RA Exercise


Rheumatoid arthritis is a chronic, inflammatory autoimmune disease that is characterized most notably by inflamed, swollen joints. For years it was thought that exercise would only increase this inflammation and/or accelerate bone damage. However, studies are finding that this is not the case. Evidence shows that, though exercise may not reverse the disease process, it does attenuate disease activity and slow the rate of cartilage and bone destruction in the joints. Below is a synthesis of the effects of exercise on RA and its various cellular components. Overall, it appears that moderate, aerobic-type exercise and resistance training are safe and beneficial activities for patients with RA from a cellular perspective.

Immune Response and Exercise in RA


Leukocytes are key players in the immune system and its subsequent response to stress. According to Goebel et al. [1], after a 15-18 minute bout of cycling at an average 75% of VO2max in healthy adults, there was a significant increase in lymphocytes, including helper T-cells (CD4+) and cytotoxic T-cells (CD8+) in the plasma. They additionally saw a plasma increase in monocytes (which develop in to macrophages, which release TNF-α), granulocytes (neutrophils, eosinophils, basophils), and NK cells. This is the result of exercise, as a stressor, stimulating β-receptors. This stimulation of the β-receptors causes an increase in cyclic adenosine monophosphate (cAMP), and an increase in intracellular cAMP inhibits cytokine production. Thus the authors suggest that this may be partially responsible for the anti-inflammatory effects of a bout of moderate-intense exercise in addition to the significant increase of IL-6. Moldoveanu et al. [2] also studied the effects of cycling exercise, as well as incline walking, on plasma levels of cytokines and leukocytes in healthy (untrained) men. They also found that leukocyte counts increased. The increase took place both during and 2 hours after a 3 hour bout of exercise at 60-65% of VO2max, with peak increase occurring at the end of the exercise bout. The authors note that this elevation is merely temporary, with counts eventually returning to baseline. Scharhag et al. [3] found a similar response with 4 hours of cycling at 70% of anaerobic threshold in healthy, competitive men.

The effects of exercise on leukocyte levels in patients with RA have also been studied. In a review by Pool and Axford [4], the authors noted that after an 8 week cycling program, there was only a temporary increase in lymphocyte production in RA patients who exercised (when compared with RA patients who were sedentary). The authors also note that regular exercise has little effect on the resting state of the immune system in RA patients (which is elevated compared to healthy individuals), and that regular exercise training is unlikely to have any permanent therapeutic benefit or adverse effect on the disease process of RA. A study by Pool et al. [5], the authors measured CD4+ and CD8+ levels of patients with RA and healthy individuals after a cycle ergometer workout to tolerance. Resistance was increased by 10 watts every minute, and plasma levels were determined at baseline, peak exercise, and 1 hour after exercise. They found that there was an increase in CD4+ at peak exercise in both groups, but it only reached significance in the RA group. There was no significant increase in CD8+ at peak exercise in RA, but there was in the healthy group. The authors concluded that more studies need to be done to determine the effects of short, intense exercise bouts on patients with RA. One limitation with exercise studies in the RA population is that often one of the inclusion criterion is that their disease must be stable with their medications<link to pharmacological agents> during the intervention. Therefore, it is difficult to discern if the results are purely from the exercise, or if anti-inflammatory medications are influencing the cellular responses.

TNF-α, IL-6, IL-1β

Stress is the body’s natural response to stimuli that try to disrupt homeostasis. Therefore, exercise is considered to be a stressor if intensity or duration of the activity is enough to push the body out of its normal resting state. As a response to acute exercise in healthy individuals, the sympathetic nervous system is activated, to which both pro-inflammatory and anti-inflammatory cytokines react [1]. The major pro-inflammatory cytokines in the immune system are TNF-α, IL-6, and IL-1β [1]. Both TNF-α and IL-1β are produced locally in response to inflammation, and they stimulate IL-6 production [6]. Curiously, IL-6 can act as both a pro-inflammatory cytokine and an anti-inflammatory cytokine [6]. Peterson and Pederson [6] note that levels of TNF-α and IL-1β do not increase with exercise. On the contrary, IL-6 produced by skeletal muscle, which is independent of the TNF-α pathway, is the first responder cytokine into the bloodstream, and it acts in an anti-inflammatory manner. Not only does IL-6 stimulate the anti-inflammatory cytokines IL-1ra (inhibits IL-1), sTNF-R (inhibits TNF-α), and IL-10 (inhibits IL-1α, IL-1β, and TNF-α), but it also inhibits TNF-α production [6].

In individuals with RA, the anti-inflammatory effect of exercise also occurs. In a position statement by Walsh et al. [7], they note that it is unknown whether the benefits of exercise in autoimmune diseases stem from an alteration in immune cell counts and cytokines, or from a change in the level of activity of the transcription of the pro-inflammatory cytokines. This is a reflection of studies that suggest the transcription rate of cytokines (i.e. IL-6) is enhanced by exercise [6], and studies that suggest exercise increases plasma levels of IL-1β, IL-6, and TNF-α, but not their gene expression [2].

Exercise recommendations and Immune Response

After reviewing the literature, it appears that moderate to moderately-intense aerobic exercise is a safe in patients with RA in terms of immune response. Though leukocyte levels increase with exercise, it is only temporary. Though exercise may not reverse disease activity, the evidence shows that it may slow progression. Regardless, aerobic exercise between 60-75% of VO2max does not appear to increase disease activity. From the standpoint of cytokine response, it appears even more evident that exercise is beneficial for patients with RA. In an anti-inflammatory role, IL-6 released from skeletal muscle has the largest response to exercise. As this response also promotes an increase in other anti-inflammatory cytokines, and pro-inflammatory cytokine responses do not significantly increase, it appears exercise would be beneficial for patients with RA. As with any exercise program, exercise interventions need to be tailored to the individual. As RA is an autoimmune disease, with inflammation being one of the biggest clinical manifestations, it is important to follow up with the patient during and after exercise. If the patient is experiencing an increase in redness, swelling, soreness, stiffness, or pain in the affected joints, exercise should be modified or discontinued until inflammation is under control.

