Are early arthritis clinics necessary?

Best Practice & Research Clinical Rheumatology Vol. 19, No. 1, pp. 1–17, 2005 doi:10.1016/j.berh.2004.08.001 available online at http://www.sciencedirect.com

1 Are early arthritis clinics necessary? Mark A. Quinn

MBChB, MRCP

Lecturer in Rheumatology

Paul Emery* BSC, MBBS, FRCP, MD Professor of Rheumatology Academic Unit of Musculoskeletal Disease, Department of Rheumatology, Leeds General Infirmary, 1st Floor, Great George Street, Leeds LS13EX, UK

Landmark studies published in the 1980s were the first to reveal the long-term consequences of rheumatoid arthritis (RA). Instead of the benign outcomes previously reported from early population studies, disability, deformity and excess mortality were evident. At this time, the conventional pyramid approach was the standard for therapeutic intervention. Patients were initially treated with non-steroidal anti-inflammatory drug with rest and splinting a approach. Disease-modifying anti-rheumatic drugs and corticosteroids were reserved for patients with joint damage and disability, who had ‘earned their treatment’ and the perceived risk of toxicity. Thus irreversible damage and disability occurred prior to effective therapy being instituted, with the consequences of poor outcomes. The late 1980s saw the introduction of the concept of early intervention in RA and, shortly after, the introduction of specialist clinics for early assessment of patients with inflammatory arthritis (IA), so-called ‘early arthritis clinics’ (EACs). Such clinics, initially in large research units, targeted patients with early RA or IA with the potential to evolve to RA, with the aim of early case definition and treatment. After all, the common sense approach to treatment of a chronic inflammatory, destructive condition would be to treat it effectively from its onset, prior to the development of irreversible damage. The detailed documentation undertaken in these clinics subsequently provided much information regarding persistence and prognosis in early IA. Since their initial introduction, EACs have become commonplace, not only in academic units, but also kas part of clinical service provision in many institutions. This review details what an EAC is, who should be referred and when, and the benefits and potential future benefits of their introduction. Key words: early arthritis clinics; rheumatoid arthritis; undifferentiated arthritis.

* Corresponding author. Tel.: C44-113-392-5068; Fax: C44-113-392-2896. E-mail address: [email protected] (P. Emery). 1521-6942/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

2 M. A. Quinn and P. Emery

WHAT IS AN EARLY ARTHRITIS CLINIC? The first early rheumatoid arthritis (RA) clinical cohort was established in Bath, UK in 19571 with further cohorts established in Middlesex, UK2 and Memphis, Tennessee, USA3 in the 1960s. However, the concept of early arthritis clinics (EACs) only emerged in the late 1980s4 with growing evidence for the destructive and debilitating natural history of RA.5,6 Conceptually, there were obvious clinical and academic benefits from this approach. From the clinical perspective, if effective therapy could be introduced prior to the development of presumed irreversible damage, outcomes could be improved. This was fuelled by the ‘window of opportunity’ hypothesis for therapeutic intervention in RA.7 The hypothesis was based on the existence of a time frame within which there is potential for a disproportionate response to therapy, resulting in long-term benefits or, more importantly, the chance of ‘cure’. From the academic perspective, in order to gain as much information as possible about a condition, it must be studied from its inception and followed over time. Through early case definition, there could be potential for advances in disease aetiology, pathogenesis and therapeutics. Initially, EACs were confined to clinical research units, but through the 1990s, EACs have emerged as part of general rheumatology services worldwide. An EAC should, by definition, offer early assessment of patients with either signs or symptoms of inflammatory arthritis (IA), i.e. patients with the potential to develop RA. Ideally, the clinic should be able to offer assessments of patients within 2 weeks of referral to the clinic, and, almost as importantly, offer early review to assess response to intervention. The exact structure and services offered at such clinics are varied and frequently determined by availability of resources. Several models exist including the Birmingham8, Leeds9 and Swedish set-ups.10 Each offers a slightly different approach based around a unifying concept of early case definition and intervention. The Birmingham system was the original model, much copied and altered. The service allowed direct referral via long-range pager to a rheumatologist for all patients with possible IA. Urgent patients could be seen on the same or next working day, and all patients could be seen within 2 weeks of referral. Assessments were undertaken in a purpose-built ward with staff including rheumatologists, trainee rheumatologists, a specialist nurse, a physiotherapist and an occupational therapist (OT), and with the capacity to cope with medical students and general practitioners for educational purposes. The Leeds system further developed the Birmingham model with an increased emphasis on imaging, with addition of ultrasound (US) and magnetic resonance imaging (MRI) assessment of joints to dual energy X-ray absorptiometry (DEXA) and X-ray for all patients. The Karolinska system in Sweden is based on a daycare programme where all patients receive comprehensive assessment and education by, in turn, a junior doctor, physiotherapist, social worker, OT and co-ordinating nurse prior to a prognostic and therapeutic decision from the senior physician. The Department of Rheumatology of the Leiden University Medical Centre established an EAC service in 1993 with close collaboration with their clinical scientist colleagues.11 The clinical service is similar to that provided by the Birmingham model, and links to basic science have had additional benefits with research in genetics11 and cellular immunology12, resulting in a greater understanding of the pathophysiology of early IA.

Are early arthritis clinics necessary? 3

Table 1. Difficulties encountered in assessment of very early patients for referral to an early arthritis clinic. Patients present late to primary care Presentation often insidious Non-steroidal anti-inflammatory drugs may mask clinical signs Pathagonomic features of disease are not obvious early Investigations may be normal

