multiple chronic conditions [6367] including OA [39].

Together with molecular products, bacterial translocation is possible and raises concerns, particularly in patients with prosthetic implants or artificial valves [68, 69]. Although emerging evidence suggests that bacterial translocation is associated with OA severity in the native joint, it is still unclear whether an increased risk for infection of the joints does exist in obese patients or in patients with inflammatory bowel disease. Ultimately, low-grade inflammation of the gut may lead to transient bacteremia, in which either direct colonization of bacteria to the joint may occur or in which leucocytes and macrophage [32, 70] may act as “Trojan Horse” [68, 69, 71] for bacteria to reach the joint. This concept needs to be further explored in subsequent studies.

Adaptive and innate immune response

The aforementioned short-term alterations of the gut microbiome composition are associated with worse patient-reported outcome measures and with a more severe radiologic OA classification [5, 19]. Our findings indicate that this is probably due the observed increase in the Firmicutes/Bacteroidetes (F/B) phyla ratio of the gut microbiome [31, 32, 35].

Bacteroidetes produce high levels of short-chain fatty acids, which regulate the differentiation of Treg cells that are the main actors of the suppression of inflammation in chronic and acute condition [72]. Thus, we propose that the observed clinical findings are due to a decrease in Bacteroidetes, as displayed in OA patients, through the impairment of adaptive immunity and the delicate balance of pro-and anti-inflammatory mechanisms. Also, the abundance of specific strains, such as Streptococcus spp. and other gram-negative strains seems to be critical to prime both local and systemic inflammation through LPS- or other metabolites-induced macrophage activation [5] via common inflammatory pathways such as NFkB or MAPK [73], or by forming active complexes with Lipopolysaccharide-binding-Protein (LBP), laminin-binding protein, and CD14 [19]. Mechanistic evidence was reported with increased CD14 levels (as a surrogate marker for macrophage activation) in the synovium of OA patients compared to healthy controls [74]. Thus, an increase in F/B ratio and Streptococcus spp. prevalence are probably able to dysregulate the delicate balance of pro-inflammatory markers [34]. However, two of the studies we included in our analysis [36, 37], reported that gut microbiome dysbiosis alleviates the progression of osteoarthritis in mice. Additionally, an antibiotic-induced gut microbiota dysbiosis in OA male mice significantly decreased the relative abundance of Bacteroidetes, but for female mice, the relative abundance of Firmicutes was increased significantly compared with that in Con-OA female mice [37]. While this further highlights the pivotal role of the microbiota as whole in the development, progression and symptomatology of OA, also induces contradictory conclusion about bacterial diversity and F/B ratio. While many conclusion can be made, we hypothesize that any change that can lead to impaired gut barrier integrity and the release of byproducts such as LPS into the bloodstream can perhaps induce immune priming and adjuvate the initiation and progression of OA disease.

We also hypothesize that bacterial molecules, that can potentially act as biomarkers, are involved in OA inflammation. We encourage further studies on microbial exotoxins, cationic bacterial peptides, and other metabolites, to assess their role in the pathogenesis of OA, their influence on quorum sensing and inter-and intra-species interactions in biofilm-based musculoskeletal infections such as periprosthetic joint infections [75].

Pain symptomatology and chemokines

Studies show that the gut microbiome is associated with joint pain [76]. The main actors are cartilage degradation products and other damage-associated molecular patterns either through direct neuronal activation of dorsal root ganglia or by indirect neuro-immune signaling acting on immune cells receptors that in turn stimulate neurons amplifying the mechanism. Apart from cytokines and LPS that may partially explain the onset of pain, chemokines are also reported as potential critical players [70]. Indeed, chemokines were consistently increased in murine animal models [32, 49].

Current treatments for OA focuses on pain relief, but do not target the pathogenesis of the disease [52]. Since the gut microbiome is modifiable by several factors (dietary intervention, fecal transplant, and future microbiome-targeted therapeutics), the gut microbiome is a promising target for future treatment strategies.


Our findings have several limitations. First of all, despite the consistent data on the association of OA with specific bacterial strains and phyla ratio in the gut microbiome, we caution the reader about the absence of large human studies other than the few reported. To intervene in the gut-joint axis as part of treatment requires a more profound mechanistic understanding of microbiome-host interactions and detailed characterization of the complex community interactions involved. Current studies, despite the overall moderate to high quality of evidence, still do not allow for such conclusions.

