Show Summary Details
Page of

Reactive arthritis and enteropathic arthropathy 

Reactive arthritis and enteropathic arthropathy
Chapter:
Reactive arthritis and enteropathic arthropathy
Author(s):

J. S. Hill Gaston

DOI:
10.1093/med/9780199642489.003.0115_update_003

Update:

Update on possible role of ERAP1 in ReA

Updated on 24 May 2018. The previous version of this content can be found here.
Page of

PRINTED FROM OXFORD MEDICINE ONLINE (www.oxfordmedicine.com). © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 18 December 2018

Definitions

The term ‘reactive arthritis’ is often used loosely to cover any form of arthritis linked to preceding infection. This produces a large group of diseases with little in common as far as their clinical features or pathogenesis are concerned, for example, post-viral arthritis, post-streptococcal arthritis, and Lyme disease. It is preferable to reserve ‘reactive arthritis’ (ReA) for the clearly identified clinical syndrome which falls naturally within the spondyloarthropathies on clinical, immunogenetic, and pathogenetic grounds. This syndrome is usually triggered by infection with a relatively small number of organisms and forms an important part of the differential diagnosis of acute oligo- and mono arthritis (Table 115.1). Other forms of arthritis seen in relation to infection are best termed ‘post-infectious’.

Table 115.1 Infections that trigger arthritis

Reactive arthritis

Post-infectious arthritis

Common triggering infections

Postviral arthritis

Salmonella

Parvovirus

Campylobacter

Rubella

Yersinia

Mumps

Shigella

HIV

Chlamydia trachomatis

Hepatitis B and C

Occasional triggering infections

Arboviruses

Chlamydia pneumoniae

Other case reports (e.g. herpesviruses)

Clostridium difficile

Bacterial infections

M. bovis BCG (intravesical)

Streptococci

Mycoplasma (e.g. U. urealyticum)

Rheumatic fever

Rare triggering infections

Post-streptococcal arthritis

Many case reports of infection, particularly of the intestine, including:

Neisseria

Gonococcal arthritis

Giardia

Meningococcal arthritis

Cryptosporidium

Lyme disease (Borrelia)

Chlamydia psittaci

Brucella

Hafnia alvei

Whipple’s disease (Tropheryma whippeli)

Vibrio parahaemolyticus

and many other single case reports

There are several examples of disorders linking the gastrointestinal tract and arthritis, ReA following gastroenteritis being an obvious example, along with the arthritis associated with Crohn’s disease or ulcerative colitis. Both the latter are classified within the spondyloarthropathies, but other forms of arthritis are sometimes included in the ‘enteropathic’ grouping such as Whipple’s disease, arthritis associated with coeliac disease, and other syndromes. This chapter is divided into two sections, the first dealing with ReA, and the second with other forms of enteropathic arthritis.

Reactive arthritis

Historical perspective

An association between urethritis and arthritis has been known for nearly 500 years, though there is a possible mention in Egyptian papyri. Many authors attribute the first description of ReA to van Forest in the 1500s, and a description in Spanish literature by Lopez de Hinojosos in Mexico City in 1578 has been recognized. By the beginning of the 19th century the syndrome was clearly recognized, but inevitably in a prebacteriological era ‘venereal arthritis’ encompassed both post-gonococcal arthritis and true ReA, mainly due to Chlamydia trachomatis. Nevertheless, some of the case descriptions, particularly those which mention prominent eye involvement, very likely represent ReA. Brodie described six cases in 1818 and a survey of early case records in London hospitals shows that the condition was commonly recognized from 1820 onwards, accounting for a significant proportion of hospital admissions.1

A link between gastrointestinal infection and reactive arthritis was noted in the early years of the 20th century and outbreaks of dysentery in the trenches of the First World War led to classical descriptions by Fiessinger and Leroy in 1916, and also by Hans Reiter. Although the term ‘Leroy–Fiessinger–Reiter syndrome’ is sometimes encountered, it is the name of Reiter which has come to be associated with ReA, and the terms ‘Reiter’s syndrome’ and ‘reactive arthritis’ have often been used interchangeably. There are several cogent reasons for bringing this practice to an end and abandoning the term ‘Reiter’s syndrome’. First, Reiter was clearly not the first to describe ReA; disease associated with dysentery was first reported by Vossius in 1904. Secondly, Reiter did not shed any useful light on its pathogenesis, erroneously attributing it to spirochaetal infection. Thirdly, and more practically, the triad of arthritis, conjunctivitis, and urethritis/cervicitis has often been taken to imply that Reiter’s syndrome is the form of ReA secondary to genitourinary infection, missing the point that the urethritis in his original description was also ‘reactive’, following bacillary dysentery. There are no prognostic implications associated with the classical triad and therefore no clinical value in singling out Reiter’s triad as a subsection of reactive arthritis. Lastly, several authors have felt the eponym inappropriate in view of Reiter’s later enthusiasm for the Nazi party.

Case definition

Classification criteria for reactive arthritis present difficulties. Since ReA is one of the spondyloarthropathies, criteria such as those devised by Amor2 or the European Spondyloarthropathy Study Group (ESSG)3 can be applied. More recently the international ankylosing spondylitis assessment group (ASAS) has put forward criteria for both axial and peripheral spondyloarthritis.4,5 The success of these criteria is variable. On the one hand many cases of ReA do not have sufficient features to satisfy the Amor criteria, whilst on the other most cases would meet the ESSG criteria, or either of the ASAS criteria. Hpwever, these criteria are very wide-ranging and do not allow ReA to be distinguished from undifferentiated spondyloarthropathy. Unfortunately all current criteria fail to deal adequately with cases with no preceding symptomatic gastrointestinal or genitourinary infection. In these cases, a link to preceding infection depends on adequate laboratory investigation (see below). While diagnostic and classification criteria continue to be debated, a suggested working definition is presented in Box 115.1. This groups:

  1. 1. patients with classical clinical features (oligoarthritis, predominantly of the lower limbs, and/or inflammatory back pain, with extra-articular signs, including enthesitis) plus proven preceding infection with one of the organisms known to trigger reactive arthritis, irrespective of preceding symptoms

  2. 2. patients with proven preceding infection plus arthritis with no diagnostic features

  3. 3. patients with classical clinical features without proven preceding infection.

This definition avoids separating patients with identical clinical manifestations on the basis of the skills of the microbiological laboratory, or failing to include patients with new inflammatory disease of joints or entheses after proven, but asymptomatic, infection with reactive arthritis-associated organisms. The latter are often seen in follow-up of cohorts of patients from outbreaks of food poisoning. Indeed, in a recent series, the proportion of patients witha clinical diagnosis of ReA in whom a triggering infection could be identified after appropriate investigations was very similar in patients with a preceding symptomatic infection and those with no such history. Practically, it is worth considering the diagnosis of ReA in any patient presenting for the first time with acute oligo- or monoarticular synovitis, in whom no other diagnosis (sepsis, crystal arthropathies, etc.) can be made. This allows adequate history, physical examination, and diagnostic tests to be performed, so that a definite diagnosis of ReA can be made.

Epidemiology

Given the difficulties in case definition, accurate epidemiological studies on incidence and prevalence present problems. There are two possible approaches. In the first, the proportion of patients with ReA can be measured in clinics designed to examine all cases of early synovitis. One such study in Oslo measured an incidence of 9.6 per 100 000, equally divided between cases due to Chlamydia trachomatis and enteric infection.6 Previous estimates of chlamydia-induced arthritis of 5 per 100 000 have also been recorded.7 The difficulty with this approach is that it cannot take account of mild cases not referred to early synovitis clinics, cases where arthritis is short lived, or cases with inflammatory back pain or enthesitis in the absence of synovitis. The alternative approach has been to take advantage of food poisoning outbreaks, when an entire population infected with an organism at a particular time can be followed up prospectively. Patients are sent questionnaires on the occurrence of arthritic symptoms subsequent to their exposure to the infection, and those reporting symptoms are examined by a rheumatologist. Although the incidence of ReA in food poisoning outbreaks varies substantially, several series report incidences ranging from 5% to 25%, with an additional percentage of patients developing arthritic symptoms but not meeting criteria for ReA. These figures are interesting in relation to a comprehensive United Kingdom study on the community incidence of enteric infections in a population of 460 000. This showed a combined incidence of campylobacter, salmonella, and yersinia infection confirmed by stool culture of 17.7 per 1000. Interestingly, particularly for yersinia, the number of cases identified in the community survey was substantially higher than those in the same population who presented to their family physician. This means that many of the cases of yersinia infection were too mild to require medical attention. Combining these results would give an estimate of ReA in the community of 100–200 per 100 000, not including cases induced by chlamydia. Although a large proportion of these cases may have mild self-limiting disease, the figures suggest that the incidence calculated on the basis of patients seen at early synovitis clinics is a serious underestimate.