Bone Destruction and Exercise in RA

Exercise can induce positive changes regarding joint health in patients with RA including improved joint movement and flexibility [8]. These changes involve the tendons, ligaments, cartilage, and bone integrity. The influx of cytokines that characterizes RA causes the tendons around joints to become less stiff [8]. It is to the benefit of the joint that tendons are stiff in order to provide more efficient force production [8]. Through strengthening exercises, tendons can increase their stiffness, although these exact changes are not clear in the RA population [8]. Ligaments are also important for optimal joint health because they provide stability and biomechanical guidance of the joint [8]. Exercise has been shown to strengthen ligaments while immobilization weakens them [8]. Because joint health is compromised in RA patients, it is in their best interest to improve ligamental structure through exercise [8]. In a healthy joint, cartilage exists to protect joints from bone-on-bone contact [8]. Cartilage integrity is compromised in RA but exercise can help it to remain strong and resilient [8]. It was previously believed that intense exercise in RA could further joint cartilage damage, but it has since been proven that exercise does indeed benefit cartilage health in RA patients [8]. A high-intensity exercise program which was implemented in a study by de Jong et al. revealed that a 2-year long exercise program, including 1.25 hour sessions every other week, consisting of aerobic and resistance exercise, joint mobility, and sport activity, did not cause further joint damage when compared to RA patients receiving conventional care [8]. Cartilage damage was measured by cartilage oligomeric matrix protein (COMP) in this study [8]. It is common for studies to emphasize that exercise does not further disease activity in RA patients; few studies aim at looking at actual improvements in joint health in the RA population [8]. However, vigorous exercise has been shown to reduce the number of involved joints in the active disease process [8]. It is believed that an increase in the pressure within the joint complex during exercise may increase synovial blood flow, which may have a beneficial effect on the inflammation within the affected joint [8].

Since joint destruction in RA is associated with an inflammation-driven action of the RANKL-RANK pathway that increases osteoclastic activity, it may be said that reducing the activity of this pathway may result in less bone destruction [9]. In fact, exercise has been shown to decrease osteoclastic activity by a reduction of the activity of RANKL through a decrease in levels of pro-inflammatory cytokines [10]. In addition, exercise stimulates osteoblastic function and bone reformation due to an increase in circulatory IGF-1 levels [11]. As IGF-1 is important for muscular optimization, it is also important for bone health as a key player in the balance between protein synthesis and break-down [12], [13]. In those that are healthy, a constant low level of cytokines at rest, specifically TNF-α and IL-1, allow for normal levels of IGF-1 that promotes optimal bone health [11]. With a rise in these cytokines with acute exercise, there is a temporary inhibition on IGF-1 levels [11]. In RA, since cytokines are above normal levels even at rest, IGF-1 levels are inhibited, which does not allow for optimal bone health [11]. With exercise in RA patients, the decrease in cytokines that occurs with exercise allows for IGF-1 to increase [11]. It can be said that because of a decrease in the RANKL activity and an increase in IGF-1 levels, osteoblastic activity is promoted while osteoclastic activity is suppressed with physical activity [9], [10], [11].

Decreased bone mineral density in patients with RA is a combined result of inactivity, the disease process itself, and use of corticosteroids [14]. Exercise that incorporates weight-bearing induces the Wnt signaling pathway and promotes bone reformation [15]. This process is the theory behind promoting weight-bearing exercise in osteoporotic patients to reverse or slow the effects of decreased bone mineral density [15]. Although the exact mechanism of exercise promoting this pathway in the RA population is not clear, it can be said that weight-bearing exercise is likely to at least slow the demineralization effects that is common with the disease [15]. Because the process of bone reformation occurs after mechanical loading and a phase of bone resorption, it is important to find a balance with exercise intensity and type because it is necessary that bones go through the reformation phase in order to promote optimal bone density and overall bone health [16]. This is especially important in RA patients because the reformation phase is already inhibited and exercise that does not allow for appropriate reformation can further deprive bones from the opportunity to build through reformation. It has been shown that exercise can maintain or even enhance bone mineral density in RA patients [14]. A study by Hakkinen et al. was conducted to examine the effects of a strength training program on bone and joint health in patients with RA [14]. This study showed that bone mineral density and joint damage were basically unchanged after a 5 year follow-up [14]. This may demonstrate that although joint health and bone density are difficult to improve with exercise in patients with RA, it may at least benefit the patient in that the detrimental effects that RA has on bone and joints may be slowed or maintained with exercise [14]. This study involved patients who took DMARDs as a therapeutic agent for RA and discussed how this may have helped the results [14].

Exercise Recommendations and Bone Destruction

Resistance training or weight-bearing exercise can help to combat the effects of bone destruction in RA. With a decrease in TNF-α during exercise, there is a decrease in osteoclastic activity that leads to break-down of bone [10]. Exercise can also increase IGF-1 levels in the bone, which reflects osteoblastic activity [11]. In addition, bone reformation is promoted through the Wnt pathway which helps to mitigate the osteoporotic processes that occur in bone in RA patients [15].

Effect of continuous passive motion on cartilage damage and inflammatory markers

Ferretti et al examined the effect of immobilization (IMM) and continuous passive motion (CPM) on inflammation in the knees of rabbits with artificially induced arthritis [17]. In this study, rabbits were prepared so that their knee joints could be induced to display arthritis-like symptoms when injected with an antigen [17]. The rabbits were then divided into two groups [17]. One received right knee immobilization by wrapping with gauze. The other received continuous passive motion of the right knee from 40 to 110 degrees, with each cycle lasting 40 seconds [17]. Subjects received either 24 or 48 hours of their assigned treatment [17]. The researchers then examined the effect of the treatment on glycosaminoglycans (GAG), IL-1, matrix metalloproteinase-1 (MMP), cyclooxygenase-2 (COX-2), and IL-10 [17].

Knee menisci consist of two areas: zone A, a well vascularized outer area consisting of 25% of the total meniscal tissue, and zone B, a poorly vascularized, fibrocartilage inner area consisting of 75 percent of the tissue [17]. GAG is a polysaccharide that is an important component of articular cartilage [17], while MMP is an enzyme that degrades joint tissue, and is found in increased levels in inflamed joints [17]. After 24 and 48 hours, treatment of arthritis induced rabbit menisci with immobilization resulted in a significant reduction in GAG expression in both zone A and zone B when compared to healthy controls [17]. CPM treatment attenuated this response, showing a decreased GAG loss at both the 24 and 48 hour period when compared to the immobilization group [17]. Healthy meniscal tissue did not show expression of MMP in control knees [17]. Immobilization resulted in an increased expression of MMP after 24 and 48 hours of treatment in the arthritic model, while continuous passive motion produced a significant decline in the presence of the enzyme at both time intervals [17]. This data suggests that joint damage occurs relatively quickly following the immobilization of an inflamed joint, while a CPM intervention may help to alleviate the damage.

COX-2 is an enzyme linked to inflammation and pain in synovial joints [17]. In human subjects, patients treated with CPM as an intervention have been found to request less pain medication than individuals without treatment [17]. In the antigen-induced arthritic knee model, researchers found significantly less expression of COX-2 enzyme expression with CPM when compared with healthy controls [17]. In contrast, immobilized arthritic knees showed a significantly increased expression of COX-2 enzyme, at both the 24 and 48 hour interval [17]. The authors believe passive motion may reduce pain in arthritically inflamed joints through the decreased expression of the COX-2 enzyme [17].

In the arthritic-induced knee, the researchers found immobilization induced a significant increase in IL-1 at both 24 and 48 hours [17], while the continuous passive motion intervention resulted in a significantly lower number of IL-1 positive cells, and a decrease in the intensity of IL-1 activation [17]. Similarly, they found the levels of IL-10 to be increased following CPM treatment [17].

