Characteristic hand deformities include ulnar deviation of the MCP joints, boutonnière deformities of the PIP joints, and swan-neck contractures of the fingers. In the wrists, a permanent loss of extension often occurs. A boggy, tender, dorsal wrist mass may result from tenosynovitis, and the median nerve can be compressed. The subsequent carpal tunnel syndrome is usually reversible, but nerve damage is permanent by the time wasting of the thenar eminence becomes obvious.

Erosion of the metatarsal heads can lead to ventral subluxation. The results are increased weight bearing on the inflamed heads and the formation of painful calluses. Erosive disease may be silent in the MTP joints.

Involvement of these joints can be a source of much disability when weight bearing causes severe pain. A loss of internal rotation is the first change noted in the hip, followed by flexion contracture. One hip may predominate, even though the process is bilateral. In the knee, distention of the suprapatellar pouch by synovial effusion is common. If pressure rises rapidly, herniation of the synovium with the formation of a popliteal Baker cyst can occur, and the cyst can cause severe pain if it ruptures into the calf. Loss of full knee extension is followed by flexion contractures and gait difficulties.

In the elbow, extension may be compromised, and olecranon bursitis is often present. Shoulder involvement presents as a subacromial or subdeltoid bursitis or as a limitation of motion. Erosion of the rotator cuff leads to painful upward subluxation of the humeral head against the acromion. In the cervical spine, atlantoaxial subluxation is common but usually asymptomatic. This development is potentially serious because it can lead to direct compression of the spinal cord or of the blood supply to the brainstem; fortunately, it is a rare event. When the temporomandibular joint is affected, pain is felt during chewing or biting, and it is difficult to open the mouth. Avascular necrosis of the femoral head and vertebral osteoporosis and collapse are usually a consequence of corticosteroid therapy.

Radiographic manifestations of early disease are limited and may be little more than soft tissue swelling around the joint and periarticular osteopenia. Relatively uniform narrowing of the joint space occurs as cartilage is destroyed, a finding that may take as much as 2 years to develop after onset of symptoms. Periarticular subchondral erosion is noted at the joint margin where pannus has developed. Finally, joint architecture is lost as the joint space is obliterated and erosion of subchondral bone progresses. Such radiologically evident disease is predictive of future disability.

A variety of autoantibodies can be found in RA. The two most characteristic are rheumatoid factor and anti-citrullinated peptide antibodies (ACPAs). Their high frequency in RA has led to their consideration regarding etiology and interest for use in diagnosis. Less specific are antinuclear antibody (ANA) and biomarkers or inflammation (C-reactive protein [CRP] and erythrocyte sedimentation rate [ESR]).

These antibodies form against synovial peptides that have been “citrullinated” (a posttranslational deimination, converting the N-terminal amino acid on targeted peptides from arginine to citrulline), which alters the immunogenicity of these peptides as
part of an immune response by autoreactive B cells. Citrullinated synovial peptides are believed to be deiminated chains of fibrin targeted by ACPAs. The ACPAs arise early in the course of the disease, often before onset of symptoms, but do not appear to be etiologic; nonetheless, they are rather unique to RA, detectible in about 60% of patients, but very uncommon in other autoimmune conditions (<2%). Their specificity and early appearance make them attractive for use in diagnosis (see Diagnosis).

Extra-articular manifestations are most prominent in patients with persistent symptoms and high titers of RF (the so-called seropositive disease), although up to 50% of patients with RA may show some form of extra-articular disease.

Pulmonary manifestations include interstitial changes, pulmonary nodules, and pleuritis. The latter is manifested by a pleural effusion, in which the levels of glucose (5 to 20 mg/dL) and complement are characteristically low, the leukocyte count (5,000/mm³) is low, and the lactate dehydrogenase level is high. Accompanying pleuritic pain may or may not be present. An asymptomatic pericardial effusion also may occur.