WHO SHOULD BE REFERRED? There are inherent difficulties in defining the patient population that is most appropriate for EAC assessment (Table 1). Of principal importance is the relationship with primary care and the awareness of the service provision offered. A letter detailing the aims and objectives of the service with the emphasis on improving patient outcomes is often useful and should be distributed to primary care colleagues. Reminding them of what services are available, developments in the field, and which patients warrant early assessment on a regular basis (annually may be sufficient) may also aid smooth running of the clinic. Unless guidance is offered, EAC services may be exploited for non-urgent cases, especially where conventional waiting lists are lengthy. This may, in turn, compromise the quality of care subsequently offered to true inflammatory patients where early intervention has the greatest potential for benefit. Ideally, all patients referred for assessment would have IA. In this case, the role of the EAC would simply be to determine persistence, prognosis and therapy. The reality is that only 56–70% of referrals will be IA in the average EAC.13,14 This highlights the fact that early IA can be very difficult to assess in its earliest stages, but also the difficulty in determining disease phenotype from the primary care referral letter. Strategies to improve the sensitivity of the referral system include direct pager access for all referrals where patients may be discussed prior to assessment8, or a preformatted referral letter system where key parameters such as early morning stiffness (EMS), joints affected, family history, non-steroidal anti-inflammatory drug (NSAID) response etc. are clearly documented (Table 2). In practice, such methods may be useful where resources are limited. Even then, much IA will be undifferentiated15, subclinical disease may be present in asymptomatic joints16,17, and investigations such as rheumatoid factor (RF), C-reactive protein (CRP) and X-rays may be normal in 35, 70 and 80% of cases, respectively.18 Also, pathagonomic features, i.e. rheumatoid nodules or erosions, etc. occur late in the disease process, and therefore clinical examination findings are often non-specific. Table 2. Symptoms and signs of patients to consider for assessment at an early arthritis clinic. Joint swelling Symmetrical symptoms Metacarpophalangeal and metatarsophalangeal involvement Significant early morning stiffness Good response to non-steroidal anti-inflammatory drugs Family history of rheumatoid arthritis

4 M. A. Quinn and P. Emery

It is important for primary care physicians to be aware of the limitations of laboratory and X-ray tests, and therefore avoid falsely reassuring themselves that pathology is absent. Primary care physicians should also be aware of key symptoms and signs that may be factors in determining persistence, and hence offer review or referral as appropriate. The evidence for early referral recommendation has been reviewed extensively.19 It would appear that there is an indication for rapid referral if the patient presents with three or more swollen joints, metacarpophalangeal (MCP) or metatarsophalangeal joint involvement and EMS of greater than 30 minutes.

WHEN SHOULD PATIENTS BE REFERRED? In order for the maximum potential to be achieved from EACs, all patients with the potential to develop RA should be referred as soon as the condition is suspected in primary care. However, two critical areas of delay can be identified. The first is the time it takes the patient to recognize the symptoms of IA and seek appropriate medical advice. The second is the time it takes the primary care physician to make the diagnosis, as the earliest features of RA may be non-specific and inconclusive.18 This may be further confounded by a profound response to NSAIDs, which may be purchased over the counter or provided in primary care, and can offer false reassurance of cure in the very early stages of IA.20 The best way to increase patient awareness is unclear. Patient education leaflets are usually available from source of care and so may have limitations. The Internet has growing potential, yet governance and quality control are varied. Movements such as the Bone and Joint Decade21 are undoubtedly beneficial and raise the profile of musculoskeletal complaints, as well as the valuable contribution from arthritis-based charities such as the Arthritis Research Campaign (www.arc.org.uk) and the National Rheumatoid Arthritis Society (www.rheumatoid.org.uk) in the UK. However, there is no evidence that such disease promotion has an impact on time to presentation to medical care for patients with new-onset IA. Creating awareness amongst primary care physicians may offer the optimal approach, as outlined in the previous section, with continued communication vital for successful running of an EAC.

WHAT SHOULD THE CLINIC OFFER? Disease assessment As already mentioned, the EAC should offer rapid access for all patients with suspected RA (Table 3). Patient assessments should be comprehensive and provide a full diagnostic and prognostic evaluation. The more complete the initial review, the more likely it is that intervention or reassurance can be provided reliably. To allow objectivity, clinical assessments may be quantified, with the use of joint counts or composite scores such as the disease activity score (DAS).22,23 Additional information may be acquired with questionnaires such as the Health Assessment Questionnaire (HAQ) for assessment of functional impairment24, the Rheumatoid Arthritis Quality of Life Questionnaire (RAQoL)25, and the Rheumatoid Arthritis Disease Activity Index (RADAI)26, among others. Such measures are not essential yet may provide objective

Are early arthritis clinics necessary? 5

Table 3. What should an early arthritis clinic offer? Rapid access Full diagnostic/prognostic assessment (off non-steroidal anti-inflammatory drugs) Early therapeutic intervention Access to allied health professionals, e.g. physiotherapy, occupational therapy and podiatry services Patient education Early re-assessment

evidence of disease progression or resolution on review. Simply documenting a good history can perform this task, with accurate detailing of distribution of symptomatic joints, duration of EMS, response to NSAIDs, any prodromal illness, and family and personal past medical histories. The examination findings are also important and a comprehensive examination of all systems should be performed. It should be remembered that articular symptoms may be the presenting manifestation of many infectious, inflammatory or malignant conditions. Documentation should include description of any skin rashes, weight, height, urinalysis and blood pressure measurement. Further investigations need to be tailored to the individual, although most suspected IA cases will warrant a measure of acute phase response, basic serology including RF and antinuclear antibodies, and assessment of renal and liver function and serum urate. If symptomatic, x rays of hands and feet should be performed and, in some centers, chest X-rays are performed routinely. The latter can be justified to exclude significant pulmonary pathology contributing to the presenting condition and also as a screening test prior to therapeutic intervention. Approximately 0.5% of all presenters to the Leeds EAC, seeing 1200 new patients a year, receive a diagnosis of carcinoma from chest X-ray screening (unpublished data). More specific testing may be directed by clinical presentation including cultures where infection may be suspected, virology, serology for atypical infections, serum angiotensin-converting enzyme, specific autoantibodies and polymerase chain reaction for Chlamydia. Assessment of human leukocyte antigen (HLA) genotype status and anti-cyclic citrullinated peptide (antiCCP) antibodies may also be useful (see later). However, the prognostic utility to date does not appear as strong as that for RF.27 Composite scores for both persistence and severity are also available: the Birmingham-developed Persistent Inflammatory Symmetrical Arthritis (PISA) score18 and the Netherlands-developed seven-parameter model28 (Tables 4 and 5). Both scoring systems use a combination of factors already described to assess a patient’s risk of poor prognosis and persistence, respectively. Imaging assessment X-rays provide the mainstay of imaging in EACs and have been used conventionally for prognostic and outcome assessment. However, up to 75% of patients will have normal radiographs at first presentation.13 The utility of X-rays therefore lies in assessment of disease progression29 and in detecting pathagonomic lesions where abnormalities are present. X-rays tell us little about the extent of inflammatory disease and so alternative modalities which simultaneously assess bone and soft tissues are of greater benefit in the initial assessment of IA.