In later work, Ferretti and colleagues re-examined the effects of CPM in arthritis induced knees [18]. Non-arthritic controls with no intervention, immobilization, or CPM treatment resulted in cartilage that was visually undisturbed after 24 hours of treatment [18]. This trend continued up to 96 hours in knees with no intervention and CPM [18]. By contrast, healthy knees subjected to IMM showed drastic changes with visual inspection with 48 to 96 hours of treatment [18]. These changes included discontinuous surface texture and increased surface abrasions [18]. At 15 days of CPM treatment in healthy knees, minor surface abrasions and surface changes began to appear, while immobilized knees displayed visual cartilage erosion and vertical cracks that were similar to the arthritis-induced knees of free-roaming rabbits [18]. IMM in arthritic knees showed profound changes in surface appearance [18]. These changes included severe disorganization of chondrocytes along with profound levels of cartilage erosion, beginning after 24 hours of IMM [18]. CPM in arthritic knees were noted to be normal after 24 hours of treatment, with mild disorganization increasing through 15 days of treatment [18]. After 15 days, arthritic knees treated with CPM displayed mild degenerative changes, drastically reduced from that of similar knees subjected to IMM [18]. In agreement with previous work, they also found a significant reduction in GAG deterioration, MMP, COX-2, and neutrophil content, and IL-1, with an increase in IL-10 levels [18].

This research indicates that CPM may produce a decrease in joint degradation and pain resulting from a decrease in COX-2 and MMP activation from a decrease in IL-1 production, and an anti-inflammatory effect from increased IL-10 activity. As such, a long duration, passive motion intervention may produce joint protecting effects in patients with acutely inflamed joints. Immediate onset of the intervention following inflammation is likely indicated.


Rheumatoid cachexia is a condition in which there is an enhanced loss of skeletal muscle mass [8], [19]. It is typically present in about 2/3 of patients with RA and affects one’s quality of life regarding function and strength [8], [19]. Rheumatoid cachexia differs from cachexia associated with cancer and AIDS in that it is not correlated with death like it is with these other diseases [19]. The exact cause of rheumatoid cachexia is not currently known, but several possible mechanisms have been studied [19]. One main contributor involves the interruptive effects of TNF-α on the balance between protein formation and degradation in the muscle, tipping the scales toward degradation due to hypermetabolism [8], [19]. In fact, patients with RA have demonstrated an increase in energy expenditure at rest compared to humans without RA [19]. As previously described, TNF-α levels are increased in the presence of RA along with other pro-inflammatory cytokines and therefore have a greater effect on the balance between the formation and break-down of protein [8]. Decreased peripheral insulin activity, decreased levels of muscular insulin-like growth factor 1 (IGF-1) and testosterone, and use of therapeutic corticosteroids also contribute to the muscular loss associated with cachexia [8]. Decreased peripheral insulin activity occurs in RA and contributes to rheumatoid cachexia [19]. Although the exact mechanism in which this occurs is not known, it is thought that TNF-α may interfere with insulin receptor pathways [19]. Insulin, an anabolic hormone, normally inhibits muscular protein degradation [19]. Thus, due to lower levels of insulin activity, more muscular degradation can take place [19]. IGF-1 located in skeletal muscle normally regulates skeletal muscle maintenance [12]. Consistent with muscle loss, decreased levels of IGF-1 are noticed in cachexia [12]. Cachexia may also partially be the result of increased apoptosis in muscular tissue caused by an increase in cytokines [20].

Cachexia often goes unnoticed because the weight that is lost due to muscular wasting is counter-balanced by an increase in fat mass due to inactivity [8]. Patients with RA are often inactive and do not participate in exercise because of the joint pain associated with joint damage [8]. Unfortunately, as with anyone, a lack of physical activity results in poorer health in RA patients [8]. Low physical activity levels contribute to the muscular wasting that characterizes cachexia as well as an increased risk for disease, especially that of the cardiovascular system, and particularly, atherosclerosis [8]. With RA patients, there tends to be an ongoing cycle consisting of the pain associated with the disease causing low physical activity levels, which further contributes to cachexia [8]. It is not uncommon for patients with RA to lose up to 70% of their muscular strength and 15% of their lean muscle mass over the course of their disease because of a lack of exercise combined with the effects of cachexia [8], [12]. There is no treatment for rheumatoid cachexia but the effects can be reversed with high intensity exercise by re-building muscle tissue [8]. Resistance training has been shown to be an effective method to combat cachexia and strength loss without further damage or disease activity [20]. It helps to restore function and build muscle mass while it may decrease disease activity and pain [20].

Resistance training exercise can help reverse or slow the effects of cachexia through increased muscle fiber cross-sectional area, resulting in muscular hypertrophy [8], [12], [20]. Muscular hypertrophy involves size or volume amplification of the cells rather than an increase in the number of cells [21]. Hypertrophy is typically attained through resistance training and involves several biological changes at the cellular level that result in an increase in muscular mass through amplified muscle fiber cross-sectional area [21]. In simple terms, hypertrophy is the result of an increased rate of protein synthesis [22]. Resistance training acts as a mechanical demand that results in gene expression modification in order to increase protein synthesis [21]. The main pathway resulting in increased protein synthesis in muscular hypertrophy is the Akt/TSC2/mTOR pathway [21]. This pathway starts with the activation of the kinase PI3K, leading to phosphorylation of PIP2 into PIP3 (PIP2/3: phosphatidylinositol-bi/triphosphate) [21]. PIP3 provides a location for kinases Akt and PDK to bond [21]. PDK phosphorylates Akt which then triggers phosphorylation of the protein TSC2 [21]. TSC2 congregates with TSC1 and leads to a decrease in the inhibition of the gene Rheb [21]. Rheb then activates the protein mTOR which stimulates ribosomal action, resulting in protein synthesis [21]. The protein IGF-1 is a key activator of the Akt/TSC2/mTOR pathway and therefore supports protein synthesis in muscular hypertrophy [21]. It has been shown that IGF-1 and IGF-binding protein-3 levels increase intramuscularly with resistance exercise, resulting in anabolical effects and thus muscular hypertrophy [12]. A study by Lemmey et al. conducted a randomized controlled trial that consisted of RA patients participating in either a resistance training program or a ROM program for 6 months [12]. IGF-1 levels were measured from muscle biopsies and were shown to increase significantly in the resistance training group compared to the ROM group [12]. IGF-1 also activates and regulates the activity of satellite cells which contributes to hypertrophy as well [21]. Satellite cells, which are simply cells in the muscle that are readily available for a specific purpose, are activated by resistance training, multiply, and join with each other and with other fibers to build muscle tissue [21], [23]. This is actually a reparative process, indicating that muscle tissue is damaged and broken down with resistance training before it is built back up [21]. Resistance training can also increase nuclear density, which results in a greater potential for mRNA transcription which may lead to greater protein accrual and muscle mass [20], [21]. Ultimately, the potential for an increased rate of protein synthesis is made possible by an increase in myonuclei [21]. With an increase in nuclei within the muscle, there is also an increase in the amount of DNA that provides as the source of gene transcription and therefore protein synthesis through the cellular mechanisms involving mRNA translation [21].