The incidence of ReA may be declining; this decline seems particularly marked for chlamydia-associated arthritis8. This is not due to any decline in the overall incidence of chlamydia infection in Western countries—quite the opposite, with the incidence in the United Kingdom doubling through the 1990s to 50 000 cases per year, with a further doubling over the next decade to 123 000 in 2008. This suggests that other factors must be affecting the incidence of chlamydia-induced arthritis. These could include differences in the organism—whole-genome sequencing of chlamydiae is now able to identify different ‘subspecies’ which do not necessarily correspond to the serovar classification, so that ‘arthritogenic’ chlamydiae might be identified. Other changes may involve the host; chlamydia infection is now commonly acquired in early teenage years. At this age few subjects will have been previously infected with related organisms, such as Chlamydia pneumoniae, an infection whose overall incidence has declined with improved socio-economic conditions. In the 20th century many subjects acquiring C. trachomatis in their third decade would previously have experienced C. pneumoniae infection in childhood or adolescence. It has been shown that such infection has the potential to prime the immune system for more vigorous responses to C. trachomatis,9 which would be relevant to the occurrence of arthritis.

Infections with salmonella and with campylobacter increased substantially in the 1990s in the United Kingdom, without a notable increase in the incidence of ReA, and then have declined in recent years as public health measures have been implemented. Again the lack of a relationship between incidence of infection and ReA is striking, and suggests that changes in immunological memory in those who now become infected has somehow altered their susceptibility to arthritis.

Pathogenesis

Since ReA is a member of the spondyloarthropathy group, pathogenesis of these conditions can be considered together, and is discussed in detail elsewhere (see Chapters 113 and 114). However, its clear association with infection by known bacteria, and its usual acute presentation, means that ReA lends itself to investigation, and much work has been done in the last 20 years. From this three factors have emerged as critical to pathogenesis: the infecting organism; the immune response elicited by infection, particularly that mounted by T lymphocytes; and genetic influences, principally HLA B27. Interestingly, all of these are important in rat HLA B27 transgenic models of spondyloarthropathy;10 in this model arthritis requires infection (gut flora rather than specific pathogens), T lymphocytes (especially CD4+ T cells), and an appropriate genetic background in addition to transgenic B27, since disease incidence varies in different inbred rats. Until recently there has been no good animal model of ReA involving ReA-associated bacteria, but three have now been reported: Chlamydia muridarum induces arthritis in a strain of mouse with a T cell receptor signalling impairment (SKG),11 Salmonella enteritidis induces synovitis in streptomycin-treated mice,12 and Yersinia induces arthritis in mice lacking the p55 TNF receptor.13 The SKG mouse develops arthritis, spondylitis, psoriaform skin lesions, and conjunctivitis, and dissemination of chlamydia antigen to peripheral lymph nodes and spleen has been shown. The latter may be due to inadequate control of the organism locally - decreased T-cell production of IL-17 and interferon-γ‎ were noted - together with increased production of TNF in response to the disseminated organisms. In contrast, lack of a TNF response in the Yersinia model, and hence control of infection, seemed to lead to enhanced IL-17 and interferon-γ‎ responses. The critical factor in each model may be the appropriate cytokine response to control initial infection and prevent dissemination of organisms; there is some evidence that macrophages from recovered ReA patients make inadequate TNF and IL-23 responses to bacterial stimuli.14

The list of bacteria which commonly cause ReA is quite short (Table 115.1). Another set of organisms has been implicated in ReA on multiple occasions, but their contribution to ReA incidence is relatively small. Lastly, there are many single case reports implicating particular infections. In these cases, there is always uncertainty about whether the organism provoked true ReA or simply a post-infectious arthritis, but where classical extra-articular features are noted (as distinct from only positivity for HLA B27), the reports are convincing. Very similar organisms, e.g. Shigella flexnerii and Shigella sonnei, have previously been associated with very different rates of ReA, with S. sonnei infection rarely being reported as a cause of ReA in series from the United States; but, interestingly, a recent survey of shigella-infected patients in Finland found a majority of the arthritis cases to be triggered by S. sonnei.15 Thus differences in the ability of an organism to trigger ReA may not be consistent in all populations, and investigations focused on the differences between such organisms to identify arthritogenic factors may not be informative.

Contrary to previous expectations, there is now substantial evidence that bacteria or their antigens reach the affected joints in ReA. Bacterial antigens were first demonstrated in phagocytic cells within the joint using immunofluorescence and immunoblotting techniques.16,17 This has been shown for salmonella, yersinia, and shigella, and since phagocytic cells in peripheral blood were also shown to contain antigen,18 it seems likely that there is traffic from the site of infection to the affected joint. Most studies to demonstrate nucleic acids of enteric bacteria in ReA joints have been negative,19 but by using the more sensitive technique of reverse transcription polymerase chain reaction (RT-PCR), ribosomal RNA from Yersinia pseudotuberculosis was unequivocally demonstrated in a ReA joint.20 The demonstration of bacterial RNA is particularly important since, unlike DNA, this has a relatively short half-life and its presence implies that transcriptionally active organisms can reach the joint, although ribosomal RNA has a longer half-life than that of messenger RNA. Evidence that intact C. trachomatis reaches the joint is much stronger, with detection by both PCR and RT-PCR21,22 and the demonstration of organisms (albeit with atypical morphology) by electron microscopy or immunofluorescence.23 Interestingly, in all these studies evidence for organisms or their antigens in the joint has been obtained long after the initial infection, strongly suggesting that these organisms can persist at low levels, perhaps in sites such as lymphoid tissue associated with the gut or genital tract. This persistence may relate to their ability to survive intracellularly. Such persistence has been demonstrated, even after ‘curative’ antibiotics, but dormant intracellular organisms that divide infrequently may be insensitive to conventional antibiotics—this reasoning has led to new attempts to treat chlamydia-associated ReA with novel combinations of antibiotics (see below).

All forms of spondyloarthropathy are associated with enthesitis (inflammation of ligamentous and tendinous insertions), and it has been postulated that this is the first site of inflammation,24,25 perhaps due to a resident population of IL-23 responsive T cells.26 Nevertheless, experimental work in ReA has been confined to joints, which have a substantial T-cell infiltrate, including both CD4+ and CD8+ T cells, some of which may have trafficked from the gut or genitourinary tract. Traffic may be more abundant to the inflamed synovium, as compared to the normal paucicellular synovial membrane, and the susceptibility of weight-bearing joints in ReA might reflect pre-existing microtrauma and hence increased monocyte and lymphocyte traffic to such joints, delivering both antigen and antigen-specific T cells. Prominent responses to arthritis-triggering bacteria have been demonstrated in both CD4+ and CD8+ populations, most work having been done on CD4+ T cells.27,28 Organism-specific clones have been obtained and their specificities identified (e.g. Goodall et al.,29 Mertz et al.30). In several studies bacterial hsp60 has emerged as a major target antigen,31,32,33 although this is a common feature of T-cell-mediated responses to intracellular bacteria. Hsp60 is a good candidate for ‘molecular mimicry’, since there are large numbers of conserved peptides in human and bacterial hsp60, so that T cells stimulated by bacterial hsp60 might conceivably cross-react with human hsp60 and generate autoimmunity. This was not found when analysed with T-cell clones,34 though the possibility was raised in a non-clonal analysis.33 Generally evidence for molecular mimicry is lacking and the significance of CD4+ T-cell responses to bacterial hsp60 for ReA pathogenesis is not known. Likewise the attractive idea that bacteria-specific CD8+ T cells restricted by HLA B27 would make an autoreactive response to a joint-specific antigen has not received experimental support, although yersinia- and chlamydia-reactive, B27-restricted T-cell clones have been isolated from ReA joints.35,36 Analysis of T-cell-receptor expression has shown expansions of multiple clones in both CD4+ and CD8+ subsets of blood and joint,37 with CD8+ expansions commonest in the joint. It may be that both subsets are involved in disease. In the ReA which frequently occurs in HIV-infected individuals in sub-Saharan Africa, the arthritis is associated with modest diminution in CD4+ T-cell counts rather than profound levels of CD4+ T-cell depletion,38 suggesting that CD4+ T cells are required for arthritis—an observation in agreement with the B27 transgenic model.39 The cytokines produced by synovial bacteria-specific T cells are generally proinflammatory, particularly interferon-gamma (IFNγ‎) and/or interleukin (IL)-17 produced by the Th1 or Th17 subsets of CD4+ T cells40,41,42; increased quantities of IL-17 have also been noted in ReA synovial fluid.43