6 M. A. Quinn and P. Emery

Table 4. The persistent inflammatory symmetrical arthritis (PISA) scoring system. Prognostic factor

Score

Rheumatoid factor positive HLA DR1/DR4/DR10 C-reactive protein O20 mg/l Female sex HAQ* raw scoreO4%11 HAQ* raw scoreO11 Total score

1 1 1 1 1 2 6

Patient score

HAQ, health assessment questionnaire. A score of R3 indicates poor prognosis.

Imaging of involved joints soon after disease onset with US or MRI may be useful for diagnostic or prognostic purposes, and serial dual-energy X-ray absorbtiometry (DXA) has been used as an outcome measure in EACs. The merits of these modalities are discussed later in this review. Access to allied health professionals EACs should not only provide a comprehensive medical assessment, but also an opportunity for assessment and education from allied health professional colleagues. The medical consultation can be complex and requires an explanation of the diagnosis and prognosis, therapeutic options available, possible side-effects and monitoring requirements, often with time at a premium. RA is a physical illness with frequently occurring psychological symptoms. Adopting a team approach allows multidisciplinary education from nurse specialists, physiotherapists, podiatrists and OTs, providing a holistic approach. The nurse specialist plays a central role in patient management, often co-ordinating treatment by other medical staff and therapists. The increasing use of combination disease-modifying anti-rheumatic drug (DMARD) therapy and biologics requires adequate patient counselling, education and monitoring for their safe administration. The nurse practitioner often provides vital psychological support for patients at all stages of their disease. The physiotherapist improves patients’ mobility, independence and self-esteem. Several studies have shown that the majority of RA sufferers will show an improvement in function and a lessening of pain if they exercise30, and a prolonged course of hand exercises induced significant improvements compared with a control group.31 Hydrotherapy may improve physical and psychological Table 5. The Netherlands model to predict persistent (erosive) disease. Prognostic factor Symptom duration at first visit Early morning stiffness for more than 60 minutes Arthritis in more than three joints Bilateral compression pain in metatarsophalangeal joints Rheumatoid factor positivity Anti-cyclic citrullinated peptide antibody positivity Erosions on hand or feet X-rays

Are early arthritis clinics necessary? 7

outcomes32, and maintain grip strength and functional outcome when compared with controls.33 The OT promotes independent living, often facilitated by the use of aids and adaptations in the home and workplace. Although many such interventions seem appropriate and use common sense, there is little supportive evidence. In addition, OTs can give advice regarding joint protection. One recent study showed significant improvements in pain, disease status and functional ability in a group receiving joint protection advice compared with those receiving standard RA education.34 Historically, the podiatrist has been utilized late in RA when deformity and damage have required orthotic intervention or specialist footwear. However, intervention can improve patient pain and mobility, and should be considered early for those with affected joints. The prevention and early treatment of foot ulceration is also extremely important, especially in the era of biological therapies.

WHAT DATA SHOULD BE COLLECTED? The information collected from the EAC will vary between clinically-orientated and research-orientated services. It has, however, been proposed that a common core set of data should be collected in order for comparisons between series to be made.35 The proposed ‘standard protocol to evaluate rheumatoid arthritis’ comprises five pages of data collection that may be completed in 15–30 minutes.36 The data collected include clinical features such as the American College of Rheumatology (ACR) classification criteria37, medications used for RA, joint counts, HAQ and radiographic scoring. Such a proposal has obvious advantages for research purposes, yet for smaller clinical units, the additional time required for data collection may be outwith the time constraints of service provision. Within hardpressed clinical services, the data collected will be less complete. However, a detailed baseline prognostic assessment may still save time in the long term, as management may be altered according to findings. Such assessment would include parameters from the prognostic models in the previous section. Also of use may be selfreported questionnaires that correlate with parameters such as joint counts, radiographic scores and laboratory tests, and require less physician time.38

WHAT CAN BE/HAS BEEN GAINED FROM AN EARLY ARTHRITIS CLINIC? Gains for the patient The principal objective of the early arthritis movement was to improve patient assessment and outcome. A review of primary trial data from 14 randomized controlled trials of DMARD therapies in RA indicated disease duration as a significant determinant of response to therapy, with patients with short disease duration responding more favourably.39 Patients presenting with a disease duration of less than 1 year showed response in 53% of cases, whereas patients presenting with a disease duration of 1–2 years, 2–5 years, 5–10 years and O10 years showed diminished response with time. This emphasizes the importance of early intervention. One of the key delays in assessing patients with IA described previously was time to referral to secondary care. Research suggests that there has been a reduction in the time from

8 M. A. Quinn and P. Emery

symptom onset to referral to secondary care, with average time to referral being 21 months before 1986 and 4 months in 1997.40 Another study compared the performance of a routine clinic with an EAC, and demonstrated a median reduction in referral time to a rheumatologist of at least 3 months.13 From the same study, 70% of all patients seen could be accurately diagnosed within 2 weeks of assessment; where RA was diagnosed, the diagnosis was unchanged at 1-year follow-up. It would appear therefore that early arthritis awareness has reduced referral times, and provision of an EAC service allows earlier and successful diagnosis. This therefore offers the patient the best chance of responding to therapy.39 Perhaps the greatest benefit has been in prognostic assessment with earlier differentiation between self-limiting and persistent disease.14,41,42 Further benefits gained by the patient are also of benefit to the physician in the development of a greater understanding of prognostic factors and improved outcome in terms of function, XX-rays and mortality. These will be discussed below. Gains for the physician The body of evidence for early intervention and assessment in early RA continues to grow and is driven by EAC-orientated research.43 Two principal areas of clinical research in RA are apparent: (i) assessment of prognostic factors; and (ii) outcome assessment in therapeutic studies. However, much has also been learned from studies in non-RA, early IA such as oligo- and undifferentiated arthritides. Assessing disease persistence With earlier case definition and treatment, it is important to determine which patients with non-specific disease phenotypes or very early presentation will have self-limiting or persistent disease. In the largest published series of undifferentiated arthritis patients, a stepwise logistic model was used to determine factors predisposing to development of RA.44 Only 1987 ACR (and ARA 1958 definite) criteria RA, RF positivity and disease duration were significant factors. In a further study where factors predicting persistent or selflimiting symmetrical polyarthritis were analysed in an EAC, RF positivity emerged as the strongest variable, accounting for 45% of the variability in outcome.45 Adding an erythrocyte sedimentation rate O30 mm/hours improved this to 49%, making patients 11 times more likely to have persistent disease. In assessing affected joints, the relative risk (RR) for persistence was highest for wrist swelling (RR 3.04, confidence intervals 1.77–5.22, sensitivity 65% and specificity 83%), with MCP swelling and proximal interphalangeal (PIP) swelling close behind. It is interesting that wrist swelling and tenderness are identified as important, as this was also a finding of the ACR when developing the classification criteria.37 Further information can be gathered from intervention studies. Green et al studied 63 patients with mild early IA, defined as synovitis in two or more joints with !12 months of symptoms, longitudinally for 6 months.41 The aims of the study were to determine the factors that predict persistence of inflammation 6 months following corticosteroid therapy, and to assess the ability of the ACR criteria to select these patients. In this group, the best predictor of persistence was disease duration R12 weeks irrespective of disease phenotype at first presentation. With disease !12 weeks, the chance of remission was increased five-fold. In analysing the seronegative