Keratoconjunctivitis sicca (Sjögren syndrome) has a strong association with RA, being found in up to 15% of patients. Splenomegaly is present in 5% to 10% of patients with RA, and lymphadenopathy is not unusual. The combination of RA, splenomegaly, and neutropenia (Felty syndrome) is noted in an occasional patient. Neutropenia may be severe, but the arthritis is often quiescent. Other features include chronic leg ulceration, lymphadenopathy, and cryoglobulinemia. The risk for sepsis is high.

Vasculitis is believed to be responsible for a number of systemic manifestations, including fever, mononeuritis multiplex, Raynaud phenomenon, chronic leg ulcers, mucosal erosions of the gastrointestinal (GI) tract, focal ischemia of the digits, and necrotizing mesarteritis. The anemia of chronic disease is seen in a large percentage of patients with RA.

Atherosclerotic disease is prevalent among persons with RA; cardiovascular morbidity and mortality are increased, and markers of preclinical disease, such as carotid artery atherosclerotic changes, are common. Although some treatments for RA increase cardiovascular risk (see later discussion), RA also appears to confer risk independent of other cardiovascular risk factors, perhaps related the generalized inflammation associated with the condition and its contribution to atherosclerosis (see Chapter 31).

Osteoporosis risk is doubled in RA and markedly exacerbated if steroid therapy is used.

In stage I, no symptoms or signs are present, just the presentation of the relevant antigen to an immunologically susceptible host. Stage II, in which the immune response becomes organized in the perivascular areas of the synovium, is characterized by increasing numbers of T cells, proliferation and differentiation of B cells, antibody production, synovial cell increase, and new blood vessel formation. Morning stiffness develops because fluid about the joints is increased. They are warm but not erythematous because superficial vessels are uninvolved. In stage III, the pathophysiologic processes of stage II continue, and extra-articular manifestations become evident. In stage IV, proliferating synovial membrane becomes invasive, injuring cartilage, bone, and tendons.

The natural history of the disease is generally one of exacerbations and remissions. About 40% of patients untreated by disease-modifying therapy become disabled after 10 years, but outcomes are highly variable. Some patients experience a relatively self-limited disease, and others suffer a chronic, progressive illness. Improvements in the detection of early joint injury have provided a previously unappreciated view of how common and important early joint damage is. The first 3 months from onset of symptoms appears particularly important to clinical course and prognosis—intervention during this period, which has been labeled a “window of opportunity,” is associated with the best treatment outcomes (see later discussion), even if applied only transiently.

It remains difficult at the outset to predict the course of an individual case, although the HLA-DRB1*04/04 genotype, a high serum titer of RF or ACPA, extra-articular manifestations, a large number of involved small joints, age less than 30 years, female gender, and systemic symptoms all correlate with an unfavorable prognosis. Development of radiologic manifestations of joint damage is predictive of future disability. Insidious onset is also an unfavorable sign. Disease that remains persistently active for more than 1 year is likely to lead to joint deformities and disability. Cases in which periods of activity lasting only weeks or a few months are followed by spontaneous remission have a better prognosis. The absence of RF does not necessarily portend a good prognosis. Outcome is compromised when diagnosis and treatment are delayed. Joint damage often begins within the first 3 months after onset of symptoms, the time when intervention with disease-modifying therapy achieves the best outcomes. Other laboratory markers of a poor prognosis include early radiologic evidence of bony injury, persistent anemia of chronic disease, and elevated levels of the C1q component of complement.

The overall mortality for patients with RA is reported to be 2.5 times that of the general population. In those with severe articular and extra-articular disease, the mortality approaches that of patients with three-vessel coronary disease or stage IV Hodgkin disease. Some of the excess mortality derives from infection (especially pneumonia), vasculitis, and poor nutrition, but most is attributable to complications of atherosclerotic disease. Mortality from cancer is unchanged.

Most data on rates of disability derive from specialty units caring for referred patients with severe disease. Little information is available on patients seen in primary care community settings. Estimates suggest that more than 50% of these patients remain fully employed, even after 10 to 15 years of disease, with one third having only intermittent low-grade disease and another one third experiencing spontaneous remission.