HLA B27 is associated with the occurrence of ReA,44,45 but strikingly with severe disease.46 Patients with prolonged course, recurrent attacks, or evolution to chronic arthropathy are very likely to be B27+.47 However, B27 is not necessary for ReA to develop, and ascertaining B27 status is not useful diagnostically, although it may help predict prognosis. How B27 acts remains frustratingly unclear. It is assumed, but not proven, that the role of HLA B27 is identical in all spondyloarthropathies. There are currently three main theories proposed for the role of HLA B27 in spondyloarthritis:48 first, that it involves its classical function in presenting antigenic peptides to T cells, and that there is an ‘arthritogenic’ peptide which binds to B27: secondly, that HLA B27 is able to form heavy chain dimers which are surface expressed, interact with receptors (KIR and LILR) on lymphocytes and antigen-presenting cells, altering their function;49 thirdly, that inefficient B27 heavy chain folding in the endoplasmic reticulum generates cellular stress which alters cytokine production, favouring proinflammatory responses, particularly those involving IL-23 and differentiation of IL-17-producing cells.50,51 Recent genome-wide association studies (GWAS) have revealed a large number of genes associated with ankylosing spondylitis (AS).52 Several involve responses to IL-23 (especially the IL-23 receptor) and genes expressed by Th17 cells. ERAP-1 is an enzyme whose function is to trim antigenic peptides to the correct length for fitting into the groove of class I MHC molecules such as HLA B27. Importantly, an association between ERAP1 polymorphisms and AS is confined to HLA B27+ patients.53 Whether there is an association between ERAP-1 and ReA remains to be determined. Its association with AS points to the importance of the antigenic peptides which are both presented by HLA B27 and form an important part of its molecular structure. One of the polymorphisms which confers a lower risk of AS is thought to be a less efficient trimmer of peptides.54 In contrast, it was reported that AS patients expressed ERAP1 alleles (combinations of polymorphisms producing amino acid changes) which are either relatively inefficient trimmers of antigenic peptides, or in some cases “over-efficient” trimmers, which could destroy antigenic peptides.55 A more recent study has not replicated these findings, and it seems more likely that the ERAP1 alleles which confer a lower risk of SpA are those which are hypofunctional in terms of peptide trimming.56 Whether ERAP1 polymorphisms play any role in the pathogenesis of ReA remians to be determined In principle, a failure to trim bacterial peptides appropriately might decrease the initial CD8 T cell response to ReA-associated organisms, or alter its proinflammatory character, so as to make ReA less likely.Bacterial products are initially detected by the innate immune system, particularly the Toll-like receptors (TLR) expressed on phagocytic cells, and a polymorphism in TLR2 (R753Q) has been reported to be associated with acute salmonella-induced ReA, with 6/48 patients having this rare variant as compared to 0/27 patients infected by salmonella without developing arthritis and 2/91 healthy controls.57 This finding needs confirmation in larger studies, but in principle polymorphisms in TLRs, or other receptors for bacteria products, could also result in altered cytokine production, alter the immune response to the organism, and thus play an important role in susceptibility to ReA.

Clinical features

Preceding illness

A history of urethritis (dysuria or discharge) or diarrhoea must be specifically sought for several reasons. The interval between these symptoms and the development of arthritis means that patients may not connect these apparently unrelated events. Moreover, preceding infection may be virtually asymptomatic—chlamydia infection in women is notoriously silent, and in men these symptoms or a sexual history are often not volunteered spontaneously. Recent reports have noted ReA secondary to rectal infection with lymphogranuloma venereum serovar of Chlamydia trachomatis in men who have sex with men.58 Of gastrointestinal infections, salmonella and shigella are likely to produce symptoms in those who develop ReA, whereas in yersinia-related arthritis many patients have subclinical or mild gastrointestinal symptoms.

Arthritis

ReA is usually an asymmetric oligoarthritis, generally involving fewer than six joints, with a tendency to affect the lower limbs preferentially. Some patients may have arthralgias at sites in addition to those affected by synovitis. Community surveys of patients infected by ReA-associated organisms also reveal patients with polyarticular arthralgia, but these tend not to present to rheumatologists since the symptoms are mild and resolve spontaneously. In rheumatology practice any joint can be affected, and a proportion of patients have monarthritis. Affected joints generally become rapidly hot and swollen, and large effusions can develop especially in the knee. The evolution of joint involvement is such as to make septic arthritis or crystal-induced arthritis likely differential diagnoses. There is a real possibility of patients being treated for a culture-negative septic arthritis if a full history is not obtained from the patient or careful examination made for extra-articular signs of ReA. The arthritis can have features seen in other forms of spondyloarthropathy. Dactylitis, resembling that seen in psoriatic arthritis, occurs in ReA, and many patients experience inflammatory low back pain, or buttock pain due to acute sacroiliitis. The arthritis is generally most severe in the first few weeks following its onset, followed by substantial improvement over the next few weeks. However, mild but significant symptoms often persist 6–12 months before full resolution. Patients need reassurance during this phase that complete resolution is still probable. In addition, not all patients experience a ‘monophasic’ disease and both exacerbations and involvement of new joints can occur, even in those in whom the disease eventually settles completely.

Extra-articular disease

One of the characteristic features of spondyloarthropathies is the presence of enthesitis, and its presence is often helpful in making a diagnosis of ReA, with the Achilles tendon insertion and plantar fascia the commonest sites of involvement. Enthesitis may be mild, particularly in relation to active synovitis at other sites, and so the patient may not report symptoms, particularly if not weightbearing. Other ligamentous insertions involving the pelvis and chest wall may be symptomatic, and some low back symptoms may represent enthesitis.

Conjunctivitis is a classical feature of ReA but is usually painless and often transitory. It may not be evident when the patient presents with arthritis, and a history should be sought from the patient (or relatives who may have noticed red eyes). Persistent eye inflammation and painful eyes raise the question of acute anterior uveitis, but this is much less common in ReA. It requires full ophthalmological assessment.

Skin and mucous membranes can be involved. Keratoderma blennorhagica is histologically identical to psoriasis, and most commonly seen on the soles of the feet, although it can also involve the hands and trunk. Its site, and the fact that it is painless, means that again it needs to be sought directly, since patients may not have noticed it. Balanitis is also asymptomatic and may not be readily apparent, particularly in the uncircumcised. The occurrence of balanitis, like urethritis, does not imply a genitourinary aetiology for the ReA. Ulcerative lesions in the mouth and soft palate are also seen but are usually asymptomatic. Lastly, erythema nodosum has been noted in ReA associated with yersinia.

In general, extra-articular disease is associated with severity of arthritis, a less favourable prognosis, and HLA B27 positivity. The most severe extra-articular manifestations are aortitis and cardiac conduction disorders but fortunately these are rare. Table 115.2 shows the frequency of joint and extra-articular involvement in a number of series of ReA patients.59

Table 115.2 Frequency (%) of extra-articular signs in reactive arthritisa

Lesion

Enteric infection

Sexually acquired infection

Keratoderma and psoriaform lesions

1–2

10–30

Circinate balanitis

5–25; higher in shigella

40– 70

Conjunctivitis

6–20; higher in shigella

25–40

Oral ulcers

2–5

10–30

Erythema nodosum

7 (mainly yersinia)

0

Reactive urethritis

10–30

N/A

a Data from Angulo and Espinoza52 combined with other published series.

