Rheumatoid Arthritis in the Elderly: A Distinct Clinical Entity?
This article is no longer available for CME credit
Release Date: January 15, 2008
Expiration Date: January 15, 2009
Internists, family practitioners, geriatricians, cardiologists, and others who care for older patients.
MEDIUM USED & METHOD OF PARTICIPATION
Read the article, complete the evaluation and post-test, and return both to: NACCME via fax at (610) 560-0502. You will receive your certificate in 6-8 weeks. If you would like to print your certificate immediately, go to http://cme.naccme.com, register as a NACCME user and access the activity test and evaluation online.
Successful completion entails participants obtaining a score of at least 70% on the post-test. A certificate of completion will be mailed to the address listed on your post-test/evaluation form within 6-8 weeks of receipt of the documents.
Take the Test On-Line Priority Code (RHE472)
This activity is sponsored by the North American Center for Continuing Medical Education (NACCME). NACCME is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. NACCME designates this educational activity for a maximum of 1 AMA
PRA Category 1 Credit™. Physicians should only claim credit commensurate with the extent of their participation in the activity. This activity has been planned and produced in accordance with the ACCME Essential Areas and Policies.
This activity is sponsored by The North American Center for Continuing Medical Education.
All those with control over the content of continuing education programs sponsored by NACCME are expected to disclose whether they do or do not have any real or apparent conflict(s) of interest or other relationships related to the content of their presentation(s). It is not assumed that these relationships will have an adverse impact on presentations; they are simply noted here to fully inform participants.
Dr. Lozada is on the Speaker’s Bureau of Amgen, Bristol-Myers-Squibb, Abbott, Wyeth, and Novartis.
Dr. Hopkins reports no relevant financial relationships.
No faculty member has indicated that the presentation will include information on off-label products.
All NACCME employees involved in the planning and editing of this educational activity have disclosed that they have no significant financial relationship with any organization that could be perceived as a real or apparent conflict of interest in the context of this educational activity.
CONFLICT OF INTEREST RESOLUTION-CONTENT VALIDATION
In compliance with ACCME Standards for Commercial Support and NACCME’s policy and procedure for resolving conflicts of interest, this continuing medical education activity was reviewed by a member of the Advisory Board in December, 2007 for clinical content validity, to ensure that the activity’s materials are fair, balanced, and free of bias toward the commercial supporters of the activity, and that activity materials represent a standard of practice within the profession in the U.S. and that any studies cited in the materials upon which recommendations are made are scientifically objective and conform to research principles generally accepted by the scientific community.
Upon completion of this educational activity, participants should be able to:
1. Identify the unique features of the clinical presentation of elderly-onset rheumatoid arthritis.
2. List the factors that predict bad prognosis in rheumatoid arthritis.
3. Discuss the importance of early treatment in elderly-onset rheumatoid arthritis.
4. List the available rheumatoid arthritis treatments for elderly-onset of rheumatoid arthritis.
Rheumatoid arthritis (RA) is the most common inflammatory, erosive arthritis. It has been estimated to affect 1% of the adult population in the United States.1 Although most commonly diagnosed during the third to fifth decades of life, the incidence and prevalence of RA continue to increase up to the ninth decade of life. Specifically, the prevalence of RA in persons age 60 years and older has been reported to be approximately 2%.2 These older adults with RA comprise two groups: individuals with long-standing disease diagnosed at a young age and individuals with disease arising late in life. This review will focus on the latter group.
Since Schnell3 first described “rheumatic infection in the old” in 1941, many studies have delineated RA in the elderly patient. The majority of studies define this older-onset RA as diagnosis after age 60,4 though some authors more stringently define it as after age 65.5 A few have used age 55 or even 50.6 Similarly, this syndrome has been variably described as late-onset RA (LORA), elderly-onset RA (EORA), or geriatric-onset RA.
