Chlamydia trachomatis and Chlamydia pneumoniae together comprise the most frequent causative pathogens that elicit reactive arthritis (ReA). Advances in our understanding of the molecular biology/molecular genetics of these organisms have improved significantly the ability to detect chlamydiae in the joint for diagnostic purposes, as well as extending our current understanding of the pathogenic processes they elicit in the joint and elsewhere. An important aspect of the latter is that synovial chlamydiae infect the joint in an unusual but metabolically active state. While some standard treatments can provide a palliative effect on the ReA disease phenotype, many reports have indicated that standard antibiotic treatment does not provide a cure. Of critical importance, however, two recent reports of controlled clinical trials demonstrated that Chlamydia-ReA can be treated successfully using combination antibiotic therapy. These observations offer the opportunity of a cure for this disease, thereby increasing the practical importance of awareness and diagnosis of the spondyloarthritis caused by Chlamydia. In this viewpoint, we provide an overview of recent key findings in the epidemiology, pathophysiology, clinical manifestations, diagnosis and treatment of Chlamydia-induced arthritis. Our intention is for these insights to be translated rapidly into clinical practice to overcome misdiagnosis and underdiagnosis of the disease, and for them to stimulate the continued development of a cure.
- Reactive Arthritis
Statistics from Altmetric.com
That chlamydial infection acts as a trigger for Reiter's syndrome and reactive arthritis (ReA) has been known for many years. However, it has become clear during the last two decades that the two species primarily responsible for causing arthritis, Chlamydia trachomatis and Chlamydia pneumoniae elicit arthritis by their persistence in the joint.1 ,2 Interestingly, evidence now exists indicating a chlamydial aetiology for at least some patients with spondyloarthritis (SpA).3 Epidemiological data suggest that Chlamydia induced reactive arthritis (CReA) is a more common condition than previously thought, and that clinicians often fail to recognise it.4 ,5 Most importantly, a recent controlled study demonstrated that CReA can be successfully treated with combination antibiotic therapy, raising the potential for a cure.6 That possibility highlights the increased significance of awareness and diagnosis of arthritis and SpA caused by chlamydiae.7 This viewpoint provides an overview on key findings in epidemiology, pathophysiology, clinical manifestations, diagnosis and treatment of CReA, all of which should be translated into clinical practice to overcome misdiagnosing/underdiagnosing of this arthritis.
Terminology and classification
An aetiological connexion with C trachomatis infection was first described in patients with Reiter's syndrome, sexually acquired reactive arthritis, and undifferentiated arthritis (cf ref 8). To date, CReA has been classified as an instance of ReA within the SpA disease group.9 Historically, CReA has been defined as a non-purulent arthritis that develops during or soon after extra-articular bacterial infections elsewhere in the body, but in which the micro-organism cannot be recovered from the joint.10 However, it is well accepted today that CReA is caused by intra-articular persistent infection and therefore is of postinfectious origin.11
We introduced the term CReA specifically to describe the rheumatological manifestations following urogenital infections with C trachomatis.12 However, other clinical designations have been employed to describe the arthritis caused by chlamydial infections in general (table 1).13–20 Recently, the specific Chlamydia species involved have been included in published terms (table 1).15 ,21–27 It may be most appropriate to designate Chlamydia induced arthritis and Chlamydia SpA specifically as C trachomatis induced arthritis, C pneumoniae induced arthritis, C trachomatis SpA and C pneumoniae SpA, thereby distinguishing these disease entities from other SpA. Such a change emphasises causation by viable, persistently infecting organisms, replacing the misleading concept of ReA as a disease characterised by absence of bacteria in the joint. Regardless, in this review we maintain the designation CReA in accordance with the current classification as part of ReA and SpA (however, see refs 28, 29).