Differential diagnoses

Differential diagnoses are listed in Box 115.2. In patients who present with an acute arthritis, the principal differential diagnoses are septic arthritis and crystal arthropathies. Other forms of post-infectious arthritis enter the differential, particularly post-gonococcal or post-meningococcal arthritis, along with arthritis due to streptococci and Lyme disease. Although streptococcal infection rarely gives rise to rheumatic fever, except under conditions of social deprivation, an inflammatory oligoarthritis strongly resembling ReA is seen,60 but classical extra-articular features are not present. In subacute disease, it is sometimes difficult to distinguish ReA from other forms of spondyloarthropathy, particularly if there is no history of preceding infection and no laboratory evidence to implicate specific infections. The term ‘undifferentiated spondyloarthropathy’ is rightly applied to these cases, with the possibility that psoriasis or inflammatory bowel disease (IBD) will eventually declare themselves.

Laboratory investigations

Investigations are directed towards excluding the major differential diagnoses, and thereafter to establishing the organism responsible for triggering the disease.

General investigations

In acute ReA severe enough to present to hospital, there is usually a major acute-phase response (ESR >100, CRP 100–200 mg/litre) and neutrophilia. Other biochemical investigations are not generally helpful, although serum urate should be checked. Classical autoantibodies such as rheumatoid factor (RF) and anti-nuclear antibodies (ANA) are absent, and although positive anti-neutrophil cytoplasmic antibodies have been described, they are not diagnostically useful. The antibodies are not directed against proteinase-3 or myeloperoxidase, unlike those seen in granulomatosis with polyangiitis (Wegener’s) and microscopic polyangiitis.

Microbiology

Tests to exclude septic arthritis or other forms of post-infectious arthritis

Microscopy, Gram stain, and culture of synovial fluid is the single most important investigation to diagnose septic arthritis, together with blood cultures and culture of any other possible site of infection, including throat swab for streptococci. Application of PCR techniques may establish the triggering agent in ReA (see below), but may also be helpful in the diagnosis of mycobacterial infection, borreliosis, and Whipple’s disease. Antibody responses to streptococcal antigens, including both anti-streptolysin and anti-DNaseB, should be sought, along with antibodies to borrelia in patients who have been in endemic areas. Viral antibodies may also be checked, particularly IgM antibodies to parvovirus, Epstein–Barr virus (EBV), and cytomegalovirus (CMV). Note that IgG antibodies will very often be present since infection with these viruses is endemic, and so only IgM antibodies are diagnostically useful. In appropriate geographic regions antibodies to Ross river virus or chikungunya can also be measured.

Identification of ReA-associated organisms

An excellent account of microbiological tests useful in the diagnosis of ReA is available.61 By definition, ReA and associated bacteria are not cultured from affected joints, but the organisms may be identified by stool cultures in enteric arthritis, and cultures of swabs from the genitourinary tract or urine in chlamydia infection. Nucleic acid amplification techniques on urine are now favoured for diagnosing C. trachomatis infection and avoid invasive swabs while maintaining high sensitivity, though rectal swabs should be obtained if indicated. The use of PCR should allow a higher proportion of patients with chlamydia-induced ReA to be positively diagnosed. C. pneumoniae can also be diagnosed by PCR on sputum. There has been interest in using PCR diagnostically in chlamydia-induced ReA to identify organisms in infected joints, but thus far reliable techniques have not emerged. Although some laboratories have reported that a high proportion of chlamydia-induced ReA cases are positive for the organism by PCR, the test can never establish 100% specificity. This is because, in populations in whom chlamydia infection is frequent, chlamydiae may well traffic to joints of those affected by e.g. rheumatoid disease in the same way that they do in ReA.

Serological techniques can also be used to establish preceding infection. Current chlamydia serology is unsatisfactory since there are difficulties in distinguishing antibodies to C. trachomatis and C. pneumoniae, and the incidence of antibodies to the latter can be very high in the general population. Specialized techniques measuring antibody bound to purified organisms, and comparing the titre of antibodies to C. trachomatis and C. pneumoniae, can establish specificity, but are not used routinely. There are also problems with the serology of enteric infections, mainly because certain populations have previously encountered salmonella and campylobacter in foods such as chicken and eggs, and developed antibodies. In acute disease, measurement of salmonella-specific IgM (the Widal test), or IgM responses to other enteric organisms, can be useful but has low sensitivity. Since high titres of IgG antibodies to organisms such as salmonella or campylobacter are not uncommon in the general population, it is preferable to demonstrate a rising titre of IgG antibodies or high levels of persistent IgA antibodies to the organism of interest. IgA has a shorter half-life than IgG, so persistence of high titres of specific IgA antibodies is often taken to indicate persistent infection.

ReA-triggering organisms also elicit prominent T-cell-mediated responses and the possibility of these being useful diagnostically has been considered. The responses are most easily detectable in synovial fluid, but because they only demonstrate T-cell memory for the organism, they cannot provide more than supporting evidence that a particular episode of arthritis is due to the organism recognized by synovial fluid T cells. They may have some negative diagnostic value—for instance, it would be very unusual for a patient to have chlamydia-triggered ReA without a T-cell response to chlamydia being detectable in the joint. However, as noted in discussion of PCR tests, patients with rheumatoid arthritis (RA) can incidentally become infected with ReA-associated organisms and will then have organism-specific T cells detectable in their affected joints.

In summary, for routine investigation of ReA, stool should be cultured, chlamydia sought by PCR, and antibodies to yersinia and camplyobacter measured. Other tests require further study for their validation.

Tissue typing

HLA B27 testing has no diagnostic value since ReA often occurs in B27-negative patients. However, it is still worth doing to help establish prognosis, since this is less favourable in B27+ patients. HLA B27+ patients may warrant earlier treatment with disease-modifying drugs (see below) or closer follow-up to detect relapse or chronic spondyloarthropathy.

Radiology

Radiological examination of acutely affected joints is not helpful in the diagnosis of ReA, but may aid in establishing differential diagnoses, for example, by showing chondrocalcinosis. Chest radiograph may reveal hilar adenopathy in sarcoidosis, or occasionally in yersinia infection which can cause a sarcoid-like illness. Although sacroiliac joint involvement is common acutely in ReA, radiographic abnormality is not. MRI is a more sensitive indicator of enthesitis and/or synovitis in the sacroiliac joint. Ultrasonography can also demonstrate enthesitis and distinguish oedema of tendon and ligament insertions from bursitis. Bone isotope scintigraphy may also be useful to demonstrate joint inflammation, especially in the sacroiliac joints, and also enthesopathy, but has mainly been superseded by MRI. In patients in whom disease persists, radiological changes can develop, with erosion of affected joints, including the sacroiliac joints, and the formation of new bone, which can be seen as either periostitis in the hands and feet or spur formation at entheses such as the plantar ligament insertion. Erosions may also be present at these same sites. In chronic ReA, paravertebral ossification may be seen in the lumbar spine; unlike ankylosing spondylitis this is often asymmetric. Atlantoaxial subluxation has also been described in chronic ReA.

Treatment

Symptomatic treatment

Patients require conventional measures to treat acutely inflamed joints, including non-steroidal anti-inflammatory drugs (NSAIDs), joint aspiration with injection of depot steroid preparations (when septic arthritis has been excluded), together with analgesia and physiotherapy to maintain range of motion and regain the muscle power and bulk which is lost during acute inflammation. In severe disease, systemic corticosteroids may be used but injection of affected joints is preferable. Extra-articular disease does not call for specific treatment, with the exception of uveitis which requires topical steroids. Of the triggering infections, only chlamydia infection generally requires specific treatment in its own right, with conventional short-term antibiotic regimes using tetracyclines or azithromycin.

Specific treatment

Current opinion on the pathogenesis of ReA would suggest two specific treatment strategies. If disease is maintained by persistent infection and trafficking of triggering organisms or their products to affected joints and entheses, long-term antibiotics should hasten resolution of the disease. On the other hand, if disease is mainly maintained by an aberrant immune response, triggered by infection but later directed against components of the joint, immunosuppressive treatment would be indicated. Unfortunately, evidence from therapeutic trials does not yet allow us to distinguish definitively between these hypotheses.