Exercise Recommendations and Cachexia

As mentioned above, resistance training can help combat the effects of rheumatoid cachexia by increasing levels of IGF-1 in the muscle which triggers hypertrophy through the Akt/TSC2/mTOR pathway [21]. The reduction of TNF-α levels with exercise also leads to a decrease in protein degradation associated with muscular wasting [8], [19]. Patients with RA may not necessarily see a change in their disease process, but through building muscle and strength, their quality of life is likely to improve with exercise, particularly resistance training and its effects on the muscle.

Mitochondrial Involvement and Exercise in RA

As previously described in the Rheumatoid Arthritis cellular biology section, mitochondrial function is disrupted in Rheumatoid Arthritis. The inflammatory component of this disease increases the accumulation of reactive oxygen species (ROS) leading to an environment of oxidative stress in the cell. Increased levels of mitochondrial derived ROS activate transcriptional factors which then activate gene expression and increase the immune response [24]. Exercise increases the synthesis of ATP through the process of oxidative phosphorylation which in turn leads to a greater production of ROS [25]. Therefore, it is imperative to consider the effects of exercise on the mitochondria in individuals with RA.

Reactive Oxygen Species

It is well documented that the mitochondria in skeletal muscle produce reactive oxygen species at rest as well as during periods of contractile activity [25], [26], [27]. Homeostasis needs to be maintained in the cell as high levels of free radicals produced by exercise-induced oxidative stress can damage cellular processes while low-to-moderate levels of free radicals are necessary for the regulation of gene expression, signaling pathways, and skeletal muscle force production [27]. Mitochondrial adaptations that occur secondary to exercise are dependent upon the frequency, intensity, type, and duration of the exercise [25]. In response to exercise, mitochondria adapt by increasing the production of adenosine triphosphate (ATP) through the process of oxidative phosphorylation (OXPHOS). Exercise also causes the mitochondria to increase the gene expression of necessary proteins and enzymes, induces mitochondrial biogenesis to create more mitochondria or increase inner membrane density for improved protein regulation and energy production, and changes the shape of mitochondria through fission and fusion processes [25]. In response to acute exercise, muscle is fueled aerobically through mitochondrial OXPHOS, but as intensity increases, glycolysis produces ATP to fuel the exercise. The switch between anaerobic glycolysis to mitochondrial oxidative metabolism results in changes of the inner membrane of the mitochondria [25]. The switch between substrates and regulation of the production of ROS is also dependent upon the balance of mitochondrial fission and fusion in which the cell either splits or comes together [25], [27]. Acute exercise-induced fission is important in intense exercise when OXPHOS is not used to produce energy. It enhances glycolysis and the uptake of pyruvate which maintains sufficient levels of energy to fuel exercise while inhibiting the accumulation of lactate [25], [27]. However, if fusion does not occur to balance mitochondrial fission and intense exercise persists, apoptosis will occur [25]. On the positive side, mitochondrial fusion is associated with improved transmission of calcium signals in the cell and helps to repair damaged mitochondrial DNA (mtDNA) which occurs secondary to increased oxidative stress [25]. The balance between these processes allows appropriate remodeling of the mitochondrial network which allows the muscle to adapt to regular physical exercise.

A study by Falone et al. [28] examined the effects of exercise performed by untrained individuals. The study concluded that exercise performed by untrained individuals was associated with an increased amount of ROS secondary to increased oxidative stress. It also concluded that elite athletes who engage in regular high-intensity exercise also demonstrate increased oxidative stress and therefore increased levels of ROS [28]. More importantly this study recognizes that individuals who engage in exercise of moderate-intensity on a regular basis demonstrate improved antioxidant defense mechanisms which decreases the oxidative damage experienced by the mitochondria. This helps to regulate the level of ROS creating a homeostatic functioning state in the mitochondria. ROS that are induced by lower intensity exercise help to promote appropriate substrate usage in the mitochondria, but higher levels of ROS from strenuous and prolonged exercise may induce apoptosis. Increased levels of ROS may also result in type II muscle fibers being utilized more during exercise thus leading to increased muscle fatigue [27]. While regular, moderate intensity maintains the homeostatic balance of the mitochondria through improved antioxidant defense mechanisms, other changes must also occur in the mitochondria to allow proper functioning during exercise. Biogenesis must occur to increase the production of mitochondria in the cell or to increase the inner membrane density for improved electron transport chain functioning which will help to respond to the increased oxygen and fuel demands induced by the exercise. Also the appropriate balance of fission and fusion must occur for the appropriate substrates to be utilized to produce energy, repair mtDNA, and reduce unwanted apoptosis in the cell [25], [27], [28]. When these processes occur and the appropriate amount, regularity and intensity of exercise is performed, the mitochondria maintain a state of homeostasis, generating well-balanced amounts of ROS which help to utilize appropriate substrate to produce ATP to fuel the exercise while avoiding apoptosis.

As previously discussed, RA is characterized by inflammation and oxidative stress, two mechanisms of which are related to muscle atrophy and weakness [29]. In order to examine the effects of exercise on oxidative stress in Rheumatoid Arthritis, it is important to consider the environment of the synovial fluids in the joints. In the joints of individuals with RA, there are low levels of oxygen in the synovial fluid accompanied by increased levels of lactic acid [30]. This signifies that the synovium utilizes anaerobic metabolism to generate ATP. To examine these effects, in vivo studies of the RA synovium have investigated the effects of weight bearing in the form of ambulation [31], [32]. These studies concluded that weight bearing through inflamed joints decreases oxygen levels in the synovium, but when the joint is unloaded reoxygenation occurs [30], [31], [32]. Oxygen levels in the joint during weight bearing decrease as a result of decreased blood flood to the joint and when the joint is unloaded, blood flow is restored to the joint. These frequent periods of hypoxia and reoxygenation result in the production of ROS and the activation of transcription factors [30], [31], [32]. This in turn promotes the inflammatory response and leads to an environment of chronic oxidative stress in the synovium of joints affected by RA [30]. Although exercise cannot reverse the physiological or pathological effects of RA, it can restore balance to disruptions in homeostasis as well as improve overall well-being and prevent the occurrence of other diseases [30]. Research regarding the effects of oxidative stress induced by exercise on individuals with RA is currently limited, but is theorized to be similar to the response produced in individuals without RA [30], [33].