Genital infections with C trachomatis are among the most common sexually transmitted infections in Europe and America, and they primarily affect young people. Studies in the USA that followed patients after venereal chlamydial infection demonstrated that 4.1–8.1% of patients developed ReA.5 ,30 In Norway and Finland, the incidence of CReA was determined to be 3–5.4/100 000 population/year.27 ,31 ,32 CReA was seen in approximately 20% of patients, with nearly equal sex ratio, in individuals with undiagnosed arthritis attending an early synovitis clinic.33 Comparable prevalence has been reported from several rheumatological clinics, with identification of C trachomatis as the causative pathogen in 16% of those with undifferentiated oligoarthritis, and a much higher prevalence of 63% in urogenic ReA.34 Importantly, the preceding genital chlamydial infection is often asymptomatic; in our own experience this is obtained in 30% of patients with CReA, and it is more frequent in women (61% of the asymptomatic patients) than in men.33 Two other studies of patients with oligoarthritis of undetermined origin reported evidence of undetected, asymptomatic C trachomatis infection in 36% and 17% of individuals studied, respectively.35 ,36 Silent infection, frequent spontaneous remission and reluctance to reveal C trachomatis genital infection may be major reasons why CReA is often underdiagnosed.
C pneumoniae typically causes acute upper respiratory manifestations, especially atypical pneumonia, but the organism may engender persistent infections as does C trachomatis. Most patients infected with C pneumoniae are asymptomatic, and the course of respiratory illness is relatively mild.37 The serological prevalence is high, with 50% at age 20 years and 70–80% at age 60–70 years.38 Despite the high seroprevalence, prospective epidemiological studies estimated that the occurrence of ReA triggered by C pneumoniae is less than by C trachomatis, with 2.2% and 13.4%, respectively.31 ,39 Other studies in patients with arthritis and undifferentiated SpA confirmed that ReA caused by C pneumoniae is much rarer than C trachomatis-induced ReA.40–42 Of note, 5/18 cases published to date with C pneumoniae ReA were asymptomatic for preceding upper or lower respiratory infection.8 ,40 ,21 In undifferentiated SpA a history of C pneumoniae infection was encountered in 0%, compared with a history of C trachomatis infection at any time in 42%.42
Pathogenesis of Chlamydia-induced reactive arthritis
Knowledge of the biology of the chlamydiae is essential in understanding the pathogenesis, diagnosis and treatment of CReA. Chlamydiae are non-motile obligate intracellular bacteria whose replication relies on intact host cells for much of their energy supply. Both chlamydial species infect a variety of human cell types, including epithelial, endothelial and smooth muscle cells, as well as macrophages and monocytes.43 They have a unique developmental cycle that takes place within membrane-bound cytoplasmic inclusions in eukaryotic host cells (figure 1).44 These organisms display two forms, the infectious elementary body (EB) and the non-infectious reticulate body. Extracellular EBs attach to permissive cells and are taken up by endocytosis. Metabolically active reticulate bodies are formed 12–24 h after uptake, and they multiply by binary fission within the inclusion. Typical inclusions can occupy up to two-thirds of the host cell at late times postinfection. Reorganisation to the EB form takes place after 24–30 h. EBs are released by host cell lysis or exocytosis to propagate infection.
The essential elements underlying the origin and pathogenesis for CReA have been elucidated.2 ,3 ,5 ,45–47 At the site of primary infection (urogenital tract, ocular conjunctivae or respiratory tract mucosae), the bacteria infect monocytes/macrophages and are disseminated via the bloodstream to settle into the joint. Thereafter the persistent chlamydiae cannot be detected using traditional culture techniques but can be located in synovial membrane and synovial fluid using various methods, including electron microscopy, immunofluorescence and PCR.
Chlamydiae persist in the inflamed joint in an unusual form, which is, in contrast with infectious or purulent arthritis, non-culturable (figure 1).1 ,44 Morphologically aberrant but viable, metabolically active persistent Chlamydiae display a panel of characteristic changes in gene expression and host cell-dependent energy uptake that are in contrast with those of acute infection (figure 1).3 ,47–50 For example, during persistence, expression of the major outer membrane protein gene (omp1) and several genes required for the cell division process are severely downregulated.3 ,48 Inflammation is elicited when monocytes persistently infected with Chlamydia elicit a Th1-type immune response, recruiting T lymphocytes for defence through a variety of proinflammatory mediators, particularly tumour necrosis factor-α (TNF-α), interleukin 1 and interferon- γ (for review of the contribution of the immune system to the pathogenesis see refs 50–54).