There have been several trials of prolonged doses of antibiotics in ReA. When tetracycline was used to treat episodes of non-gonococcal urethritis very promptly, there was a decrease in the number of episodes of ReA, but in this instance treatment was administered prior to the onset of arthritis.62 The first trial to treat established ReA using lymecycline did not show efficacy overall, but a subgroup of patients with chlamydia-induced arthritis had a decreased disease duration.63 Patients with chronic yersinia infection, diagnosed by showing organisms in intestinal mucosa or gut lymphoid tissue, and with wider symptomatology than ReA, were reported to respond to ciprofloxacin in an uncontrolled trial. These findings prompted controlled trials but those in enteric ReA have shown no therapeutic advantage of 3 months’ treatment with ciprofloxacin,64,65,66 nor did similar trials of azithromycin, which has activity against both enteric organisms and chlamydia, or of roxithromycin plus olfloxacin.67,68 Likewise 4 months’ doxycycline was no more effective than 10 days’ in chlamydia-induced disease,69 This does not necessarily imply that disease is not maintained by persistent infection, since organisms may be relatively resistant to ciprofloxacin or azithromycin when not actively dividing. This argument led to a recent trial aiming to target both bacterial gene transcription and protein synthesis using combinations of antibiotics—rifampicin and azithromycin/doxycycline for 6 months. The patients chosen had longstanding disease (>10 years) and had RT-PCR evidence of persistent chlamydia infection (C. trachomatis or C. pneumoniae). In a double blind placebo-controlled trial impressive improvements were noted at 6 months.70 This intriguing result in a relatively small trial requires has not yet been reproduced.

Disease-modifying agents used in rheumatoid disease have also been tested in ReA, although in many trials ReA patients have been included along with patients with other forms of spondyloarthropathy.71 One difficulty in these trials is that the rate of spontaneous remission in ReA is high, so that large numbers of patients would need to be tested to have adequate power to show a therapeutic effect. If extrapolation from other forms of spondyloarthropathy is justified, both sulfasalazine and methotrexate should be useful, but the effect of sulfasalazine is relatively modest.72 It is reasonable to withhold disease-modifying drugs for the first 3 months in ReA and reserve their use for those whose disease is not settling (particularly if they are B27+), or who have involvement of new joints. Therapies directed against TNFα‎, having proved useful in both AS and psoriatic arthritis, have been shown to be effective in patients with intractable chronic ReA,73 with no evidence of recrudescence of infection by ‘latent’ triggering organisms. There is little experience of other biologics in the treatment of ReA; again extrapolation from AS and psoriatic arthritis suggests that antibodies targeting the IL-23/IL-17 pathway might be effective.

Prognosis

Many patients with ReA are young, otherwise healthy, and used to an active lifestyle. To these individuals ReA, even when self-limiting, is a major life event, particularly since cases seen in hospital practice commonly take months to resolve. Patients therefore need considerable reassurance that there is an 80% chance of complete resolution of their symptoms within the first year, with a further 10% settling in the following year. Prognosis is less favourable in those who have disease severe enough to require hospitalization and who are B27+. Progression to chronic disease in 16% was recorded in one series, with recurrent attacks of reactive disease in a further 22%,74 and some eventually have disease which resembles AS.75 The tendency to trivialize the disease because its prognosis is generally so much better than that of RA needs to be resisted. Although many ReA patients may have future episodes of gastroenteritis or urethritis without recurrence of arthritis, it is advisable for patients to minimize their risk of recurrent infection, particularly younger patients who may need advice on foreign travel and barrier contraception.

Enteropathic arthropathy

Although ReA due to enteric infection falls within the definition of enteropathic arthritis, this section discusses forms of arthritis other than ReA which are associated with gut inflammation. These include others within the spondyloarthropathy group, along with the arthritis associated with coeliac disease, and Whipple’s disease.

Spondyloarthropathies associated with gut inflammation

Crohn’s disease and ulcerative colitis

The association between IBD and arthropathy has long been recognized.76 Although estimates of the incidence of arthritis vary, several large series from different geographical areas suggest approximately 10% of IBD patients have peripheral arthritis, with a somewhat higher incidence in Crohn’s disease as compared to ulcerative colitis. An additional proportion of patients (~5%) have axial disease indistinguishable from AS, while asymptomatic sacroiliitis is even commoner (15–20%). Arthritis in IBD is associated with other extragastrointestinal features, and is present in 30–60% of patients with uveitis or skin lesions such as erythema nodosum and pyoderma gangrenosum. Arthritis is also commonly accompanied by enthesopathy, as in other spondyloarthropathies. The idea of linkage between spondyloarthopathies and IBD is given greater credence by ileocolonoscopy studies which have shown that up to 60% of patients with ankylosing spondylitis or undifferentiated spondyloarthopathy have subclinical inflammatory lesions in the gut;77,78 a proportion of these patients progress to overt clinical IBD. The situation has been further clarified by subclassification of the peripheral arthritis of IBD into patients with fewer than five involved joints (type I) and those with five or more involved joints (type II).79 Type I patients mainly have asymmetric involvement of predominantly lower limb joints (often a monoarthritis, particularly at onset), whereas type II patients have symmetric polyarticular small joint involvement in a rheumatoid-like distribution. Further analysis suggests that only type I patients can be regarded as falling within the spondyloarthopathy group. Like ReA, 80% with type I arthritis have self-limiting disease and active arthritis is associated with relapse of IBD in 80%. In contrast, 80% of type II patients have persistent disease and a much less marked association between active arthritis and IBD relapse. These clinical observations are underpinned by the associations with the relatively rare MHC class II allele, HLA DRB*0103, seen in both type I arthritis (relative risk 12) and ReA due to enteric infection. The same allele has also been found to be increased in IBD patients with AS, or with IBD and uveitis. An association with B27 was also described, but this was weaker for type I disease than for ReA. In contrast, type II disease showed no association with either B27 or DRB*0103, but instead a weak association with HLA B44. The pathogenesis of the peripheral arthritis in IBD likely has many features in common with ReA. Whereas specific pathogens are required to provoke the gastrointestinal inflammation which produces ReA in the presence of susceptibility genes (HLA B27 and others), the same genes predispose to arthritis in response to the gastrointestinal inflammation of IBD. IBD is considered to be an aberrant response to normal gut bacteria—loss of tolerance of the host microbiome.80 As in ReA, further genetic influences, again including B27, may lead to chronicity with established sacroiliitis and axial disease, rather than self-limiting peripheral arthritis. Unlike classical AS, HLA B27 is only found in 50% of the IBD patients with spondylitis, suggesting that other genetic factors can substitute for B27. In a different genetic background, other patterns of arthritis are seen, including the rheumatoid-like type II arthritis.

Treatment is directed at the underlying IBD, and for type I peripheral arthritis effective control of IBD allows resolution of arthritis. This is not the case for type II arthritis or for spondylitis, and these require treatment in their own right. NSAIDs and methotrexate, which are generally appropriate in the management of inflammatory arthritis, can both exacerbate IBD, although methotrexate is also sometimes used to treat IBD, and COX-2 selective NSAIDs are suggested not to have the same risks as non-selective drugs. In contrast, sulfasalazine, and more recently, anti-TNFα‎ antibodies, have been found to be effective for both IBD and its associated arthritis.

Pouchitis

Patients treated by proctocolectomy for ulcerative colitis commonly have a pouch fashioned by ileo-anal anastomosis to avoid ileostomy. The pouch develops inflammation in 20–40% of patients, possibly due to bacterial overgrowth, and this ‘pouchitis’ responds to antibiotics, particularly metronidazole. A proportion of patients with pouchitis also develop an arthritis which has features of spondyloarthopathy including lower limb involvement, enthesitis, and sacroiliitis.81,82 Other extra-articular features such as iritis and erythema nodosum have been described. The development of pouchitis and associated arthritis is much commoner in patients operated on for ulcerative colitis than in those requiring colectomy for familial polyposis, perhaps indicating a genetic predisposition to make inflammatory responses to gut flora. Some authors have described a close relationship between the severity of pouchitis and arthritis (similar to type I arthritis in IBD) with recovery following pouch removal in some cases. Clinically significant arthritis associated with pouchitis is relatively uncommon, but a larger proportion (30%) of patients complain of arthralgia involving hands and knees; in these cases a relationship with pouch inflammation is not evident.83