Are early arthritis clinics necessary? 9

group for RF separately, patients in possession of the shared epitope (SE) were significantly more likely to have persistent disease. No RFCSE-positive patients entered remission. The ACR classification criteria were only predictive after 12 weeks’ duration. A further study examined the use of a similar protocol of intra-articular corticosteroid injections in patients with an early oligoarthritis, i.e. four or less joints, followed by an early review to assess for the presence of persistent synovitis.42 This paper showed that at least 50% of patients with oligoarthritis have complete response at 2 weeks and the best predictor of response at 12 and 26 weeks is the presence or absence of synovitis on examination at 2-weeks follow-up. Failure to respond by 2 weeks indicated a high likelihood of persistent disease and the need for DMARD therapy. Also, in a large cohort of undifferentiated IA with hand symptoms, the best predictor of persistent disease at 1 year was persistence or resolution of synovitis at 3 months after use of a standard treatment protocol.14 Interestingly, undifferentiated arthritis with hand involvement appears to carry a poor prognosis compared with other undifferentiated cohorts, and this has been confirmed by others.46 There are no therapeutic guidelines for the treatment of patients identified in these studies. A large proportion of these patients may have self-limiting disease and, traditionally, management has involved a watch-and-wait policy, perhaps leading to a delay in intervention. From the above data, a more informed decision regarding the use of DMARD therapy may be possible following a single dose of corticosteroid, and decisions may be made with a degree of certainty within the first few months of assessment. The alternative approach is to look at patients more likely to enter spontaneous early remission. Remission rates vary according to prognostic indicators. A favourable outcome is associated with seronegativity for RF and fewer active joints at baseline in early RA.47 Other studies have shown a relationship with male sex, younger age and absence of erosions with remission rates.48 Prognostic assessment Once persistence is established, it is important to assess disease severity. EAC research has added to our knowledge of prognostic factors in RA. C-reactive protein Although time-integrated inflammation has been known to carry a poor prognosis in RA, EAC research has further emphasized the importance of this. If CRP is used as a surrogate marker of inflammation, the principal outcome measures of progression of radiographic damage49–51, loss of function52 and bone mineral density53 correlate well with persistent elevation of CRP. Equally, suppression of CRP results in at least stabilization of the respective parameters.54–56 This has re-inforced the paradigm, inflammation!timeZdamage, and provides further evidence for the importance of early case definition and adequate disease control. However, a high CRP at presentation may confer good prognosis if patients are evaluated at a very early stage (!12 weeks of symptoms) and this is likely to be accounted for by patients with acute-onset reactive arthritis who have a more favourable outcome than the more insidious onset RA.41

10 M. A. Quinn and P. Emery

Rheumatoid factor RF is perhaps the most consistent prognostic marker in early IA and correlates with both radiographic and functional outcome in the majority of case series. In a large early polyarthritis cohort of 486 patients, RF seropositivity predicted both radiological erosions and poor functional outcome, and seronegativity best predicted remission at 3 years.57 In early RA, RF seropositivity has been shown to have a relative risk for the development of erosions at 1 year of 13.49 (sensitivity 95%, specificity 39%).58 These data correspond with established RA studies where RF predicted both short-59 and long-term radiological damage60 and also functional disability.61 Since the introduction of isotype-specific assays for RF, a number of studies have been published concerning the clinical value of isotype analysis. All isotypes of RF have been shown to be associated with a more progressive disease course, but IgA RF appears to be more specifically associated with the development of erosions in early disease and a persistent disease course, although with a reduced sensitivity.62 RF seropositivity at the onset of RA is therefore associated with a more progressive disease course resulting in increased disability and radiological progression. However, there is some evidence to suggest that once disease is established, the differences between seronegative and seropositive patients may diminish.63 Anti-cyclic citrullinated peptide antibodies Recently, data have supported the use of serological tests for RA in diagnosis and prognosis. Anti-CCP antibodies are a collective term for antibodies that recognize filagrin. Not only have they shown a high specificity of 98%64, but also a prognostic utility in terms of radiographic damage.27 In addition, they have also shown discriminative ability in a large cohort of undifferentiated arthritis.12 A secondgeneration assay has shown specificity of up to 98% and sensitivity of 81%. Such findings mean that, as a screening test, anti-CCP antibodies are likely to provide more information than RF. The clinical role is yet to be confirmed, but it is likely to be of greatest use in seronegative patients, and for use in conjunction with RF as the latter has greater prognostic importance. Function By far the best predictor of functional outcome, measured using the HAQ, is baseline functional assessment.47 Using functional disability as the primary outcome measure at 1 year, the Norfolk Arthritis Register (NOAR) group has analysed predictors of disability in a cohort of 381 patients with inflammatory polyarthritis.65 The strongest predictors were a high baseline HAQ [likelihood ratio (LR)Z2.3], large joint involvement (LRZ2.2), female sex and longer disease duration (LRZ1.6). Also of importance is that reversibility of functional impairment may be lost with time. Patients treated early (!2 years) showed a significant improvement in function, using the HAQ, compared with patients with longer symptom duration.66 The reason for this is likely to be that disability in late disease is determined by joint damage rather than inflammatory activity, and is therefore less amenable to therapy.