The clinical course is no longer one of inevitable joint destruction. The timely application of treatment capable of blunting or halting the immunoinflammatory process can greatly slow disease progression and prevent or limit permanent joint damage, especially if initiated shortly after the onset of symptoms (see Principles of Management).

The suboptimal accuracy of the RF for diagnosis of RA (see next section) has stimulated a search for other serologic markers that would enable more accurate early confirmation of the RA diagnosis. The discovery of citrullinated peptides derived from the synovium and rather unique antibodies (ACPAs) that target them has led to a search for assays that would aid in early diagnosis.

Second-generation assays (labeled anti-cyclic citrullinated peptide antibodies [anti-CCP₂] because they add cysteine to the antigen preparation [which causes the target proteins to become cyclic and thus more antigenic]) are now widely available. Systematic review finds test accuracy when applied during early phases of RA superior to that of the RF (summary sensitivity, 58% vs. 56%; summary specificity, 96% vs. 86%; positive likelihood ratio, 12.7; negative likelihood ratio is 0.45). Such test characteristics indicate the test is most useful for ruling in the diagnosis, but cannot be used to rule it out. For the typical patient with early disease presenting as new onset of synovitis, the pretest probability of RA is about 45%; having a positive anti-CCP₂ test result raises the posttest probability to over 90%, sufficient in the minds of many to begin disease-modifying treatment.

Anti-CCP₂ can detect ACPAs before symptoms arise and during the critical first 3 months of symptomatic disease, a period when treatment has the greatest potential to limit joint damage (see later discussion). Their early appearance, slightly better assay sensitivity, and much enhanced specificity for diagnosis of RA compared to RF have made their detection the preferred approach to serologic testing for RA. The typical patient presenting with synovitis has roughly a 40% pretest probability of RA. With a positive predictive likelihood value of 12.7, a positive test increases the posttest probability to 90%, sufficient to start disease-modifying treatment. Although useful for ruling in the diagnosis, the low sensitivity and modest negative likelihood ratio make the test of little help in ruling out RA.

Being more accurate than RF testing, the anti-CCP₂ test can probably be used in its stead. Combination testing has not been found to enhance sensitivity or specificity. The test may also have some prognostic significance, with positivity and higher titers associated with increased risk of radiologic progression. Anti-CCP₂ is more costly than testing for RF, but the benefits derived from early treatment should make such testing quite cost-effective. Newer assays for detection of ACPAs are being developed (e.g., anti-CCP₃) and hopefully will offer improved tests characteristics at lower cost.

The test for RF detects IgM antibodies against immunoglobulins (IgGs) altered by autoimmunity. Testing for RF can help to support diagnosis if the pretest probability of RA is at least intermediate, but the reported summary test sensitivity of 56% and summary specificity of 86% make for only modest test accuracy. In general, 70% to 80% of all patients meeting strict clinical criteria for RA eventually become RF positive. However, because specificity is lacking, the test cannot rule in disease as well as the anti-CCP₂ assay. As noted, the presence of RF is not unique to RA: RF positivity also occurs in a host of other inflammatory conditions (e.g., systemic lupus, subacute bacterial endocarditis, vasculitis, and even viral infection); 5% to 15% of “normal” persons (the percentage increases with age) also manifest RF positivity, albeit at low titers. RF negativity does not rule out RA because 20% to 30% of patients with definite RA remain RF negative. The higher the RF titer, the
more likely is the diagnosis of RA and the greater the risk of more aggressive disease. IgM RF has been found to appear years before the onset of symptoms and may facilitate early diagnosis, but the utility of the test is being superseded by anti-CCP₂ testing, due to the latter’s better specificity.

Although certain genotypes have been associated with poor prognosis (e.g., HLA-DRB1*04/04) and other non-HLA gene mutations have been linked to RA (e.g., PTPN22, STAT4, TRAF1-C5), routine genetic testing is not recommended for diagnostic purposes. Most autoimmune conditions are polygenic, involving a complex interplay of susceptibility genes. Too little is known about these genetic determinants and their interactions for such testing to be diagnostically useful at this time, but their use as prognostic indicators is promising (e.g., HLA-DRB1*04/04 for overall prognosis and TRAF1-C5 for the risk of anti-CCP-positive RA).