Most studies were conducted on animal models, not fully mimicking the complexity of the human microbiome. In addition, the design of all the studies included do not allow for causality conclusions. However, all pieces of evidence analyzed have shown concordance of findings and clinical studies show similar results [5]. Another major limitation of the analyzed studies is the use of different methods to induce obesity in animal models and the heterogeneity of the studied animal models itself. However, all the different animals and the different models used (high fat diet, high sugar diet, sex steroid deficiency) show the same results. Finally, most animal studies only described the alterations of the gut microbiome after specific diet regimens, and by design are not able to describe a clinically relevant microbiome description. Also, despite the studies reported a control group with different diet regiment (e.g. chow-fed), further studies should investigate the alteration of the bacterial strains, and their association with OA, regardless of the diet, maybe using fecal transplantation as validated in other studies [77]. Finally, bacterial genomic plasticity is missing in most of the studies, where under antibiotic—although subinhibitory—pressure, influence of heavy metals from food or drinking water, and nutritional influence, mobile genetic elements drive in-vivo evolutionary events and adaptation processes in the microbiome, its composition and inflammatory activity.


In conclusion, our systematic review provides evidence for the existence of a gut-joint axis in the pathogenesis OA. The proposed concept starts with disruption and dysbiosis of the normal gut homeostasis, the continuous change of the microbial composition and genomic plasticity for optimal bacterial adaptation to the host environment, followed by both adaptive and innate immune responses due to translocation of bacteria and bacterial products into the circulation toward to the joint. This cascade ultimately leads to low-grade inflammation in the joint and contributes to the pathogenesis of OA. Future studies are needed to further strengthen this hypothesis.

Supporting information

S1 Table. CAMARADES CHECKLIST of the included studies.


S2 Table. SIRCLE bias assessment of the included studies.


S3 Table. ROBINS bias assessment of the included studies.


S4 Table. Rob 2.0 bias assessment of the included studies.



