Rheumatoid arthritis: status of drug therapies.

[Moncur C, Williams HJ. Rheumatoid arthritis: status of drug therapies. Phys Ther. 1995; 75:511-525.]

Key Words: Drug therapy, Physical therapy, Rheumatoid arthritis.

Most physical therapists will likely be involved in treating one or more individuals with rheumatoid arthritis (RA) at some time in their practice. The perception that nothing much can be done to alter what happens to these patients is unfortunate. Over the past 25 years, considerable options have been opened in the way of medications, total joint replacement, and self-management techniques. In particular, since the discovery and use of corticosteroids in the treatment of RA, a plethora of nonsteroidal anti-inflammatory and other drugs have been created to treat RA. it is often difficult for the physical therapist to know what the attributes of the medications the patient is taking are and how these characteristics will influence a physical therapy program.

Unfortunately, there is a paucity of .reports in the literature describing how these drugs influence rehabilitation of individuals with RA. Most of what is known has been derived from anecdotal data or from data reporting the side effects or adverse reactions of the drug during clinical trials. This information provides clues from which conclusions are drawn as to how a rehabilitation program might be affected by a given drug. The purposes of this article are to review and describe the current medications used in treatment of individuals with RA and to discuss how these medications might affect physical therapy interventions.

Etiology of Rheumatoid Arthritis

To fully appreciate the challenges presented by RA from both a pharmacological perspective and a rehabilitation perspective, it is important for the physical therapist to gain an understanding of what is currently known about the etiology, pathology, pathogenesis, and clinical features of the disease. Although considerable research is ongoing regarding the etiology of RA, the cause remains obscure. For example, it has been determined that RA occurs more frequently in identical twins than is predicted from the frequency of RA in the general population.[1] Furthermore, investigations of class II genes and molecules of the major histocompatibility complex have demonstrated an increased susceptibility to seropositive RA in patients with the HLA-DR4 haplotype.[2] Schiff et al[3] and Woodrow et al[4] have determined that the DR-1 antigens are present in greater frequency in selected groups of patients with adult-onset RA. Although the relative risk of developing RA in people with the DR-4 antigen seems greater, the majority of individuals with this marker are unaffected. Likewise, patients with RA may have a haplotype other than DR-4 or DR-1.

Any infectious agent that may be a cause of RA has thus far eluded investigators. Models used to study these phenomena include an infectious form of arthritis in rodents[5] and bacteria, virus, and spirochete models in humans.[6] Unfortunately, a firm connection between these models and the cause of RA has not been established.

Patients are often curious to know whether improper nutrition contributes to the cause of RA. No investigations thus far conducted have established specific nutritional factors as causative agents.

Pathology of Rheumatoid Arthritis

The normal synovial lining of the joint capsule is usually two or three cells in thickness. It has considerable vascularity, whereas the underlying stromal elements are relatively avascular and acellular. The subsynovial tissue consists of connective tissues composed of areolar, fatty, and fibrous tissues.[7,8] Schrieber[9] suggests that, contrary to past impressions, the vascular endothelium plays an active and pivotal role in a variety of inflammatory and immunological processes occurring in patients with RA. Examples of the endothelial contribution to the inflammatory processes in the joint are angiogenesis, coagulation, mediator and cytokine production, antigen presentation, and tissue recruitment of leukocytes.[9]

Microvascular Alterations

Early in the course of the disease, the synovial microvasculature undergoes substantial changes, including high endothelial venule (HEV) formation, which potentially increases the movement of lymphocytes to the synovial lining, and new abnormal blood vessel formation, which is thought to contribute to cartilage destruction.[9] The capillaries become distended with the accumulation of plasma and fluid and potentially recruit disease-causing agents to the synovium. The sequela of this process is to damage the vascular endothelium, thus limiting the efflux of soluble and cellular constituents of the vasculature. The end result to the capillary can be edema, hemorrhage, thrombosis, ischemia, and infarction.[10,11]

Synovial Proliferation

The synovial cells proliferate and hypertrophy, and the surface of the synovium may be covered with compact fibrin. The synovial fluid contains inflammatory products, with an increase in the leukocyte count and a decrease in the normal viscosity of the fluid. The latter may result from decreased production and/or dilution of the hyaluronic acid. The majority of inflammatory cells present in the synovial fluid in early RA are polymorphonuclear leukocytes, namely neutrophils.[12]

In the chronic stages of RA, synovial hypertrophy and hyperplasia continue, forming villus projections into the joint capsule. The dense aggregates of lymphocytes around vessels increase and are surrounded by plasma cells resembling the form of a lymphoid nodule.[7] This inflamed tissue extends into the joint space, joint capsule, ligaments, tendons, bursal elements, and other soft tissues surrounding the joint. The proliferating synovium begins to extend over the surfaces of the articular cartilage and is called pannus formation.

Because the tendon sheaths and bursae are lined with synoviocytes, it is possible for these linings to become inflamed and hypertrophied. By infiltrating the tendon, weakening can occur that can lead to tendon rupture. Eventual fibrosis of the inflamed tendon sheath can lead to a decrease in function without there being a ruptured tendon.

Lymphocytic Activity

The majority of lymphocytes in synovial inflammation are T cells.[13] The factors facilitating the movement of lymphocytes to the synovia are poorly understood; however, Paulus et al[14] conducted the pioneering studies that demonstrated the role of recirculating lymphocytes in the pathogenesis of RA. In chronic RA, the T lymphocyte accumulates in the inflamed synovial membrane, activating a cell-mediated immune response and contributing to the lesions seen in disease.

Pathogenesis

The intent of this discussion is to present a brief overview of the complex pathogenic processes of RA. Many of the attributes of the process are still under investigation and are not well understood.

Immunologic Processes

Immune complexes are found in the synovial fluid and these are often the IgG/anti-IgG antigen-antibody complex; however, in RA the joint constituents such as collagen, cartilage, and nucleoproteins may serve as antigens.[15] There may be other unidentified antigens involved in triggering the disease process. During the inflammatory process, the complement cascade is activated, which eventually attracts neutrophils that ingest the immune complexes and release lysosomal enzymes (including elastase and collagenase), oxygen free radicals, and arachidonic acid metabolites (prostaglandins, thromboxane, and leukotrienes).[16] These substances appear to have damaging effects on the joint structures; however, their role is not fully understood.[17]

Characteristics of Destruction

Tissue damage occurs in sites contiguous to the cells of the inflamed synovium. Jasin[18] reported finding immune complexes in the superficial articular cartilage. Interleukin-1 (IL-1) is released by the macrophages and other cells in the proliferating synovium, which stimulates the synoviocytes, chondrocytes, and osteoclasts to release cytokines that are destructive to the cartilage and bone. Inactive collagenase is produced by rheumatoid synovial cells.[19] The inactive collagenase is activated, leading to collagen destruction and thus destroying the framework of the cartilage of the joint.[20]

T and B lymphocytes and macrophage interactions stimulated by antigens release cytokines and immunoglobulins.[21] Local tissue damage occurs as a result of antibody-dependent cell-mediated cytotoxic reactions contributed by cells in the tissue.[22] The results of these inflammatory processes in the synovial fluid and synovium contribute to the pain, swelling, and destruction of cartilage, bone, and periarticular tissues of the joint.

Clinical Features

Rheumatoid arthritis is more than a disease of the joints. It is a systemic disease that can involve many organs and present systemic symptoms. There is a wide variability in patients' complaints and physical findings. Some patients may have an insidious onset over several weeks or months, whereas others may exhibit acute symptoms appearing over days. The; disease is commonly accompanied by fatigue and malaise. The fatigue is proportionally greater than is expected from nutritional anemia or similar conditions.

