Exercise and inflammatory disease


  • Ronenn Roubenoff

    Corresponding author
    1. Tufts University, Boston, Massachusetts
    • Nutrition, Exercise Physiology, and Sarcopenia Laboratory, Jean Mayer USDA Human Nutrition Research Center on Aging, Tufts University, 711 Washington Street, Boston, MA 02111
    Search for more papers by this author

  • The contents of this article do not necessarily reflect the views or policies of the US Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government.


Chronic inflammation affects the body's composition and metabolism in several ways, which ultimately favors the loss of body protein and the accretion of fat. These changes in turn lead to reduced function, loss of the ability to mount an effective response to acute illness, and increased risk of fat-related chronic diseases such as obesity, heart disease, and diabetes. Ultimately, these changes may be responsible for much of the increased mortality reported with diseases such as rheumatoid arthritis (RA) (1), which are not intrinsically fatal. Current treatment, although effective in suppressing inflammation and controlling symptoms, does not appear to correct the metabolic and body composition changes that occur with inflammatory arthritis. It appears that a rehabilitative approach is required to reverse the loss of protein and the increase in fat, and that structured exercise programs offer the only proven effective method of achieving this goal.

Statement of the Problem

Great progress has been made in understanding the pathophysiology of target organ damage in inflammatory diseases, such as the changes that lead to joint destruction in RA, skin damage in systemic sclerosis, or visceral organ damage in systemic lupus erythematosus. However, few clinicians understand the profound effects of chronic inflammation on metabolism, body composition, and function. These changes, while driven by the immune system, occur in parallel to the joint damage but often are ignored because they are clinically silent, without pain or other cardinal symptoms. As a result, treatment, although properly focused on the target organs, fails to address the “collateral damage” done to muscle, which leads to weakness, dysfunction, and loss of independence. There is a growing body of literature that shows that exercise can reverse many of the effects of chronic inflammation on muscle, and often have direct benefits on joint function as well. A major challenge for the future is translating this new-found understanding into routine clinical practice.

Review of the Evidence

Effects of chronic inflammation on body composition.

Normal aging causes loss of muscle protein stores and increases in body fat, even in master athletes who continue to train regularly (2). Most people experience an increase in body weight until their 60s, followed by a slow decline in weight (3). Both the loss of muscle and the gain in fat have important effects on health and function. The loss of muscle is termed sarcopenia, from the Greek “poverty of flesh,” and loss of >40% of baseline muscle mass is fatal (4, 5). However, lesser degrees of muscle mass reduction are associated with loss of strength. When strength falls below a threshold required for function, disability ensues. Thus, although sarcopenia appears to be a relatively smooth, continuous decline, its functional consequences are more typically a step function because the ability to perform certain tasks, such as rising from a chair, are generally dichotomous—either one can do them or one cannot. It is important to recognize the role of sarcopenia because the effects of inflammatory arthritis are superimposed on it (6). The increasing difficulty with function and independence that arthritis causes with age is no doubt in part due to the additive effects of sarcopenia and inflammation.

It is important to understand that sarcopenia is not a disease, but a normal process of aging. Nevertheless, it can be mitigated with exercise, diet, and perhaps with hormone replacement therapy (3). Diseases, on the other hand, can accelerate the effects of sarcopenia through 2 distinct processes, which we have previously called wasting and cachexia (7). Wasting is the loss of weight—both lean and fat—that occurs when there is inadequate dietary intake of energy and protein. This is the typical course in starvation, as well as in advanced cancer, acquired immunodeficiency syndrome (AIDS), tuberculosis, and other illnesses that cause anorexia. Wasting is often very aggressive, and can be fatal in weeks if not treated with nutritional support.

In contrast, cachexia is a term used to define the loss of lean body mass—muscle and viscera, but primarily muscle —that occurs despite relatively little loss of body weight. Often in cachectic conditions, while the muscle compartment is being consumed, other compartments of the body are increasing, such as fat mass in RA or extracellular water in human immunodeficiency virus infection, congestive heart failure, or renal failure. Cachexia can occur in the presence of adequate dietary intake, and appears to be driven primarily by inflammation-mediated changes in metabolism. These changes are, at least proximately, controlled by inflammatory cytokines, such as interleukin (IL)-1β, tumor necrosis factor α (TNF-α), and IL-6 (Figure 1) (8–10).

Figure 1.

Pathophysiology of rheumatoid cachexia. TNF = tumor necrosis factor; IL = interleukin.

Aside from severe complications, such as rheumatoid vasculitis or multi-organ system failure in systemic lupus, most patients with inflammatory disease suffer from cachexia. Cachexia causes loss of muscle mass over months to decades. In our experience, patients with RA, successfully treated, have body cell mass (composed of muscle plus viscera) that is approximately 15% lower than age-, sex-, and race-matched controls (Figure 2) (8). If we accept that 40% loss of cell mass is fatal, these patients have lost about one-third of what they can afford to lose. Helliwell and Jackson (11) showed that muscle mass accounted for about one-third of the variability in grip strength, independent of joint pain and deformity, which explained only an additional 6% of the variability. Stucki et al (12) found that a muscle strength index based on testing of extension and flexion strength of the knees and elbows correlated strongly with physical function and radiographic damage in patients with RA. This muscle strength index was superior to grip strength as a measure of disease activity.

Figure 2.

Body composition in patients with rheumatoid arthritis and age-, sex-, race-, and weight-matched controls. BMC = bone mineral content; ECW = extracellular water; BCM = body cell mass.