  1. 1. Glyn-Jones S. et al. Osteoarthritis. in The Lancet (2015).
  2. 2. Martel-Pelletier J. et al. Osteoarthritis. Nat. Rev. Dis. Prim. 2, 16072 (2016). pmid:27734845
  3. 3. Guilak F. Biomechanical factors in osteoarthritis. Best Practice and Research: Clinical Rheumatology (2011) pmid:22265263
  4. 4. Panoutsopoulou K. & Zeggini E. Advances in osteoarthritis genetics. Journal of Medical Genetics (2013) pmid:23868913
  5. 5. Boer C. G. et al. Intestinal microbiome composition and its relation to joint pain and inflammation. Nat. Commun. 10, 4881 (2019). pmid:31653850
  6. 6. Rothschild D. et al. Environment dominates over host genetics in shaping human gut microbiota. Nature 555, 210–215 (2018). pmid:29489753
  7. 7. Almeida A. et al. A new genomic blueprint of the human gut microbiota. Nature 568, 499–504 (2019). pmid:30745586
  8. 8. Rinninella E. et al. What is the healthy gut microbiota composition? A changing ecosystem across age, environment, diet, and diseases. Microorganisms (2019) pmid:30634578
  9. 9. Arumugam M. et al. Enterotypes of the human gut microbiome. Nature (2011) pmid:21508958
  10. 10. Frank D. N. et al. Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases. Proc. Natl. Acad. Sci. U. S. A. (2007) pmid:17699621
  11. 11. Mayer E. A., Tillisch K. & Gupta A. Gut/brain axis and the microbiota. Journal of Clinical Investigation (2015) pmid:25689247
  12. 12. Manfredo Vieira S. et al. Translocation of a gut pathobiont drives autoimmunity in mice and humans. Science (80-.). (2018) pmid:29590047
  13. 13. Pesesse L., Sanchez C. & Henrotin Y. Osteochondral plate angiogenesis: A new treatment target in osteoarthritis. Joint Bone Spine (2011) pmid:20851653
  14. 14. Zhu Hongyi, Jin Hanqiang, Zhang Changqing, T. Y. Intestinal methicillin-resistant Staphylococcus aureus causes prosthetic infection via ‘Trojan Horse’ mechanism: Evidence from a rat model. Bone Jt. Res. 9, 152–161 (2020). pmid:32431806
  15. 15. Krezalek M. A. et al. Can Methicillin-resistant Staphylococcus aureus Silently Travel from the Gut to the Wound and Cause Postoperative Infection? Modeling the ‘trojan Horse Hypothesis’. Ann. Surg. 267, 749–758 (2018). pmid:28187042
  16. 16. Richards R. L. et al. Persistent Staphylococcus aureus Isolates from Two Independent Cases of Bacteremia Display Increased Bacterial Fitness and Novel Immune Evasion Phenotypes. Am Soc Microbiol (2015) pmid:26056388
  17. 17. Cuevas-Sierra A., Ramos-Lopez O., Riezu-Boj J. I., Milagro F. I. & Martinez J. A. Diet, Gut Microbiota, and Obesity: Links with Host Genetics and Epigenetics and Potential Applications. in Advances in Nutrition (2019). pmid:30721960
  18. 18. Dunn C. M. et al. Identification of cartilage microbial DNA signatures and associations with knee and hip osteoarthritis. Arthritis Rheumatol. (2020) pmid:31961065
  19. 19. Huang Z. Y., Stabler T., Pei F. X. & Kraus V. B. Both systemic and local lipopolysaccharide (LPS) burden are associated with knee OA severity and inflammation. Osteoarthr. Cartil. (2016) pmid:27216281
  20. 20. Liberati A. et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 339, b2700–b2700 (2009). pmid:19622552
  21. 21. Kolodziejczyk A. A., Zheng D. & Elinav E. Diet–microbiota interactions and personalized nutrition. Nature Reviews Microbiology (2019) pmid:31541197
  22. 22. Blaser M. J. Antibiotic use and its consequences for the normal microbiome. Science (2016) pmid:27126037
  23. 23. Imhann F. et al. Proton pump inhibitors affect the gut microbiome. Gut (2016) pmid:26657899
  24. 24. Richardson W. S., Wilson M. C., Nishikawa J. & Hayward R. S. The well-built clinical question: a key to evidence-based decisions. ACP journal club vol. 123 (1995). pmid:7582737
  25. 25. Sterne J. A. et al. ROBINS-I: A tool for assessing risk of bias in non-randomised studies of interventions. BMJ 355, i4919 (2016). pmid:27733354
  26. 26. Sterne J. A. C. et al. RoB 2: A revised tool for assessing risk of bias in randomised trials. BMJ 366, (2019). pmid:31462531
  27. 27. Hooijmans C. R. et al. SYRCLE’s risk of bias tool for animal studies. BMC Med. Res. Methodol. 14, 43 (2014). pmid:24667063
  28. 28. Guyatt G. H., Oxman A. D., Schünemann H. J., Tugwell P. & Knottnerus A. GRADE guidelines: A new series of articles in the Journal of Clinical Epidemiology. J. Clin. Epidemiol. 64, 380–382 (2011). pmid:21185693