Surprisingly, a recent study demonstrated that synovial tissue samples from patients with documented CReA were positive for ocular (trachoma), not genital, serovars, the latter of which have been assumed to elicit CReA following genital infections.55 This observation led to the suggestion that ocular strains are arthritogenic and probably more likely to disseminate compared with genital strains; this might also account for the fact that many patients develop eye disease such as conjunctivitis and/or iritis/uveitis.5
Clinical spectrum of chlamydial infections
Infection with Chlamydia usually precedes the occurrence of arthritis by 1–4 weeks. Initial clinical manifestations are variable, given the genitourinary and respiratory portal of entry of the two chlamydial species (table 2).8 ,12 ,21 ,33 ,39 ,40 ,42 A specific medical history querying the highly variable spectrum of infection-related signs and symptoms is of paramount diagnostic importance, since the patient usually is not familiar with the causative relationship. Therefore, the rheumatologist should inquire in detail concerning any incidence of urethritis and cervicitis and ascending infections by intraluminal spread to detect preceding C trachomatis infection (table 2). In some cases, a positive history of such signs in the sexual partner, as well as the disclosure of new or changing sexual partners, may be important hints.
Other less easily assessed sites of primary infection with C trachomatis include the conjunctivae, a result of transmission from genitalia to eye by smear infection; this can lead to follicular conjunctivitis as well as epithelial and subepithelial infiltration of the cornea with mucopurulent or serous reaction.56 In some patients C trachomatis has been isolated from pharyngeal swabs, nasal and sinusoidal discharges, and bronchoalveolar lavage probably due to transmission of the infection by orogenital sexual contact. Clinical manifestations include tonsillitis, sore throat, sinusitis, bronchitis and pneumonia, the latter especially in immunocompromised subjects.13 ,57–60 Finally, anorectal C trachomatis infections due to anogenital sex are reported primarily in men who have sex with men.61 The infection usually causes few or no symptoms but may be associated with proctitis.62
Typical respiratory manifestations of C pneumoniae infection include community-acquired atypical pneumonia, bronchitis, pharyngitis and sinusitis (table 2).38 Exacerbations of chronic bronchitis have been associated with C pneumoniae infections, and the pathogen is also involved in acute and chronic asthma; some data link this agent to new-onset asthma in adults.38
Importantly, C trachomatis and C pneumoniae infections are often asymptomatic or mild and therefore only discovered by targeted investigation.33 ,63 ,64 Moreover, the urogenital infection can sometimes precede the arthritis by several years, or it may be clinically apparent long after appearance of rheumatic signs and symptoms.65 This can result from reinfection or reactivation of persistent infection, either of which can engender arthritis.66 Evidently, C trachomatis is present and metabolically active during the remitting phase in synovial tissues from patients with chronic CReA.67 Moreover, simultaneous C pneumoniae and C trachomatis infection was described in a HLA-B27 positive subject with ReA.66 This observation extends an earlier finding of simultaneous detection of DNA from both chlamydial species in synovial tissue of patients with arthritis and undifferentiated SpA.41 ,42 Another coinfection was described in two patients with ReA in which their tonsils were concomitantly infected with C trachomatis and Streptococcus spp.68 The clinical relevance of these coinfections is unknown.
Overall, rheumatologists should be cognisant of the high degree of variability and broad spectrum of presentation, the frequently asymptomatic manifestation and the possible coinfection by chlamydiae; with these in mind, the clinician will not overlook chlamydial infections in individual patients presenting with early arthritis and SpA.
Clinical presentation, manifestations of Chlamydia-induced reactive arthritis and spondyloarthritis
Description of the specific demographics and clinical spectrum of CReA in patients given a definite diagnosis of chlamydial infection has been rare (table 3).10 ,33 ,42 CReA, like other ReA and SpA, combines four syndromes: peripheral arthritis, enthesopathy, axial involvement and extramusculoskeletal manifestations. Combination of these four varies from patient to patient and in any given patient during the disease course. The arthritis usually manifests itself in CReA as monoarthritis or oligoarthritis; polyarticular involvement is rare. Large joints, especially of the lower extremities, are most commonly affected, and the pattern of disease is usually asymmetrical.33 ,42 Other manifestations are enthesopathy and dactylitis. Sacroiliitis and axial manifestations of SpA also are frequent.