Unfortunately, simply instituting disease-modifying treatment did not reverse cachexia: one year after starting methotrexate, for example, patients who had elevated serum TNF-α levels at baseline were found to have no detectable serum TNF-α, but no improvement in their body composition either (although it did not continue to deteriorate) (13).

As a consequence of both muscle loss and weakness, and of the joint pain that inflammatory diseases often cause, patients with inflammatory arthritis show a marked reduction in physical activity and aerobic capacity (14–16). This inactivity further reduces the anabolic input to muscle, and reinforces a cycle of inactivity leading to reduced fitness, leading to more inactivity. Thus, there are important behavioral, as well as immunologic and metabolic, causes of weakness and disability in patients with inflammatory diseases. These changes serve to reinforce the loss of function caused directly by joint damage and pain. However, it is unlikely that even joint replacement can reverse the muscle effects of inflammatory disease without specific anabolic interventions.

Thus, it appears that a direct anabolic signal to muscle must be given to reverse rheumatoid cachexia. What could this signal be? Based on the experience with AIDS wasting, the available pharmacologic agents are recombinant growth hormone and several synthetic androgens. However, growth hormone causes side effects, such as carpal tunnel syndrome and joint swelling, which may be particularly difficult for patients with RA to tolerate. Androgens are more promising, because low testosterone levels have been reported in patients with RA (17, 18). However, these drugs are more difficult to dose in women because of their masculinizing effects, and because they may be proinflammatory. In addition, it is not known if newer anticytokine treatments for inflammatory arthritis will have a more beneficial effect on cachexia than older drugs, such as methotrexate. To date, there have been no clinical trials of pharmacologic intervention in rheumatoid cachexia. In contrast, there is now a fairly large body of literature suggesting that exercise can be used safely to provide the needed anabolic signal to muscle.

Exercise for inflammatory arthritis.

Despite early evidence that strength training could improve muscle function in RA (19, 20), most physicians recommended rest and reduced physical activity for patients with arthritis until the mid-1970s, in keeping with similar recommendations for patients with myocardial infarction and other severe illnesses. This “era of physical inactivity,” as it has been called (21), was challenged in rheumatic diseases by researchers in Scandinavia, who showed that patients treated with aerobic exercise and modest strength training had improvements in strength, function, and joint symptoms (22–27). Moreover, Ekblom et al (22) showed that patients who continued to exercise regularly (4 or more times per week) maintained these gains to a greater extent than those who trained 2 or fewer times per week; Nordemar et al (27) even suggested a reduced rate of radiographic progression of RA in patients who exercised.

A limitation of these studies is that the training programs were only sketchily described and were carried out in inpatient settings, in keeping with the prevailing treatment methods of those times. In the 1980s, Harkcom et al (28) and Minor et al (15) demonstrated the benefit of relatively low-intensity aerobic exercise in RA. Improvements were noted in aerobic capacity and functional status, and some patients also reported improvements in some aspects of joint symptoms, such as swelling and stiffness. More importantly, perhaps, the improvements in aerobic, functional, and performance changes occurred without exacerbation of joint symptoms, in contrast to the dire warnings of the physical inactivity era (28, 29).

The 1990s saw continued progress in the area of exercise and inflammatory disease, and growing recognition that almost any type of exercise is superior to physical inactivity for patients with arthritis. Such different approaches as low-intensity isokinetic training (30), low-intensity physical therapy (31), intermediate-intensity circuit training (in which participants move quickly from one machine to another in an effort to increase both strength and aerobic conditioning) (32), and high-intensity strength training (16, 33) have been shown to improve muscle function and joint function in patients with RA. High-intensity strength training in recent-onset arthritis (33) was comparable to a similar intervention in well-established disease (16).


Almost any form of exercise, if done carefully, will help to reverse the catabolic effects of inflammatory disease on muscle, improve function, and decrease the risk of disability. A combined approach using both strength training and aerobic exercise is likely to be optimal, and can be accomplished by most persons in less than 1 hour per day. The major barrier now before us is to translate research results into clinical practice, especially in a society where fewer than 20% of persons in the general population meet current recommendations for exercise or physical activity (see http://www.cdc.gov/nccdphp/sgr/pdf/sgraag.pdf).

The existing data clearly demonstrate the efficacy of exercise training, when carried out in laboratory settings. There is enough evidence to consider a position statement from the American College of Rheumatology in favor of strength training and aerobic exercise—distinct from symptom-directed physical therapy—as a routine part of care of all patients with inflammatory diseases. Such a statement could be modeled on recent position statements from the American College of Sports Medicine (34), while specifically tailoring its message to the population of patients with rheumatic diseases and the professionals who care for them.

More information is needed on the effectiveness of exercise, as performed in the real world, to gain acceptance for third-party payment for this treatment. In addition, more evidence regarding the optimal long-term training programs (1 year and longer) is needed to test life-long strategies of exercise and physical activity. Finally, we must develop grass-roots, low-tech, low-cost delivery systems that will allow patients with inflammatory arthritis to exercise without constraints of age, socioeconomic status, or personal exercise history. Exercise is an explosively powerful stimulus to the human body, and failure to harness it in the service of antirheumatic treatment would be a great disservice to our patients. Research has pointed the way in this field, but much hard work lies ahead to realize its promise. Nevertheless, it is likely that with ongoing efforts from the American College of Rheumatology and others, more and more patients will reap the benefits of this extraordinary intervention.