  29. 29. Landis S. C. et al. A call for transparent reporting to optimize the predictive value of preclinical research. Nature vol. 490 187–191 (2012).
  30. 30. Macleod M. R., O’Collins T., Howells D. W. & Donnan G. A. Pooling of animal experimental data reveals influence of study design and publication bias. Stroke vol. 35 1203–1208 (2004). pmid:15060322
  31. 31. Rios J. L. et al. Protective effect of prebiotic and exercise intervention on knee health in a rat model of diet-induced obesity. Sci. Rep. (2019) pmid:30846801
  32. 32. Schott E. M. et al. Targeting the gut microbiome to treat the osteoarthritis of obesity. JCI insight (2018) pmid:29669931
  33. 33. Panicker S., Borgia J., Fhied C., Mikecz K. & Oegema T. R. Oral glucosamine modulates the response of the liver and lymphocytes of the mesenteric lymph nodes in a papain-induced model of joint damage and repair. Osteoarthr. Cartil. (2009) pmid:19364543
  34. 34. So J. S. et al. Lactobacillus casei enhances type II collagen/glucosamine-mediated suppression of inflammatory responses in experimental osteoarthritis. Life Sci. (2011) pmid:21167838
  35. 35. Collins K. H. et al. Relationship between inflammation, the gut microbiota, and metabolic osteoarthritis development: Studies in a rat model. Osteoarthr. Cartil. (2015) pmid:26521745
  36. 36. Ulici V. et al. Osteoarthritis induced by destabilization of the medial meniscus is reduced in germ-free mice. Osteoarthr. Cartil. 26, 1098–1109 (2018). pmid:29857156
  37. 37. Guan Z. et al. Gut microbiome dysbiosis alleviates the progression of osteoarthritis in mice. Clin. Sci. 134, 3159–3174 (2020). pmid:33215637
  38. 38. KH C., DJ S., KL L., CA H. & F G. Taxonomic changes in the gut microbiota are associated with cartilage damage independent of adiposity, high fat diet, and joint injury. Sci. Rep. 11, (2021).
  39. 39. RF L. et al. Association of Increased Serum Lipopolysaccharide but not Microbial Dysbiosis with Obesity-related Osteoarthritis. Arthritis Rheumatol. (Hoboken, N.J.) (2021) pmid:34423918
  40. 40. XH Y., YQ Y., RR C., L B. & SF L. The causal role of gut microbiota in development of osteoarthritis. Osteoarthr. Cartil. (2021) pmid:34425228
  41. 41. 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) pmid:22821424
  42. 42. J J. et al. Oral Administration of Lactobacillus rhamnosus Ameliorates the Progression of Osteoarthritis by Inhibiting Joint Pain and Inflammation. Cells 10, (2021).
  43. 43. M L., C G., D W., C Z. & L H. The effect of probiotic Lactobacillus casei Shirota on knee osteoarthritis: a randomised double-blind, placebo-controlled clinical trial. Benef. Microbes 8, 697–703 (2017). pmid:28726510
  44. 44. Huang Z. et al. Fecal microbiota transplantation from metabolically compromised human donors accelerates osteoarthritis in mice. Ann. Rheum. Dis. 79, 646 (2020). pmid:32205337
  45. 45. Chen J., Wang A. & Wang Q. Dysbiosis of the gut microbiome is a risk factor for osteoarthritis in older female adults: a case control study. BMC Bioinforma. 2021 221 22, 1–11 (2021). pmid:34082693
  46. 46. Wang Z., Zhu H., Jiang Q. & Zhu Y. Z. The gut microbiome as non-invasive biomarkers for identifying overweight people at risk for osteoarthritis. Microb. Pathog. 157, 104976 (2021). pmid:34023440
  47. 47. ZY Huang. et al. Biomarkers of inflammation—LBP and TLR- predict progression of knee osteoarthritis in the DOXY clinical trial. Osteoarthr. Cartil. 26, 1658–1665 (2018).
  48. 48. Li J. Y. et al. Sex steroid deficiency-associated bone loss is microbiota dependent and prevented by probiotics. J. Clin. Invest. (2016) pmid:27111232
  49. 49. Griffin T. M., Huebner J. L., Kraus V. B., Yan Z. & Guilak F. Induction of osteoarthritis and metabolic inflammation by a very high-fat diet in mice: Effects of short-term exercise. Arthritis Rheum. (2012) pmid:21953366
  50. 50. Mooney R. A., Sampson E. R., Lerea J., Rosier R. N. & Zuscik M. J. High-fat diet accelerates progression of osteoarthritis after meniscal/ligamentous injury. Arthritis Res. Ther. (2011) pmid:22152451
  51. 51. Metcalfe D. et al. Does endotoxaemia contribute to osteoarthritis in obese patients? Clin. Sci. (2012) pmid:22888972
  52. 52. Kolasinski S. L. et al. 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. Arthritis Rheumatol. (2020) pmid:31908163
  53. 53. Messier S. P. et al. Intentional Weight Loss in Overweight and Obese Patients With Knee Osteoarthritis: Is More Better? Arthritis Care Res. (2018) pmid:29911741