Articular Manifestations

The articular manifestations of RA result in pain on motion and/or tenderness to pressure and swelling with redness or warmth. Loss of motion and function can result from the effects of swelling and pain and can also be the consequence of bone, cartilage, or soft tissue damage. The joint involvement is usually polyarticular and symmetrical; however, a patient can present monarticular joint symptoms. Any synovial joint can be affected, but the proximal interphalangeal (PIP) and metacarpophalangeal (MCP) joints of the hand are most commonly involved. The wrists, knees, and metatarsophalangeal joints are also frequently involved. If a patient complains of hoarseness, the cricoarytenoid joint may be involved, as it is a synovial joint.

Chronic inflammation of the joints can result in joint destruction and the associated complications. Typical hand deformities include ulnar deviation of the MCP joints, boutonniere and swanneck deformities of the fingers, and joint subluxation. Muscle atrophy and contractures are not uncommon. Rupture of the transverse ligaments and erosions of the odontoid process can lead to atlanto-axial subluxation of the neck. Hip, knee, and subtalar joint deterioration are not uncommon.

Extra-articular Manifestations

Extra-articular disease is infrequent in patients with RA but can be life threatening. Some common manifestations are subcutaneous nodules of the skin, vasculitis, pericarditis, pulmonary nodules, pleural effusions, or chronic interstitial pulmonary fibrosis. Nerve entrapment, most commonly carpal tunnel syndrome, can be seen as can mononeuritis multiplex with vasculitis or a peripheral sensory neuropathy. Inflammation of the eye is seen, with episcleritis and scleritis. The latter can result in scleral perforations. Other manifestations might be sicca syndrome (dry eyes and mouth), Felty's syndrome, and synovial cysts. A common site for a synovial cyst is in the popliteal space and is known as a Baker's Cyst.[23]

Course of the Disease

The course of RA is variable among patients. Some patients may have several years of polyanatis with very little destructions of the bone, whereas others may rapidly progress to total disability. There are some patients who go into spontaneous remission during the early phase of their disease.[24] More frequently, the disease has periods of exacerbations and improvements, with a slowly progressive course of impaired functional ability and increased mortality.[25]

Drug Therapy

Traditional drug therapy in RA is approached in a pyramid formation, with nonsteroidal and inflammatory drugs (NSAIDs) serving as the base of the pyramid. The next progression is to slow-acting antirheumatic drugs (SAARDs), also called disease-modifying antirheumatic drugs (DMARDs) by some authors. We have selected DMARDs as the classification for drugs such as antimalarials, gold, D-penicillamine, methotrexate, azathioprine, and immunosuppressive cytotoxic agents. Corticosteroids are used selectively as an adjunct between NSAIDs and DMARDs. A description of the current medications is used by rheumatologists follows.

Nonsteroidal Anti-inflammatory Drugs

Nonsteroidal anti-inflammatory drugs, including aspirin, are some of the most widely used drugs in the world. The NSAIDs are well recognized for their analgesic and anti-inflammatory properties. As a group, the NSAIDs are inhibitors of the enzyme cyclooxygenase, which catalyzes the conversion of arachidonic acid to prostaglandins, prostacyclin, and thromboxanes. The NSAIDs that are currently available seem to have little or no effect on lipoxygenase, which transforms arachidonic acid to leukotrienes. A simplified version of the enzymatic pathway of arachidonic acid metabolism is depicted in Figure 1. By impairing the natural mediators of inflammation found in the arachidonic acid path, NSAIDs presumably interfere with the final expression of inflammation. These drugs do not prevent tissue injury and they are not the agents of choice for disease-modification. Other characteristics demonstrated by NSAIDs are the suppression of bradykinin release, alterations of lymphocyte responses, and decreasing of granulocyte and monocyte migration and phagocytosis.[26,27]

Aspirin has been one of the first-choice NSAIDs used to treat patients with RA. To appreciate the amount of aspirin prescribed for an individual with RA, the usual over-the-counter preparation contains from 200 to 400 mg per aspirin tablet. The analgesic properties of aspirin seem to result from interference with the synthesis of prostaglandins and the effects of bradykinin. The anti-inflammatory effects are advantageous in the management of RA. Analgesia can be achieved from doses of 600 to 900 mg of aspirin, whereas much larger doses (3,900-6,500 mg daily) are required for anti-inflammatory effects. Serum salicylate levels in steady state of 20 to 30 mg/dL are desired for anti-inflammation.[29] Aspirin differs from acetaminophen, which possesses no clinically useful anti-inflammatory property but does have analgesic capabilities.[28]

The large doses of aspirin required to achieve an anti-inflammatory effect border on toxic levels; however, many patients are able to offset the gastric distress created by aspirin by taking enteric-coated preparations.[30] The important property of the enteric coating is to prevent the breakdown of the aspirin in an acid medium, as found in the stomach; however, the coating will break down in a base medium, as occurs in the duodenum. Another concern generated by aspirin use is its ability to irreversibly inhibit platelet function, thus interfering with platelet aggregation and prolonging bleeding time.[28,31,32] Other adverse reactions to aspirin are dose related and are reversible with reduction of the amount of the drug, as in the case of tinnitus and diminished hearing, vertigo, dizziness, loss of balance, metabolic acidosis, and renal and hepatic toxicity. Despite the efficacy of aspirin, in our experience, the majority of patients with RA who begin treatment with this agent eventually discontinue its use because of one or more adverse reactions.[29,33] Some reactions are dose related and can be managed by simply reducing the daily dose; however, with the creation of newer NSAIDs, the belief exists that there are safer alternatives to aspirin, even though these medications are more costly for the patient.[28]

Some of the common NSAIDs used, their dose range, doses per day, and duration of action are listed in Table 1. The nonsalicylate NSAIDs were created specifically for patients unable to tolerate therapeutic doses of aspirin and other salicylates. Although the nonsalicylate NSAIDs are pharmacologically similar to salicylates and influence natural body processes, as shown in Figure 2, there is considerable variability in patient response to each medication. Decisions regarding NSAID use are made between the patient and the physician, giving consideration to dosing schedules, side effects, and preference. Which NSAID will be most effective in a which patient is not always predictable. The NSAID that is most effective in a patient may be identified only by multiple challenges with several drugs.[28] There are a large number of drugs in this classification, with new drugs proliferating rapidly. Brooks[34] suggests that although there are subtle differences among NSAIDs in terms of pharmacokinetics, adverse reactions, and the mechanism of action, none of these factors explain the wide variability of response noted in patients with RA.

[TABULAR DATA 1 OMITTED]

Because of the better tolerability of NSAIDs, patients with long-standing RA will continue to use them, even though the risk of side effects will increase with aging and long term-use. Roth and Bennett[35] have reported that elderly women with end-stage RA are particularly at risk for gastropathy. A second major concern in the elderly individual with end-stage RA is the adverse effects of NSAIDs on renal function.[36] Selected potential adverse reactions of the NSAIDs as a class of drugs are presented in Table 2.

Table 2. Selected Adverse Reactions
to Nonsteroidal Anti-inflammatory Drugs
Site                     Symptom
Gastrointestinal         Indigestion
                Gastroesophageal reflux
                Peptic ulcer
                Gastrointestinal hemorrhage
                  and perforation
                Small- and large-bowel
                  ulceration
Hepatic              Hepatocellular changes
                Transaminase elevation
Renal                Transient rise in serum
                creatinine
                Acute renal failure
                Interstitial nephritis
                Nephrotic syndrome
Hematologic          Thrombocytopenia
                Neutropenia
                Red cell aplasia
                Hemolytic anemia
                Prolongation of bleeding
                  time
Cutaneous       Photosensitivity
                Erythema multiforme
                Urticaria
                Drug-induced rashes
Respiratory          Bronchospasm
                Pneumonitis
Central nervous           Headache
  system             Dizziness
                Personality change
                Aseptic meningitis
                Tinnitis
                Confusion

Some discussion of drug interactions is in order, although an in-depth discussion of these interactions is outside the scope of this article. Because NSAIDs are taken primarily by elderly patients with RA and concomitant diseases, it is likely they will be taking other medications. The most troublesome drug interactions occur between NSAIDS and diuretic and hypotensive agents, anticonvulsants, lithium, and anticoagulants. Careful dose planning and drug selection are required to avoid severe drug interactions.

Corticosteroids

Because immune-driven inflammation is a central feature of RA, the therapeutic administration of corticosteroids produces reliable, effective, and rapid suppression of synovitis. George and Kirwan[37] concluded in clinical trials of both short- and long-term administration of corticosteroids that these are effective anti-inflammatory agents and that they are much superior to NSAIDs. Unfortunately, in order to sustain the benefits of these drugs, there is an increased risk of side effects and adrenal suppression with long-term use.[37] The adverse effects of corticosteroids are outlined in Table 3.

Table 3. Adverse Effects of Systemic
Corticosteroid Therapy
Metabolic               Obesity, glucose/protein
                        metabolism,
                        electrolyte imbalance,
                        enzyme induction,
                        corticosteroid
                        withdrawal syndrome,
                        hypothalamic-pituitary
                        -adrenal axis
                        suppression, growth
                        retardation
Immunity                Inclination for
                        opportunistic
                        infections
Musculoskeletal         Myopathy,
                        osteoporosis,
                        osteonecrosis, tendon
                        rupture
Gastrointestinal        Peptic ulcer disease,
                        pancreatitis
Central nervous         Psychosis, depression,
  system                benign intracranial
                        hypertension
Ophthalmic              Cataract, glaucoma
Dermatologic            Acne, striae, alopecia,
                        bruising, skin atrophy

The efficacy of corticosteroid use in RA depends on the dose, dosing schedule, and length of treatment. Because of the numerous corticosteroid preparations available and the variety of dosing regimens suggested, astute clinical judgment is required to prescribe their use in patients with RA. It is known that by increasing the dosage and dosing frequency, inflammatory suppression is enhanced, more rapid onset of therapeutic benefit occurs, and the risk of side effects increases.[38] For example, doses given several times a day are more potent than doses given once a day. Often the choice of therapeutic regimen is driven by the need to minimize toxicity yet provide sufficient anti-inflammatory effect to modify disease activity. Corticosteroid preparations may be given orally, intramuscularly, or intravenously. When disease control is required but not urgent, daily oral administration in the morning may be sufficient. If the disease activity is such that more rapid control is required, split doses given two to four times a day may be recommended t6 achieve a greater degree of anti-inflammatory effect more quickly.[38] There are occasions when RA requires immediate intervention for damaging inflammation. In this case, treatment can be initiated with a short course of high dosages of corticosteroid. Pulse therapy with, intravenous methylprednisolone, 1,000 mg infused over 30 to 45 minutes, given daily for 1 to 3 days, has been demonstrated to be effective in curtailing the inflammatory process.[39,40] Table 4 presents an overview of options of dosing regimens used in RA. How corticosteroids achieve their profound effect on the inflammatory process is not entirely understood. It has been suggested[38-40] that corticosteroids may interrupt the inflammatory and immune cascade at several levels such as (1) by impairing antigen opsonization (opsonization is the rendering of antigens susceptible to phagocytosis), (2) by interfering with inflammatory cell adhesion and migration through vascular endothelium, (3) through interruption of cell-cell communication by alteration of release or antagonism of cytokines, (4) by impairment of leukotriene and prostaglandin synthesis, and (5) by inhibition of neutrophil superoxide production. Furthermore, it appears that corticosteroids decrease immunoglobulin generation, inhibit immune clearance of sensitized erythrocytes, and impair the transport of immune complexes through cell membranes.[40]

[TABULAR DATA 4 OMITTED]

Several considerations must be taken into account when prescribing the use of corticosteroids, including the general health and age of the patient, the specific disease process to be treated, concurrent illnesses such as diabetes or hypertension, the advantages or disadvantages of one corticosteroid preparation versus another, and the dosing regimen. if long-term usage of corticosteroids is necessary, a key point that has to be considered is what minimum dosage can be used to achieve disease suppression. This can be a difficult decision to make because RA is noted for fluctuations in the disease process and requires the drug therapy to be lengthened more than previously anticipated.

The preferred protocol for drug withdrawal following the administration of high dosages or prolonged dosing of corticosteroids is gradual tapering of the drug in such a fashion as to avoid a major flare of the disease requiring another increase in the corricosteroid dosage or frequency. Abuptly decreasing or withdrawing corticosterolds from a patient with RA frequently facilitates a marked flare in the disease. When dosages greater than 40 mg/d of prednisone or an equivalent are given, dose reductions can be made in decrements of 10 mg every few days to weeks. More rapid decrements invite exacerbation of the disease.[38] In drug dosages below 40 mg/d, serial reductions of 2.5 to 5.0 mg/d every few weeks or months, or even less in some patients with RA, are preferable. When the patient has received doses below 10 mg/d, the reduction interval must be sufficient to allow the adrenal glands to begin functioning. The length of this interval is proportional to the amount of time the patient has used corticosteroids: The longer the therapeutic use, the longer the reduction intervals must be. Corticosteroid use of less than 4 weeks usually does not result in persistent adrenal suppression.

A worst-case scenario from the exogenous administration of corticosteroids is hypothalamic-pituitary-adrenal suppression of a magnitude that the individual lacks the ability to increase corticosteroid production during such demands as stressors, infection, trauma, or surgery.[38] The end result may be an Addisonian crisis and adrenocortical shock, which can be life threatening. The symptoms of adrenal crisis are profound asthenia; severe pain in the abdomen, lower back, and legs; peripheral vascular collapse; and finally renal shutdown. All of these symptoms can result in death.

Occasionally, intra-articular injections of corticosteroids are necessary to control acute inflammation in a joint that is otherwise uncontrollable. Injection of corticosteroids into a joint can provide prolonged relief and markedly improve the mobility and function of the patient. The rationale for use in the joint is to suppress the synovitis, because there is currently no evidence that intra-articular injections retard the progression of erosive disease. Intraarticular injections must be carefully selected and must not be abused. No single joint should have more than three or four injections before other procedures are pursued.[38] Some authors have recommended that the patient should not bear weight on an injected joint for several days after injection[41]; however, others have reported there is no need for a such a conservative approach. The less conservative approach would permit nonstrenuous activity.[42] Vigorous exercise may speed resorption of the steroid from the joint and reduce the intra-articular efficacy.

Previous adverse reactions to corticosteroids or concomitant illnesses may influence the choice of the corticosteroid and the dosing regimen. Corticosteroids are known to decrease collagen synthesis and impair wound healing, augment hepatic gluconeo-genesis and glycogen deposition while inhibiting the action of insulin (detrimental in diabetes mellitus), impair lipogenesis and stimulate lipolysis in adipose tissue, and increase liver synthesis of protein while enhancing peripheral catabolism. The profound effects of these drugs on bone metabolism by interfering with intestinal resorption of calcium, inhibition of osteoblast collagen synthesis, elevation of parathormone levels that amplify osteoclastic bone resorption, and enhancing renal excretion of calcium are well known.[3,44]

Disease-Modifying Antirheumatic and Cytotoxic Drugs

Fortunately, over the past 20 years, the armamentarium of drugs that can be used in the treatment of patients with RA has changed considerably. There is an increased agreement among rheumatologists to begin aggressive treatment early with medications that have the potential of preventing cartilage damage. Brook and Corbett[45] demonstrated that bony erosions occur early in RA, indicating the need for early treatment with previously considered second-line drugs.[46] Disease-modifying antirheumatic drugs and cytoxic agents are drugs that might be considered in this case. Table 5 presents the current profile of the DMARDs and the availability of their use. other descriptors used to name these drugs are "slow-acting antirheumatic drugs" and "second-line agents." Although rheumatologists intuitively believe that these drugs alter the natural course of RA, it is difficult to document disease modification. Characteristically, these medications are slow in onset of action and require several months before a response is seen. All of the DMARDs are toxic and require close monitoring for side effects.[31]

[TABULAR DATA 5 OMITTED]

The first choice of DMARDs is those that are the safest to use. Many of these drugs were first used empirically and, as in the case of gold salts (chrysotherapy), somewhat serendipitously. Chrysotherapy has been used since 1927, when it was first recognized by Lande[47] that the use of aurothioglucose in patients with bacterial endocarditis and other conditions resulted in relief of joint pain. Forestier[48] reported 550 cases of RA he had treated over a 6-year period and believed that between 70% to 80% of his patients had responded favorably to the treatment with gold. Although the initial reports of the use of gold described positive results, the introduction of corticosteroids derailed the interest in using gold for treatment of patients with RA. In 1960, the Empire Rheumatism Council of Great Britain reported the results of a double-blind, prospective, controlled trial comparing gold with placebo in 200 patients with RA.

The results astounded the rheumatology community by demonstrating that those individuals treated with gold had a substantial reduction of inflamed joints, increased functional capacity scores, and improvement in their global condition and general strength as compared with a placebo group. Equally important was the finding that the treatment regimen used produced a low level of toxicity in the subjects. Hence, 50 years after the initial reports of the efficacy of gold in managing joint pain, chrysotherapy is an important DMARD in the treatment of patients with RA.[49] The relative safety of the DMARDs is shown in Table 6.

Table 6. Relative Safety and Toxicity
of Disease-Modifying Antirheumatic
Drugs(31,a)
Safest           More Toxic Very Toxic
Auranofin        Azathioprine    Chlorambucil
Hydroxychloro-   Cyclosporin     Cyclophosphamide
  quine
Sulfasalazine    Gold salts
                 Methotrexate
                 Penicillamine
(a) From the Primer on the Rheumatic Diseases,
10th edition, copyright 1993. Used by permission
of the Arthritis Foundation.

Unfortunately, the mechanism of action of gold and the other DMARDs remains obscure. For those DMARDs approved by the Food and Drug Administration (FDA) for use in RA, the efficacy versus a placebo has been reported in the literature.[50] Some possible mechanisms of action of the commonly prescribed DMARDs are depicted in Table 7. Various factors influence the decision to use one DMARD over another. For example, in an individual with low-grade synovitis in early RA, the choice might be an antimalatial or sulfasalazine, whereas in an individual with unrelenting disease with synovitis and extra-articular manifestations, the drug of choice might be injectable or parenteral gold or low-dose methotrexate.[51] Felson et al[52] reported the results of a meta-analysis of randomized controlled trials of DMARDs in RA, demonstrating that there is little difference in efficacy between methotrexate, injectable gold, D-penicillamine, and sulfasalazine. The investigators also reported that auranofin (oral gold) and hydroxychloroquinine (antimalarial) are significantly less effective than those DMARDS mentioned.[52]

Because the reader may be unfamiliar with DMARDs, a brief description of each seems in order. It is important to recall that some DMARDs have been approved for use by the FDA and others are still undergoing clinical trials. The current FDA status of DMARDs used in the treatment of patients with RA is outlined in Table 5.

Auranofin (Ridaura). This DMARD is an oral gold preparation whose mechanism of action, like the other drugs in this classification, is not known and is subject to conjecture. The majority of opinion is that therapy has the ability to retard the radiologic progression of RA; however, auranofin is less effective than injectable gold or other DMARDS.[52] Champion et al[53] have suggested that gold compounds may be transported to cells or tissues and form monomeric aurocyanide from cyanide released during polymorph phagocytosis. The theory is that auranocyanide is able to enter activated polymorphs and macrophages in RA. The end result would be the reduction of levels of rheumatoid factor and the erythrocyte sedimentation rate (ESR).

The route of administration is by mouth, and auranofin is given in a dosage of 3 to 9 mg/d, with the usual dosage being 6 mg/d. Improvement of symptoms is not usually expected for about 3 months. The usual trial period is 6 months to determine efficacy. To monitor the efficacy of auranofin, the patient usually has blood tests and a urinalysis monthly while taking this medication.[31] Some patients experience gastrointestinal discomfort and diarrhea, which are generally dose limited. Toxicity is less common with the oral drug than with the injectable form of gold; however, symptoms of drug toxicity are mucocutaneous reactions, conjunctivitis, proteinuria, and thrombocytopenia.[54] Parenteral gold salts. The two most commonly used gold compounds in the United States are gold sodium thiomalate (aurothiomalate) and aurothioglucose. The mechanism of action, as with auranofin, is open for debate. Champion et al[53] have indicated that gold is transported to and from the synovial fluid as a complex with albumin and is closely bound to inflamed synovial tissue, being found in synovial lining cells and deeply placed macrophages. A new finding suggests that aurothiomalate may regulate gene expression by inhibiting the binding of gene transcription factors to their response elements in deoxyribonucleic acid (DNA).[55]

The route of administration is by deep intramuscular injection. Generally, a test dose of 10 mg is given, followed by a second dose of 25 mg the next week. Subsequent injections of 50 mg are given weekly until a cumulative dose of 1 g is reached or the patient experiences toxicity. If no improvement is demonstrated after a cumulative dose of 1 g, some rheumatologists will discontinue using the drug, whereas others will continue use until the dose is equal to 2 g or toxicity occurs. If a favorable response is observed in terms of decreased synovitis and pain and improved functional measures (eg, increased grip strength or walking time), the dosing interval is increased to 2 weeks and eventually to 3 or 4 weeks if the disease does not flare. Complete blood counts and urinalyses are required before the next injection to monitor the effects of the drug.

The expected results demonstrating efficacy should appear within 6 to 12 months of treatment. Gold appears to improve the symptoms and signs of RA, to decrease the ESR and rheumatoid factor, and to slightly retard radiologic progression of the disease.[53] Toxicity with injectable gold is not uncommon. Some of the more obvious complaints by patients are mouth sores, rashes, and a syndrome of sweating, flushing, and faintness. This syndrome can occur within minutes of having an injection. Other reactions are proteinuria, neutropenia, and thrombocytopenia.[56,57]

Antimalarials. Hydroxychloroquine and chloroquine have been used in the treatment of patients with RA since the 1950s and are classified with the DMARDS. The two drugs seem to be no different in reducing the inflammatory symptoms in RA; however, the influence of these drugs on the disease process is not as dramatic as that of other DMARDS. The mechanism of action of these drugs is better understood than that of most of the drugs in this group. Antimalarials inhibit multiple-function phagocytes, including reactive oxygen species release.[58] Antimalarials might inhibit the antigenprocessing ability of monocytes or macrophages, thus inhibiting the response of lymphocytes to antigens. These drugs appear to inhibit the release of interleukin-1 from monocytes, thus interfering with the inflammatory process in RA.[59]

Antimalarials are taken by mouth. The recornmended doses are 5 mg/kg/d for chloroquine diphosphate and 7.5 mg/kg/d for hydroxychloroquine sulfate (400 mg/d in most patients). Response to the medication generally requires several months, and 6 months is required for a fair trial. The expected response takes several months (4-6 months is usual) and includes, as reported for 60% to 80% of patients compared with a placebo in a drug trial, 60 decreased disease activity and increased functional abilities. Antimalarials have not been proven to retard radiological progression of the disease.

When compared with other DMARDS, the incidence of side effects associated with antimalarials is transient and cessation of the drug is unnecessary. The major adverse reactions are skin rashes, indigestion, visual disturbances, and retinopathy.[51] To reduce the risk of visual impairment, a fundoscopic examination and visual field testing are recommended every 3 to 6 months.

D-penicillamine. D-penicillamine has been used in the treatment of patients with RA for over 25 years. The mechanism of action is unknown, although is seems likely that the drug modulates the immune system, acting as an immunosuppressant.[59] The drug should be taken in a nonenteric form about 1 to 2 hours before meals or ingestion of antacids and iron preparations. A dose of 250 mg is gradually increased to a maintenance dose of 750 to 1,000 mg/d in 4 to 8 weeks. The drug should be continued for 4 to 6 months before deciding whether it is efficacious.

A positive response to D-penicilamine should show improvements in both disease manifestations and functional abilities. There are conflicting data whether D-penicillamine prevents joint erosions in RA[52] or prevents the formation of nodules or extra-articular manifestations. Monitoring of blood and urine are required. Twenty-five percent or more of patients using D-penicillamine will discontinue use of the drug within 12 months of beginning its use due to side effects. Adverse reactions to this medication include skin rashes, proteinuria, hematuria, neutroperga, thrombocytopenia, and a variety of autoimmune phenomena, including myasthenia gravis, polymyositis, pemphigus, and Goodpasture's syndrome.[31]

Methotrexate. The mechanism of action of methotrexate demonstrates that it interrupts purine biosynthesis, thus producing an anti-inflammatory effect. Some of the effects are reduced IgM-rheumatoid factor production; reduced polymorphonuclear chemotaxis; decreased IL-1 and IL-2 production, secretion, or binding; and decreased IL-6 activity.[61] Methotrexate is a folic acid antagonist that impairs DNA synthesis, The anti-inflammatory efficacy of the drug has been demonstrated, including a study showing that the effect is dose related.[62]

The drug can be orally administered or intramuscularly injected, starting with a usual dosage of 7.5 mg/wk, either as a single dose or divided over 24 hours. Maximum doses are 15 to 20 mg/wk. Tugwell et al[63] have reported that methotrexate works rapidly (within 1-2 months), reaching peak efficacy in about 6 months. Wolfe et al[64] Suggested that patients with RA remain on methotrexate for a longer period of time than with other SAARDs. Response to the drug is generally seen in 6 to 8 weeks.

Concurrent therapy, with sulfacontaining antibiotics or the presence of human immunodeficiency virus infections are contraindications for use. Alcohol consumption, gross obesity, and diabetes mellitus may contribute to hepatic toxicity. While taking the drug, the patient should have complete blood counts monthly and liver enzyme studies every 2 to 3 months. Recent recommendations regarding liver biopsy suggest the need for biopsy only in certain circumstances. Common side effects of methotrexate use are skin rashes, gastrointestinal symptoms, transaminase liver enzyme elevation, hepatic fibrosis, proteinuria, increased risk of infections, and hematologic abnormalities.

Azathioptine. This DMARD is a cytotoxic purine analog with the potential for severe side effects. In RA, azathioprine reduces the amount of circulating B and T lymphocyctes,[65,66] mixed lymphocyte reactivity,[67] IgM and IgG synthesis,[66] and interleukin IL-2 secretion.[68]

Azathioprine is taken orally. The usual starting dose is 1.25 to 1.5 mg/kg/d, with a maximum dose of 2.5 mg,/kg/d. The efficacy is delayed for several months and may require a trial of 6 months before changes in disease activity and functional ability are noted. Common adverse reactions include nausea and vomiting, rash, bone marrow suppression, and hepatitis. The drug requires careful monitoring, including blood counts monthly and liver function tests every 3 months. Because renal excretion is the major route of elimination, the kidneys must be monitored closely.

Sulfasalazine. As a DMARD, sulfasalazine interferes with the inflammatory process in RA, and there is some suggestion that it slows bone erosion. [69] Symmons et al[70] have reported that the numbers of activated lymphocytes circulating were reduced in 12 weeks of sulfasalazine therapy.

Taken orally, the adult dosage is around 2 g/d (range=0.5-3 g/d) of an enteric-coated preparation, usually taken with meals. The starting doses are generally 500 mg/d to 1 g/d to help the patient develop a tolerance for taking the drug. Sulfasalazine exhibits antirheumatic actions in terms of modifying disease activity in about 4 weeks following the initiation of the drug. There are conflicting reports whether the drug slows radiologic progression of the disease.

Up to 50% of patients who take sulfasalazine develop side effects, which generally occur within the first 4 months of use. Only about one half of those who develop side effects, however, will be required to stop the medication.51 Side effects include skin rashes, abdominal pain, liver enzyme abnormalities, central nervous system disturbances, and blood dyscrasias. Close monitoring of the liver and blood counts is necessary early, but is needed less frequently with prolonged use.

Chlorambucil. Chlorambucil crosslinks DNA and proteins, thus preventing cell replication; however, it is not considered cytoxic.[67,71] Its active metabolite is phenylacetic acid mustard, which appears to inhibit an established cellular and humoral response, similar to cyclophosphamide. Chlorambucil can be administered orally in doses between 0.1 and 0.2 mg/kg/d.[72,73] The drug is rapidly absorbed and is particularly beneficial for patients with RA who are faced with a life-threatening complication such as systemic vasculitis or rampant disease. The beneficial effect of decreased symptoms is usually seen in 1 to 2 months.

The most common adverse reactions with chlorambucil use are dose-related bone marrow suppression and infertility. Cumulative bone marrow toxicity often requires discontinuation of the drug. With increasing doses and duration of therapy, there is an increased risk of infertility, azoospermia, and amenorrhea.[74,71] Chromosomal damage can be induced by chlorambucil, and the major adverse reaction is an increased risk of developing leukemia and other malignancies.[31,51]

Cyclophosphamide. Cyclophosphamide is a derivative of nitrogen mustard and is cytoxic to dividing and resting lymphocytes.[67,71,76] The drug effectively suppresses a variety of cell-mediated responses and has anti-inflammatory properties.[71,76] Oral administration of cyclophospharnide in clinical trials has demonstrated that the drug is effective in the treatment of severe RA when used in doses of >1.5 mg/kg/d.[75] The drug reaches its peak effectiveness after approximately 16 weeks of use.[75]

Because of the toxicity produced by cyclophosphamide, this drug is recommended for use only in patients for whom all other DMARDS have been ineffective and who have severe, unrelenting disease.[31] Oral cyclophosphamide at a dose of 2 mg/kg/d might be considered in these patients. Patients taking this drug should be cautioned to force fluids for effective kidney function because cyclophosphamide can cause hemorrhagic cystitis that can lead to bladder fibrosis. Because cyclophosphamide is an immunosuppressive agent, the adverse events associated with its usage are increased risk of infection, suppression of gonadal function in both men and women, and the long-term effect of an increased incidence of lymphoma and other hematologic malignancies.[31,51]

Cyclosporin. Cyclosporin A is an immunomodulating agent thought to selectively block the synthesis and release of interleukin-1 from monocytes and interleukin-2 from T-helper cells.48 It is a drug that is used in the suppression of organ transplant rejection. Although it is considered an experimental drug for use in RA, it has been reported in three placebo controlled trials to be an effective agent for treatment of severe RA.-[77,78]

Taken orally, patients taking 2.5 to 5.0 mg/kg/d should demonstrate a response within 3 to 4 months and in some cases as early as 1 month. When used at low doses in treatment of RA, side effects are generally mad and reversible. The maximum dose is 5 mg/kg/d. There is risk, however, for irreversible loss of renal function and for hypertension, To ascertain the levels of cyclosporin for toxicity, it is best to monitor by frequent serial serum creatinine levels every 7 to 14 days with dose manipulation to keep the serum creatinine within 25% of baseline.[77,78]

General Approach to Medication Use

The objectives of drug therapy in RA are pain relief, decreased joint and tissue inflammation, maintenance and restoration of joint function, and prevention of bone and cartilage destruction. An important aim is to modify the course of the disease, although research has yet to prove this aim is accomplished. Interrupting the inflammatory process is an important goal of treatment with medications. In general, medications for use in treatment of RA can be categorized into two groups: (1) those that provide symptomatic relief and (2) those that have the potential for modifying the course of the disease.[79]

Management Plan by Pyramid Approach

The choice of medications to be used in the treatment of RA is made after determining whether the disease is in the early, established, or late stage of progress and the rapid of bone and cartilage damage. Figure 3 depicts how decisions might be made regarding the drug of choice. Although symptomatic medications such as aspirin and NSAIDs reduce pain and swelling, they are not suited to modifying the process of joint and tissue destruction.

Historically, rheumatologists have approached traditional drug therapy by using a pyramid approach similar to that depicted in Figure 4. The base of the pyramid begins with the use of NSAIDs and progresses to experimental and cytoxic drugs. Decisions of how to progress through the pyramid are variable among physicians, as there are no data to substantiate one method versus another.[31] Recent criticism of the pyramid approach has appeared in the literature.[46,80] These authors propose that rheumatologists use an approach to drug therapy similar to that used by oncologists who initiate the use of multiple medications and then gradually remove drugs one at a time until a single or foundation regimen is achieved, thus reducing the necessity of multiple drug use. There have been no controlled clinical trials in RA to support the rationale for multiple second-line drug usage.

Discussion

Unfortunately, there is a paucity of research literature describing the impact of medications taken by persons with RA on the success or failure of their rehabilitation program. Our experience has been that individuals with pharmacologically uncontrolled RA will have a difficult time cooperating with physical therapy intervention. Although there are palliative measures the physical therapist can teach the patient to do to care for himself or herself, finding the appropriate medication; encouraging the patient to cooperate by taking it; and keeping close contact between the patient, the physician, and the physical therapist are imperative for a successful physical therapy program.

Medications may influence the ability of individuals to participate in an exercise program by modifying their mood or perception of pain. The side effects of individual drugs may cause easy bruising, increased bleeding time, myopathy, delayed healing, capillary fragility, or osteoporosis, or the drugs may not be effective as a pain medication. Some medications may interact with each other to cause adverse reactions. Corticosteroids may cause psychosis in elderly patients, promote infections, or cause skin and capillary disruption. All of these conditions can influence the progress of the physical therapy program. This possibility, however, does not preclude a treatment program from being instituted and carefully monitored for any of the reactions described. Careful use of medications can optimize an exercise program by reducing pain and inflammation so that the patient can maintain his or her mobility. The challenge to the physical therapist is to reduce the potential of exercise-induced overuse or strain of muscles and joints because the patient's perception of pain is altered.

It is important to be realistic about what can be achieved by the patient. Individuals with long-standing use of steroids may make very slow progress because of steroid myopathy and osteoporosis. Rolling over on a mat may induce a compression fracture of the spine. We have had cases of patients developing neuropapthies due to interference of blood flow the vasa xiasorum to the nerve. Elderly patients with RA may be taking multiple drugs, including steroids and NSAIDS, which may impair their ability to complete a physical therapy regimen.

Therapists should be well versed in the side effects and adverse reactions of the medications their patients with RA are taking. Rehabilitation can be impaired by gastric bleeding, hypertension, fatigue, nausea, and a host of other symptoms patients may experience as a result of taking their medicine. Patients may not understand that not taking their medication, not taking their medication correctly, or the adverse effects of their medication may contribute to their lack of success in physical therapy. Many patients are taking multiple medications, and their symptoms may be be attributable to polypharmacy.

Certainly, we intuitively believe that without these medications, patients could not perform as well as they do; however, the rehabilitation literature does not report any clinical trials between medication use and rehabilitation measures in patients with RA. Therein lies a challenge for researchers in physical therapy. Knowing the duration of action and peak effect of a medication will provide a window of time when physical therapy intervention can be optimized. This, of course, is with the proviso that correct judgment is applied by both the patient and the therapist to avoid repetitive trauma, overuse, or harmful activities to already compromised joints. The goal of bringing RA under control is achieved in most patients by combining several "measures, including medications, physical therapy and occupational therapy, rest, and psychosocial support. These individuals, although not cured of RA, manage to live successful lives within the framework of their disease. About 5% to 10% of individuals with RA have unrelenting disease that requires the institution of powerful medications and methods of care to try and abort what will otherwise be serious progressive disease and disability. A certain level of physical wellness and health is necessary for a meaningful and satisfactory life. Those patients who lose the capacity to have their health or establish meaningful life experiences are endangered as pain and deformity threaten to destroy their quest for an improved quality of life. Yelin et al[181] have reported that 50% of patients with RA who were previously working become too disabled to continue their work within 10 years after the onset of the disease. The main factors contributing to their inability to work were the duration and severity of the disease, social status, and working conditions. Our experience has been that patients with unrelenting disease become permanently disabled and are unable to work at the job they had when they were diagnosed with RA.

These considerations, along with others, have led scientists to conduct drug trials with medications that have the potential of altering the course and duration of the disease. Once it became established that RA was an immunogenic phenomenon, the introduction of cytoxic drugs in the treatment of intractable RA became important. With the introduction of these considerably powerful drugs has come the phenomena of considerable powerful side effects and adverse reactions. The seriousness of these side effects, including the possibility of cancer, is worrisome for both the patient and the physician. Combined chemotherapies are being studied to treat difficult RA on the principle that these combined drugs would suppress inflammation, synovial proliferation, pannus formation, and joint destruction and may be used in lower doses with lesser toxicity.

As previously mentioned regarding use of a pyramid approach to drug therapy, however, there are those who prefer to use a single efficacious agent so that the patient will not be subjected to so many potentially harmful agents. Because of this controversy, considerable thought, planning, and communication must occur between the physician and the patient. In a disease such as RA that can fluctuate over time, the patient, the physician, and the physical therapist have to ask the questions: When are too few or too many drugs making the situation worse? When is too much treatment ethical or unethical?

Over the last 25 years, the introduction of new drugs and the incorporation of behavioral, lifestyle, and activity changes are allowing individuals, who might not otherwise do so, to become active and live meaningful lives in spite of having RA. Physical therapists have something to offer these patients to help empower them and, to a certain unsubstantiated degree, decrease the necessity of relying on medication as the only source of pain and discomfort relief.

The implication of the current treatment of RA for physical therapy researchers might be to scientifically assess and substantiate the extent to which physical therapy measures can reduce the need of the patient for drugs to relieve the symptoms of RA. The primary author (CM) is often asked to provide research data that demonstrate physical therapy measures alter the onset and course of synovitis in patients with RA. It is difficult to find corroborating literature, thus presenting another challenge for physical therapy researchers. For example, when a patient is taking a NSAID and using heat treatments concurrently, as recommended by a physical therapist, to what extent does the heat interact with the NSAID to influence the progress of die patient on clinical and functional measures? Well-controlled clinical trials might help to determine the answers to such questions.

Conclusion

The purpose of this article was to present the current medications used in the treatment of RA and to facilitate for physical therapists consideration of how the use of these drugs can enhance or deter physical therapy intervention. Optimum treatment for the person with RA involves a combination of medications; physical therapy and occupational therapy; rest; psychosocial support, including techniques of self-efficacy; and lifestyle modifications.

[Figure 1 to 4 ILLUSTRATION OMITTED]

References

[1] Vaughan JH, Kouri T, Petersen J, et al. On the etiology of rheumatoid arthritis. Scand J Rheumatol Suppl. 1988;74:19-28. [2] Stastny P. Association of the B-cell alloantigen Drw4 with rheumatoid arthritis. N Engl J Med. 1978.:298:869-871. [3] Schiff B, Mizrachi Y, Orgad S, et al. Asscociation of HLA-AW3L and HLA-DR1 with rheumatoid arthritis. Ann Rheum Dis. 1982.:41:403-404. [4] Woodrow JC, Nichol FE, Zaphiropoulos G. DR antigens and rheumatoid arthritis: a study of two populations. Br Med J 1981;283:1287-1288. [5] Moreland LW, Koopman WJ. Infection as cause of arthritis. Curr Opin Rheumatol. 1991; 3:639-649. [6] Tuckwell JE, Esparza L, Carson DA, et al. The susceptibility sequence to rheumatoid arthritis is a cross-reactive B cell epitope shared by the Escherichia coli heat shock protein DNA and histocompatibility leukocyte antigen DRB10401 molecule. J Clin invest. 1992;89:327-331. [7] Cooper NS. Pathology of rheumatoid arthritis. Med Clin North Am. 1968;52:607-621. [8] Simkin PA. The musculoskeletal system: joints. In: Schumacher HR, Klippel JH, Koopman WJ, eds. Primer on the Rheumatic Diseases. 10th ed. Atlanta, Ga: Arthritis Foundation; 1993:5-8. [9] Schrieber L. Rheumatoid arthritis: the endothelium in rheumatoid arthritis. In: Klippel JH, Dieppe PA, eds. Rheumatology. London, England: Mosby-year Book Europe Ltd; 1994: 3.10.1-3.10.6. [10] Schumacher HR. Synovial membrane and fluid morphologic alterations in early rheumatoid arthritis: microvascular injury and viruslike particles. Ann NY Acad Sci. 1975;256:39-64. [11] Stevens CR, Blake DR, Merry P, et al. A comparative study by morphometry of the microvasculature in normal and rheumatoid synovium. Arthritis Rbeum. 1991;34:1508-1513. [12] Harris ED jr. Mechanisms of disease: rheumatoid arthritis-pathophysiology and implications for therapy. N Engl J Med. 1990;322: 1277-1289. [13] Van Boxel JA, Paget SA. Predominately T-cell infiltrate in rheumatoid synovial membranes. N Engl J Med. 1975;293:517-524. [14] Paulus HE, Machleder HI, Clements PJ. Prolonged thoracic duct drainage in rheumatoid arthritis and systemic lupus erythematosus. West J Med. 1979;130:309-324. [15] Paget SA, Gibofsky A. Immunopathogenesis of rheumatoid arthritis. Am J Med. 1979; 67:961-970. [16] Weissmann G. Activation of neutrophils and the lesions of rheumatoid arthritis. J Lab Clin Med. 1982;100:322-333. [17] Dingle JT. Articular damage in arthritis and its control. Ann Intern Med. 1978;88:821-826. [18] Jasin HE. Autoantibody specificities of immune complexes sequestered in articular cartilage of patients with rheumatoid arthritis and osteoarthritis. Arthritis Rbeum. 1985;28:241-248. [19] Woolley DE, Harris ED, Mainardi CL, Brinkerhoff CE. Collagenase immunolocalization in cultures of rheumatoid synovial cells. Science. 1978;200:773-775. [20] Werb Z, Mainardi CL, Vater CA, Harris ED. Endogenous activation of latent collagenase in rheumatoid synovial cells: evidence for a role of plasminogen activator. N Engl J Med. 1977; 296:1017-1023. [21] Ishikawa H, Ziff M. Electron microscope observations of immunoreactive cells in the rheumatoid synovial membrane. Arthritis Rheum. 1976;19:1-14.

[22] Abrahamsen TG, Froland SS, Natvig JB, Pahle J. Antibody-dependent cytoxcity mediated by cells eluted from synovial tissues of patients with rheumatoid and juvenile rheumatoid arthritis. Scand J Immunol. 1977;6: 1251-1261. [23] Katz RS, Zizic TM, Arnold WP, Stevens MB. The pseudothrombophlebitis syndrome. Medicine. 1977;56:151-164. [24] Pinals RS, Masi AT, Larsen RA. Subcommittee for Criteria of Remission in Rheumatoid Arthritis of the America Rheumatism Association Diagnostic and Therapeutic Criteria Committee: Preliminary criteria for clinical remission in rheumatoid arthritis. Arthritis Rheum. 1981;24:1308-1315. [25] Pincus T, Cal]ahan LF, Sale WG, et al. Severe functional declines, work disability, and increased mortality in seventy-five rheumatoid arthritis patients studied over nine years. Arthritis Rheum. 1984;27:864-872. [26] Abramson SB, Cherksey B, Gude D, et al, Nonsteroidal anti-inflammatory drugs exert differential effects on neutrophil function and plasma membrane viscosity: studies in human neutrophils and liposomes. Inflammation. 1990; 14:11-30. [27] Goodwin JS, Ceuppens JL. Effect of nonsteroidal anti-inflammatory drugs on immune function. Semin Arthritis Rheum. 1983; 13(suppl):134-143. [28] Dahl SL. Nonsteroidal anti-inflammatory agents: clinical pharmacology/adverse effects/ usage guidelines. In: Willkens RF, Daw SL, eds. Therapeutic Controversies in the Rheumatic Diseases. Orlando, Fla: Grune Stratton Inc; 1987:27-68. [29] Dromgoole SH, Furst DE, Paulus HE. Rational approach to the use of salicylates in the treatment of rheumatoid arthritis. Semin Arthritis Rheum. 1981;11:257-283. [30] Orozco-alcala JJ, Baum J. Regular and enteric coated aspirin: a re-evaluation. Arthritis Rheum. 1979;22:1034-loi". [31] Williams HJ. Rheumatoid arthritis: treatment. In: Schumacher HR, Klippel JH, Koopman WJ, eds. Primer on the Rheumatic Diseases. Atlanta, Ga: Arthritis Foundation; 1993: 96-99. [32] Patromo C, Ciabattoni G, Pinca E, et al. Low dose aspirin and inhibition of thromboxane B2 production in healthy subjects. Thrombosis Res. 1980; 1 7:317-327. [33] Katz WA. Modem management of rheumatoid arthritis. Am J Med. 1985;79(suppl 4C): 24-31. [34] Brooks PM. Pharmacologic approaches: NSAIDs. In: Klippel JH, Dieppe Pa, eds. Rheumatology. London, England: Mosby-year Book Europe Ltd; 1994:8.10.1-8.10.6. [35] Roth SH, Bennett RE. Nonsteroidal anti-inflammatory drug gastropathy. Arch Intern Med. 1987;147:2093-2100. [36] Brooks PM, Day RO. Nonsteroidal anti-inflammatory drugs: differences and similarities. N Engl J Med. 1991;324:1716-1725. [37] George E, Kirwan JR. Corticosteroid therapy in rheumatoid arthritis. Baillieres Clin Rheumatol. 1990;4:621-647. [38] Paulus HE, Bulpitt KJ. Nonsteroidal anti-inflammtory agents and corticosteroids. In: Schumacher HR, Klippel JH, Koopman WJ, eds. Primer on the Rheumatic Diseases. Atlanta, Ga: Arthritis Foundation; 1993:298-303. [39] Neumann V, Hopkins R, Dixon J, et al. Combination therapy with pulsed methylprednisolone in rheumatoid arthritis. Ann Rheum Dis. 1985;44:747-751. [40] Iglehart IW, Sutton JD, Bender JC, et al. Intravenous pulsed steroids in rheumatoid arthritis: a comparative dose study. Rheumatol. 1990;17:159-162. [41] Neustadt DH. Intra-articular therapy in rheumatoid synovitis of the knee: effects of post injection rest regimen. Clin Rheumatol Practice. 1985;3:65-68. [42] Chatham WW, Williams GV, Moreland LW, et al. Intra-articular injections: Should you rest joints? Arthritis Rheum. 1989;32(suppl 1) :R45. [43] Weiss MM. Corticosteroids in rheumatoid arthritis. Semin Arthritis Rheum. 1989;19:9-21. [44] Parrillo JE, Fauci AS. Mechanisms of glucocorticoid actions on immune processes. Ann Rev Pharmacol Toxicol. 1979; 19:179-210. [45] Brook A, Corbett M. Radiographic change in early rheumatoid disease. Ann Rheum Dis. 1971;36:71-73. [46] Wilske KR, Healey LA. Remodeling the pyramid: a concept whose time has come. Rheumatol. 1989;16:565-567. [47] Landd K. Die gunstige Beeinfulssung schleichender Dauerinfekte durch Solganal. Munchen Med Wochenschr. 1927;74:1132-1134. [48] Forestier J. Rheumatoid arthritis and its treatment by gold salts: the results of six years' experience. J Lab Clin Med. 1935;20: 827-840. [49] Ward JR, Dahl SL. The use of gold containing drugs. In: Willkens RF, Dahl SL, eds. Therapeutic Controversies in the Rheumatic Diseases. Orlando, Fla: Grune Stratton inc; 1987:133-158. [50] Furst DE. Rational use of disease-modifying antirheumatic drugs. Drugs. 1990; 39:19-37. [51] Brooks PM. Slow-acting antirheumatic drugs and cytotoxic agents. In: Schumacher HR, Klippel JH, Koopman WJ, eds. Primer on the Rheumatic Diseases. Atlanta, Ga: Arthritis Foundation; 1993:303-306. [52] Felson D, Anderson J, Meenan R. The comparative efficacy and toxicity of secondline drug in rheumatoid arthritis. Arthritis Rheum. 1991;33:1449-1461. [53] Champion G, Graham G, Zeigler J. The gold complexes. In: Brooks P, ed. Slow-acting antirheumatic drugs and immunosuppressives. Bailieres Clin Rheumatol 1990;4:491-534. [54] Ward JR, Williams HJ, Egger MJ, et al. Comparison of auranofin, gold sodium thiomalate, and placebo in the treatment of rheumatoid arthritis: a controlled clinical trial. Arthritis Rheum. 1983;26:1303-1315. [55] Handel ML, defazio A, Watts CK, et al. Inhibition of DNA binding and transcriptional activity of a nuclear receptor transcription factor by aurothiomalate and other metal ions. Mol Pharmacol. 1991;40:613-618. [56] Heuer M, Pietrusko R, Morris, Scheffler B. An analysis of worldwide safety experience with auranofin. Rheumatol. 1985;12:695-699, [57] Sambrook P, Browne C, Champion G, et al. Terminations of treatment with gold sodium thiomalate in rheumatoid arthritis, J Rheumatol, 1982;P:93-934. [58] Krogstad D, Schlesinger M. Acid-vesicle function, intracellular pathogens, and the action of chloroquine against Pasmodium falciparum. N Engl J Med. 1987;317:542-559. [59] Saimeron G, Lipsky P. Immunosuppressive potential of antimalarials. Am J Med. 1983; 75(suppl IA):19-24. [60] O Day R. Pharmacological approaches: SAARDs-I. In: Klippel JH, Dieppe RA, eds. Rheumatology. London, England: Mosby-year Book Europe Ltd; 1994:8.12.1-8.12.10. [61] Segal R, Yaron M, Tartakovsky B. Methotrexate: mechanism of action in rheumatoid arthritis. Semin Arthritis Rheum. 1990;20: 190-199. [62] Songsiridey N, Furst DE. Methotrexate: the rapidly acting drug. Clin Rheumatol. 1990;4: 575-594. [63] Tugwell P, Bennett K, Ghent M. Methotrexate in rheumatoid arthritis. Ann Intern Med. 1987;107:358-366. [64] Wolfe F, Hawley DJ, Lathem MA. Temination of slow-acting antirheumatic therapy in rheumatoid arthritis: a 14-year prospective evaluation of 1,017 consecutive starts. Rheumatol. 1990;17:994-1002. [65] McKendry RJ. Purine analogs. In: Dixon J, Furst DE, eds. Second-line Agents in the Treatment of Rheumatic Diseases. New York, NY: Marcel Dekker Inc; 1991:223-243. [66] Trotter JL, Rodey GE, Gebel HM. Azathioprine decreases suppressor T cells in patients with multiple sclerosis. N Engl J Med. 1982; 3o6:365-366. [67] Cannon GW, Ward JR. Cytoxic drugs and sulfasalazine. In: McCarty DJ, ed. Arthritis and Allied Conditions, IIE. Philadelphia, Pa: Lea and Febiger; 1989:563-591. [68] Bacon PA, Salmon M. Modes of action of second line agents. Scand J Rheumatol. 1987; 64(suppl): 17-24. [69] Van Der Heijde D, Van Riel P, Nuver-Zwart I Gribnau F, Van De Putte L. Effects of hydroxychloroquine and sulfasalazine on progression of joint damage in rheumatoid arthritis. Lancet. 1989;1:1036-1038. [70] Symmons D, Salmon M, Farr M, et al. Sulfasalazine treatment and lymphocyte function in patients with rheumatoid arthritis. J Rheumatol. 1988;15:575-579. [71] Klinenburg JR, Reichman R, Clements PJ. Investigational theraphy for rheumatoid arthritis. In: Cohen AS, ed. Progress in Clinical Immunology, Volume 1. Orlando, Fla: Grune Stratton Inc; 1984:111-144. [72] Cannon GW, Jackson CG, Samuelson CO Jr, et al. Chlorambucil therapy in rheumatoid arthritis: clinical experience in 28 patients with literature review. Semin Arthritis Rheum. 1985;15:106-118. [73] Clements PJ, Davis J. Cytotoxic drugs: their application to the rheumatic diseases. Semin Arthritis Rheum. 1986;4:231-234. [74] Clements PJ. Cytotoxic immunosuppressive drugs. In: Paulus HE, Furst DE, Dromgoole SH, eds. Drugs for Rheumatic Disease. New York, NY: Marcel Dekker Inc; 1987:135-156. [75] Clements PJ. Alkylating agents. In: Dixon J, Furst DE, eds. Second-line Agents in tbe Treatment of Rheumatic Diseases. New York, NY: Marcel Dekker Inc; 1991. [76] Tsoko GC. Immunomodulatory treatments in patients with rheumatic diseases: mechanism of action. Semin Arthritis Rheum. 1987; 17:24-38. [77] Tugwell P, Bombardier C, Geut M, et al. Low dose cyclosporin versus placebo in patients with rheumatoid arthritis. Lancet. 1990; 335:1051-1055. [78] Yocum DE, Klippel JH, Wilder RL, et al. Cyclosporin A in severe, treatment refractory rheumatoid arthritis. Ann Intern Med. 1988; 109:863-839. [79] Edmonds JP, Scott DL, Furst DE, et al. Antirheumatic drugs: a proposed new classification. Athritis Rheum. 1993;36:336-339. [80] Bensen WG, Bensen W, Adachi JD, et al. Remodeling the pyramid: the therapeutic target of rheumatoid arthritis. J Rheumatol. 1990; 17:987-989. [81] Yelin E, Meenan R, Nevitt M, Epstein W. Work disability in rheumatoid arthritis: effects of disease, social, and work factors. Ann Intern med. 1980;93:551-556.

C Moncur, PhD, PT, is Professor and Co-Director, Division of Physical Therapy, College of Health, Adjunct Professor of Internal Medicine, Division of Rheumatology, Department of Internal Medicine, School of Medicine, and Adjunct Professor of Bioengineering, College of Mines and Engineering, University of Utah, Salt Lake City, UT 84112 (USA). Address all correspondence to Dr Moncur.

HJ Williams, MD, is Professor and Chief, Division of Rheumatology, Department of Internal Medicine, and Associate Chair, Department of Internal Medicine, School of Medicine, University of Utah.