Exercise Recommendations and Reactive Oxygen Species

After reviewing the current literature regarding the effects of exercise on the production of reactive activity species, it can be concluded that individuals with RA should engage in regular, moderate intensity exercise. To summarize the literature, a sedentary lifestyle characterized by inactivity leads to muscle atrophy because of the increased production of ROS in the mitochondria [33], [34]. Exercise also increases the production of ROS, but this production is often considered beneficial since it regulates gene expression, signal transmission and helps to induce changes in the mitochondria which improves substrate usage and contributes to the advantageous adaptations of muscle [27], [34]. Acute exercise, exercise performed by sedentary aged individuals, and intense, strenuous exercise increase oxidative stress in the mitochondria [33], [34]. However, as an individual continues to perform regular exercise at moderate intensity, the mitochondria adapt, the production of ROS is advantageous and homeostasis is maintained [33], [34]. Therefore, individuals with RA should engage in regular, moderate intensity exercise to decrease oxidative stress which will maintain homeostasis for proper functioning of the mitochondria, leading to improved overall health.

Transcription Factors

Nuclear Factor-kappa Beta (NFkB)

NFkB and inflammatory markers

Nuclear Factor-kappa Beta (NFKB) is a transcription factor which transcribes the genes that are responsible for immune system function [35]. Its activation and subsequent transcription of the modulators of immune cells is important in the pathogenesis of RA [35]. In its inhibited form, NFkB is found in the cytosol bound to an inhibitory protein known as IkB. It is activated by IKK, an enzyme that catabolizes IkB [35]. IKK can be stimulated by a number of cellular signaling factors, most importantly, TNF-α, IL-1, IL-6, and reactive oxygen species [35].

NO, or nitric oxide, is used to increase the toxicity of macrophages, neutrophils, and other innate immune system cells [36]. The immune cells use this reactive oxygen species to destroy pathogens, typically viruses, bacteria, and tumor proteins [36]. NO produced by macrophages enter into the extracellular matrix, where they are converted into NO metabolites, a more potent form of NO [37]. NO is converted into its metabolites by an enzyme called nitric oxide synthase (NOS) [37]. There are three types of NOS; neuronal NOS (nNOS), inducible NOS (iNOS), and endothelial type NOS (eNOS) [37]. Research indicates that NO metabolites used by macrophages are created by iNOS [37]. Stimulation of iNOS is dependent on the expression of NFkB [37], and presence of NO and iNOS is a commonly used marker indicating the presence of inflammation [17], [18], [40], [41], [42], [43].

Effect of resistance exercise on NFkB in muscle tissue

Durham and associates examined the effect of a maximal, fatiguing exercise and the expression of NFkB in human subjects [38]. The volunteers were subjected to an protocol which consisted of an hour of lower extremity resistance exercise, followed by 8 reps of leg press at 70% of the subject’s 1 rep max and 8 reps at 80 percent of their 1 rep max [38]. This protocol produced fatigue in all but one of 12 volunteers. Muscle biopsies were taken from the vastus lateralis muscle either 1 hour before exercise, immediately following the exercise, or following 1 hour of bed rest after exercise [38]. NFkB was found to be active in all subjects before the fatiguing exercise, decreased immediately following the exercise, and returned to baseline in an hour after the exercise in all subjects studied [38]. The researchers then studied the levels on NFkB activity in murine tissues, including heart, diaphragm, soleus, liver, gastrocnemius, and extensor digitorum longus muscles at rest [38]. Heart, diaphragm, soleus and liver muscles were found to have a similar baseline NFKB level, while gastrocnemius and extensor digitorum longus muscles were found to be lower [38]. Removed diaphragm muscle was then contracted to determine the effect of fatigue-inducing exercise on NFkB activity [38]. The muscle was repeatedly contracted by electrical stimulation in-vitro for 10 minutes to induce fatigue, as evidenced by a 75 percent force reduction [38]. The tissue was then left passive for 1 hour to determine the effects of recovery [38]. Fatigue exercise induced activation of the transcription factor, and after 1 hour NFkB activation levels had decreased by a statistically significant 44% [38]. To determine the time effect on NFkB, they subjected diaphragm muscle fiber bundles to 1 minute of cyclic contractions to avoid fatigue [38]. This protocol showed a slight decrease in NFkB levels which was determined to be non-significant [38]. From this research, it was determined that fatiguing exercise does not cause a dramatic increase in transcription factors in striated muscle [38].

Ho et al. examined the activation of IKK during in-vivo and in-vitro muscle contractions in the skeletal muscle of mice [39]. In their research in-vivo, they examined the effect of 5, 15, 30, or 60 minutes of moderate intensity treadmill exercise at .8 mph at a 12 percent incline [39]. The researchers then removed the gastocnemius and soleus of the mice muscles to examine the levels of IKK and IkB levels [39]. For in-vitro study, the excised extensor digitorum longus and soleus muscle were subjected to contraction by electrical stimulation at 2 contractions per minute, for 10 seconds per contraction, for 10 minutes [39]. In-vivo, no significant difference was found in IKK levels following a single bout of exercise [39]. Significant differences were found in IKK activity in type I gastrocnemius fibers and soleus muscles after 60 minutes following exercise when compared to baseline activity [39]. All levels of IKK activity returned to baseline by 3 hours following activity [39]. No increase in IKK activity was seen in type II gastrocnemius fibers after any of the exercise parameters [39]. The resulting NFkB binding activity was also increased in soleus and type 1 gastrocnemius fibers, reaching a maximum level at 3 hours post exercise, and returning to baseline in 5 hours post exercise [39]. Similar results were found with in-vitro exercise, with IKK and NFkB activity increasing after 10 minutes of contractile activity and no change in the protein levels of IKK and IkB with exercise [39]. The authors concluded that sub-maximal exercise creates a local effect on NFkB activation in skeletal muscle, which returns to baseline quickly, and is not likely to induce a global immune response [39].

Effect of tensile strain on NFkB

NFkB activity in response to cartilage strain is a possible source for the changes seen in the arthritic knee following continuous passive motion and immobilization treatment. In-vitro cyclic tensile strain (CTS) has been shown to affect the IL-1 activation with a subsequent effect on NFkB activity and inflammatory markers [40], [41], [42]. Dossumbekova et al. examined the effect of cyclic tensile strain on in-vitro murine type II knee chondrocytes and its effect on NFkB expression [40]. These cells were subjected to repeated tensile strain of 3% of the total length of the tissue and a frequency of 25 hertz [40]. Groups of cells were subjected to the intervention for 10, 30, 60, and 90 minutes intervals [40]. In unstimulated chondrocytes and cells subjected to repeated tensile strain, no NFkB activation was noted at any time interval [40]. Cells treated with IL-1 cytokine showed NFkB activation within 10 minutes of application, which continued through 90 minutes of observation [40]. Cells treated with IL-1 and cyclic strain showed near complete inhibition of NFkB activity in the first 10 minutes of treatment, and a more than 90 percent inhibition of binding in the 90 minutes of intervention [40]. Madhavan et al. examined in-vitro cyclic strain on murine TMJ articular disc chondrocytes subjected to a 12% magnitude strain at .05 hertz [41]. Cells were stimulated with a known pro-inflammatory cytokine IL-1 or TNF-α [41]. Cells treated with an inflammatory cytokine showed an increase in iNOS and MMP mRNA production within three hours [41]. Control cells not treated with intervention and cells treated with CTS alone showed no iNOS or MMP activity [41]. CTS treatment in cells subjected to a pro-inflammatory cytokine injection produced a 50 percent reduction in iNOS activity and a 70 percent reduction in MMP mRNA levels when compared to cells treated with pro-inflammatory alone [41]. Cells treated with IL-1 showed an increase in NFkB activity between 10 and 90 minutes of treatment. CTS in IL-1 treated cells showed a decrease in NFkB activity throughout the 90 minute duration of intervention when compared to these cells [41].