Extra-articular organ manifestations most commonly affect skin and eye, more rarely the heart, kidneys or muscle, and central nervous systems (table 2). This includes the well-known Reiter's syndrome with the triad of urethritis, conjunctivitis and arthritis; this term should no longer be used because of the Nazi past of Hans Reiter and the obvious lack of evidence for a separate disease entity.69 ,70
No specific recommendations currently exist at the national or international level defining specific clinical and/or laboratory investigations for tests to diagnose CReA.71 The following proposals are based on available literature and personal experience.
The search for Chlamydia or its components at the site of the primary infection or in the joint is the optimal approach to confirm chlamydial aetiology of the arthritis.The most specific diagnosis of CReA is made by the detection of the pathogen in the joint itself (synovial fluid, synovial biopsy), using PCR and/or other molecular amplification assays.41 ,42 ,55 ,72–74 Importantly, we were able to detect C trachomatis also in the sacroiliac joint.75 Of note, three patients taking TNF-α antagonists for rheumatoid arthritis developed psoriasiform lesions resembling keratoderma blennorrhagicum which were found to be positive for C trachomatis DNA.76 Commercially available tests for chlamydial DNA which are adapted for joint samples are not available, however, thus only a few specialised centers around the world perform the examination with the required sensitivity using inhouse systems. Finally, C trachomatis and C pneumoniae can appear in the joints of patients who have other clinical diagnoses, including psoriatic arthritis, undifferentiated oligoarthritis, synovitis acne pustulosis hyperostosis osteitis syndrome, rheumatoid arthritis, sarcoidosis, osteoarthritis, and in asymptomatic, clinically normal joints.76–83
Joint material for appropriate analyses often cannot be obtained or is not available in clinical practice. Moreover, data are lacking which provide information on diagnostic efficiency and cost of the recommended molecular testing for Chlamydia in the joint. Therefore, direct detection of Chlamydia usually relies on the examination of specimens collected from the site of primary infection (table 4).33 ,74 ,84 ,85 The method of choice is nucleic acid amplification, which is more sensitive than cell culture, direct immunofluorescence assay, enzyme immunoassay, and nucleic acid hybridisation tests.85 Urogenital specimens and/or rectal smears are investigated for C trachomatis and respiratory secretions for C pneumoniae. However, positive findings without other diagnostic hints cannot prove the causal role of these pathogens for arthritis because of the frequent unrelated, asymptomatic chlamydial infections in healthy people.
Serology may be of some value in cases where laboratory testing to identify Chlamydia in the joint are not available. For interpretation of chlamydial serology, the temporal occurrence and combination of IgG, IgA and IgM antibodies are important. In particular, the simultaneous detection of IgG and IgA antibodies is typical for fresh or persistent infections and indicates the diagnosis of probable C trachomatis ReA in patients with corresponding clinical history and symptoms (figure 2).15 ,16 ,86–88 The detection of IgM antibodies is of diagnostic relevance in primary infection with C pneumoniae, indicating a fresh, acute infection.38 Interestingly, recent C pneumoniae infections occur frequently in patients with ankylosing spondylitis (AS) and C pneumoniae IgM antibody is correlated with active disease, as reported in a Chinese cohort.89
Commercially available serological diagnostics, however, are limited by gender-specific cross-reactivity test systems, non-specific stimulation of antichlamydial antibodies, poor sensitivity, and in cases of simultaneous or consecutive exposure to both chlamydial species. Further, diagnostic accuracy is limited by an increasing prevalence of antibodies against C trachomatis and/or C pneumoniae with age in the healthy population.90 ,91 The best association of sensitivity (76%) and specificity (85%) was obtained when IgG and/or IgA reactivity to two C trachomatis antigens was determined87; these were derived from species-specific epitopes in variable domain IV of the major outer membrane protein and a recombinant polypeptide encoded by open reading frame 3 on the chlamydial plasmid. The diagnosis of CReA by serology alone is always debatable and should be supplemented and confirmed by search for Chlamydia at the entrance gate whenever possible (figure 2).