Acute-phase reactants are proteins synthesized by the liver in response to inflammation or tissue necrosis. They include fibrinogen, haptoglobin, coagulation factors, complement, ferritin, and albumin. Measures of them, such as the ESR and CRP, can provide sensitive but nonspecific evidence of inflammation, helping to both diagnose and monitor inflammatory diseases such as RA. However, they show little use in predicting the risk of developing clinical RA, even when measured shortly before the onset of symptoms and signs. Despite these limitations, their sensitivity has led to their inclusion in the new consensus criteria for diagnosis of early RA (see earlier discussion), and they are useful for monitoring disease activity. The CRP level responds more quickly than does the ESR to changes in inflammatory disease activity, making it the more sensitive test; however, its performance does not match the 1-hour convenience and simplicity of the ESR.

Early initiation of disease-altering therapy has become the standard of care—outcomes are best when treatment is started within 3 months of the onset of symptoms, the so-called window of opportunity for RA management. Response rates and outcomes are maximized when intensive immunomodulating therapy is implemented during this period, which represents the earliest phase of joint injury leading to possible joint destruction if not halted.

Two classes of pharmacologic agents capable of altering the course of RA have emerged: (a) conventional, small-molecule immunosuppressive agents, often referred to as the diseasemodifying antirheumatic drugs (DMARDs, e.g., methotrexate, hydroxychloroquine, sulfasalazine), and (b) large-molecule drugs, mostly humanized antibodies, labeled biologic response modifiers or biologics, which inhibit specific mediators or promoters of the immunoinflammatory process. The latter are usually divided into tumor necrosis factor (TNF) inhibitors and non-TNF inhibitors (e.g., interleukin-1 inhibitors, anti-B-cell agents, JAK inhibitors, and costimulation signal modifiers). The glucocorticosteroids, with their rapid onset of both immunosuppression and anti-inflammatory action, can serve as both a bridge to diseasealtering therapy and a means of treating acute disease flares.

The biologic agents have markedly enhanced RA treatment, significantly raising response rates and improving outcomes. Being well tolerated, they have become a mainstay of treatment, administered to as many as 50% of RA patients. However, their extremely high cost, typical need for parenteral administration, and immunosuppressive side effects have rekindled interest in the conventional DMARDs. When the latter are applied early as monotherapy or in multidrug regimens (either as a combination DMARD regimen or in conjunction with a biologic agent), they can have a significant effect on disease activity and progression. Their low cost and ease of use (i.e., oral administration) make them a cost-effective first-line therapy.

Treatment initiated early with the goal of inducing a “remission” (i.e., absence of symptoms and signs) is associated with enhanced short- and long-term outcomes. In a “tight-control” strategy, patient and physician mutually agree upon remission targets for symptoms, signs, and daily functioning. Treatment response is evaluated monthly and treatment intensified (monotherapy dose increased or additional agents added) until the desired goals are reached. Such a tight-control strategy improves physical functioning, reduces joint damage, and produces better outcomes compared to conventional treatment that lacks preset treatment goals, especially when instituted within the first 3 months of illness.

Both nonsteroidal anti-inflammatory drugs (NSAIDs) and oral glucocorticosteroids have been used for decades to provide symptomatic relief from the discomfort of joint inflammation. Although effective for this purpose, only the corticosteroids influence disease progression, relegating the NSAIDs to a second-tier role while corticosteroids are reserved for supplementation of firstline, disease-modifying therapy. An increasing appreciation for the contribution of oral corticosteroids to early disease control has sparked a renewed interest in their use as part of a comprehensive disease-modifying program.

The effective management of RA requires a collaborative effort that engages specific contributions from a host of professionals both within and outside the primary care practice, including the consulting rheumatologist, nurse educator, and occupational and physical therapists as well as members of the primary care medical home team.