  54. 54. Goh S. L. et al. Relative Efficacy of Different Exercises for Pain, Function, Performance and Quality of Life in Knee and Hip Osteoarthritis: Systematic Review and Network Meta-Analysis. Sports Medicine (2019) pmid:30830561
  55. 55. Ruff W. E. & Kriegel M. A. Autoimmune host–microbiota interactions at barrier sites and beyond. Trends Mol. Med. 21, 233–244 (2015). pmid:25771098
  56. 56. Lee Y. K. & Mazmanian S. K. Has the microbiota played a critical role in the evolution of the adaptive immune system? Science 330, 1768–73 (2010). pmid:21205662
  57. 57. Chakaroun R. M., Massier L. & Kovacs P. Gut microbiome, intestinal permeability, and tissue bacteria in metabolic disease: Perpetrators or bystanders? Nutrients (2020) pmid:32295104
  58. 58. Lamprecht M. et al. Probiotic supplementation affects markers of intestinal barrier, oxidation, and inflammation in trained men; a randomized, double-blinded, placebo-controlled trial. J. Int. Soc. Sports Nutr. (2012) pmid:22992437
  59. 59. Aleksandrova K., Romero-Mosquera B. & Hernandez V. Diet, Gut Microbiome and Epigenetics: Emerging Links with Inflammatory Bowel Diseases and Prospects for Management and Prevention. Nutrients 9, (2017).
  60. 60. Saltiel A. R. & Olefsky J. M. Inflammatory mechanisms linking obesity and metabolic disease. J. Clin. Invest. 127, 1–4 (2017). pmid:28045402
  61. 61. Johnson A. M. F. et al. High fat diet causes depletion of intestinal eosinophils associated with intestinal permeability. PLoS One (2015) pmid:25837594
  62. 62. Araújo J. R., Tomas J., Brenner C. & Sansonetti P. J. Impact of high-fat diet on the intestinal microbiota and small intestinal physiology before and after the onset of obesity. Biochimie (2017) pmid:28571979
  63. 63. Rathinam V. A. K., Zhao Y. & Shao F. Innate immunity to intracellular LPS. Nat. Immunol. 20, 527–533 (2019). pmid:30962589
  64. 64. McAleer J. P. & Vella A. T. Understanding how lipopolysaccharide impacts CD4 T-cell immunity. Critical Reviews in Immunology vol. 28 281–299 (2008). pmid:19166381
  65. 65. Monguió-Tortajada M., Franquesa M., Sarrias M. R. & Borràs F. E. Low doses of LPS exacerbate the inflammatory response and trigger death on TLR3-primed human monocytes article. Cell Death Dis. 9, 1–14 (2018).
  66. 66. Günther J. et al. Lipopolysaccharide priming enhances expression of effectors of immune defence while decreasing expression of pro-inflammatory cytokines in mammary epithelia cells from cows. BMC Genomics 13, (2012). pmid:22235868
  67. 67. Jeljeli M. et al. Trained immunity modulates inflammation-induced fibrosis. Nat. Commun. 10, 1–15 (2019).
  68. 68. Alverdy J. C., Hyman N. & Gilbert J. Re-examining causes of surgical site infections following elective surgery in the era of asepsis. The Lancet Infectious Diseases vol. 20 e38–e43 (Lancet Publishing Group, 2020). pmid:32006469
  69. 69. Thwaites G. E. & Gant V. Are bloodstream leukocytes Trojan Horses for the metastasis of Staphylococcus aureus? Nat. Rev. Microbiol. 9, 215–222 (2011). pmid:21297670
  70. 70. Miller R. J., Malfait A. M. & Miller R. E. The innate immune response as a mediator of osteoarthritis pain. Osteoarthritis and Cartilage (2020) pmid:31862470
  71. 71. Muraille E., Leo O. & Moser M. Th1/Th2 paradigm extended: Macrophage polarization as an unappreciated pathogen-driven escape mechanism? Frontiers in Immunology vol. 5 (2014). pmid:25505468
  72. 72. Honda K. & Littman D. R. The Microbiome in Infectious Disease and Inflammation. Annu. Rev. Immunol. (2012) pmid:22224764
  73. 73. Lawrence T. & Natoli G. Transcriptional regulation of macrophage polarization: Enabling diversity with identity. Nature Reviews Immunology (2011) pmid:22025054
  74. 74. Daghestani H. N., Pieper C. F. & Kraus V. B. Soluble macrophage biomarkers indicate inflammatory phenotypes in patients with knee osteoarthritis. Arthritis Rheumatol. (2015) pmid:25544994
  75. 75. Saeed K. et al. Bacterial toxins in musculoskeletal infections. J. Orthop. Res. (2020) pmid:32255540
  76. 76. Wu H. J. et al. Gut-residing segmented filamentous bacteria drive autoimmune arthritis via T helper 17 cells. Immunity (2010) pmid:20620945
  77. 77. Wrzosek L. et al. Transplantation of human microbiota into conventional mice durably reshapes the gut microbiota. Sci. Rep. 8, 1–9 (2018).