Introduction: Infections such as Chlamydia trachomatis and Yersinia enterocolitica are recognized triggers of reactive arthritis, but their role in chronic spondyloarthritis (SpA)—including ankylosing spondylitis (AS) and psoriatic arthritis (PsA)—remains incompletely defined.

Aim: To evaluate the potential role of selected urogenital and enterocolitic infections in the onset and clinical evolution of AS and PsA.

Materials and methods: This prospective observational study included 1202 patients (709 PsA, 493 AS) followed between 2019 and 2025. Clinical subtypes, disease activity, and imaging features were assessed alongside multiplex PCR and serological screening for C. trachomatis, Mycoplasma spp., Ureaplasma spp., and Y. enterocolitica. Comparisons were made between infection-triggered and idiopathic cases.

Results: Infection was found in 6.2% of PsA and 8.1% of AS patients, which was slightly higher than the control group. Infection-triggered cases presented more often with acute onset, oligoarthritis, and peripheral joint involvement (notably in AS, p=0.002). Over time, PsA showed a shift from oligoarticular to polyarticular and axial forms; axial PsA increased from 2.1% to 21.1% in 2 years. Radiographic and treatment outcomes were comparable between groups. A minority of infection-triggered cases showed remission following antibiotic therapy.

Conclusions: Urogenital and enterocolitic infections may precipitate SpA in a small subset of genetically susceptible individuals, particularly with HLA-B27. While long-term disease trajectories resemble idiopathic forms, early identification of infectious triggers may aid in personalized management strategies. Further research is needed to clarify their role in chronic disease propagation and treatment responsiveness.

ankylosing spondylitis, Chlamydia trachomatis, infection-triggered arthritis, spondyloarthritis, psoriatic arthritis

AS: ankylosing spondylitis

ASAS: Assessment of SpondyloArthritis international Society

ASAS20: 20% improvement criteria of the ASAS

BASDAI: Bath Ankylosing Spondylitis Disease Activity Index

BASFI: Bath Ankylosing Spondylitis Functional Index

BASMI: Bath Ankylosing Spondylitis Metrology Index

CG: control group

CRP: C-reactive protein

DIP: distal interphalangeal (joints)

DMARDs: disease-modifying antirheumatic drugs

ELISA: Enzyme-Linked Immunosorbent Assay

ESR: erythrocyte sedimentation rate

GI: gastrointestinal

GU: genitourinary

HLA-B27: human leukocyte antigen B27

IBD: inflammatory bowel disease

IgA/IgG: immunoglobulin A / immunoglobulin G

IL: interleukin

IQR: interquartile range

MASES: Maastricht Ankylosing Spondylitis Enthesitis Score

MASEI: Madrid Sonographic Enthesitis Index

MRI: magnetic resonance imaging

NAAT: nucleic acid amplification test

NSAIDs: nonsteroidal anti-inflammatory drugs

PCR: polymerase chain reaction

PsA: psoriatic arthritis

RA: rheumatoid arthritis

ReA: reactive arthritis

SD: standard deviation

SpA: spondyloarthritis

STI: sexually transmitted infection

TNF: tumor necrosis factor

UTI: urinary tract infection

YOPs: Yersinia Outer-Membrane Proteins

Ankylosing spondylitis (AS) and psoriatic arthritis (PsA) are two major forms of seronegative spondyloarthritis (SpA)—a group of chronic inflammatory diseases affecting the joints and entheses. AS primarily involves the axial skeleton (sacroiliac joints and spine), whereas PsA often presents with peripheral arthritis associated with psoriasis, though it can also affect the spine.^[1,2] These conditions impose substantial morbidity; they typically begin in young adulthood and can lead to chronic pain, stiffness, and disability. AS has a prevalence of approximately 0.1%–1.4% in the general population, with males more frequently affected, and more than 80%–90% of AS patients carry the HLA-B27 allele.^[2] PsA develops in an estimated 14%–30% of patients with cutaneous psoriasis and contributes significantly to impaired quality of life in this population. The pathogenesis of SpA is multifactorial, involving a complex interplay of genetic predisposition (HLA-B27 and others), immune dysregulation, and environmental factors.^[1,3]