Of particular importance is arranging early consultation with a rheumatologist for design and implementation of a comprehensive disease-modifying program. The complexity, cost, and potentially serious adverse effects of disease-modifying agents necessitate expert guidance in their prescription and use, similar that for cancer chemotherapy. Periodic return visits are needed for systematic assessment of disease activity and response to treatment, utilizing validated instruments such as the Disease Activity Score (DAS) and the American College of Rheumatology Response Criteria (e.g., ACR 20 and 50, which specify a 20% or 50% reduction in key symptoms, standards for response to treatment). These instruments help guide management decisions (see later discussion).

Specially trained nurse educators can improve outcomes when included in the patient care team for RA patients. Their special role in patient education leads to enhanced patient selfsufficiency and daily functioning as well as reductions in pain, depression, and number of physician visits.

Patients with problematic joint injury or pain can benefit greatly from working with occupational and physical therapists, who teach useful methods and exercises for maximizing functionality and minimizing disability. The contributions from members of the medical-home team are essential to provision of emotional support and education, fostering compliance with the treatment program, and monitoring for outcomes and adverse effects.

Despite the major advances in drug therapy, nonpharmacologic measures continue to be critically important to the maintenance of daily function and prevention of disability. The basic components of a program must include thorough patient education, adequate rest, and proper exercise.

Likely any treatment program for chronic disease, its design needs to be personalized, taking into account the patient’s personal goals, lifestyle, and home/work environments and responsibilities. Attention to the patient’s emotional makeup and attitudes toward treatment is also essential. Cost is an increasingly important consideration, given the very high cost of biologic therapies. A balanced, multifaceted, personalized approach to therapy helps maximize compliance and achieve best outcomes. In the future, genetic and biomarker profiling should help in personalizing the treatment program, aiding in selection of treatments most likely to benefit the patient and avoiding unnecessary exposure to others.

Early initiation (within 3 months of onset of symptoms) of firstline disease-modifying therapy provides the best opportunity to achieve remission and minimize risk of permanent joint damage. Symptoms, joint function, and quality of life all improve; survival can be prolonged. Response rates approaching 75% are attainable; many patients experience complete remission of symptoms.

First-line therapy usually begins with a conventional DMARD (most often methotrexate). For patients with especially aggressive disease or failure to respond, the addition of a biologic or use of combination DMARD therapy (using agents with complementary modes of action) can enhance control without greatly increasing toxic side effects (analogous to combination cancer chemotherapy). Adding a low dose of prednisone to methotrexate at the outset of treatment enhances outcomes and reduces need for combination therapy and biologics. Optimizing safe and effective use of these potent drugs requires consultation and coordination with the rheumatologist as well as an understanding by the primary care physician and medicalhome team of their proper use and potential for adverse effects.

Disease-modifying therapy often begins with a small-molecule, orally active, conventional synthetic agent—a so-called diseasemodifying antirheumatic drug. The first-line DMARDs include methotrexate, hydroxychloroquine, sulfasalazine, and leflunomide. Methotrexate is typically the initial drug of choice, being effective, orally active, titratable, well tolerated, and relatively inexpensive. Patients unable to tolerate methotrexate may be started on another first-line DMARD agent such as leflunomide, which has comparable efficacy, though more side effects, or hydroxychloroquine, which is better tolerated but less potent—used for persons with mild disease.

Since the clinical onset of DMARD action can take anywhere from 6 weeks to a few months to become apparent, initial DMARD treatment is often complemented by adding a low dose of prednisone (e.g., 10 mg/d) at the outset to provide prompt symptomatic relief and minimize early joint injury while awaiting DMARD activity to take hold. Some argue that starting low-dose prednisone at the outset of initial DMARD treatment should become a standard of care because it improves response rates, improves outcomes, and reduces need for additional therapy (see later discussion on prednisone use).

Indefinite continuation of DMARD therapy appears necessary because disease activity returns when treatment is stopped, but the use of intensive therapy (treating to remission
of symptoms) at the outset may allow tapering over time. Nonetheless, even limited courses of treatment (especially during the first 3-month window-of-opportunity period) can provide some lasting benefit (e.g., better joint preservation). More data are needed to determine the optimum duration of therapy.