Whether the inflammatory arthritides, including RA, have a distinct presentation and course in older as compared to younger subjects as first postulated in 1941 remains unsettled. The issue has gained significant importance as the proportion of the population who are over the age of 60 continues to grow in developed nations. As multiple newer treatment strategies emerge for RA, the efficacy and safety of these in elderly populations need to be evaluated as a distinct clinical issue.
There have been various—sometimes conflicting—reports regarding the differences between EORA and younger-onset RA (YORA). These have focused on differences with regard to female-to-male ratio, mode of onset, frequency of associated systemic features, and the prevalence of subcutaneous nodules. Published studies include older, descriptive studies and later, largely cross-sectional studies with a YORA control group. It appears likely that several important distinctions between EORA and YORA exist.
Whereas a clear female predominance of 3:1 exists in YORA, numerous studies of patients diagnosed after age 60 demonstrate a near equal sex ratio ranging from 1.1:1 to 1.9:1.4 Many studies have demonstrated the tendency for YORA to be insidious versus a more abrupt onset of EORA.3 Corrigan et al7 determined that most cases with RA onset over the age of 60 present similarly as compared to younger patients; however, they highlighted a subgroup of rheumatoid factor (RF)–positive patients who presented with distinctive features, such as an abrupt onset. The distribution of joint involvement has drawn interest from various investigators. Although small- and medium-sized joints are the most frequently involved joints in the overall RA population,8 numerous investigators have noted relatively prominent large-joint involvement, in particular shoulder synovitis, in EORA.9 Most studies have found no significant differences in the incidence of rheumatoid nodules between EORA and YORA patients.10,11
Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are nonspecific acute-phase reactants that can rise in the face of an inflammatory condition, infection, trauma, or malignancy. In RA, the ESR and CRP correlate with disease activity and radiographic damage.12,13 In numerous studies, the ESR and/or CRP of EORA patients were noted to be higher than those observed in younger patients.14 Consequently, greater elevation of these acute-phase reactants has been touted as a distinguishing feature from YORA. However, the normal range for ESR and CRP increases with age.15-17 Furthermore, it has recently been reported that in patients with EORA and YORA with similar disease duration and severity, the discrepancies in elevation of both the baseline ESR and CRP disappear after age adjustment.18,19 Nonetheless, the cohort tested only included RF-positive patients; therefore, these results may not necessarily extend to seronegative patients.
A lower frequency of positive serum test results for RF has been reported in EORA.20 Narayanan et al8 observed that seronegativity was significantly higher, specifically among elderly female patients. This observation seems counterintuitive since the prevalence of RF increases with age.
There is increased citrullination of proteins in the synovium of patients with RA through increased activity of the enzyme peptidylarginine deiminase (PAD). Antibodies to these citrullinated proteins, called anti-cyclic citrullinated peptide (anti-CCP) antibodies, have been detected in the sera of more than 70% of patients with RA and have become an increasingly useful tool in the clinician’s armamentarium to diagnose RA, particularly early in the disease course. This is especially true in the geriatric population, because not only are anti-CCP antibodies more sensitive than RF in the early stages of RA, but they are also potentially more specific than RF in the elderly in whom, as previously mentioned, the rate of RF false positives increases with age. Furthermore, anti-CCP antibodies are particularly valuable to facilitate the diagnosis of patients who present with an atypical clinical picture, such as a polymyalgic syndrome, and/or are seronegative for RF. Unfortunately, no controlled study to date has looked at anti-CCP antibodies in an EORA population. Like RF, these antibodies are not useful in following disease activity over time.