The magnitude of applied tensile strain plays a key role in the inhibition of inflammation in articular tissue. Agarwal and associates examined various levels of cyclic tensile strain on cultured rabbit articular chondrocytes [42]. Cells were treated with pro-inflammatory cytokine and subjected to a symmetrical radial strain of .05 hertz at a variety of magnitudes, ranging from 1 to 18 percent [42]. In agreement with previous studies, untreated control cells showed no increase in NO activity [42]. Cells treated with CTS alone also showed a lack of NO activity with magnitudes of strain of up to 8 percent [42]. However, CTS alone increased NO production with strain magnitudes of 12, 14, and 18 percent [42]. IL-1 treatment in CTS treated cells resulted in a dramatic increase in NO production and iNOS activity, which was attenuated by 4 and 8 percent strain magnitude CTS [42]. Levels of strain above 12% in IL-1 and CTS treated cells resulted in a statistically significant increase in NO activity and iNOS activity [42]. As expected, NFkB levels increased with IL-1 treatment alone [42]. Low strain (4 and 8 percent magnitude) resulted in a decrease in NFkB levels after 30 minutes of treatment, with and this effect lasted for 30 minutes after CTS treatment [42]. With high levels of strain magnitude (greater than 15 percent), NFkB levels increased significantly over 30 to 180 minutes of treatment [42].

Effect of compressive strain on NFkB

Dynamic compressive strain (DCS) appears to have a similar threshold for the induction of NFkB and pro-inflammatory cytokines in articular cartilage. Nam et al examined the effect of DCS on cartilage cells in-vitro [43]. Cells were subjected to a DCS of 1 hertz at a variety of strain magnitudes [43]. As with CTS, control cells showed no NFkB, iNOS or NO production [43]. Low levels of DCS resulted in no change in iNOS production [43]. iNOS production increased proportionally with compression, reaching a 300 fold magnification above a 30% strain [43]. IL-1 similarly increased iNOS expression in non-strained cells, and this effect was attenuated by the application of DCS of up to 15 percent [43]. Compressive strains of higher magnitude had no inhibitory effect on the production of iNOS in cytokine treated cells [43]. Rather, 30 percent magnitude DCS resulted in an increase in iNOS production over that induced by IL-1 treatment alone [43]. IKK activity showed a similar reaction, with decreased activity at lower levels (10 percent) of strain and increased activity at higher (30%) strain [43]. This indicates NFkB activation is subject to a threshold similar to that of CTS, with lower levels of strain being inhibitory and higher levels of strain resulting in a dramatic expression of NFkB activity and an increase in inflammatory markers [43].

Exercise recommendations and NFkB

Based on this research, resistance exercise is not likely to cause an inflammatory response initiated by NFkB production in skeletal muscle. While NFkB has been found to increase following moderate exercise [39], it was significantly decreased following fatigue-inducing resistance exercise [38]. In both of these situations, levels of transcription factor returned to baseline quickly. Further, levels of NFkB induced in skeletal muscle are not believed to be of great enough magnitude to induce a global immune response [39]. This data, combined with the effects of exercise on pro-inflammatory cytokines, the hormonal response, and the benefits of resistance exercise on cachexia, and bone regeneration as listed on this page, indicate that muscle activity as a result of resistance training is not likely the cause of symptoms in patients with RA.

In contrast, strain experienced by articular cartilage has been shown to have a large influence on inflammation, joint damage and repair in tissues affected by inflammation [40], [41], [42], [43]. Cyclic tensile strain at low magnitudes (up to 12 percent of tissue deformation) has been shown to decrease the expression of NFkB, and with it the presence of inflammatory markers in arthritic joints [40], [41], [42]. Similar decreases have been noted with compressive forces of up to 15 percent of tissue deformation [43]. The therapeutic effect of cartilage strain appears to happen rather quickly (within 10 minutes of application), and has been shown to last at least 90 minutes in duration [40], [41], [42], [43]. Excess compressive and tensile forces have a contradictory effect, increasing both NFkB and inflammation [40], [42], [43]. No duration has been determined for this effect. Due to the possible damage that can result from an inflammatory episode on joint tissue, high strain activities are best avoided.

The difficulty in making exercise recommendations based on this information lies in the variability in joint forces experienced by articular cartilage during activity. That is, it is difficult to quantify the amount of cartilage tissue deformation experienced during typical exercise paradigms. However, interventions involving movement that are likely to induce at least some minimal level of tissue deformation have been studied [17], [18]. Continuous passive motion has been shown to have a beneficial effect on cartilage and inflammatory markers in arthritic knees. It has been demonstrated that CPM can induce the preservation of cartilage structure, reduce the presence of derogatory enzymes, limit inflammatory and pain-provoking cytokines, and increase the production of anti-inflammatory regulators [17], [18]. As such, joint motion may be critical to the preservation of acutely inflamed arthritic joints. Damage to cartilage has been shown to occur rather quickly following the onset of inflammation and subsequent immobilization (within 24 hours), and low grade movement should begin without haste to best attenuate the detrimental affect on joint tissue.

In consideration of this information, the following exercise recommendations can be suggested for patients with RA. Activity should be encouraged to avoid the damaging effects of immobilization on arthritic joints. Interventions that produce a large amount of joint strain should be avoided. Exercise mediums that unload the joints and decrease weight bearing should be considered to optimize the strain on articular tissue. Examples may include aquatic based programs and stationary cycling without resistance. Low grade joint mobilizations and passive and non-resisted stretch may also be utilized for a similar purpose.

These interventions should take place daily due the rapid destruction that can take place with immobilization. They should take place for a minimum of 10 minutes to maximize the benefits of cartilage deformation on inflammation, and should continue in a best case scenario for at least 90 minutes. Due the variability of joint strains experienced by individuals, the patient should be monitored for signs of inflammation, such as redness, increased temperature, and pain in the joints. If available, the clinician should monitor levels of ACPA to determine the level of disease activity. If signs of inflammation should appear, activity levels should be altered to reduce the strain on joints. However, inactivity should not be encouraged. Low grade joint mobilizations and passive motion is strongly suggested to help attenuate inflammation by taking advantage of the therapeutic effects of low grade articular tissue deformation, even in acutely inflammed joints.