One hypothesized environmental factor is infection. It has long been observed that reactive arthritis can follow certain gastrointestinal or genitourinary infections, and reactive arthritis is classified within the spondyloarthritis spectrum. Urogenital infections with Chlamydia trachomatis are the classic example: C. trachomatis is the most strongly linked pathogen in sexually acquired reactive arthritis, accounting for the majority of non-venereal ReA cases.^[3–5] In retrospective studies, 30%–50% of patients with Chlamydia-induced reactive arthritis develop a chronic or relapsing course of arthritis, sometimes progressing to a clinical picture indistinguishable from primary AS.^[5,6] Chlamydia pneumoniae has also been implicated, though less frequently.^[6] Among enterocolitic infections, Yersinia enterocolitica and Y. pseudotuberculosis are well-documented triggers in Northern Europe—for example, outbreaks of Yersinia food contamination have led to reactive arthritis in ~22% of exposed adults (especially HLA-B27 carriers).‌^[4,7] Similarly, Salmonella, Shigella, and Campylobacter infections can precipitate reactive arthritis, typically within 1-4 weeks of the inciting gastroenteritis. Mycoplasmas and Ureaplasmas are less common pathogens in this context, but case reports and small series have suggested Mycoplasma genitalium (a cause of non-gonococcal urethritis) can lead to reactive arthritis in HLA-B27-positive individuals.‌^[7–9] Ureaplasma urealyticum has been isolated in some patients with reactive arthritis or undifferentiated arthritis, though establishing causality has been challenging.^[8,10] Overall, these infections are thought to act as triggers that induce an immune response, which, in a genetically susceptible host, results in autoimmune or autoinflammatory arthritis even after the pathogen is cleared from the initial site.

Mechanistic theories to explain the link between infection and spondyloarthritis include molecular mimicry (microbial antigens resembling self-proteins, leading to cross-reactive immune responses) and persistent bacterial antigens or occult infection in joint tissues.^[11,12] Notably, DNA or antigen from Chlamydia has been detected in the synovium of patients with chronic reactive arthritis and even in some patients with undifferentiated or axial SpA.‌^[6] This suggests that Chlamydia can evade complete eradication and reside in an intracellular, metabolically altered state in host tissues, perpetuating inflammation. Yersinia and other enteric bacteria have been shown to alter host immune responses; for instance, Yersinia infection can modify HLA-B27 antigen expression on leukocytes, potentially enhancing inflammatory arthritis in B27-positive hosts.^[7] Furthermore, gut microbiome changes have been linked to AS pathogenesis – subclinical gut inflammation is present in up to 50% of AS patients, hinting that chronic exposure to enteric microbial products might drive axial inflammation.^[13] In PsA, chronic skin inflammation and trauma (Koebner phenomenon) are known triggers, but infection may also play a role in certain cases.^[14,15] Streptococcal infection, for example, is associated with guttate psoriasis flares and has been proposed as a trigger for PsA onset in some patients.^[16,17]

Despite these insights, the frequency and clinical impact of specific infections in AS and PsA remain areas of active investigation. Not all patients with AS or PsA have a history of infection, and in those who do, it is often unclear whether the infection was causal or coincidental. The progression from an acute reactive arthritis to chronic AS is reported only in a subset (an estimated 15%-30% of reactive arthritis cases persist or recur chronically).^[4,12] In PsA, infection triggers are less well documented aside from the noted streptococcal association in psoriasis. There is a need to better delineate which patients with SpA have an infection-associated onset and whether their disease course differs (for example, in terms of joint involvement pattern or disease severity).^[3,6,14] This could inform early therapeutic strategies—for instance, the use of antibiotics or more aggressive immunomodulation in those cases.

The study’s aim was to determine the potential pathogenic role of specific urogenital and enteric infections in causing or modifying the clinical course of spondyloarthritis. This could enhance the understanding of infection-driven immunopathogenesis and identify patients who may benefit from early targeted interventions.