In addition to serologic studies, plain films may play a role in differentiating EORA from YORA. Reports from early RA cohorts have found higher baseline radiographic scores, as assessed by either the Larsen scale, the total Sharp score (TSS), or the joint space narrowing (JSN) component of the Sharp score, in EORA patients as compared with YORA patients.21 This difference was mainly due to a higher baseline JSN score and not attributable to differences in erosion scores.21 Accordingly, Pease et al10 and Ferraccioli et al22 did not find a difference in baseline joint erosions when comparing the radiographs of EORA and YORA patients. Khanna and colleagues19 have demonstrated that the increased JSN seen at baseline in EORA is most likely a result of age-related changes due to osteoarthritis (OA), hence the statistically significant correlation between the OA-JSN score, presence of osteophytes, and RA-JSN score. Therefore, initial differences in radiographic presentation are likely related to age-associated changes secondary to OA, rather than inherent differences between EORA and YORA.19
In accordance with the known prognostic advantage of RF seronegativity in RA, seronegative EORA patients also have a better disease course when compared to late-onset RF-positive patients. Seronegative EORA patients have been reported to have significantly lower joint counts than their seropositive counterparts.23 Seropositive patients are also more likely to develop joint erosions, are less likely to go into clinical remission, and have poorer functional outcomes.10 Furthermore, the relative risk of death is fivefold higher in RF-positive than RF-negative patients.24
Another factor that may contribute to prognosis of this geriatric population is comorbidity. This issue takes on much importance in an older population, as the prevalence of comorbidities such as hypertension, cardiovascular disease, diabetes, neurologic disorders, and even other rheumatologic disorders such as OA and osteoporosis, increase. These conditions can alter function, change the structure of bones, joints, and muscles, as well as require numerous pharmacologic interventions. In addition, other factors such as an increased propensity to trauma secondary to neurologic conditions, polypharmacy, and atrophied soft tissues and fat pads contribute to increased risk.25
Mikuls et al’s26 study of female EORA patients from the Iowa Women’s Health Study determined that they had significantly lower physical health–related quality-of-life scores as compared to controls, even when controlling for depression, recent fracture, and other comorbidities including cancer, hypertension, heart disease, lung disease, neurologic disease, diabetes, and kidney disease.
In the elderly, the differential diagnosis of RA is quite extensive because conditions that can have similar presentations, such as polymyalgia rheumatica (PMR), OA, gout, pseudogout, and occult malignancy, are prevalent in this population. It is critical to distinguish RA from those with PMR, OA, and other similarly presenting diagnoses.1
Polymyalgia rheumatica, in particular, often has overlapping symptoms and manifestations with EORA, making clinical differentiation between the two diseases difficult at times, particularly early in the disease course. Specifically, RA-like PMR onset, seronegative EORA, and PMR with peripheral synovitis constitute the subgroups presenting the greatest difficulties.27 The group of elderly RA patients that poses the greatest clinical challenge in terms of differentiation from PMR is seronegative EORA.
Clinically, the arthritis of EORA and that presenting in PMR with peripheral synovitis can be similar. Notably, anti-CCP antibody testing is becoming an increasingly useful diagnostic tool for identifying EORA patients with a polymyalgic onset of disease. The significance of this serologic marker is attested to by Lopez-Hoyos et al28 and Ceccato et al,27 who both determined that the presence of anti-CCP antibodies in a patient presenting with clinical symptoms of PMR must be interpreted as highly suggestive of EORA. Thus, anti-CCP antibodies have become a very useful tool in narrowing the differential diagnosis.
Imaging with ultrasonography or magnetic resonance imaging (MRI) may be helpful in differentiating PMR from EORA. McGonagle and colleagues29 found that, although patients with RA and PMR had a comparable degree of joint effusion, bursitis, and tenosynovitis, only patients with PMR had inflammatory changes outside the joint cavity when evaluating their shoulders with fat suppression MRI.
It has been estimated that in 10% of patients presenting with polymyalgic symptoms, the correct diagnosis of EORA, PMR, and temporal arteritis is delayed due to similarities in initial clinical presentation.30 Therefore, long-term follow-up is essential to establish the correct diagnosis.