The process of programmed cell death, apoptosis, occurs with exercise and is dependent upon the presence of disease as well as frequency, intensity, type, and duration of the exercise. Apoptosis is regulated by genes that code for proteins that either inhibit (Bcl-2) or facilitate (Bax) the process [44]. The ratio of Bcl-2 relative to Bax expressed during exercise will either inhibit or initiate apoptosis. To highlight this point, if the ratio of Bcl-2 to Bax is high then cells will survive, but if the genes increase the expression of Bax relative to Bcl-2 then cell death will be promoted. Immediately after an acute bout of exercise there is a decreased amount of Bcl-2 present compared to Bax, therefore indicating increased apoptosis which occurs secondary to the increased release of ROS. However, a few days after exercise, homeostasis is restored and Bcl-2 increases in comparison to Bax [44]. As exercise is performed on a regular basis, the appropriate ratio of Bcl-2 to Bax is maintained thus promoting homeostasis [44]. Also playing a role in apoptosis are neutrophils. As discussed previously, neutrophils are a critical component of the innate immune response and are regulated through spontaneous apoptosis. Inhibition of neutrophil apoptosis increases the inflammatory response which results in the accumulation of inflammation seen in individuals with RA [45]. Neutrophil apoptosis is regulated by the release of ROS and can either prolong or decrease the death of neutrophils [45], [46]. Exercise may play a role in modulating this response. As described in the section on reactive oxygen species, acute strenuous exercise increases oxidative stress which results in the release of the pro-inflammatory cytokines, TNF-a, IL-6 and IL-1 and induces apoptosis while regular, moderate exercise maintains homeostatic function of the mitochondria, decreases release of pro-inflammatory cytokines and modulates apoptotic processes [30], [33], [46].

In order to examine the effects of acute, severe and chronic, moderate exercise on the regulation of neutrophil apoptosis and the redox status in healthy subjects, Syu et al. [46] conducted a study involving thirteen sedentary young males. All subjects participated in the acute, severe portion of the exercise protocol which involved cycling with increases in workload every 3 minutes until exhaustion was reached at 90% of the predicted maximal heart rate. Subjects in the exercise group labeled chronic moderate exercise (CME) engaged in exercise for 30 min/day, 5 days/week at 60% maximal workload for 2 months and then participated in 2 months of detraining (ie no regular exercise). Exercise subjects performed the acute severe exercise (ASE) of cycling to exhaustion once per month while the control group (sedentary) was tested every 2 months. Results demonstrated that ASE increased oxidative stress which increased apoptosis of the neutrophils while CME delayed apoptosis and maintained the homeostatic state of the mitochondria [46]. Even after the detraining phase, CME continued to block the effects of ASE which delayed apoptosis and maintained mitochondrial homeostasis [46]. This study provides evidence that healthy individuals should engage in regular physical activity of moderate intensity to maintain a homeostatic state in the mitochondria which will provide protection from premature apoptosis and decrease the release of pro-inflammatory cytokines. Although strong evidence is still somewhat lacking in regards to the role of exercise and apoptosis in individuals with RA, these results can still be synthesized and applied to this population of patients. While the disease process of RA cannot be reversed through exercise, regular physical activity of moderate intensity may be beneficial to maintain mitochondrial homeostasis for the appropriate expression of apoptosis and modulation of pro-inflammatory cytokines.

Hormonal Influence and Exercise in RA


Cortisol is an anti-inflammatory hormone in the body that is secreted through a cascade of hormonal catalysts, triggered by the inflammatory cytokine IL-1β in the hypothalamus – pituitary – adrenal (HPA) axis [47]. A review of studies of exercise’s effect on cortisol by Pool et al turned up evidence on the differences in healthy individuals versus those with rheumatoid arthritis. In the healthy individual, exercise of both moderate and vigorous intensities increases the amount of cortisol produced as well as inducing efficient break down and use of the hormone through its catabolism [4]. During exercise to an exhaustive state, cortisol levels have been shown to rise temporarily followed by a fast decrease to at or below the levels at baseline, within the first few hours post-exercise [4]. The extent of cortisol rise during and fall after exercise are similar for those who exercise regularly and those who do not. However during exercise, it increases a little more for those who are unfit and a little less for those who are [4]. In general, exercising regularly does not significantly effect resting levels of cortisol [4]. In a study using aerobic exercise to exhaustion, compared to healthy controls, patients with an established diagnosis of RA, who were taking nonsteroidal medication for at least 2 years, had a significant reduction in cortisol response to physical exercise with no differece in baseline cortisol levels pre-exercise in both groups [5]. Also, those with RA had lower tolerance of exercise measured by peak VO2 during and post-exercise, despite no change in how the muscles expended energy compared to healthy individuals [5].


Prolactin (PRL) is a hormonal peptide that is pro-inflammatory and helps with immune system regulation by supporting the role of macrophages and increasing the production of antibodies [4], [48]. Among the factors that aid in the release of prolactin centrally are oestrogen, oxytocin, substance P, arginine vasopressin (AVP) and thyrotrophin-releasing hormone (TRH) [49]. The main factor that inhibits or controls the central secretion of prolactin is dopamine (DA) [49]. Additionally, certain stimuli promote the release of prolactin, exercise being among them as it stimulates the increase of prolactin releasing factors (PRF) [49]. Even though exercise has been shown to increase both the production and use of dopamine, the main prolactin inhibitory factory (PIF), serum prolactin increases as the rise in PRFs outweighs the inhibitory effect of dopamine [49]. The specific PRF responsible for this remains unknown, but there is speculation the excitatory pathways of serotonin may be the catalyst that stimulates their release [49].

In a review conducted by Vega et al. on exercise’s effect on prolactin levels, various forms of exercise produced different results. Intense anaerobic exercise is a strong PRL secreting stimulus where higher intensities are correlated with higher levels of PRL [49]. Further synthesis of the research lead to the finding that when exercising at a low intensity, 50% VO2max, more than 60 minutes of duration may pass before a rise in PRL is noted, while if one exercises at 75%VO2max for the same amount of time, there are significantly greater increases noted [49]. In regards to exercising at a % of VO2max, research has shown that short duration aerobic exercise of 30 minutes at 50% VO2max an increase in PRL was not noted, at 70% VO2max PRL increased slightly, and increased the most at 90% VO2max [50]. The same study injected lactate into a resting control group at concentrations found exercising between 70-90%VO2max and a significant rise in PRL was noted, yet not as large as those exercising [50]. The idea of increasing lactate playing a role in increasing PRL would in part support the correlation of exercise intensity with level of PRL. In a study by Daly et al., PRL peaked at 30 minutes post high intensity (100% VO2) endurance (until fatigue) runs; it began to decrease an hour post-exercise and returned to baseline at 90mintues post-exercise [51]. Literature on PRL levels in those who perform strength training revealed that all forms of resistance training from a single session to chronic training to heavy chronic training, do not produce significant changes in PRL level [49].

