 |
|||||||||
|
|||||||||
| CECIL |
| TEXT BOOK of MEDICINE |
Section XXII Rheumatic Disease
| 286 THE SPONDYLOARTHROPATHIES Robert D. Inman • |
|
|
Definition
The spondyloarthropathies (SpAs) encompass a group of clinical syndromes that are linked in terms of disease manifestations and genetic susceptibility. The clinical subsets most commonly recognized are ankylosing spondylitis (AS), reactive arthritis (ReA), psoriatic arthritis (PsA), and enteropathic arthritis (EA) (Fig. 286-1). In addition, a sizable population of patients do not fit into one of these distinct diagnostic categories but share some of the common clinical features described in this chapter. The syndrome associated with this subset is termed undifferentiated spondyloarthropathy (USpA), which over time may evolve into a classic pattern such as AS but may remain in an undifferentiated pattern in long follow-up studies.
Pathobiology
Family studies in which there are multiple individuals with SpA have emphasized some of the common features between the four distinct subsets just mentioned. The impression from such studies is that there is a shared common path of immunogenetic susceptibility on which further genetic and environmental influences lead to characteristic clinical subsets. Thus, if EA occurs in such a family, in another family it may be PsA. In this sense, the SpAs seem to “breed true.” It should be recognized, however, that some distinct clinical features can be very restricted in their manifestations (e.g., guttate psoriasis and keratodermia blennorrhagica), thus making simple discrimination sometimes difficult.
Clinical Manifestations
There are several common features in the family of SpAs that at once link them and serve to distinguish them from the other major contributor to chronic polyarthritis, rheumatoid arthritis (RA) (Chapter 285). The SpAs have a strong predilection for the spine, in particular the sacroiliac joints. There is a shared tendency for new bone formation at sites of chronic inflammation, with joint ankylosis as a consequence. When peripheral arthritis occurs, it is commonly lower extremity and asymmetrical. There is a predilection for sites of tendon insertion into bone (entheses), so enthesitis becomes one of the most specific clinical manifestations of the SpAs. Theories postulating the basis for this target organ involvement have invoked biomechanical factors, innervation, local vascularity, and bone marrow–derived inflammatory mediators, but the precise mechanism for this relationship remains incompletely defined. Whatever the reason, inflammation in the enthesis and contiguous subchondral bone is a characteristic feature of this arthritis, and the appearance of this inflammation on magnetic resonance imaging (MRI) is distinct enough to lead some investigators to use such imaging for diagnostic purposes.
Predilection for ocular inflammation, particularly acute anterior uveitis, is a common feature of all SpA subsets. Indeed, some investigators consider anterior uveitis to be a feature of SpAs in its own right because it may occur in the same susceptible populations of patients even in the absence of joint involvement and may have a unique genetic predisposition. Finally, all SpA subsets have an association with the class I human leukocyte antigen (HLA) allele B27, with the strength of the association varying somewhat between them.
Diagnosis
Increasingly, diagnostic criteria (Table 286-1) are being used that emphasize the common clinical features, namely, inflammatory spinal pain or asymmetrical, lower extremity synovitis. Several distinct features differentiate the SpAs from RA, the other main contributor to the differential diagnosis of chronic polyarthritis (Table 286-2). These features include sex predilection, HLA association, pattern of joint involvement, and the presence of rheumatoid factor, which becomes the serologic distinction between seropositive disease (RA) and seronegative disease (SpAs).
Click here to view this table....
Click here to view this table....
At the level of joint histopathology, sites of chronic inflammation in RA are associated with erosions, but in the SpAs such sites are associated with new bone formation. This distinction suggests a fundamental difference in the cytokine profile in the microenvironment of the joint, and although there is some evidence that SpAs reflect more of a type 2 T-helper cell (TH2) cytokine profile as opposed to the TH1 profile of RA, this issue has not been resolved and the mediators of neo-ossification await identification. Synovial histopathology in SpA is characterized by abundant neutrophils, macrophages, and hypervascularity, whereas in RA the prominent features are lymphoid aggregates, dendritic cells, lining cell hyperplasia, and citrullinated proteins.
Treatment
General Measures
The SpA group of diseases necessitates a global approach to management in which education of patients is the cornerstone. With the typical onset during young adulthood and with a male preponderance, these patients may express significant frustration or depression if their acute arthritis evolves into a chronic course that may significantly impair their functional capabilities and quality of life. A clinician managing patients with SpA should be aware that these psychosocial aspects of the disease are an important part of the burden of illness. Similarly, there may be important implications for the workplace, particularly if the job demands significant bending or twisting.
Exercise is an important part of the treatment plan for patients with AS. Generally, high-impact sports should be avoided, whereas swimming is an ideal exercise. Stretching to maintain mobility and maintenance of posture should be emphasized, and an experienced physiotherapist can greatly assist in instructing patients in daily exercises. Long car trips and air travel should include periodic stretching. Sleep position should emphasize a straight back position rather than one curled on the side. Deep breathing exercises and avoidance of cigarettes should be stressed.
One key area of concern for patients is prognosis because the SpAs, particularly ReA, often occur in young, active individuals for whom athletic activity is a priority. There is general recognition that ReA has a greater propensity for chronicity than was previously appreciated, and this should temper an overly optimistic projection on the natural history of the disease. At a 5-year follow-up of a cohort of patients with point source Salmonella-induced ReA, two thirds continued to have subjective complaints and one third demonstrated objective changes in their joints. The variability in prognosis for the large group of patients falling into the diagnostic category of USpA is perplexing for patients in this category. At present, there is a lack of reliable predictors of progression in patients with this heterogeneous cluster of articular and extra-articular features.
Medical Therapy
Nonsteroidal Anti-inflammatory Drugs
In general, joint inflammation improves significantly after the introduction of nonsteroidal anti-inflammatory drugs (NSAIDs), the exception being salicylates, which seem less effective in control of pain and stiffness. Indomethacin and diclofenac, up to 200 mg daily in divided doses, are generally well tolerated in this population of patients. These agents have to be used with caution in EA because of concern about exacerbating the underlying inflammatory bowel disease. In the case of AS, the goal with anti-inflammatory treatment is to achieve sufficient control of pain and stiffness to allow an active, sustained program of exercise and physical activity that maintains posture in addition to improvements in quality of life. Whether NSAIDs have disease-modifying capability has not been definitively resolved.
Corticosteroids
The response to the commonly used intra-articular steroid injection of peripheral joints in patients with SpA is often neither as dramatic nor as sustained as in patients with RA. Corticosteroid injection into the sacroiliac joints is usually performed under imaging guidance (fluoroscopy or computed tomography [CT]). One study observed that such injections gave a good response in 79% of 24 patients and that the improvement could persist for many months. Systemic corticosteroids (either orally or via an intravenous bolus protocol) have been used for severe symptomatic flares, but controlled trials to validate effectiveness are lacking. The goal should be prompt tapering of the dose when symptomatic control is achieved. The recognition that osteoporosis (Chapter 264) is a significant problem in AS provides further impetus to use corticosteroids sparingly. Topical steroids are usually effective for treatment of the mucous membrane and skin manifestations of reactive arthritis (ReA). For uveitis, topical corticosteroid eye drops are an integral component of management, and treatment should be monitored jointly with an ophthalmologist.
Sulfasalazine
Randomized, placebo-controlled trials have provided evidence to support some role for sulfasalazine (SSZ), particularly in PsA.1 Three 36-week, randomized, double-blind multicenter studies of patients with AS, PsA, or ReA, respectively, were undertaken to compare SSZ (2 g/day) with placebo in each case. The different response rates in the placebo arms of these trials highlight the variability in clinical course of the SpAs. An analysis of these studies with SSZ stratified the patients into those having axial or peripheral disease. In patients with only axial disease, response criteria were met equally in the SSZ group and the placebo group. In the patients with peripheral arthritis, responses were seen in 59% of the SSZ group and 43% of the placebo group (P < .0005). These findings are useful in guiding selection of patients for SSZ treatment options.
Methotrexate
Concurrent with the widespread use of methotrexate (MTX) in patients with RA, there has been increasing use of MTX in patients with SpA. Generally, responses have been good, particularly for peripheral joint disease, but few controlled trials have been conducted to substantiate these clinical impressions. There is no evidence that MTX changes the course of axial diseases in AS. Experience with long-term MTX therapy in 38 patients with PsA has been reported, and despite an improvement in joint count, there was no evidence of slowing of radiographic progression. Long-term follow-up may be required to resolve whether MTX has a joint-sparing effect in PsA.
Other Disease-Modifying Agents
There have been several therapeutic approaches to control of PsA, including chloroquine, intramuscular gold, and cyclosporine, all of which show some clinical efficacy, but long-term studies are needed to evaluate them more comprehensively. The response of SpA patients to SSZ may be attributable to the antibiotic moiety of this compound (sulfapyridine) or to the anti-inflammatory moiety (5-aminosalicylic acid [5-ASA]). A study reported that of SpA patients who had been taking SSZ and were then switched to 5-ASA, most maintained their response profile, thus supporting the notion that 5-ASA may be the active moiety in SSZ. An open study of intravenous pamidronate for refractory AS showed significant improvement in disease activity scores; this agent is under investigation at several centers.
Antibiotic Therapy
The current concept of the pathogenesis of ReA postulates that a bacterial infection, usually gastrointestinal (GI) or genitourinary (GU), is the triggering event in an immunogenetically susceptible host. For the other SpAs there is less compelling evidence to implicate infection in a causal role. It is sound clinical practice to treat any culture-proven chlamydial urethritis in conjunction with treatment of the sexual partner. For this indication, azithromycin, 1 g as a single dose, is as effective as doxycycline, 100 mg twice a day for 7 days. The role of antibiotics in the management of SpA has been controversial. An earlier retrospective review concluded that 37% of episodes of urethritis not treated with anti-Chlamydia agents were associated with subsequent ReA whereas only 10% of such episodes when treated with tetracycline progressed to ReA. In a 3-month, double-blind, placebo-controlled study of chronic ReA, lymecycline significantly decreased the duration of illness in patients with Chlamydia-induced ReA but not in patients with ReA triggered by enteric pathogens. Two controlled trials of ciprofloxacin demonstrated no significant difference from placebo in ReA and undifferentiated oligoarthritis. Ciprofloxacin was not effective in Yersinia- or Salmonella-induced ReA, but there was a trend toward response in Chlamydia-induced ReA. A recent study examining azithromycin in the treatment of SpA found no convincing evidence of disease-modifying potential.2
Anti–Tumor Necrosis Factor Therapy
The role of immunomodulatory cytokines in the pathogenesis of SpA has been controversial. Some studies have implicated as an AS susceptibility marker a genetic polymorphism associated with relative impairment of tumor necrosis factor-α (TNF-α) production, but not all studies have supported this notion. Despite the uncertainties about the role of pro-inflammatory cytokines in AS, some of the newer biologic agents such as monoclonal antibodies to TNF-α (infliximab and adalimumab) or the soluble TNF receptor (etanercept) have been used in the treatment of SpA.3 At present, the three anti-TNF agents have been shown to be effective in short-term trials of AS and PsA.4 These studies have generally reported a prompt response in clinical outcome measures, as well as laboratory indicators of inflammation, and MRI evaluations have shown improvement in local inflammation in the sacroiliac joints and spine. The anti-TNF treatments have been well tolerated with no significant incidence of serious adverse events, but patients appear to relapse when treatment is discontinued. Experience with longer-term treatment intervals with anti-TNF agents (>2 years) has been encouraging in the persistence of the therapeutic effect and the infrequency of late adverse events. Whether anti-TNF treatments can alter the progressive ankylosis of this disease over the long course of the disease awaits further study. The significant improvements seen on MRI provide rational hope that these agents may alter the long-term outcomes of AS, both structurally and functionally.
GENETIC SUSCEPTIBILITY
HLA-B27
Definition
The major histocompatibility complex (MHC), on the short arm of chromosome 6 in humans, is one of the most polymorphic regions of the human genome (Chapter 43). This is particularly so for the B locus, which constitutes part of the class I MHC genes in this complex. More than 200 different alleles have been detected at this locus, of which B27 is just 1. As with all HLA alleles, there is codominant expression of B locus genes such that most individuals who are “B27 positive” are heterozygous for the B locus. There appears to be little clinical or prognostic significance associated with the less common homozygous B27 state. The conventional role of class I HLA antigens is to present a processed peptide to the T-cell receptor of a specific CD8+ cytotoxic T cell, thereby initiating an immune response against the pathogen from which that peptide was derived by intracellular proteolysis and processing. This function places the HLA antigens in a critical role in host defense against pathogens, and it is the heterogeneity of cellular immune responses that is alleged to be an advantage for a species (such as humans) with an extensive polymorphism region of the genome. This idea has led to the concept that infectious diseases have driven allelic polymorphisms in the MHC. Such a hypothesis postulates a selective advantage in the extensive peptide-binding capabilities conferred by different alleles of the B locus, but there may also be disadvantages.
Epidemiology
It is clear that HLA is strongly associated with the SpAs, yet the prevalence of HLA-B27 varies widely in different racial and ethnic clusters around the world. It is virtually absent in aboriginal populations in Australia, occurs in 1% of the population in Japan, in 7% in northern European countries, and in 50% in some of the native tribes in western Canada. The environmental-genetic interaction that may account for the expansion or restriction of this gene in human populations is unknown, but some evidence indicates that B27 may confer more effective host response to some viruses. There is a practical impact of this variability for the clinician. Because the relative risk conferred by a gene reflects the prevalence of the gene in the affected individuals versus the prevalence of the gene in the normal population, the relative risk for SpAs is higher in a population in which the gene is uncommon (e.g., Japan) than in a population in which B27 is more common (e.g., Scandinavia). In the North American white population, the prevalence of the gene is approximately 7%. Thus, there is a 7% “false-positive” rate if one is attempting to use the gene as a diagnostic marker to decipher the cause of chronic back pain in an unselected population of patients. On the other hand, 90% of patients with AS are B27 positive, so there is a 10% “false-negative” rate in the use of the test diagnostically. The key factor is pretest probability. In a patient with chronic back pain that is clearly inflammatory in character, the addition of B27 positivity combines to strengthen the likelihood of AS accounting for the back problem. The presence of distinctive extra-articular features (such as uveitis) further increases this likelihood.
It had been held that the prevalence of AS in various parts of the globe closely parallels the prevalence of B27 in that population, and in general this pattern is valid. What has introduced complexity into this concept has been the recognition that there is not just one B27 but in fact more than 24 subtypes of B27 (ranging from B⋆2701 to B⋆2724 at present). In terms of the evolution of molecular variability, B⋆2705 is regarded as the primordial subtype with variability developing over time on the basis of alterations in genomic DNA. The amino acid substitutions occurring as a result are generally reflected in the peptide-binding cleft of the B27 molecule, thus supporting the argument that there has been a functional change concurrent with the structural change. This notion has been bolstered by the observation that some subtypes, notably B⋆2706 and B⋆2709, do not seem to confer increased susceptibility to the development of AS. As an extension of genetic epidemiology, this observation has led to a search for “arthritogenic peptides” that can be presented by the disease-associated subtypes such as B⋆2705 and B⋆2704 but not by the non–disease-associated subtypes. To date, no simple peptide-susceptibility relationship has been demonstrated, but this is an important clue to the pathogenic role of B27, and studies are ongoing to explore this relationship.
Other approaches have also been taken to define the mechanism whereby B27 confers disease susceptibility in addition to that of uniquely presenting an arthritogenic peptide to T cells. According to the theory of molecular mimicry, an autoimmune response can ensue after an infection if the immune response against the pathogen cross-reacts with host antigens. There is a degree of sequence homology between B27 and several candidate gram-negative enteric bacteria and there is evidence for cross-reacting monoclonal antibodies, but the significance of such homology for disease pathogenesis remains unresolved. It has also been argued that B27 is distinctive in its propensity to misfold in the endoplasmic reticulum, which may induce a pro-inflammatory cascade called the unfolded protein response. Furthermore, B27 may have a distinct tendency to form heavy chain homodimers at the cell surface, and the possible consequences of this change for the immune response are under investigation. There has also been investigation into alteration of primary host-pathogen interactions, such as modulation of invasion, intracellular replication, and pathogen clearance, but no definitive allele-specific relationships have been demonstrated in these studies, although this topic remains an area of active study. In B27 transgenic rats, the spontaneous development of pathology strikingly similar to human SpAs has supported the notion that B27 itself is the critical genetic factor in disease pathogenesis. These animals demonstrate pathology similar to that of Crohn's disease in the GI tract, spondylitis, peripheral arthritis, uveitis, and psoriasiform skin and nail changes. Of interest, if such animals are raised in a germ-free environment, there is a marked reduction in joint and gut disease, thus implying a dynamic interrelationship between microbial triggers and background host genes that seems to recapitulate the situation seen clinically.
Genome-wide screening studies of multiplex families with SpAs, particularly AS, are ongoing in several countries in order to identify other important genes that are involved in predisposing to these diseases. Several candidate loci have been identified, although considerable variability between different studies is apparent. The common observation, however, is that the strongest association of SpAs is with the HLA complex, so at least in familial AS, B27 may to a certain extent be necessary, if not sufficient to confer disease susceptibility. There are estimates of an additional four to nine genes acting in concert with B27; their identity remains a topic of active investigation. MRI studies in asymptomatic B27-positive individuals indicate that there is a much higher prevalence of sacroiliitis than previously recognized, and studies are continuing to define that prevalence and indeed the prevalence of SpAs in the general population. Some investigators have concluded that SpAs are as common as RA in the population.
▪ CLINICAL SUBSETS OF THE SPONDYLOARTHROPATHIES
▪ Ankylosing Spondylitis
Epidemiology
AS is the most common inflammatory disorder of the axial skeleton. The following is a useful rule of thumb: AS occurs in 0.2% of the general population, in 2% of the B27-positive population, and in 20% of B27-positive individuals with an affected family member. There is a male preponderance in the disease, with the male-to-female ratio ranging from 2.5:1 to 5:1; however, recent epidemiologic studies have found less female involvement than these estimates. The basis for the gender bias is not resolved. It is held, however, that AS is under-recognized in women, perhaps because of milder axial disease and a more delayed disease onset, but alternative diagnoses of pelvic and low back pain in women may hinder clinician awareness of the disease in female patients.
Clinical Manifestations
AS typically begins in young adulthood, but symptoms may arise in adolescence or earlier. Up to 15% of children with juvenile chronic polyarthritis are classified as having juvenile AS. Such children may have a pauciarticular pattern, with a predilection for the tarsal joints and frequently minimal axial complaints. During the adolescent years there is an increasing prevalence of radiographic sacroiliitis, with a significant proportion of patients manifesting this feature by the end of the teenage years. At the other end of the age spectrum, a small number of patients with late-onset AS may have sacroiliitis and oligoarthritis. The axial involvement and asymmetrical lower extremity involvement may serve to differentiate such patients from those with late-onset RA, although clinically there may be overlapping features. Recent studies indicate that long-term functional disability may be more marked in juvenile-onset AS than in adult-onset AS.
The classical manifestation of AS is the onset of low back pain that persists for more than 3 months, is accompanied by early morning stiffness, and is typically improved by exercise (Table 286-3). An extensive discussion of the differential diagnosis of low back pain is presented in Chapter 423. Some investigators would also include a response to NSAID therapy as a feature further differentiating AS from mechanical low back pain. Back pain that awakens the patient from sleep is often a clue to inflammatory back pain that may have previously been misdiagnosed as the pain of degenerative disc disease, the latter being the much more common cause of low back pain in the population at large. The pain typically occurs in the region of the sacroiliac joints, with or without slight radiation to the buttock area. Midthoracic pain and cervical pain, particularly at night, are less common but strongly suggest inflammatory back pain when they occur. Fatigue is also a suggestive symptom and is often highlighted by these patients because the typical young male patient sets a high functional norm in terms of sports and recreation. If the inflammation is inadequately controlled, there is increasing stiffness, which may persist most of the day, as well as progressive loss of mobility and flexibility.
Click here to view this table....
Peripheral oligoarthritis is seen in up to 30% of patients with AS. Typically, it is an asymmetrical oligoarthritis with a predilection for the lower extremities. It is important to ask about concurrent or previous tendinitis (e.g., Achilles tendinitis) or heel pain (e.g., plantar fasciitis) because either may reflect an enthesitis that is part of the clinical picture. Involvement of the hip can occur at any point in the course of AS and can follow a course to joint destruction. A hip flexion contracture on this basis may contribute to increasing stoop on standing and walking, which may otherwise be attributed to spinal involvement in the disease.
Extra-articular features primarily involve the eye. Ocular involvement may occur in up to 40% of AS patients, most typically acute anterior uveitis (iritis). The uveitis is often manifested by a slight impairment in visual acuity, and photophobia and eye pain may accompany the uveitis. Typically, it is unilateral and recurrent. Uncommon manifestations include aortic insufficiency, cardiac conduction defects, and pulmonary fibrosis.
Diagnosis
Physical Examination
Physical examination of the spine characteristically indicates restricted movement, which in the early stages may reflect paraspinal muscle spasm in part but late in the course reflects ankylosis of the zygapophyseal joints and syndesmophyte bridging of the vertebral bodies. Forward flexion is restricted and can be monitored by measuring the finger-to-floor distance on bending forward. Schober's test is used to measure mobility in the lower part of the back: with the patient standing upright, a 10-cm span is marked from the fifth lumbar vertebra upward. On maximal forward flexion, the distance between the marks is remeasured. With normal spinal mobility the flexed distance should register as 15 cm, or an increment of 5 cm. Thoracic involvement is measured in chest expansion, with the chest circumference, at maximum inspiration, being more than 5 cm greater than the circumference at maximal expiration. Changes in cervical mobility can be measured as the occiput-to-wall distance, with the patient's heels against the wall and the patient attempting to touch the back of the head to the wall. Inflammation in the sacroiliac joint may be reflected by joint line tenderness to direct pressure or by the Fabere or Gaenslen maneuver. In the former, the patient lies supine while the examiner flexes and externally rotates the hip. In the latter, the examiner extends the hip by letting the leg dangle off the side of the examining table. In both cases, stress is placed on the sacroiliac joint and may reproduce the back pain if it derives from this site.
Laboratory Findings
Laboratory tests in the evaluation of inflammatory back pain are relatively nonspecific. The erythrocyte sedimentation rate and C-reactive protein are typically elevated, but normal levels do not exclude inflammatory back pain, and the degree of elevation is typically less than would be seen in acute RA. The anemia of chronic disease may be observed if the condition is long-standing. Serum immunoglobulin A levels may be elevated, but autoantibodies are notable for their absence. HLA-B27 is rarely the definitive factor for diagnosis, and the false-positive and false-negative rates have already been discussed, but in the setting of characteristic back symptoms, the test has reasonably high sensitivity and specificity.
Imaging
Radiographic assessment is important for confirmation of disease, but early in the course of the disease there may be no radiographic changes in the sacroiliac joints. If the clinician has a high index of suspicion in such cases, the use of a bone scan, CT scan, or especially MRI may improve the sensitivity of the plain radiograph. Views of the sacroiliac joints should be requested because the obliquity of the joint renders it difficult to assess on a routine anteroposterior pelvic radiograph. The classic findings are bilateral changes in the sacroiliac joints (Fig. 286-2). Abnormalities include erosions in the joint line, pseudowidening, subchondral sclerosis, and finally, ankylosis reflected as complete bony replacement of the sacroiliac joints.
 |
| FIGURE 286-2 Bilaterally symmetrical sacroiliitis in ankylosing spondylitis. |
Radiographs of the spine may reveal squaring of the vertebral bodies (loss of the normal anterior concavity of the lumbar vertebra) and “shiny corners” (subchondral sclerosis at the upper edge of the vertebral body), both being manifestations of enthesitis. Syndesmophytes, which represent marginal bridging of the vertebrae (Figs. 286-3 and 286-4), eventually develop and make the diagnosis clear. Because ankylosis of the apophyseal joints may occur without syndesmophyte formation, it is important to assess the posterior joints on the lateral lumbosacral spine views, as well as the anterior margin of the vertebrae. Eventually, the changes may result in a “bamboo spine,” so called because the bridging syndesmophytes can mimic the appearance of bamboo. It is now appreciated that osteoporosis is a significant feature of AS, probably reflecting both the local chronic inflammation and the abnormal biomechanical loading of the vertebrae as the disease progresses.
Differential Diagnosis
The differential diagnosis of AS includes the other SpAs; osteitis condensans ilii; diffuse idiopathic skeletal hyperostosis; the syndrome of synovitis, acne, pustulosis, hyperostosis, and osteomyelitis (SAPHO); and some induced hyperostotic states (vitamin A intoxication, fluorosis). New bone formation occurs in degenerative disc disease, but the bulky, horizontal appearance of osteophytes is easily distinguished from that of syndesmophytes, and narrowing of the disc space is not a feature of AS.
The clinical course and severity of AS are highly variable. Inflammatory back pain and stiffness dominate the picture in the early stages, whereas chronic pain and deformity may develop over time. In both early and late phases of the disease there may be a significant impact of work disability on quality of life. In only a minority of patients may the full-blown picture of a bamboo spine eventually develop, but at present there are few variables that can reliably aid in prognosticating the course. In AS patients in whom new, refractory spinal pain develops, an intervertebral fracture should be considered and may occur after minimal trauma.
Additional late complications may include cauda equina syndrome, osteoporotic compression fractures, spondylodiscitis, and restrictive lung disease.
▪ Reactive Arthritis
Definition
ReA is an aseptic arthritis that occurs subsequent to an extra-articular infection, most typically of the GI or GU tract. In the former, the key pathogens are Salmonella typhimurium, Yersinia enterocolitica, Shigella flexneri, and Campylobacter jejuni. For the latter, Chlamydia trachomatis is the most common offender.
Epidemiology
The true incidence and prevalence of ReA are not well defined. In epidemics involving Salmonella (Chapter 329) or Yersinia (Chapter 333), it is estimated that ReA develops in 2 to 7% of infected individuals but in as many as 20% of B27-positive infected individuals. In such epidemic studies, B27 confers risk not only for the onset of arthritis but also for axial involvement and chronicity. The variability in the rate of ReA is determined by the heterogeneity of the cohorts reported, which introduces confounding variables of different genetic backgrounds in the population and different species of pathogens. Even in the setting of an epidemic point source outbreak, the inoculum varies widely among the exposed individuals, and the genetic makeup of the population at risk (e.g., the prevalence of B27) may differ greatly between different studies. Case ascertainment and relative risk are even more difficult to determine for post-Chlamydia ReA. Young adults in the United States have a high prevalence of asymptomatic Chlamydia carriage in the GU tract, and establishing a causal link between Chlamydia and synovitis can be difficult. Nevertheless, it is with Chlamydia that ReA has been most intensively studied.
Pathobiology
Although immunofluorescence studies have identified bacterial antigens in the joints of patients with ReA after both GI and GU infections, it is primarily in post-Chlamydia ReA that polymerase chain reaction results on synovial tissues have most consistently been positive, thus suggesting that viable Chlamydia may persist in the joints of such patients, albeit in a metabolically altered state.
Typically, the arthritis has its onset 1 to 3 weeks after the GI or GU infection, but the temporal details are often difficult to define precisely.
Although the definition of aseptic arthritis after an extra-articular infection may include a broader range of pathogens (e.g., Chlamydia pneumoniae), sites of infection (e.g., streptococcal pharyngitis), and types of infections (e.g., Giardia infections of the GI tract), these clinical scenarios have not generally been included in the ReA diagnosis. They lack the other associated clinical features of the SpA group of diseases, and they lack an association with B27.
Diagnosis
The pattern of joint involvement in ReA is one of asymmetrical oligoarthritis with a predilection for the lower extremity, a pattern shared by most SpA syndromes. Enthesitis may arise as Achilles tendonitis or plantar fasciitis. Dactylitis, arising as a sausage digit, may also be seen. Dactylitis is the net result of inflammatory changes affecting the joint capsule, entheses, periarticular structures, and periosteal bone. Sacroiliitis may be seen in the acute phase, but radiographic changes are seen largely in the patients with a more chronic course.
When ReA is accompanied by certain extra-articular features such as urethritis, conjunctivitis, or mucocutaneous lesions, the term Reiter's syndrome has been applied historically, but this term is not used any more. The urethritis may be manifested as dysuria or discharge and the rash as circinate balanitis, which appears as vesicles or shallow ulcerations on the glans penis. Painless lingual or oral ulcerations may also be seen. The fact that the cervicitis may be less symptomatic may in part account for the underdiagnosis in women. The classic skin manifestation of ReA is keratoderma blennorrhagicum, a painless papulosquamous eruption on the palms or soles (Fig. 286-5). Occasional nail dystrophy with pitting and onycholysis or subungual keratosis can be seen. The conjunctivitis can be bilateral and painful, in contrast to the acute anterior uveitis that can also be seen in this setting, which tends to be painless and unilateral.
 |
| FIGURE 286-5 Keratoderma blennorrhagicum of the feet in reactive arthritis. |
Radiographic findings in ReA can be seen in the involved peripheral joints, with soft tissue swelling and juxta-articular osteopenia being the early findings. Areas of periostitis and new bone formation may develop in peripheral joints. When sacroiliac changes are seen, they are typically asymmetrical, in contrast to the symmetrical pattern seen in AS (Fig. 286-6). In the chronic phase, syndesmophytes may develop, but they are described as bulky, nonmarginal, often asymmetrical formations that differ from the classic syndesmophytes of AS. The frequency with which AS can evolve into bona fide AS has not been resolved definitively.
 |
| FIGURE 286-6 Bilaterally asymmetrical sacroiliitis in reactive arthritis. Erosions, pseudowidening, and ileal sclerosis are present. |
Differential Diagnosis
The most important differential diagnosis for such reactive arthropathies is septic arthritis. Both Yersinia and Salmonella can induce septic arthritis, so an appropriate culture of synovial fluid should precede the diagnosis of ReA whenever possible. The course of ReA is variable, and few prognostic markers are available for the clinician to predict the course in any individual case. The majority of patients have an initial episode lasting 2 to 3 months, but synovitis may persist for a year or longer. In one 5-year follow-up of a point source cohort of post-Salmonella ReA, 20% of patients had ongoing inflammatory joint disease at this time, and some degree of functional disability was observed in 30% of patients 5 years after the onset of disease.
▪ Reactive Arthritis and Human Immunodeficiency Virus
An aggressive form of SpA may be seen in patients who are concomitantly infected with human immunodeficiency virus (HIV) (Chapter 409). There is no increased frequency of ReA in patients infected with HIV, but HIV may alter the course of these arthropathies, with a tendency for a more aggressive and more refractory course of joint disease. Aggressive skin and joint disease may be seen in patients in whom PsA develops in the setting of HIV infection. Most North American patients with the HIV-ReA constellation are B27 positive, but studies of comparable patients in Africa have found a sizable B27-negative component in such groups of patients. The arthritis in these patients falls into two clinical patterns: (1) an additive, asymmetrical polyarthritis or (2) an intermittent oligoarthritis that most commonly affects the lower extremities. Enthesitis, fasciitis, conjunctivitis, and urethritis can all be seen in such patients. Sacroiliitis can occur, although extensive spinal syndesmophyte formation is not common.
▪ Psoriatic Arthritis
Epidemiology
PsA develops in 5 to 7% of patients with psoriasis. Although most cases arise in patients with established cutaneous disease, some patients (particularly children) have arthritis that antedates the appearance of the skin lesions. Although the extent of psoriatic skin disease correlates poorly with the development of arthritis, the risk for PsA increases with a family history of SpA. The age at onset can range from 30 to 55 years, with an equal predilection for women and men. Psoriatic spondylitis has a slight male preponderance.
Pathobiology
The genetic associations with PsA are complex. Psoriasis itself is associated with HLA-B13, HLA-B16, HLA-B17, and HLA-Cw6. By contrast, HLA-B39 and HLA-B27 have been associated with sacroiliitis and axial involvement. No etiologic agent has been proved in PsA, although some investigators have proposed that the disease process represents ReA in response to cutaneous bacteria. The histopathology of the synovitis of PsA is comparable to that of the other SpAs, with the absence of local production of immunoglobulin and rheumatoid factor being differentiating features from RA. There is the potential for aggressive osteolysis, fibrous ankylosis, and heterotopic new bone formation to occur in PsA. As mentioned earlier, the coexistence of HIV and PsA seems to set the stage for an aggressive course of joint destruction in some patients.
Diagnosis
PsA has a variable manifestation and disease course, but several clinical patterns have been identified in prospectively monitored cohorts of patients. The clinical subsets are not mutually exclusive, nor are they static over time. The most common form, in which 30 to 50% of patients are affected, is an asymmetrical oligoarthritis that may involve both large and small joints. Dactylitis, arising as sausage digits, can be seen in fingers and toes and actually represents an enthesitis. In the second subset there is selective targeting of the distal interphalangeal joints, seen in 10 to 15% of patients. These changes are strongly associated with nail dystrophy, of which the features are onycholysis, subungual keratosis, pitting, and oil drop–like staining (Fig. 286-7). The third subset (15 to 30% of patients) has a symmetrical polyarthritis that mimics RA in many ways except for the absence of rheumatoid nodules and rheumatoid factor. The fourth clinical variant is psoriatic spondylitis, which occurs in 20% of patients; 50% of such patients are B27 positive. Finally, arthritis mutilans (5% of patients) is a destructive, erosive arthritis that affects large and small joints. It can be associated with marked deformities and significant disability.
 |
| FIGURE 286-7 Nail pitting in psoriasis. The pits are more discrete and regular compared with pits affecting the nail plate in dermatitis. |
Radiographic changes in PsA involve soft tissue swelling (particularly in the case of dactylitis), erosions, and periostitis. Axial involvement may lead to the appearance of asymmetrical sacroiliitis with syndesmophytes that are bulky, asymmetrical, and nonmarginal. The classic “pencil-in-cup” deformity may be seen in patients with distal interphalangeal joint disease or arthritis mutilans. Acro-osteolysis is noted in a minority of patients and reflects an aggressive erosive process.
Differential Diagnosis
The diagnosis of PsA depends on finding the typical skin or nail changes in association with one of the articular variants described previously. The differential for the skin lesions can include seborrheic dermatitis, dyshidrotic eczema, fungal infection, keratodermia blennorrhagica, and palmoplantar pustulosis.
▪ Enteropathic Arthritis
Definition
EA refers to the arthritis associated with Crohn's disease (CD) or ulcerative colitis (UC) (Chapter 145) (Table 286-4).
Click here to view this table....
Pathobiology
All extra-enteric manifestations, including arthritis, occur more commonly in CD than in UC. Peripheral arthritis occurs in 10 to 20% of CD patients and in 2 to 7% of UC patients. This pattern of arthritis occurs more commonly in patients with other extra-enteric features (e.g., erythema nodosum, iritis). It is typically an inflammatory, nonerosive polyarthritis, predominantly of large joints. In general, the clinical activity of the peripheral arthritis parallels the activity of the gut inflammation, and measures that control the GI disease usually control the joint disease concomitantly. The peripheral arthritis of EA is not associated with B27.
In contrast, the sacroiliitis or spondylitis of EA follows a pattern in which the joint inflammation waxes and wanes independently of the bowel inflammation. Axial disease occurs in 2 to 7% of CD patients and 2 to 7% of UC patients. HLA-B27 is found in 50% of patients with axial arthritis. The course tends to be chronic, as opposed to the transient course of peripheral arthritis.
The association of bowel inflammation and arthritis is supported by ileocolonoscopic studies in which subclinical inflammation of the bowel covering the entire spectrum of the SpAs has been demonstrated. Histologic evaluation demonstrates that changes of acute ileitis are seen in postdysenteric ReA whereas chronic inflammatory changes are more likely to be seen in patients with AS. As mentioned earlier, the abnormalities in the bowel of B27 transgenic rats have strong similarity to lesions of CD, and a germ-free environment minimizes inflammatory changes in both the gut and the joints. This finding argues that altered bowel permeability, with enhanced bacteremia or antigenemia, may provide the link in both cases.
Diagnosis
It is important to recognize that the musculoskeletal features of EA may precede any GI symptoms or signs. Conversely, the diarrhea preceding the onset of peripheral or axial arthritis in a young patient could as likely represent a food-borne pathogen (such as Salmonella or Yersinia) with secondary ReA as inflammatory bowel disease and accompanying EA. In the initial assessment of such a patient, it is important to carry out careful and complete stool cultures. If the GI symptoms persist, diagnostic colonoscopy is often required to resolve the issue.
▪ Undifferentiated Spondyloarthropathy
Despite careful clinical and radiographic assessment, there are still a substantial number of patients who do not fall easily into one of the classic diagnostic subsets of SpA outlined previously. These patients are often defined as having USpA with peripheral enthesitis, with asymmetrical arthritis or sacroiliitis, or with iritis in the absence of identifiable antecedent infection or concurrent inflammatory bowel disease or psoriasis. The natural history of USpA has not been well defined, and case heterogeneity and diagnostic dilemmas plague a systematic or multicenter approach to the problem. When the clinical course has been examined, a number of such patients finally meet the diagnostic criteria for AS, but many remain with a distinct USpA pattern for prolonged periods.
|
|