In this study, we looked at how specific urogenital and gastrointestinal infections, particularly Chlamydia trachomatis, Mycoplasma spp., Ureaplasma spp., and Yersinia enterocolitica, affected the onset and clinical progression of AS and PsA. While the majority of AS and PsA cases appear to develop in the absence of a recognizable infectious trigger, our findings suggest that in a small but significant subset (approximately 3-5%), infection may serve as a precipitating event. This aligns with previous data indicating that C. trachomatis is the leading identifiable cause of sexually acquired reactive arthritis and may contribute to chronic SpA phenotypes in genetically predisposed individuals, particularly those positive for HLA-B27.^[5,6]

Our detection of C. trachomatis DNA in approximately 4-5% of AS and ~3% of PsA patients—rates modestly above the population baseline—supports the hypothesis that persistent or subclinical chlamydial infection may contribute to disease pathogenesis. These findings are notably lower than those reported by Carter et al.^[5], who identified C. trachomatis in the synovial tissue of 62% of patients with undifferentiated SpA using PCR-based analysis of joint biopsies. The disparity is likely attributable to differences in the sampling method, as our screening utilized non-invasive urogenital swabs or urine samples, which may underestimate localized intra-articular infection. Nevertheless, our results lend support to the “hit-and-run” model of spondyloarthritis pathogenesis, in which an acute or subacute infection initiates a self-perpetuating immune cascade that sustains inflammation even after the pathogen has been cleared.

In the case of Yersinia enterocolitica, our observed seroprevalence (~5% in SpA vs. ~4% in controls) is in line with background rates in non-epidemic settings. However, previous outbreaks in Scandinavia, particularly those documented by Gérard et al.^[6], have established that Yersinia can trigger reactive arthritis in a significant proportion of exposed individuals, especially those carrying HLA-B27. Our cohort included several patients with AS whose disease onset followed an acute gastrointestinal illness attributed to Yersinia or Y. pseudotuberculosis – a scenario echoed in the work of Hannu et al.^[7], who reported that 3.5% of patients with ReA progressed to AS over time. These observations strengthen the notion that a subset of idiopathic AS may, in fact, represent post-infectious spondyloarthritis with chronic evolution. This concept is also supported by experimental data from HLA-B27 transgenic rats, which develop spondylitis-like pathology upon enteric bacterial exposure.^[7]

A key strength of our analysis is the detailed comparison between infection-triggered and idiopathic disease presentations. In PsA, infection-triggered cases typically presented with an abrupt, oligoarticular pattern—clinically mimicking reactive arthritis—and showed a higher frequency of enthesitis and involvement of lower-extremity joints (e.g., knees, ankles, and mid-foot). This presentation contrasts with idiopathic PsA, where the disease more often developed insidiously and was characterized by a broader range of patterns, including DIP-predominant and polyarticular forms. Importantly, these differences suggest that early immune activation by microbial antigens may shape disease phenotype at onset. Similar trends were seen in AS: infection-triggered cases more frequently exhibited a mixed axial-peripheral pattern at onset, with 45% showing swollen peripheral joints compared to ~15% in idiopathic AS (p=0.002). These patients were often initially misdiagnosed with reactive arthritis before developing full-blown axial SpA.

Another significant finding from our cohort is the high frequency and progression of enthesitis and dactylitis, which are hallmarks of PsA but are significantly less common in rheumatoid arthritis patients. Importantly, these features may be potentiated by infection: reactive arthritis is classically associated with heel enthesitis and toe dactylitis, suggesting a role for microbial antigens in targeting enthesis-rich anatomical zones – possibly via molecular mimicry with bacterial heat-shock proteins or other conserved structures.^[9]

From a pathogenic perspective, our data support the model in which an environmental trigger, such as a urogenital or enteric infection, initiates immune activation in a genetically susceptible host. While the trigger may be transient, it can initiate an autoinflammatory loop sustained by persistent T-cell activation, local cytokine production, and possibly dysbiosis of gut or skin microbiota.^[13] In PsA, studies have shown increased abundance of Escherichia coli and other microbial shifts that may modulate mucosal immunity.^[4] These mechanisms likely play a larger role in early disease initiation than in later propagation, where the immune response becomes self-sustaining. The similarity in long-term disease activity between infection-triggered and idiopathic SpA in our cohort supports this transition from exogenous to endogenous pathogenic drivers.^[2]

Clinicians should therefore maintain a high index of suspicion for infection in new-onset SpA, especially in patients with abrupt presentations, HLA-B27 positivity, and recent genitourinary or gastrointestinal symptoms. Screening (e.g., for C. trachomatis, Yersinia, and stool pathogens) and prompt treatment may reduce inflammatory burden or prevent further antigenic stimulation. Beyond acute infections, chronic low-grade sources (e.g., periodontal disease) may also amplify systemic inflammation in PsA and should be managed proactively.^[14,15] While some reports have linked vaccinations (including influenza and COVID-19) to PsA flares, our data did not support this association, and the overall benefit-risk ratio of vaccination remains strongly in favor of immunization.^[7,16]

Interestingly, patients in our cohort more often cited psychological stress (14.6%) and physical trauma (~8%) than infection (6.4%) as putative disease triggers. These findings are consistent with previous research on the Koebner phenomenon and support a multifactor model of PsA initiation.^[14] The convergence of environmental, genetic, and immunologic factors complicates attempts at prevention but highlights the diagnostic importance of recognizing recent triggers – be it stress, trauma, or infection – as part of the clinical history.

In comparison to existing literature, our results reinforce key concepts. First, infection-triggered SpA is real but relatively uncommon in PsA. Second, axial PsA increases with disease duration and is frequently underrecognized in early stages. Third, chronic infection may aggravate entheseal inflammation and perpetuate immune dysregulation. And fourth, once chronic SpA is established, infection appears to play a diminishing role in disease activity.

This study has several limitations. First, infection identification was based on patient history and peripheral testing (PCR/serology), which may miss asymptomatic or past infections. Joint fluid or synovial tissue PCR was not performed, potentially underestimating microbial persistence. Second, only selected pathogens (Chlamydia trachomatis, Yersinia enterocolitica, and Mycoplasma, Ureaplasma) were screened; other relevant microbes (e.g. Salmonella and Shigella) were not systematically evaluated. Third, the follow-up period (median 3 years) may be too short to capture long-term outcomes such as axial fusion or arthritis mutilans. Fourth, treatment was not standardized; although care followed clinical guidelines, variability in timing and type of therapy could influence results. Finally, the single-center design and regional epidemiology may limit generalizability. Broader, multicenter studies with immunological profiling are needed to confirm and expand these findings.

Our findings demonstrate that urogenital and enterocolitic infections—particularly Chlamydia trachomatis and Yersinia enterocolitica—can act as triggers in a subset of patients with SpA, initiating an acute, reactive-like disease that may evolve into chronic AS or PsA. These infection-associated forms often begin with prominent peripheral arthritis and enthesitis, but over time, their clinical trajectory and treatment needs converge with idiopathic cases, highlighting the self-sustaining nature of the inflammatory process.

In PsA, we observed a predictable evolution from peripheral oligoarthritis to polyarticular and axial involvement, often accompanied by progressive dactylitis and enthesitis. In AS, while axial involvement predominates from the outset, infection-triggered cases more commonly exhibit a mixed axial-peripheral phenotype. In both conditions, infections appear to act as initiators in genetically susceptible individuals, particularly those carrying HLA-B27.

These findings emphasize the importance of early screening for infections in new-onset SpA. Timely identification and treatment of infectious triggers may alleviate acute symptoms and limit immune activation, though their impact on long-term disease control remains uncertain. Early diagnosis and aggressive management, especially in PsA, are critical to preventing structural damage and functional decline.

Ultimately, recognizing infection-triggered SpA supports a precision medicine approach. Further research should explore microbial biomarkers, the utility of anti-infective therapies in defined subgroups, and the long-term benefits of integrated infection and inflammation control strategies in spondyloarthritis.

The authors have no funding to report.

The authors have declared that no competing interests exist.

The authors have no support to report.