Are early arthritis clinics necessary? 11

Imaging US and MRI have been used in the assessment of early IA. Both modalities have demonstrated greater sensitivity for detection of synovitis compared with clinical examination67, and also detection of erosions in MCP joints compared with conventional radiography.68 From longitudinal MRI studies in early RA, synovitis appears to precede bone oedema and subsequent erosions, with erosions only occurring in the presence of synovitis.69 More importantly, if synovitis is suppressed adequately using intra-articular corticosteroids, MRI shows reduced bone oedema and absence of new bony lesions over 3 months.69 It therefore appears that adequate suppression of synovitis prevents progression of bone damage. MRI erosions have also been shown to correlate anatomically with subsequent radiographic erosions and precede them by an average of 2 years70, and also predict long-term functional outcome (McQueen). It has also been described that the site of primary inflammation detected by MRI may have prognostic importance with respect to persistence, with capsular pathology having a good prognosis, whereas true intra-articular disease appears to predict persistence.71 In inflammatory oligoarthritis, US detects synovitis in more joints than is apparent on clinical examination, i.e. subclinical. To date these tools have been used in research centres, but there is a growing number of rheumatologists using US in clinical practice.72 Whilst MRI may offer the gold standard for imaging IA, it is limited in practice by cost, availability and time, and is limited to one joint area at any one sitting, whereas US is cheap, non-invasive, and allows for many joints to be assessed at any one time. The sensitivity of the latter, however, is operator dependent, but standards are now being defined (Brown) and courses teaching musculoskeletal US are now available through the major rheumatology societies including the British Society of Rheumatology, ACR and European League Against Rheumatism (EULAR).73,74 DXA has also been used in EACs. There is good correlation of bone density loss in the spine with disease activity53 and also in the hand.54 DXA also has the advantage of assessing the effects of persistent inflammation and toxicity of therapies, e.g. corticosteroids. Currently, this remains a research-based assessment tool. Gains by the researchers Much clinical scientific work has evolved from EACs. Through serial sample storage of urine, serum, plasma, whole blood and synovium, scientists have been able to combine laboratory research allied to long-term clinical outcomes. By studying the diseases from their inception, it is possible to identify pathagonomic features early. Studies using arthroscopy have confirmed imaging findings of subclinical synovitis when asymptomatic knees of newly diagnosed RA patients are examined.17 Further arthroscopy studies have identified distinct macroscopic vascular patterns seen in early RA and psoriatic arthritis.75 Histological examination of synovium in these patients has demonstrated differential expression of the pro-angiogenic factors vascular endothelial growth factor (VEGF)76 and matrix metalloproteinase-9 (MMP-9)77, and may explain the resulting differences in macroscopic appearance. This may therefore have future use for disease classification purposes. Factors have also been identified to differentiate between self-limiting and persistent IA from studies in synovium.78 Although not new, HLA typing remains a research parameter in most areas. Many studies have looked at the prognostic utility of the SE or HLA-DRB1*04 alleles. Several longitudinal studies suggest a correlation between possession of the SE and both

12 M. A. Quinn and P. Emery

radiographic progression58 and persistence.79 However, the literature can be conflicting.80 Such variability between studies can be explained by the selection of patients and new therapeutic approaches. Patients with early, untreated disease will differ from those with chronic disabling disease. For example, patients seen late in disease have already been selected for poor prognosis and subsequently the prevalence of the SE will be high, up to 80%. Many of these patients will also have poor prognostic clinical features and the comparator group without the SE will be small and therefore have little power to show a difference. Paradoxically, it has also been shown that possession of the SE is a predictor of a favourable response to triple therapy.81 The latter suggests that new approaches to treatment may influence the natural progression of RA. Subsequently, as therapies become more effective, the prognostic utility of the SE may diminish. Recent studies have consistently shown that the SE has particular prognostic utility in seronegative patients for both disease severity80,82 and persistence.79 Perhaps the greatest benefit from the EAC movement has been clinical intervention trials in well-defined early RA cohorts. Extensive clinical trials have compared strategies, individual agents and combinations of agents.43 Through iterative research, clinical outcomes are constantly improving. Research is now emerging with biological therapies83 and there may be inherent differences between the early use of these agents rather than reserving them for recalcitrant patients. Long-term data from the etanercept study shows greater long-term benefits than those seen at 1 year, and a pilot study has suggested sustained response for a further year after initial induction therapy.84 For health economic reasons, if biologics are to be used early in the disease process, EACs with their structured assessment and follow-up will be important for the purposes of clinical governance. Gains by the funder of health care It would appear common sense that an approach to the management of a patient that reduces long-term disability and joint damage would reduce long-term healthcare costs. A problem in researching such outcomes, however, is that most clinical intervention trials are short term and the majority of costs occur late, e.g. joint replacements, care provision. It is also difficult to quantify cost as this can be divided into direct (e.g. cost of treatments), indirect (e.g. work loss) and intangible/psychosocial (e.g. quality of life). Previously it has been reported that indirect costs were two to three times greater than direct costs85, but more recent research suggests that direct costs are catching up.86,87 A review of direct costs published in 2000 estimated the average annual medical cost to be $5720.88 Medication contributed between 8 and 24% of the total medical cost, doctor’s visits contributed 8–21% and hospital admissions contributed 8–88%. There is also a correlation of pain and functional status with cost.89 With the arrival of biological therapies, drug costs may increase significantly in the years to come, although early health economic studies suggest that cost-effectiveness ratios are within limits set for treatments to be recommended for use.90,91

SUMMARY The clinician is now better informed than ever regarding the assessment of early IA. Factors predicting persistence in early IA, in addition to new therapeutic options in

Are early arthritis clinics necessary? 13

previously unresearched areas, have been proposed. As well as assessing persistence where a diagnosis of RA is made, predicting severity is important for therapeutic decision making. Research tools, e.g. US and MRI, are being introduced into the clinic that may allow more specific case definition and severity assessment.