Triggering chlamydial infection
The triggering acute or persistent urogenital C trachomatis infection should be treated with azithromycin or doxycycline according to established recommendations for sexually-acquired chlamydial infections; these include 1-day treatment with azithromycin of sexual partners to prevent reinfection of patients and complications in patients and partners.92 Because of the high prevalence of reinfection, retesting for Chlamydia is recommended in both genders approximately 3 months after treatment, regardless of whether they believe that their sex partners were treated.93 Tetracyclines and erythromycin are the most commonly employed drugs in the treatment of C pneumoniae infections, while new macrolides, ketolides and fluoroquinolones are other potentially effective drugs38 We note that use of newly developed genetic tools to elucidate the function of currently uncharacterised gene products in C trachomatis and C pneumoniae should identify novel therapeutic approaches for treatment (e.g., refs. 94 95 96).
Antichlamydial therapy of the arthritis
Identification of Chlamydia and/or other bacteria in joints initiated several studies to test various antibiotics for their elimination of pathogens from that site; all trials using antibiotic monotherapy were unsuccessful (cf refs 97, 98). However, a recent trial demonstrated positive results using an antibiotic combination in chronic SpA, with a special focus on Chlamydia.99 This was followed by a study in patients with demonstrated CReA which showed that a 6-month course of combination therapy with rifampicin (300 mg/day) plus doxycycline (200 mg/day), or plus azithromycin (500 mg/day followed by 5 days of 2–500 mg once/week) is effective in eliminating pathogens, giving improvement of arthritis; patients in this study were shown to be PCR-positive either in blood or joint fluid for C trachomatis or C pneumoniae.5 ,6 A response was observed in 63% versus 22%, and complete remission was observed in 20% versus 0% under active treatment compared with placebo, respectively. The combination of azithromycin and rifampin was most effective, although the study was not powered to determine which combination of antibiotics is superior. These results open for the first time the prospect for curative treatment. However, the effectiveness of this approach must be confirmed in additional studies, especially in patients diagnosed only by serology and clinical manifestations for chlamydial infection.7
Conventional treatment of the arthritis
Non-steroidal anti-inflammatory drugs, physical modalities and physiotherapy are applied as first-line therapy, as usual for cases of ReA from other causes.97 Intra-articular glucocorticoid injections can be employed in patients with monoarticular or oligoarticular disease; enthesopathy also responds to local glucocorticoid injections.97 Systemic use of glucocorticoids may be indicated in severe polyarthritis if a high level of systemic inflammation is present, if the patient is febrile, or if the patient has carditis or atrioventricular conduction disturbances.97 However, no controlled trials have assessed these medication or treatment modalities specifically in CReA.
Few therapeutic studies of ReA lumping together patients with posturethritic and postenteric arthritis are available. Two comparative trials demonstrated efficacy of the non-steroidal anti-inflammatory drugs azapropazone, ketoprofen and indometacin in Reiter's disease (ref 5 for review). Among disease modifying antirheumatic drugs, sulphasalazine compared with placebo seemed to improve only the short-term outcome if started during the first 3 months of ReA but has been more effective when used in chronic disease.100 ,101 In individual cases, the anti-TNF-α biological modifiers infliximab and etanercept were successful for the treatment of CReA refractory to methotrexate and sulphasalazine.102–104 However, these agents might promote persistence or dissemination of Chlamydia, as reported in patients with rheumatoid arthritis.76 Thus, more studies are needed to evaluate the efficacy and safety of these biological modifiers in CReA.
In recent years, our understanding of the persistent state and molecular biology of CReA have been advanced considerably. Diagnosis of the disease is optimised by molecular testing for presence of the bacteria in joints, even though in most cases careful checking of medical history and more easily available laboratory testing are indicative. Extremely promising are recent positive data indicating the efficacy of combination antibiotic treatment for CReA. However, these latter studies have been performed only in patients tested positive for Chlamydia in the joint or in peripheral blood cells. Hence reliable commercially available molecular tests for joint specimens are needed urgently. Moreover, development of new practical and sensitive methods for detection of Chlamydia in blood samples will allow a simpler, more rapid and specific diagnosis in patients where joint samples cannot be obtained. Nonetheless, the recent and expected advances in diagnosis and treatment of CReA point towards a cure in the near future.
Handling editor Tore K Kvien
Contributors Both authors equally contributed to the manuscript.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.