Reactive arthritis (ReA) can be defined as the development of sterile inflammatory arthritis as a sequel to remote infection, often in the gastrointestinal or urogenital tract. Although no generally agreed-upon diagnostic criteria exist, the diagnosis is mainly clinical, and based on acute oligoarticular arthritis of larger joints developing within 2–4 weeks of the preceding infection. According to population-based studies, the annual incidence of ReA is 0.6–27/100 000. In addition to the typical clinical picture, the diagnosis of ReA relies on the diagnosis of the triggering infection. Human leucocyte antigen (HLA)-B27 should not be used as a diagnostic tool for a diagnosis of acute ReA. In the case of established ReA, prolonged treatment of Chlamydia -induced ReA may be of benefit, not only in the case of acute ReA but also in those with chronic ReA or spondylarthropathy with evidence of persisting chlamydia antigens in the body. In other forms of ReA, there is no confirmed evidence in favour of antibiotic therapy to shorten the duration of acute arthritis. The outcome and prognosis of ReA are best known for enteric ReA, whereas studies dealing with the long-term outcome of ReA attributable to Chlamydia trachomatis are lacking.
The term ‘reactive arthritis’ (ReA) was introduced in 1969 . It was defined as arthritis, which develops during or soon after an infection elsewhere in the body, but in which the microorganism does not enter the joint cavity. Since then, it has been shown that antigens of the triggering microbe can be detected in the synovial fluid or synovial tissue of affected joints , and, rarely, even replicating microbes have been demonstrated in the joints of patients with typical clinical features of ReA . Despite this, the original definition covers reasonably well the clinical entity of ReA.
Despite several attempts, no universal agreement of validated classification or diagnostic criteria of ReA have been achieved. Based on discussions at the 4th International Workshop on Reactive Arthritis in Berlin in 1999, a consensus was achieved that the term ‘ReA’ should be confined to patients, who present with clinical features typical of ReA, and in whom the preceding infection is caused by a microbe, which commonly associates with ReA . Further, preliminary, non-validated classification criteria for ReA were proposed ( Table 1 ). The limitations and uncertainties considering different classifications have been discussed elsewhere .
|
The primary focus of infection is through mucosal membrane, usually in the gut or in the urogenital tract. The classical microbes in the gastrointestinal tract include Yersinia , Salmonella , Shigella and Campylobacter ( Table 2 ). Interestingly, recent epidemiological investigations from Denmark demonstrate that a considerable number of patients (1–16%) report arthralgia and, in some cases, also arthritis in association with gastroenterical infection due to diarrhogenic strains of Escherichia coli . E. coli has been connected with ReA also in clinical settings . Based on case studies, Clostridium difficile has also been attributable to ReA .
| Enteric infections |
| Salmonella |
| Various serovars |
| Shigella |
| S. flexneri |
| S. dysenteriae |
| S. sonnei |
| Yersinia |
| Y. enterocolitica (especially O:3 and O:9) |
| Y. pseudotuberculosis |
| Campylobacter |
| C. jejuni |
| C. coli |
| Clostridium difficile |
| Escherichia coli |
| Diarrhogenic strains |
| Urogenital infections |
| Chlamydia trachomatis |
| Ureaplasma urealyticum |
| Mycoplasma genitalium a |
| Respiratory infections |
| Chlamydia pneumoniae |
| Group A beta-hemolytic Streptococcus b |
a Causes urethritis, but the role in reactive arthritis still unsettled.
b Typically causes acute rheumatic fever, but has been described to cause “reactive-like” arthritis.
Chlamydia trachomatis is by far the most common cause of genital infection associated with the development of ReA . Ureaplasma urealyticum is associated with ReA, but the role of Mycoplasma genitalium , another microbe connected with urethritis, as an independent trigger of ReA (in the absence of Chlamydia infection), is not settled .
Of the respiratory pathogens , Chlamydophila (Chlamydia) pneumoniae can be involved in nearly 10% of patients with ReA . Of other respiratory infection pathogens, beta -haemolytic Streptococcus is a classical trigger of acute rheumatic fever. However, ReA has been described as a sequel to streptococcal sore throat .
Incidence
Most cases of ReA arise sporadically. Single-source epidemics, in which patients with joint symptoms during and/or after the infection have been traced with questionnaires and/or with clinical investigations, have been reported in association with different gastrointestinal microbes ( Tables 3 and 4 ). In these studies, estimates of the incidence of ReA and frequency with which it occurs following an episode of infection depend, in large part, on how ReA is defined and on the epidemiologic investigation methods used .
| Microbe | Subjects in the outbreak, n | Reactive arthritis, % | Frequency of HLA-B27 | Reference |
|---|---|---|---|---|
| Campylobacter jejuni | 347 | 0.7 | 0/1 | Eastmond |
| Campylobacter jejuni | 106 | 1.8 | 1/1 | Bremell et al. |
| Shigella | 602 | 1.5 | NA | Noer |
| Shigella sonnei | 184 | 0 | – | Simon et al. |
| Shigella flexneri 2a | 636 | 1.5 | 2/3 | Simon et al. |
| Shigella flexneri 1b | 709 | 1.5 | 3/3 | Simon et al. |
| Shigella flexneri 2a | 205 | 2.3–6.9 a | a | Finch et al. |
| Yersinia enterocolitica | 117 | 0 | – | Lindholm &Visakorpi |
| Yersinia pseudotuberculosis O:1a | NA | 3 | 1/1 | Tertti et al. |
| Yersinia pseudotuberculosis O:3 | NA | 21 | 3/4 | Tertti et al. |
| Yersinia pseudotuberculosis O:3 | 38 | 12 | 3/3 | Hannu et al. |
| Serotype | Subjects in the outbreak, n | Reactive arthritis, % | Frequency of HLA-B27 | Reference |
|---|---|---|---|---|
| Enteritidis | 217 | 29 | NA | Dworkin et al. |
| Enteritidis | 286 | 0 | NA | Rudwaleit et al. |
| Enteritidis | 94 | 19 | NA | Locht et al. |
| Typhimurium | 78 | 8 | 2/4 | Hannu et al. |
| Typhimurium | 261 | 14.6 | 5/30 | Lee et al. |
| Enteritidis | 592 | 19.2 a | 5/37 | Rohekar et al. |
| Hadar | NA | 10 | NA | Arnedo-Pena et al |
In the outbreak studies of C. jejuni , ReA has been reported to occur in 0.7–1.8% of the subjects . For Shigella spp., the occurrence was 1.5% in the classic study by Noer from the 1960s . The occurrence was 1.5–6.9% in the studies, in which the serotype was reported to be either Shigella flexneri 2a or S. flexneri 1b , whereas, in a study with serotype S. sonnei , no cases of ReA occurred . An outbreak study dealing with Yersinia enterocolitica did neither find any cases . For Yersinia pseudotuberculosis , occurrence of ReA has been 3% in serotype 0:1a and 12–21% in serotype 0:3 .
There are nearly 20 outbreak studies dealing with S. enterica . The first dozen up to the year 2000 are reviewed elsewhere . According to the results, the frequency of ReA varies greatly among these studies (1.8–15%). Moreover, the frequency of human leucocyte antigen (HLA-B27), although differing between studies, is usually considerably lower than in hospital-based series. Since 2000, outbreak studies have been published from USA , Germany , Denmark , Finland , Canada , Australia and Spain . These studies are reviewed in Table 4 . As in the previous studies, the incidence of ReA had a wide variety, depending greatly on the study methodology and the diagnostic criteria of ReA applied. Interestingly, no cases of ReA were observed in the German study dealing with children up to 9 years . However, both adults and children developed ReA, although a lesser proportion in the latter group (12% vs. 24%), in an Australian study, in which joint pain, not synovitis, was required for the diagnosis of ReA . As in the previous outbreak studies , the frequency of HLA-B27 was considerably lower than in hospital series.
The other approach to estimate incidence of ReA is to conduct population-based studies on the annual incidence of ReA ( Table 5 ). In these studies, the total incidence has been estimated to be 0.6–27/100 000. The two recent studies come from United States and from Czech Republic and reviewed here in brief.
| Country | Year | Incidence per 100 000 | Reference | ||
|---|---|---|---|---|---|
| Enteric | Urogenital | Total | |||
| Finland | 1978 | 14 | 13 | 27 | Isomaki et al. |
| Norway | 1994 | 5 | 5 | 10 | Kvien et al. |
| Finland | 2000 | 7 | 3 | 10 | Savolainen et al. |
| Sweden | 2002 | 18 | 1 | 19 | Söderlin et al. |
| USA | 2008 | 0.6–3.1 | NA | NA | Townes et al. |
| Czech Republic | 2010 | 6 | 3 | ∼9 | Hanova et al. |
In the US study, telephone interviews were performed of individuals with Campylobacter , E. coli O157, Salmonella , Shigella and Yersinia infections confirmed by positive culture reported to FoodNet . A case of possible ReA was defined as a subject with new onset of rheumatological symptoms. Subjects with possible ReA were invited to complete a more detailed medical history and physical examination. A case of definite ReA was defined as a subject with possible ReA and a history or physical examination findings consistent with monoarthritis, oligoarthritis, dactylitis, enthesitis or inflammatory back pain without other rheumatological explanation. A total of 6379 culture-confirmed infections were reported to FoodNet during the study. The estimated incidence of ReA was 0.6–3.1 cases/100 000. Campylobacter and Salmonella infections were the most important identified triggers of ReA. No association between risk of acquiring ReA and HLA-B27 positivity was found.
In the study from the Czech Republic, the patients with suspect symptoms were referred by their general physicians and other physicians and relevant hospital departments to the rheumatologists for a clinical investigation, including laboratory tests and radiographs . The diagnostic criteria proposed at the Third International Workshop on Reactive Arthritis in 1995 were used to diagnose ReA . Seventeen patients were identified as incident cases, which gave the age-standardised incidence of 9.3/100 000. Eleven patients had post-enteric and six genitourinary ReA. HLA-B27 was not investigated.
Clinical picture
There is usually an interval of 1–2 weeks from the start of infection to the onset of musculoskeletal symptoms; in Chlamydia infection, the interval can extend up to 4 weeks. The triggering infection can also be asymptomatic, which can hamper the diagnosis. The ReA patients are usually young adults, with a mean age of nearly 30–40 years. The disease is uncommon in children . Male and female patients have similar risk for the development of ReA induced by gastrointestinal infection, while ReA triggered by C. trachomatis is more frequently diagnosed in male patients.
The arthritis is typically characterised by asymmetrical oligoarthritis, often in large joints of the lower extremities. Patients can also have arthritis in the upper extremities. A mild polyarticular form, particularly in the small joints, can also occur. Patients can have dactylitis (sausage digit).
Typical extra-articular manifestation consists of enthesitis, tendinitis and bursitis. Importantly, these manifestations can occur not only in association with arthritis but also as the only reactive complication . Inflammatory low back pain, which is typically worse during night and radiates to buttocks, is not rare.
Other extra-articular features include eye disease, where conjunctivitis is most prevalent, followed by acute anterior uveitis, and skin changes, such as erythema nodosum, keratodermia blenorrhagica (a pustular skin lesion usually observed in the soles of the feet) and circinate balanitis. Erythema nodosum is usually associated with Yersinia infection, whereas circinate balanitis is most frequently associated with Chlamydia arthritis.
There are no specific laboratory tests for the diagnosis of ReA. Erythrocyte sedimentation rate and C-reactive protein are usually elevated. Synovial fluid leucocyte count is between 2000 and 64 000 mm −3 . Benign changes in electrocardiogram (ECG) can be observed indicating a sub-clinical carditis. Mildly abnormal urine analysis (increased number of leucocytes, haematuria and mild proteinuria) is common during the acute period, but frank glomerulonephritis and decreased kidney function is very rare.
Related posts:
Introduction – Microorganisms and the locomotor system
Chlamydia-induced reactive arthritis: Hidden in plain sight?
Tuberculosis and other infections in the anti-tumour necrosis factor-alpha (anti-TNF-α) era
Septic arthritis
Behçet’s syndrome and micro-organisms
The SAPHO syndrome – Are microbes involved?
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