Practice points † features of successful EACs are: – public awareness – good relations and service awareness in primary care – ability to offer early patient assessment – comprehensive prognostic assessment – early re-assessment – continued development through audit and research

Research agenda † † † † †

prognostic factors in non-RA IA the importance of subclinical synovitis the optimal role of biological therapies in RA cytokine profiling and its role in disease assessment radiographic erosion healing

REFERENCES 1. Jacoby RK, Jayson MIV & Cosh JA. Onset, early stages, and prognosis of rheumatoid arthritis: a clinical study of 100 patients with 11 year follow-up. British Medical Journal 1973; 2: 96–100. 2. Fleming A, Crown JM & Corbett M. Early rheumatoid disease. 1. Onset. Annals of the Rheumatic Diseases 1976; 35: 357–360. 3. Masi At M-CJ, Kaplan SB, Feigenbaum SL & Chandler RW. Prospective study of the early course of rheumatoid arthritis in young adults: comparison of patients with and without rheumatoid factor positivity at entry and identification of variables correlating with outcome. Arthritis and Rheumatism 1976; 4: 299–326. 4. Emery P & Gough A. Why early arthritis clinics? British Journal of Rheumatology 1991; 30: 241–242. 5. Pincus T, Callahan LF, Sale WG et al. Severe functional declines, work disbaility and increased mortality in seventy-five rheumatoid arthritis patients studied over nine years. Arthritis and Rheumatism 1984; 27: 864– 872. 6. Scott DL, Symmons DP, Coulton BL & Popert AJ. Long term outcome of treating rheumatoid arthritis: results after 20 years. Lancet 1987; 1: 1108–1111. 7. Emery P. The optimal management of early rheumatoid disease: the key to preventing disability. British Journal of Rheumatology 1994; 33: 765–768. 8. Gough AKS. Disease Severity and Prediction of Outcome in Early Rheumatoid Arthritis, thesis. London 1993. 9. Quinn MA, Green MJ & Emery P. Evaluation and management of early inflammatory polyarthritis. In Hochberg MC, Silman AJ, Smolen JS, Weinblatt ME & Weisman MH (eds.) Rheumatology, third edn. St Louis, MO: Mosby, 2003, pp. 885–891. 10. Klaresklog L, Nordmark B & Lindblad S. On the organisation of an early arthritis clinic. Best Practice and Research. Clinical Rheumatology 2001; 15: 1–15.

14 M. A. Quinn and P. Emery * 11. van Aken JHM, van Bilsen CF et al. The Leiden Early Arthritis Clinic. Clinical and Experimental Rheumatololgy 2003; 21(supplement 31): S100–S105. 12. van Gaalen FA, Linn-Rasker SP, van Venrooij WJ et al. Autoantibodies to cyclic citrullinated peptides predict progression to rheumatoid arthritis in patients with undifferentiated arthritis: a prospective cohort study. Arthritis and Rheumatism 2004; 50: 709–715. * 13. Van der Horst-Bruinsma H et al. Diagnosis and course of early-onset arthritis: results of a special early arthritis clinic compared to routine patient care. British Journal of Rheumatology 1998; 37: 1084–1088. * 14. Quinn H et al. Prognostic factors in a large cohort of early undifferentiated inflammatory arthritis patients after application of a structured management protocol. Arthritis and Rheumatism 2003; 48: 3039–3045. 15. Schumacher HR. Early arthritis clinics. Much early arthritis is unclassified. Journal of Rheumatology 2002; 29: 2258–2259. 16. Wakefield RJ, Green MJ, Gibbon WW et al. High resolution ultrasound defined subclinical synovitis-a predictor of outcome in early oligoarthritis? Arthritis and Rheumatism 1998; 41: S246. 17. Kraan MC, Versendaal H, Jonker M et al. Asymptomatic synovitis precedes clinically manifest arthritis. Arthritis and Rheumatism 1998; 41: 1481–1488. 18. Emery P. Prognosis in inflammatory arthritis: the value of HLA genotyping and oncological analogy. The Dunlop Dotteridge Lecture. Journal of Rheumatology 1997; 24: 1436–1442. * 19. Emery M et al. Early referral recommendation for newly diagnosed rheumatoid arthritis: evidence based development of a clinical guide. Annals of the Rheumatic Diseases 2002; 61: 290–297. 20. Marzo-Ortega H, Green MJ, Karim Z et al. Non-steroidal anti-inflammatory drugs alter the presentation of early inflammatory arthritis. Rheumatology 2000; 39(supplement 1): 42. 21. Lidgren L. The Bone and Joint Decade and the global economic and healthcare burden of musculoskeletal diease. Journal of Rheumatology 2003; 67: 4–5. 22. Prevoo MLL, van’tHof MA, Kuper HH et al. Modified disease activity scores that include twenty-eight joint counts: development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis and Rheumatism 1995; 38: 44–48. 23. van der Heijde DMFM, van’t Hof MA, van Riel PL et al. Judging disease activity in clinical practice in rheumatoid arthritis: first step in development of a disease activity score. Annals of the Rheumatic Diseases 1990; 49: 916–920. 24. Fries JF, Spitz P, Kraines RG & Holman HR. Measurement of patient outcome in arthritis. Arthritis and Rheumatism 1980; 23: 137–145. 25. de Jong Z, van der Heijde D, McKenna SP & Whalley D. The reliability and construct validity of the RAQoL: a rheumatoid arthritis—specific quality of life instrument. British Journal of Rheumatology 1997; 36: 878–883. 26. Fransen J, Langenegger T, Michel BA & Stucki G. Feasability and validity of the RADAI, a self-administered rheumatoid arthritis disease activity index. Rheumatology 2000; 39: 321–327. 27. Kroot EJ, de Jong BAW, van Leeuwen MA et al. The prognostic value of anti-cyclic citrullinated peptide antibody in patients with recent-onset rheumatoid arthritis. Arthritis and Rheumatism 2000; 43: 1831– 1835. 28. Visser H, le Cessie S, Vos K et al. How to diagnose rheumatoid arthritis early: a prediction model for persistent (erosive) arthritis. Arthritis and Rheumatism 2002; 46: 357–365. 29. van der Heijde D. Plain X-rays in rheumatoid arthritis: overview of scoring methods, their reliability and applicability. Balliere’s Clinical Rheumatology 1996; 10: 435–453. 30. Lyngberg KD-SB & Halskov O. The effect of physical training on patients with rheumatoid arthritis: changes in disease activity, muscle strength and aerobic capacity. A clinically controlled minimized crossover study. Clinical and Experimental Rheumatololgy 1988; 6: 253–260. 31. Brighton SW, Lubbe JE & van der Merwe CA. The effect of a long-term exercise programme on the rheumatoid hand. British Journal of Rheumatology 1993; 32: 392–395. 32. Hall J, Skevington SM, Maddison PJ & Chapman K. A randomised and controlled study of hydrotherapy in rheumatoid arthritis. Arthritis Care Research 1996; 9: 206–215. 33. Stenstrom CH, Lindell B, Swanberg E et al. Intensive dynamic training in water for rheumatoid arthritis functional class II—a long-term study of effects. Scandinavian Journal of Rheumatology 1991; 20: 358–365. 34. Hammond A & Freeman K. One-year outcomes of a randomised controlled trial of an educational– behavioural joint protection programme for people with rheumatoid arthritis. Rheumatology 2001; 40: 1044–1051.

Are early arthritis clinics necessary? 15 35. Pincus T & Sokka T. Uniform databases in early arthritis: specific measures to complement classification criteria and indices of clinical change. Clinical and Experimental Rheumatololgy 2003; 21: S79–S88. * 36. Pincus RH & Brooks LF. A proposed standard protocol to evaluate rheumatoid arthritis (SPERA) that includes measures of inflammatory activity, joint damage and long-term outcomes. Journal of Rheumatology 1999; 26: 473–480. 37. Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis and Rheumatism 1988; 3: 315–324. 38. Pincus T, Callahan LF, Brooks RH et al. Self-report questionnaire scores in rheumatoid arthritis compared with traditional physical, radiographic, and laboratory measures. Annals of Internal Medicine 1989; 110: 259–266. * 39. Anderson JJ et al. Factors predicting response to treatment in rheumatoid arthritis. The importance of disease duration. Arthritis and Rheumatism 2000; 43: 22–29. 40. Irvine S, Munro R & Porter D. Early referral, diagnosis, and treatment of rheumatoid arthritis: evidence for changing medical practice. Annals of the Rheumatic Diseases 1999; 58: 510–513. * 41. Green D et al. Persistence of mild early inflammatory arthritis: the importance of disease duration, rheumatoid factor and the shared epitope. Arthritis and Rheumatism 1999; 42: 2184–2188. * 42. Green H, Marzo-Ortega RJ et al. Predictors of outcome in patients with oligoarthritis: results of a protocol of intraarticular corticosteroids to all clinically active joints. Arthritis and Rheumatism 2001; 44: 1177–1183. * 43. Quinn PG & Conaghan P. The therapeutic approach of early intervention for rheumatoid arthritis: what is the evidence? Rheumatology 2001; 40: 1211–1220. * 44. Wolfe F et al. The prognosis of rheumatoid arthritis and undifferentiated polyarthritis syndrome in the clinic: a study of 1141 patients. Journal of Rheumatology 1993; 20: 2005–2009. 45. Tunn EJ & Bacon PA. Differentiating persistent from self-limiting symmetrical synovitis in an early arthritis clinic. British Journal of Rheumatology 1993; 32: 97–103. 46. Jansen LMA, van Schaardenburg D, van der Horst-Bruinsma IE & Dijkmans BAC. One year outcome of undifferentiated polyarthritis. Annals of the Rheumatic Diseases 2002; 61: 700–703. 47. Eberhardt K & Fex E. Clinical course and remission rate in patients with early rheumatoid arthritis: relationship to outcome after 5 years. British Journal of Rheumatology 1998; 37: 1324–1329. 48. Wolfe F & Hawley DJ. Remission in rheumatoid arthritis. Journal of Rheumatology 1985; 12: 245–252. 49. van Leeuwen MA, Van Rijswijk MH, Sluiter WJ et al. Individual relationship between progression of radiological damage and the acute phase response in early rheumatoid arthritis. Towards development of a decision support system. Journal of Rheumatology 1997; 24: 20–27. 50. Fex E, Eberhardt K & Saxne T. Tissue derived macromolecules and markers of inflammation in serum in early rheumatoid arthritis: relationship to development of joint destruction in hands and feet. British Journal of Rheumatology 1997; 36: 1161–1165. 51. Amos RA, Constable TJ, Crockson RA et al. Rheumatoid arthritis: relationship of C-reactive protein and erythrocyte sedimentation rates and radiographic change. British Medical Journal 1977; 1: 195–197. 52. Devlin J, Gough A, Huissoon A et al. The acute phase and function in early RA. CRP levels correlate with functional outcome. Journal of Rheumatology 1997; 24: 9–13. 53. Gough AKS, Lilley J, Eyre S et al. Generalised bone loss in patients with early rheumatoid arthritis. Lancet 1994; 344: 23–27. 54. Devlin J, Lilley J, Gough A et al. Clinical associations with DXA measurement of hand bone mass in rheumatoid arthritis. British Journal of Rheumatology 1996; 35: 1256–1262. 55. Stenger AAME, Van Leuewen MA, Houtman PM et al. Early effective suppression of inflammation in rheumatoid arthritis reduces radiographic progression. British Journal of Rheumatology 1998; 37: 1157– 1163. 56. Dawes PT, Fowler PD, Clarke S et al. Rheumatoid arthritis: treatment which controls the C-reactive protein and erythrocyte sedimentation rate reduces radiological progression. British Journal of Rheumatology 1986; 25: 44–49. 57. Harrison BJ, Symmons DPM, Brennan P et al. Natural remission in inflammatory polyarthritis: issues of definition and prediction. British Journal of Rheumatology 1996; 35: 1096–1100. 58. Gough A, Faint J, Salmon M et al. Genetic typing of patients with inflammatory arthritis at presentation can be used to predict outcome. Arthritis and Rheumatism 1994; 37: 1166–1170.

16 M. A. Quinn and P. Emery 59. van der Heide A, Remme CA, Hofman DM et al. Prediction of progression of radiological damage in newly diagnosed rheumatoid arthritis. Arthritis and Rheumatism 1995; 38: 1466–1475. 60. Mo¨tto¨nnen T, Paimela L, Leirisalo-Repo M et al. Only high disease activity and positive rheumatoid factor indicate poor prognosis in patients with rheumatoid arthritis treated with saw-tooth strategy. Annals of the Rheumatic Diseases 1998; 57: 533–539. 61. van der Heide A, Jacobs JW, Haanen HC & Bijlsma JW. Is it possible to predict the first year extent of pain and disability for patients with rheumatoid arthritis? Journal of Rheumatology 1995; 22: 1466–1470. 62. van Zeben D, Hazes JM, Zwinderman AH et al. Clinical significance of rheumatoid factors in early rheumatoid arthritis: results of a follow up study. Annals of Rheumatic Diseases 1992; 51: 1029–1035. 63. Panayi G, Celinska E, Emery P et al. Seronegative and seropositive rheumatoid arthritis: similar diseases. British Journal of Rheumatology 1997; 12: 172–180. 64. Schellekens GA, Visser H, de Jong BAW et al. The diagnostic properties of rheumatoid arthritis antibodies recognising a cyclic citrullinated peptide. Arthritis and Rheumatism 2000; 43: 155–163. * 65. Harrison P et al. Inflammatory polyarthritis in the community is not a benign disease: predicting functional disability one year after presentation. Journal of Rheumatology 1996; 23: 1326–1331. 66. Munro R, Hampson R, McEntergart A et al. Improved functional outcome in patients with early rheumatoid arthritis treated with intramuscular gold: results of a five year prospective study. Annals of the Rheumatic Diseases 1998; 57: 88–93. 67. Conaghan PG, Wakefield RJ, O’Connor P et al. MCPJ assessment in early RA: a comparison between Xray, MRI, high-resolution ultrasound and clinical examination. Arthritis and Rheumatism 1998; 41: S246. 68. Wakefield RJ, Gibbon WW, Conaghan PG et al. The value of sonography in the detection of bone erosions in patients with rheumatoid arthritis: a comparison with conventional radiography. Arthritis and Rheumatism 2000; 43: 2762–2770. * 69. Conaghan D et al. Elucidation of the relationship between synovitis and bone damage: a randomized magnetic resonance imaging study of individual joints in patients with early rheumatoid arthritis. Arthritis and Rheumatism 2003; 48: 64–71. 70. Østergaard M, Hansen M, Stoltenberg M et al. New radiographic bone erosions in the wrists of patients with rheumatoid arthritis are detectable with magnetic resonance imaging a median of two years earlier. Arthritis and Rheumatism 2003; 48: 2128–2131. 71. McGonagle DGW, O’Connor P et al. An anatomical explanation for good prognosis rheumatoid arthritis. Lancet 1999; 353: 123–124. 72. Wakefield RJ, Goh E, Conaghan PC et al. Musculoskeletal ultrasound in Europe: results of a rheumatology based survey at EULAR. Rheumatology 2003; 42: 1251–1253. 73. Wakefield RJ, Brown AK, O’Connor PJ et al. Rheumatological ultrasound. Rheumatology 2003; 42: 1001. 74. Speed CA & Bearcroft PWP. Musculoskeletal sonography by rheumatologists: the challenges. Rheumatology 2002; 41: 241–242. 75. Reece RJ, Canete JD, Parsons WJ et al. Distinct vascular patterns of early synovitis in psoriatic, reactive and rheumatoid arthritis. Arthritis and Rheumatism 1999; 42: 1481–1484. 76. Fearon U, Reece RJ, Smith J et al. Synovial cytokine and growth factor regulation of MMPs/TIMPs: implications for erosions and angiogenesis in early rheumatoid and psoriatic arthritis patients. Annals of New York Academic Sciences 1999; 878: 619–621. 77. Fraser A, Fearon U, Reece R et al. Matrix metalloproteinase-9, apoptosis and vascular morphology in early arthritis. Arthritis and Rheumatism 2001; 44: 2024–2028. 78. Ali M, Veale DJ, Reece R et al. Overexpression of transcripts containing line-1 in the synovia of patients with rheumatoid arthritis. Annals of the Rheumatic Diseases 2003; 62: 663–666. 79. Harrison B, Thomson W, Symmons D et al. The influence of HLA-DRB1 alleles and rheumatoid factor on disease outcome in an inception cohort of patients with early inflammatory arthritis. Arthritis and Rheumatism 1999; 42: 2174–2183. 80. Rau R, Herborn G, Zeuger S & Fenner H. The effect of HLA-DRB1 genes, rheumatoid factor and treatment on radiographic disease progression in rheumatoid arthritis over 6 years. Journal of Rheumatology 2000; 27: 2566–2575. 81. O’Dell JR, Nepom BS, Haire C et al. HLA-DRB1 typing in rheumatoid arthritis: predicting response to specific treatments. Annals of the Rheumatic Diseases 1998; 57: 209–213.

Are early arthritis clinics necessary? 17 82. Mattey DL, Hassel AB, Dawes PT et al. Independent association of rheumatoid factor and the HLA-DRB1 shared epitope with radiographic outcome in rheumatoid arthritis. Arthritis and Rheumatism 2001; 44: 1529–1533. 83. Bathon J, Martin RW, Fleischmann RM et al. A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis. New England Journal of Medicine 2000; 343: 1586–1593. 84. Quinn MA, Conaghan PG, Greenstein A et al. Sustained response in early poor prognosis RA after withdrawal of infliximab therapy. Arthritis and Rheumatism 2002; 46: 3416. 85. Yelin E & Callahan L. The economic cost and psycological impact of musculoskeletal conditions. Arthritis and Rheumatism 1995; 38: 11351–11362. 86. Clarke AE, Zowall H, Lebington C et al. Direct and indirect medical costs incured by Canadian patients with rheumatoid arthritis: a 12 year study. Journal of Rheumatology 1997; 24: 1051–1060. 87. Lanes SF, Lanza LL, Radensky PW et al. Resource utilisation and cost of care for rheumatoid arthritis and osteoarthritis in a managed care setting. Arthritis and Rheumatism 1997; 40: 1475–1481. 88. Cooper NJ. Economic burden of rhematoid arthritis: a systematic review. Rheumatology 2000; 39: 28–33. 89. Gabriel S, Crowson CS, Champion ME et al. Indirect and non-medical costs among people with rheumatoid arthritis and osteoarthritis compared to non-arthritic controls. Journal of Rheumatology 1997; 24: 43–48. 90. Kobelt G, Jonsson L, Young A & Eberhardt K. The cost effectiveness of infliximab in the treatment of rheumaoid arthritis in Sweden and the United Kingdom based on the ATTRACT study. Rheumatology 2003; 42: 326–335. 91. Wong JB, Singh G & Kavanaugh A. Estimating the cost effectiveness of 54-weeks of infliximab for rheumatoid arthritis. American Journal of Medicine 2002; 113: 400–408.