Coeliac disease

Inflammatory arthritis complicating coeliac disease has been recognized for two decades,84 but is relatively rare. The disease is generally an oligo- or monoarthritis, with preferential involvement of lower limb joints. Lumbar spine and sacroiliac joint involvement have also been described, but there is no clear association with HLA B27. In fact the genes which increase susceptibility to coeliac disease are those involved in other autoimmune disorders such as RA, type I diabetes, and autoimmune thyroid disease rather than those associated with spondyloarthritis.85 The best evidence that the arthritis is causally related to coeliac disease, and not an incidental finding, is the observation that both gut and joint disease respond to a gluten-free diet. Joint symptoms can precede gastrointestinal symptoms; indeed, coeliac disease can be silent in one-third of cases with arthritis, so investigation of unexplained seronegative arthritis by measuring anti-endomysial antibodies is appropriate, followed by endoscopy and biopsy in antibody-positive patients. This approach is justified since untreated coeliac disease carries an increased risk of lymphoma. In keeping with genetic predisposition to various forms of autoimmunity, an increased incidence of coeliac disease (~6%) among patients with juvenile idiopathic arthritis (JIA) has been reported.86

Whipple’s disease

Understanding of this disease was advanced enormously by the discovery of the causative organism, using the molecular approach of amplifying bacterial ribosomal RNA (rRNA) genes using “universal” primers targeting sequences conserved in all bacteria, followed by sequencing the product to allow particular bacteria to be identified, since rRNA genes also possess species-specific sequences.87 In the case of the Whipple’s bacillus, a previously undescribed unique sequence was identified in the rRNA PCR product, and consistently found in jejunal biopsies from Whipple’s disease patients. The organism, named Tropheryma whipplei, is related to actinomycetes, and has now been cultured, though it grows very slowly. It is not uncommon in the environment, and can be found in normal individuals’ saliva or duodenal secretions (but not normal jejunal biopsies). It does not usually establish infection, suggesting that Whipple’s patients have some impairment of their immune response. Although bacteria have long been implicated in Whipple’s disease, identification of a specific organism has allowed more accurate diagnosis and indicates that there are patients, including some with arthropathy, who do not have the histological changes in the gut previously required for diagnosis.88 PCR has shown the presence of the organism in the joints,89 and culture from synovial fluid has also been achieved.90

Whipple’s disease is a rare (incidence 1:106 systemic disorder, and the commonest triad of symptoms is weight loss, diarrhoea, and arthralgia or arthritis.91 Oligo- or polyarthritis are seen, often with involvement of large joints. Spondyloarthropathy features are generally absent, although a high frequency of HLA B27 has been observed in some series, and uveitis is seen. The arthritis can precede other features by up to 8 years, and PCR diagnosis may allow cases to be identified and treated before they develop systemic symptoms. Diagnosis can also be made by immunohistology using organism-specific antibodies. The diagnosis should be considered in patients with persistent seronegative oligoarthritis. The infection can also involve the central nervous system, heart valves, and lungs; treatment is with long courses of trimethoprim/sulphamethoxazole. Its efficacy can be monitored by determining whether patients remain PCR positive. These have persistent infection which may require a change in antibiotics. Given that some cases of Whipple’s disease will not be recognized, especially in the absence of gut symptoms, such patients may be classified as seronegative RA and eventually receive TNFα‎-blocking drugs when they fail to respond to disease-modifying anti-rheumatic drugs (DMARDs). Since the disease is driven by an intracellular bacterium, inhibition of TNFα‎ is potentially disastrous, and dissemination of the organism with severe endocarditis has been reported in these circumstances.92

References

1. Storey GO, Scott DL. Arthritis associated with venereal disease in nineteenth century London. Clin Rheumatol 1998;17(6):500–504.Find this resource:

2. Amor B, Dougados M, Mijiyawa M. Critères de classification des spondylarthropathies. Rev Rhum Mal Ostéoartic 1990;57:85–89.Find this resource:

3. Dougados M, van der Linden, S, Juhlin R, et al. The European Spondylarthropathy Study Group preliminary criteria for the classification of spondylarthropathy. Arthritis Rheum 1991;34(10):1218–1227.Find this resource:

4. Sieper J, Rudwaleit M, Baraliakos X, et al. The Assessment of SpondyloArthritis international Society (ASAS) handbook: a guide to assess spondyloarthritis. Ann Rheum Dis 2009;68(Suppl. 2):ii1–ii44.Find this resource:

5. Rudwaleit M, van der Heijde, D, Landewe R, et al. The Assessment of SpondyloArthritis international Society classification criteria for peripheral spondyloarthritis and for spondyloarthritis in general. Ann Rheum Dis 2011;70(1):25–31.Find this resource:

6. Kvien TK, Glennas A, Melby Kn, et al. Reactive arthritis: incidence, triggering agents and clinical presentation. J Rheumatol 1994;21(1):115–122.Find this resource:

7. Kvien T, Glennas A, Melby K, Husby G. The natural history of reactive arthritis. Rheumatol Eur 1995;24:15–19.Find this resource:

    8. Mason E, Wray L, Foster R, et al. (2015). Reactive arthritis at the Sydney Sexual Health Centre 1992-2012: declining despite increasing chlamydia diagnoses. Int J STD AIDS 2016 Sep;27(10):882–889.Find this resource:

      9. Telyatnikova N, Gaston JSH. Prior exposure to infection with Chlamydia pneumoniae can influence the T-cell-mediated response to Chlamydia trachomatis. FEMS Immunol Med Microbiol 2006;47(2):190–198.Find this resource:

      10. Tran TM, Dorris ML, Satumtira N, Richardson JA, Taurog JD. Additional human beta-2 microglobulin curbs HLA-B27 heavy chain misfolding and promotes arthritis and spondylitisbut not colitis in male B27 transgenic rats. Arthritis Rheum 2005;52:S447.Find this resource:

        11. Baillet AC, Rehaume LM1, Benham H1, et al. High chlamydia burden promotes tumor necrosis factor-dependent reactive arthritis in SKG MICE. Arthr Rheumatol 2015;67(6):1535–1547.Find this resource:

          12. Noto Llana, M., Sarnacki SH, Aya Castañeda Mdel R, et al. Consumption of Lactobacillus casei fermented milk prevents Salmonella reactive arthritis by modulating IL-23/IL-17 expression. PLoS One 2013;8(12): e82588.Find this resource:

            13. Cargnelutti E, Arias JL, Valdez SR, Rabinovich GA, Di Genaro MS. TNFRp55 controls regulatory T cell responses in Yersinia-induced reactive arthritis. Immunol Cell Biol 2013;91(2):159–166.Find this resource:

              14. Valimaki E, Aittomäki S, Karencko L et al. Normal inflammasome activation and low production of IL-23 by monocyte-derived macrophages from subjects with a history of reactive arthritis. Scand J Rheumatol 2013;42(4):294–298.Find this resource:

                15. Hannu T, Mattila L, Siitonen A, Leirisalo-Repo, M. Reactive arthritis attributable to Shigella infection: a clinical and epidemiological nationwide study. Ann Rheum Dis 2005;64(4):594–598.Find this resource:

                16. Granfors K, Jalkanen S, von Essen, R, et al. Yersinia antigens in synovial fluid cells from patients with reactive arthritis. N Engl J Med 1989;320:216–221.Find this resource:

                17. Granfors K, Jalkanen S, Lindberg A, et al. Salmonella lipopolysaccharide in synovial cells from patients with reactive arthritis. Lancet 1990;335:685–688.Find this resource:

                18. Granfors K. Host-microbe interaction in reactive arthritis: Does HLA- B27 have a direct effect? J Rheumatol 1998;25(9):1659–1661.Find this resource:

                19. Nikkari S, Rantakokko K, Ekman P, et al. Salmonella-triggered reactive arthritis: use of polymerase chain reaction, immunocytochemical staining, and gas chromatography-mass spectrometry in the detection of bacterial components from synovial fluid. Arthritis Rheum 1999;42(1):84–89.Find this resource:

                20. Gaston JSH, Cox C, Granfors K. Clinical and experimental evidence for persistent Yersinia infection in reactive arthritis. Arthritis Rheum 1999;42(10):2239–2242.Find this resource:

                21. Taylor-Robinson, D, Gilroy C, Thomas B, Keat A. Detection of Chlamydia trachomatis DNA in joints of reactive arthritis patients by polymerase chain reaction. Lancet 1992;340:81–82.Find this resource:

                22. Schumacher H, Arayssi T, Branigan P, et al. PCR study for bacterial DNA in normal synovium. Arthritis Rheum 1997;40:S269.Find this resource:

                  23. Nanagara R, Li F, Beutler A, Hudson A, Schumacher HR, Jr. Alteration of Chlamydia trachomatis biologic behavior in synovial membranes. Suppression of surface antigen production in reactive arthritis and Reiter’s syndrome. Arthritis Rheum 1995 Oct;38(10):1410–1417.Find this resource:

                  24. McGonagle D, Stockwin L, Issacs J, Emery P. (2001). An enthesitis based model for the pathogenesis of spondyloarthropathy. Additive effects of microbial adjuvant and biomechanical factors at disease sites. J Rheumatol 2001;28(10):2155–2159.Find this resource:

                    25. McGonagle D, Lories RJU, Tan AL, Benjamin M. The concept of a ‘synovio-entheseal complex’ and its implications for understanding joint inflammation and damage in psoriatic arthritis and beyond. Arthr Rheum 2007;56(8):2482–2491.Find this resource:

                    26. Sherlock JP, Joyce-Shakih B, Turner SP, et al. (2012). IL-23 induces spondyloarthropathy by acting on ROR-gammat+ CD3+CD4-CD8- entheseal resident T cells. Nat Med 2012;18(7):1069–1076.Find this resource:

                      27. Gaston JSH, Life PF, Merilahti Palo, R, et al. Synovial T lymphocyte recognition of organisms that trigger reactive arthritis. Clin Exp Immunol 1989;76:348–353.Find this resource:

                      28. Sieper J, Braun J, Wu P, Kingsley G. T-cells are responsible for the enhanced synovial cellular immune response to triggering antigen in reactive arthritis. Clin Exp Immunol 1993;91(1):96–102.Find this resource:

                      29. Goodall JC, Yeo G, Huang M, Raggiaschi R, Gaston JSH. Identification of Chlamydia trachomatis antigens recognized by human CD4+ T lymphocytes by screening an expression library. Eur J Immunol 2001;31:1513–1522.Find this resource:

                      30. Mertz AK, Wu P, Sturniolo T, et al. Multispecific CD4+ T cell response to a single 12-mer epitope of the immunodominant heat-shock protein 60 of Yersinia enterocolitica in Yersinia-triggered reactive arthritis: overlap with the B27-restricted CD8 epitope, functional properties, and epitope presentation by multiple DR alleles. J Immunol 2000;164(3):1529–1537.Find this resource:

                      31. Gaston J, Life P, Bailey L, Bacon P. In vitro responses to a 65kD mycobacterial protein by synovial T cells from inflammatory arthritis patients. J Immunol 1989;143:2594–2600.Find this resource:

                        32. Hermann E, Lohse AW, van der Zee, R, et al. Synovial fluid derived Yersinia reactive T cells responding to human 65 kDa heat shock protein and heat stressed antigen presenting cells. Eur J Immunol 1991;21(9):2139–2143.Find this resource:

                        33. Thiel A, Wu P, Lanowska M, et al. Identification of immunodominant CD4+ T cell epitopes in patients with Yersinia-induced reactive arthritis by cytometric cytokine secretion assay. Arthritis Rheum 2006;54(11):3583–3590.Find this resource:

                        34. Deane KH, Jecock RM, Pearce JH, Gaston JS. Identification and characterization of a DR4-restricted T cell epitope within chlamydia heat shock protein 60. Clin Exp Immunol 1997;109(3):439–445.Find this resource:

                        35. Appel H, Kuon W, Kuhne M, et al. Use of HLA-B27 tetramers to identify low-frequency antigen-specific T cells in Chlamydia-triggered reactive arthritis. Arthritis Res Ther 2004;6(6):R521–R534.Find this resource:

                        36. Matyszak MK, Gaston JS. Chlamydia trachomatis-specific human CD8+ T cells show two patterns of antigen recognition. Infect Immunol 2004;72(8):4357–4367.Find this resource:

                        37. Allen RL, Gillespie GMA, Hall F, et al. Multiple T cell expansions are found in the blood and synovial fluid of patients with reactive arthritis. J Rheumatol 1997;24(9):1750–1757.Find this resource:

                        38. Njobvu P, McGill P. Human immunodeficiency virus related reactive arthritis in Zambia. J Rheumatol 2005;32(7):1299–1304.Find this resource:

                        39. Taurog JD, Dorris ML, Satumtira N, et al. Spondylarthritis in HLA-B27/human beta2-microglobulin-transgenic rats is not prevented by lack of CD8. Arthritis Rheum 2009;60(7):1977–1984.Find this resource:

                        40. Schlaak J, Hermann E, Ringhoffer M, et al. Predominance of Th1 type T cells in synovial fluid of patients with Yersinia induced reactive arthritis. Eur J Immunol 1992;22:2771–2776.Find this resource:

                        41. Thiel A, Wu PH, Lauster R, Braun J, Radbruch A, Sieper J. Analysis of the antigen-specific T cell response in reactive arthritis by flow cytometry. Arthritis Rheum 2000;43(12):2834–2842.Find this resource:

                        42. Shen H, Goodall JC, Gaston JSH. Frequency and phenotype of T helper 17 cells in peripheral blood and synovial fluid of patients with reactive arthritis. J Rheumatol 2010;37(10):2096–2099.Find this resource:

                        43. Singh AK, Misra R, Aggarwal A. Th-17 associated cytokines in patients with reactive arthritis/undifferentiated spondyloarthropathy. Clin Rheumatol 2011;30(6):771–776.Find this resource:

                        44. Aho K, Ahvonen P, Lassus A, Sievers K, Tiilikainen A. HL-A antigen 27 and reactive arthritis. Lancet 1973;ii:157–159.Find this resource:

                        45. Brewerton DA, Caffrey M, Nicholls A, et al. Reiter’s disease and HL-A 27. Lancet 1973;2(7836):996–998.Find this resource:

                        46. Ekman P, Kirveskari J, Granfors K. Modification of disease outcome in Salmonella-infected patients by HLA-B27. Arthr Rheum 2000;43(7):1527–1534.Find this resource:

                        47. Leirisalo-Repo, M, Helenius P, Hannu T, et al. Long-term prognosis of reactive salmonella arthritis. Ann Rheum Dis 1997 Sep;56(9):516–520.Find this resource:

                        48. Gaston JSH. Mechanisms of disease: the immunopathogenesis of spondyloarthropathies. Nat Clin Pract Rheumatol 2006;2(7):383–392.Find this resource:

                        49. Kollnberger S, Bowness P. The role of B27 heavy chain dimer immune receptor interactions in spondyloarthritis. Adv Exp Med Biol 2009;649:277–285.Find this resource:

                        50. DeLay ML, Turner MJ, Klenk EI, et al. HLA-B27 misfolding and the unfolded protein response augment interleukin-23 production and are associated with Th17 activation in transgenic rats. Arthr Rheum 2009;60(9):2633–2643.Find this resource:

                        51. Goodall JC, Wu C, Zhang Y, et al. Endoplasmic reticulum stress-induced transcription factor, CHOP, is crucial for dendritic cell IL-23 expression. Proc Natl Acad Sci USA 2010;107:17698–17703.Find this resource:

                        52. Reveille JD, Sims AM, Danoy P, et al. Genome-wide association study of ankylosing spondylitis identifies non-MHC susceptibility loci. Nat Genet 2010;42(2):123–127.Find this resource:

                        53. Evans DM, Spencer CC, Pointon JJ, et al. Interaction between ERAP1 and HLA-B27 in ankylosing spondylitis implicates peptide handling in the mechanism for HLA -B27 in disease susceptibility. Nat Genet 2011;43(8):761–767.Find this resource:

                        54. Kochan G, Krojer T, Harvey D, et al. Crystal structures of the endoplasmic reticulum aminopeptidase-1 (ERAP1) reveal the molecular basis for N-terminal peptide trimming. Proc Natl Acad Sci USA 2011;108(19):7745–7750.Find this resource:

                        55. Reeves E, Colebatch-Bourn A, Elliott T, Edwards CJ, James E. (2014). Functionally distinct ERAP1 allotype combinations distinguish individuals with ankylosing spondylitis. Proc Natl Acad Sci USA 2014;111(49):17594–17599.Find this resource:

                        56. Roberts, A. R., Appleton, L. H., Cortes, A., Vecellio, M., Lau, J., Watts, L., Brown, M. A., Wordsworth, P. ERAP1 association with ankylosing spondylitis is attributable to common genotypes rather than rare haplotype combinations. P.N.A.S. 114: 558–561 (2017)Find this resource:

                          57. Tsui FW, Xi N, Rohekar S, et al. Toll-like receptor 2 variants are associated with acute reactive arthritis. Arthr Rheum 2008;58(11):3436–3438.Find this resource:

                          58. Perry ME, White JA. Three cases of reactive arthritis secondary to lymphogranuloma venereum. J Clin Rheumatol 2015;21(1):33–34.Find this resource:

                          59. Angulo J, Espinoza L. The spectrum of skin, mucosa and other extra-articular manifestations. Bailliere’s Clin Rheumatol 1998;12(4):649–664.Find this resource:

                          60. Deighton C. beta-Haemolytic Streptococci and Reactive Arthritis in Adults. Ann Rheum Dis 1993;52(6):475–482.Find this resource:

                          61. Wollenhaupt J, Schnarr S, Kuijpers JG. Bacterial antigens in reactive arthritis and spondarthritis. Rational use of laboratory testing in diagnosis and follow-up. Bailliere’s Clin Rheumatol 1998;12(4):627–647.Find this resource:

                          62. Bardin T, Enel C, Cornelis F et al. Antibiotic treatment of venereal disease and Reiter’s syndrome in a Greenland population. Arthr Rheum 1992;35:190–194.Find this resource:

                          63. Lauhio A, Leirisalo Repo, M, Lahdevirta J, Sikku P, Repo H. Double-blind, placebo-controlled study of three-month treatment with lymecycline in reactive arthritis with special reference to chlamydia arthritis. Arthr Rheum 1991;34:6–14.Find this resource:

                          64. Sieper J, Fendler C, Laitko S, et al. No benefit of long-term ciprofloxacin treatment in patients with reactive arthritis and undifferentiated oligoarthritis—A three-month, multicenter, double-blind, randomized, placebo-controlled study. Arthr Rheum 1999;42(7):1386–1396.Find this resource:

                          65. Wakefield D, McCluskey P, Verma M, et al. Ciprofloxacin treatment does not influence course or relapse rate of reactive arthritis and anterior uveitis. Arthr Rheum 1999;42(9):1894–1897.Find this resource:

                          66. YliKerttula T, Luukkainen R, YliKerttula U, et al. Effect of a three month course of ciprofloxacin on the outcome of reactive arthritis. Ann Rheum Dis 2000;59(7):565–570.Find this resource:

                          67. Kvien TK, Gaston JS, Bardin T, et al. Three month treatment of reactive arthritis with azithromycin: a EULAR double blind, placebo controlled study. Ann Rheum Dis 2004;63(9):1113–1119.Find this resource:

                          68. Kuuliala A, Julkunen H, Paimela L, et al. Double-blind, randomized, placebo-controlled study of three-month treatment with the combination of ofloxacin and roxithromycin in recent-onset reactive arthritis. Rheumatol Int 2013;33(11):2723–2729.Find this resource:

                          69. Putschky N, Pott HG, Kuipers JG, et al. Comparing 10-day and 4-month doxycycline courses for treatment of Chlamydia trachomatis-reactive arthritis: a prospective, double-blind trial. Ann Rheum Dis 2006;65(11):1521–1524.Find this resource:

                          70. Carter JD, Espinoza LR, Inman RD, et al. Combination antibiotics as a treatment for chronic Chlamydia-induced reactive arthritis: a double-blind, placebo-controlled, prospective trial. Arthr Rheum 2010;62(5):1298–1307.Find this resource:

                          71. Dougados M, Vanderlinden S, Leirisalorepo M, et al. Sulfasalazine in the treatment of spondylarthropathy: a randomized, multicenter, double-blind, placebo-controlled study. Arthr Rheum 1995;38(5):618–627.Find this resource:

                          72. Egsmose C, Hansen TM, Andersen LS, et al. Limited effect of sulphasalazine treatment in reactive arthritis. A randomized double blind placebo controlled trial. Ann Rheum Dis 1997;56(1):32–36.Find this resource:

                          73. Meyer A, Chatelus E, Wendling D, et al. Safety and efficacy of anti-tumor necrosis factor alpha therapy in ten patients with recent-onset refractory reactive arthritis. Arthr Rheum 2011;63(5):1274–1280.Find this resource:

                          74. Leirisalo-Repo, M. Prognosis, course of disease, and treatment of the spondyloarthropathies. Rheum Dis Clin N Am 1998;24(4):737–751, viii.Find this resource:

                          75. Kaarela K, Jantti JK, Kotaniemi KM. Similarity between chronic reactive arthritis and ankylosing spondylitis. A 32-35-year follow-up study. Clin Exp Rheumatol 2009;27(2):325–328.Find this resource:

                          76. Haslock I, Wright V. The musculoskeletal complications of Crohn’s disease. Medicine 1973;52:217–225.Find this resource:

                          77. Leirisalorepo M, Turunen U, Stenman S, Helenius P, Seppala K. High frequency of silent inflammatory bowel disease in spondylarthropathy. Arthr Rheum 1994;37(1):23–31.Find this resource:

                          78. Mielants H, Veys E, Cuvelier C, De Vos, M. Course of gut inflammation in spondylarthropathies and therapeutic consequences. Baillieres Clin Rheumatol 1996;10:147–164.Find this resource:

                          79. Orchard T, Wordsworth B, Jewell D. Peripheral arhropathies in inflammatory bowel disease: their articular distribution and natural history. Gut 1998;42:387–391.Find this resource:

                          80. Kaser A, Zeissig S, Blumberg RS. Inflammatory bowel disease. Annu Rev Immunol 2010;28:573–621.Find this resource:

                          81. Abi Karam, G, Awada H, Nasr F, Uthman I. Ileal pouchitis and arthritis. Semin Arthr Rheum 2003;33(3):215.Find this resource:

                          82. Balbir-Gurman, A, Schapira D, Nahir M. Arthritis related to ileal pouchitis following total proctocolectomy for ulcerative colitis. Semin Arthr Rheum 2001;30(4):242–428.Find this resource:

                          83. Thomas PD, Keat AC, Forbes A, Ciclitira PJ, Nicholls RJ. Extraintestinal manifestations of ulcerative colitis following restorative proctocolectomy. Eur J Gastroenterol Hepatol 1999;11(9):1001–1005.Find this resource:

                          84. Bourne J, Kumar P, Huskisson E, et al. Arthritis and coeliac disease. Ann Rheum Dis 1992;44:592–598.Find this resource:

                          85. Lettre G, Rioux JD. Autoimmune diseases: insights from genome-wide association studies. Hum Mol Genet. 2008;17(R2):R116–R121.Find this resource:

                          86. Stagi S, Giani T, Simonini G, Falcini F. Thyroid function, autoimmune thyroiditis and coeliac disease in juvenile idiopathic arthritis. Rheumatology (Oxf) 2005;44(4):517–520.Find this resource:

                          87. Relman DA, Schmidt TM, MacDermott RP, Falkow S. Identification of the uncultured bacillus of Whipple’s disease. New Engl J Med 1992;327:293–301.Find this resource:

                          88. Puechal, X. Whipple’s disease. Ann Rheum Dis 2013;72(6):797–803.Find this resource:

                          89. O’Duffy, JD, Griffing WL, Li CY, Abdelmalek MF, Persing DH. Whipple’s arthritis—Direct detection of Tropheryma whippelii in synovial fluid and tissue. Arth Rheum 1999;42(4):812–817.Find this resource:

                          90. Puechal X, Fenollar F, Raoult D. Cultivation of Tropheryma whipplei from the synovial fluid in Whipple’s arthritis. Arthr Rheum 2007;56(5):1713–1718.Find this resource:

                          91. Schneider T, Moos V, Loddenkemper C, et al. Whipple’s disease: new aspects of pathogenesis and treatment. Lancet Infect Dis 2008;8(3):179–190.Find this resource:

                          92. Ansemant T, Celard M, Tavernier C, et al. Whipple’s disease endocarditis following anti-TNF therapy for atypical rheumatoid arthritis. Jt Bone Spine 2010;77(6):622–623.Find this resource: