|Home | About | Journals | Submit | Contact Us | Français|
Rheumatoid arthritis (RA) is a chronic systemic inflammatory disease that affects about 1% of the population. It leads to irreversible joint damage and systemic complications, and the age-adjusted mortality of those affected exceeds that of the general population.1–3 When joint damage was seen to be an early feature of the disease,4–12 rheumatologists put forward the point at which they prescribed disease-modifying antirheumatic drugs (DMARD), in the hope of slowing or even arresting disease progression. Patients in whom DMARD therapy is introduced early have better function and radiological outcome in the long-term than those in whom it is delayed.13–19 It was for these reasons that a SIGN (Scottish Intercollegiate Guideline Network) guideline20 in 2000 indicated that a patient with inflammatory arthritis lasting >6–8 weeks should be referred for a specialist (rheumatology) opinion. However, a recent audit in our unit showed that such patients were referred after a mean of 16 weeks (interquartile range 6–34) from onset of symptoms.21 Other studies suggest that the long lag between symptom onset and the diagnosis of RA is mainly due to late referral rather than patient delay in reporting symptoms or long waits for outpatient appointments.22 Moreover, in most referral letters from general practitioners, a tentative rheumatological diagnosis is either not stated or stated wrongly.23,24 The reason is clear: RA has no disease-specific diagnostic features25 and patients can present with a wide range of manifestations. In this article I discuss the difficulties of early diagnosis, taking an illustrative case of polyarthritis (inflammation of more than four joints), the commonest presentation.
The onset of polyarthritis in RA is insidious in about three-quarters of patients and initially affects the small joints of the hands and feet (metacarpophalangeal, proximal interphalangeal and metatarsophalangeal joints) before spreading to the larger joints. The following are atypical manifestations.
A woman schoolteacher, aged 30, has for twelve weeks been troubled by pain and swelling in the small joints of her hands and feet. She says that her hands are stiff in the early mornings for a couple of hours and then improve gradually in the course of the day. Her sleep is sometimes disturbed by pain, and she feels very tired during the day. Previously she was fit and well: there is no personal or family history of psoriasis or inflammatory bowel disease and she was not knowingly exposed to infections before the onset of this illness. She lives with her husband and two children, aged 6 and 4, and her symptoms interfere with some activities of daily living. On examination her metacarpophalangeal joints (MCP) are swollen and tender bilaterally, as are a few of the proximal interphalangeal joints (PIP) in both hands. Compression of metatarsophalangeal joints (MTP) causes pain. All her other joints are clinically normal and the examination reveals nothing else of note. Blood investigations, including liver function tests, give normal results apart from an ESR of 28 mm/h. Antibodies to parvovirus are not found, and rheumatoid factor and antinuclear antibody are likewise absent. On plain radiographs of the hands and feet the only abnormality is periarticular soft tissue swelling around a few PIP joints.
Box 1 Features suggestive of inflammatory arthritis
The history of swelling in joints, early morning stiffness lasting >30 minutes, systemic symptoms such as tiredness combined with objective evidence of synovitis would favour a diagnosis of inflammatory arthritis (Box 1). However, reality can be more complex:
The most important question, in our patient, is whether she has a potentially damaging disease such as RA. The answer is not always obvious since RA in its early stages tends not to fit the textbook description. For example, seropositivity for rheumatoid factor, radiographic erosions and subcutaneous nodules are all absent. As mentioned above, at the time of presentation many patients with RA have normal inflammatory markers; moreover, about 60% are seronegative for rheumatoid factor and more than 70% have normal plain radiographs.26 Thus negative results with these do not exclude the diagnosis. In our patient, reasons for strongly suspecting RA are the longstanding inflammatory symptoms (twelve weeks) and the symmetrical involvement of MCP, PIP and MTP joints—joints that are commonly affected in RA.
The other difficulty is the wide differential diagnosis of polyarthritis (Box 2). A good history and physical examination combined with a few simple laboratory or radiological investigations should help in excluding most of the conditions that mimic RA. In our patients, the following need to be considered:
Box 2 Conditions that can present as polyarthritis and mimic RA
Postviral arthritis—e.g. parvovirus, mumps, rubella, hepatitis B and C
Seronegative spondyloarthritis—e.g. psoriatic arthritis, inflammatory bowel disease
Connective tissue diseases—e.g. systemic lupus erythematosus, scleroderma, vasculitis
Crystal arthritis—e.g. polyarticular gout, pseudogout
Miscellaneous—e.g. sarcoidosis, thyroid disease, infective endocarditis, malignant disease
Box 3 Investigations that may help in diagnosing common underlying causes of polyarthritis
The other conditions listed in Box 2 are much less likely in view of the patient's age and sex and the absence of other systemic manifestations. Since her clinical picture is not consistent with any of the specific entities she cannot yet be diagnosed as having RA; at this stage the label of 'early polyarthritis' is more appropriate.
When a patient with inflammatory arthritis cannot definitely be labelled as having RA, it becomes important to decide whether the arthritis is likely to remit or to persist. Clearly, if spontaneous remission seems likely, the patient should be spared potentially toxic DMARD therapy. On the other hand, a patient with persistent inflammation should be started promptly on DMARDs since the condition may represent RA in evolution. From the Norfolk Arthritis Register27 there is evidence that an overwhelming majority of patients with persistent polyarthritis in due course come to satisfy diagnostic criteria for RA28 (from 47% at baseline to 93% after 5 years).13 Thus, since joint damage and functional loss occur early,3–12 most patients develop these irreversible changes before a definite diagnosis of RA can be made.
How can the clinician predict persistence of disease? Several research groups have tried to identify pointers in patients with early arthritis29–34 but their results are not easily combined because of heterogeneity in populations, predictive factors used and duration of follow-up. Among the predictive factors suggested, the most useful seems to be disease duration exceeding 12 weeks: a patient who has had inflammatory joint symptoms for this long is very unlikely to experience a spontaneous remission. Other features suggesting the unlikelihood of remission are positive tests for rheumatoid factor or cyclic citrullinated peptide antibodies and the presence of erosions on radiographs.
Early treatment of RA can be a realistic goal only if general practitioners and other non-rheumatologists recognize the clinical picture of early inflammatory arthritis and refer patients promptly for a specialist opinion. Patients with joint pains that have persisted for more than 6–8 weeks should be referred especially in the presence of the following features:
Normal inflammatory markers, negative serology and normal plain radiographs are not valid reasons for delaying referral since RA is diagnosed on the basis of symptoms and signs.
Should general practitioners start DMARDs themselves? According to one survey many are reluctant, preferring to get a specialist opinion first.36 The results of our audit21 have reinforced this impression: only 10% of eligible patients had a DMARD prescribed by the general practitioner before their first clinic appointment; moreover, there is evidence that patients managed by rheumatologists do better than those strictly managed by non-rheumatologists.37 Thus, in a patient with suspected RA or 'RA-like' polyarthritis, the message of this paper is: refer early to a rheumatologist.