Aging is accompanied by changes in pharmacokinetics, hepatic and renal function, and tissue responsiveness. Age-related changes such as decreased saliva production, decreased gastric acid, increased gastric emptying time, and decreased gastrointestinal (GI) motility may affect the rate of absorption of pharmacologic agents. Other changes such as decreased cardiac output, decreased glomerular filtration rate, decreased hepatic blood flow, decreased total body water, and decreased serum albumin levels can affect the distribution of drugs. These variations can potentially lead to transient toxic or subtherapeutic drug levels. Declines in hepatic blood flow, decreased liver size, and decreased production of proteins, lipids, and glucose may also lead to a drop in phase I drug metabolism. Additionally, decreases in renal plasma flow, renal tubular clearance, and creatinine clearance can influence drug elimination.31 Other complicating comorbidities and additional issues such as decline in cognitive function, polypharmacy, and compliance may also affect therapy.25 Thus, the efficacy and toxicity of commonly used RA medications may differ in younger and elderly patient populations.32 Regardless, the goals of therapy in RA in any age group are to limit pain, slow or stop disease progression, and improve functional status in a safe and effective manner.
Nonsteroidal Anti-inflammatory Drugs
Nonsteroidal anti-inflammatory drugs (NSAIDs) are effective in alleviating the pain and stiffness of RA, but use in the elderly is restricted by adverse effects on the central nervous system (CNS), cardiovascular, renal, and GI systems. The risk of adverse events is heightened in the older population secondary to comorbidity, polypharmacy, and changes in drug pharmacodynamics and clearance. For example, NSAIDs can adversely affect the CNS, triggering psychotic reactions, cognitive dysfunction, and depression. Nonsteroidal anti-inflammatory drugs can cause fluid and sodium retention, which can lead to peripheral edema, hypertension, and worsen congestive heart failure. The inhibition of renal prostaglandin synthesis by NSAIDs decreases renal blood flow and glomerular filtration and can thus precipitate acute renal failure, especially in patients with glomerular disease, pre-existing chronic renal failure, hypercalcemia, or heart failure and decompensated cirrhosis.33 Additionally, patients who have a history of GI bleeding or peptic ulcer disease and/or concomitant use of corticosteroids or anticoagulants are at increased risk of upper GI bleeding.33 Cyclooxygenase-2 (COX-2)-specific inhibitors were introduced to the market as a safer GI alternative.
Recent evaluation of available data by the U.S. Food and Drug Administration (FDA) has prompted warnings in the labeling of all NSAIDs/COX-2 inhibitors for possibly increased risk of thrombotic and cardiovascular events.33 Ultimately, when using NSAIDs/COX-2 inhibitors in an elderly patient, the clinician must continuously weigh the balance between efficacy and tolerability.34
Although efficacious in the short term, with long-term use, corticosteroids have many harmful side effects that are particularly troubling for an elderly population. The elderly are particularly vulnerable to adverse effects such as diabetes mellitus, cardiovascular effects, cataracts, glaucoma, mood disturbances, and osteoporosis with long-term use. On the other hand, a short course of low-dose steroids can relieve the signs and symptoms of RA while improving function, with low risk of side effects. Regardless, any patient taking prednisone 5 mg or more per day for more than 3 months should be started on a bisphosphonate and appropriate calcium supplementation (if not already in place) to prevent accelerated bone loss, and have bone mineral density measured at regular intervals.35
Disease-Modifying Antirheumatic Drugs
The use of traditional disease-modifying antirheumatic drugs (DMARDs) such as methotrexate (MTX), sulfasalazine (SSZ), and leflunomide alters the course of RA—improving pain, joint damage, and disability. At times, the benefits of these medications have not been fully extended to the elderly population due to concerns of potential toxicity.
Multiple studies have determined that age does not affect MTX efficacy or the rate of side effects.36 However, since renal clearance accounts for 50-80% of MTX elimination,31 renal impairment increases the risk of side effects. Close follow-up of hepatic and renal function can eliminate most risk. However, because elderly patients have an age-related decrease in renal function,31 the dosing regimen may need to be adjusted, especially in patients with renal insufficiency.
Both the large size of SSZ pills and GI side effects such as nausea and vomiting are significant drawbacks for the elderly.37 Pharmacologic interactions can occur, such as a decrease in serum digoxin levels by 25%.
Leflunomide has been shown to have similar efficacy to MTX in RA. With regard to the elderly, however, specific pharmacodynamic, pharmacokinetic, and toxicity studies are absent from the literature.31 As with MTX, there should be frequent monitoring of hepatic function.
Hydroxychloroquine (HCQ) has a relatively modest symptom-relieving effect and slow onset of action compared with many other DMARDs. It has not been shown to retard radiographic progression in RA. This medication is generally well tolerated, and the most serious side effect of retinal toxicity, which can lead to blindness, is rare.33 Notably, HCQ can interact with various medications. This is particularly troublesome in the geriatric population, where patients are often taking more than one medication. Specifically, HCQ can increase free digoxin levels by displacing digoxin from binding sites and by decreasing renal clearance. Also, HCQ can increase the bioavailability of metoprolol. Furthermore, cimetidine doubles the elimination half-life of HCQ.31
Tumor Necrosis Factor-Alpha Antagonists
Currently, three tumor necrosis factor (TNF)-alpha antagonists are FDA-approved and available for the treatment of RA: (1) etanercept, a soluble TNF receptor construct; (2) infliximab, a chimeric anti-TNF-alpha monoclonal antibody; and (3) adalimumab, a human anti-TNF-alpha monoclonal antibody.32 These agents have been shown to reduce disease activity, slow down or arrest structural damage, and improve disability scores. Adverse events reported have included injection site reactions, infusion reactions, development of antibodies to the molecules themselves, development of antinuclear and anti-dsDNA antibodies, demyelinating disease, infections, and worsening of heart failure. In particular, anti-TNF therapies should not be given to patients with active infection and should be used with caution in patients with recurrent infections or those predisposed to infection.33
The available data on safety and efficacy of anti-TNF therapies in elderly RA have been obtained mostly through subgroup analysis and not in randomized clinical trials designed to evaluate this population. Given the post hoc nature of such analyses, a descriptive approach replaces inferential hypothesis testing. Moreover, only a small percentage of patients in a typical clinical trial are elderly, even though the elderly consume more than one-third of prescribed medications in the United States.38 As such, the elderly population is not generally well represented in clinical trials of new agents.
Most of the available literature is on etanercept, the first TNF-alpha antagonist approved for RA. Fleischmann and colleagues39 were the first to report on etanercept treatment in patients age 65 years or older. In a retrospective analysis of patients age 65 and older enrolled in four double-blind, randomized and five open-label trials with a total of 1128 patients (17% age 65 yr or older), it was reported that clinical response with etanercept did not differ between age groups. Specifically, at 1 year, 69% of younger patients and 66% of patients age 65 years or older met the American College of Rheumatology 20 response criteria. Although injection site reactions, headache, and rash occurred somewhat more frequently in younger patients, the overall rates and types of other adverse events were comparable in both groups.39
A variety of safety and efficacy measures have been evaluated through further post hoc analysis of the etanercept clinical trial database. In terms of safety, the rates of serious adverse events, the rates of serious infectious episodes, and the number of cases of cancer have been evaluated. The rates of the aforementioned events and cancer in the elderly etanercept-treated subjects were similar when compared with the control arms of elderly patients treated with either MTX or placebo.40 However, the overall rates of adverse events, infectious episodes, and cancer were higher in the elderly population as compared to the younger subjects. This is attributable to the increased risk of these related to age since the rates were similar to those for this age group within the general population.
In addition, the changes in HAQ disability scores in response to etanercept treatment were similar in the two age groups, even though mean HAQ disability scores at baseline and at study endpoints were higher in the elderly patients. This functional benefit was increased with combination therapy (etanercept and MTX), a difference not observed in younger subjects. Finally, the combination therapy arm of the trials produced the greater reduction in rate of progression of joint damage in both age groups. In an analysis of safety data from 18 RA trials, subjects over age 65 years were at no greater risk of adverse events than younger subjects.41
There are many limitations to these studies. In addition to the previously mentioned weakness of subset analysis, the majority of patients included in these studies were seropositive; therefore, the findings may not be generalizable to older seronegative RA patients. Additionally, safety events may be underestimated due to the inclusion of healthier older patients (ie, less comorbidity in clinical trials than the general RA patient population). The same can be said of efficacy if the clinical trials selected patients with less comorbidity. Finally, efficacy results from the extension studies may be suspect based on the fact that those patients who did not respond to therapy would have discontinued treatment prior to inclusion in the study.
There has been less publication on treatment of the elderly with the other anti-TNF therapies, adalimumab and infliximab. One study analyzed patients enrolled in a large longitudinal observational study of biological therapies in rheumatic diseases in order to identify specific demographic and clinical factors that correlate with the response to etanercept or infliximab at 6 months in patients with RA. Although this study did not aim specifically to evaluate the effect of age, it was found that age did not have an effect on treatment response with either infliximab or etanercept.42 Another observational cohort study found that there was a trend toward a greater percentage of elderly patients who stopped infliximab having done so due to severe infections. Unlike the previously discussed studies, this study defined “older patients” as greater than or equal to age 70 and “younger patients” as less than 70.43 A Japanese study that evaluated infliximab in an older population determined that efficacy was consistent across all age groups, and that the rates of serious infusion reactions were also similar across all age groups. However, tuberculosis and adverse drug reactions were higher among patients over 65 years of age.44
In one study, adalimumab use led to clinically significant improvement in disease activity and physical function, as measured by DAS and HAQ scores, in all age groups, including patients with LORA. In addition, the drug was well tolerated, with a generally positive benefit-to-risk ratio across all age groups.45
Although studies are needed that independently evaluate older subjects with comorbidities that mirror the general elderly population, RA subset analyses preliminarily indicate that elderly patients can benefit from TNF antagonists. Most recently, Schneeweiss et al46 found that in a large cohort of low-income elderly patients there was no increase in the rate of bacterial infections among patients on TNF antagonists as compared with patients using MTX.
Data delineating the effectiveness and adverse effects of DMARDs and biologics in an elderly RA population are limited. The paucity of trials specifically designed to answer questions pertaining to outcomes in the elderly and concerns of increased toxicity in this group has resulted in decreased use of combination DMARD therapy and TNF-alpha antagonists in the elderly population as compared to younger patients with identical disease duration and comparable disease severity and activity.47 Likewise, Calvo-Alen et al’s48 subset analysis of a larger RA study determined that not only was the mean number of DMARDs higher in younger patients than in the older ones, but that DMARD therapy was also less commonly used among the older patients. Although clinicians may simply be exercising caution in a population in which they sense a greater risk of adverse effects, patients should not be excluded from treatment options merely because of their age. Each therapeutic decision should be made on an individual case basis, using the available data to determine the safest and most efficacious therapeutic regimen to control pain and inflammation, halt disease progression, and prevent disability. Only with increased knowledge of the risks and benefits of therapy in the elderly population can a true risk-to-benefit profile for anti-TNF-alpha therapies and RA treatments in general be determined by the physician and the patient.
From the currently available literature, a tenuous profile of EORA has emerged—a more equal sex distribution, a higher frequency of abrupt-onset disease, and more large-joint complaints, specifically shoulder synovitis.
Differences such as disease duration, patient selection, the percentage of seropositive versus seronegative patients included in a study, accidental inclusion of non-RA patients, and even the inherent variation of design among cross-sectional, retrospective, and prospective studies must all be taken into account when approaching the literature from an evidence-based perspective. As diagnostic precision is improved with laboratory tests such as anti-CCP antibodies, or other as-yet undiscovered markers, some of the previously reported heterogeneity of EORA may be settled and the groups better characterized.
With improvements in healthcare and increasing life expectancy, recognizing and appropriately managing these patients will become ever more important. The available data on management of RA suggest that patients who present with a clinical picture suggestive of RA later in life must be evaluated and treated just as aggressively as younger patients with therapeutic regimens that include traditional DMARDs and the use of biologics in incomplete or nonresponders in order to improve morbidity and mortality and prevent disability.