In a study conducted by Pool et al., serum prolactin levels were measured before and after a controlled experiment of increasing intensity during cycle ergometry to exhaustion in those with rheumatoid arthritis compared to healthy individuals [5]. The level of prolactin at the peak of exercise, near exhaustion, in healthy subjects was 66% higher, lowering to baseline 1hr post; in subjects with RA was 3% higher at peak, lowering 11% post [5]. The results for prolactin levels in the RA group were not significant, but they correlated at every point of measurement with the third group of the study made up of subjects with systemic lupus erythematosus (SLE), another autoimmune disease [5]. Another point to consider in this study is that the VO2 capacity of the subjects with RA was significantly lower and their lactate threshold (LT) occurred closer to their VO2max compared to healthy subjects [5]. It is possible that because the healthy individuals reached their LT sooner than those with RA in the exercise bout, they produced more prolactin before reaching their peak exhaustion. It is also possible that those with RA truly do have an abnormal prolactate response to exercise, which if that is the case may be a positive effect counteracting the lack of rise in cortisol as well.


Estrogen, one of the sex hormones, has been established as playing a role in enhancing humoral immunity and the inflammation response [52]. The level found in the blood with those with RA is similar to the level in healthy individuals [52]. However, the level of estrogen found in rheumatic joints is significantly higher than normal due to increased presence of inflammatory cytokines (IL-1, IL-6, and TNF-α) and their interaction with aromatase that results in the production of estrogen [52]. As an enhancer of immunity, 17-beta estradiol activates fibroblast and macrophage proliferation in synovial tissue and 16 alpha-hydroxyesterone stimulates mitosis and transformation of lymphocytes [52]. Furthermore, an anti-inflammatory form of estrogen: 2-hydroxyestrogen has lower levels in the synovial fluid of those with RA [53]. So, the imbalance of various forms of estrogens, along with the overall concentration of estrogen within the synovium are responsible for the increase in joint inflammatory symptoms in those with RA [54]. It was also found that estrogen plays a role in initiating NFkB which is one of the ways it acts as an enhancer of the inflammation response [52].

The form of estrogen found in greatest abundance in the synovial fluid of joints with rheumatoid arthritis compared to osteoarthritis and healthy joints is 16 alpha-hydroxyesterone and has been found to be the only form of estrogen that does not inhibit the secretion of tumor necrosis factor (TNF) [55]. Androgens have anti-inflammatory properties, however their low level in rheumatic joints is due to TNF inhibiting the necessary change of the inactive dehydroepiandrosterone sulfate (DHEAS) to the active form (DHEA) of the androgen [56].

In a study performed by Schmitz et al., premenopausal women participated in an aerobic exercise program for 15 weeks with estrogen levels tested prior to and after the training program through urine analysis and gas spectrometry [57]. The exercise protocol included 5 times a week of: 30 minutes on the treadmill or elliptical progressing from 70-85% of max HR over the weeks of the program [57]. The study found that there was no significant change in the levels of estrogen, specifically: 2 and 4 hydroxyestrogens and 16 alpha-hydroxyestrone [57]. In postmenopausal women whose serum level of estrogen was tested and matched up with their amount of physical activity over the course of 4 weeks, 3 times, 2 years apart each [58]. A non-linear relationship was found as in lower estrogen levels were found in participants who were the least and most physcically active and higher levels in those whose physical activity level was in between [58].

Exercise Recommendations based on Hormonal Response

Studies have shown that when patients who are earlier in the disease process, are aerobically exercised to exhaustion, the hormonal response of cortisol and prolactin are abnormal [5]. Cortisol the anti-inflammatory hormone does not increase above baseline and Prolactin the proinflammatory hormone also does not reach a clinically significant increase [5]. Therefore, for those with RA the effects cancel eachother out. It has been shown that those with RA have a lower VO2max and that their lactate threshold is closer to their VO2max than healthy individuals [5]. Higher levels of lactate are linked to increasing prolactin [50]; it can therefore be seen that even with aerobic exercise to exhaustion, the minimal rise in prolactin in those with RA is not as great as the healthy individuals who reach the lactate threshold sooner and continue to increase their lactate and prolactin levels until they reach their higher VO2max. So, for individuals with RA who have been recently diagnosed (but long enough to be on anti-inflammatories for 2 years) aerobic exercise to exhaustion has not proven to be detrimental inasfar as the inflammatory properties of prolactin and cortisol are concerned.

Among the various forms estrogen, 16-hydroxyesterone are found in high concentrations in rheumatic joints and stimulate proliferation of immune cells in the synovium [52]. 16-hydroxyesterone is created in response to the interaction between aromatase and inflammatory cytokines (IL-1, IL-6, and TNF-α) in the joint [52]. It is also the only form of estrogen in the body that does not inhibit TNF-α [55]. Research has confirmed that IL-1 and TNF-α do not increase with aerobic exercise between 60-75% of VO2max and that the IL-6 released is of an anti-inflammatory nature, from the muscles [6]. Therefore, as the inflammatory cytokines are not increased with aerobic exercise, in theory this would not affect the level of 16-hydroxyesterone in the joints which in turn would not further promote the funtion of TNF-α, and again, lead to a recommendation of this form of exercise for those with RA.

Research has linked estrogen to the activation of NFkB [52], however it did not specify which form of estrogen. A non-linear relationship has been found in estrogen after exercise in healthy post-menopausal women [58], again without the form specification as to if they are the pro or anti-inflammatory metabolites which could be the wild card in varying individual response to exercise especially after menopause, as estrogen levels do not rise after exercise in healthy premenopausal women [57]. More research is needed in this area to confirm that theory, however as stated previously it is always important to monitor patient response for increase in symptoms in response to exercise and to proceed or adjust activity accordingly. When pre-menopausal, a moderate to intense aeobic program consisting of 30 minutes, 5 times a week at 70-85% of max HR [57] can be recommended cautiously as this has not been confirmed in those with RA, but if the same holds true, this frequency and intensity would not carry the risk of increasing the inflammatory effects of 16-hydroxyesterone.

General Exercise Recommendations

Activity should be encouraged in patients with RA to promote the general health benefits of exercise. Exercise including weight bearing, resistance training, stretching, joint movement, and aerobic activity up to 70-85 percent of maximum heart rate has been shown to be beneficial in slowing the progression of RA. These benefits include offsetting the effects of rheumatoid cachexia, optimal regulation of pro and anti-inflammatory cytokines, promotion of bone formation, long-term reduction of reactive oxygen species, improvements in mitochondrial function, promotion of apoptosis of inflammatory immune system cells, preservation of joint tissue, and reduction of pain. As discussed above, several research studies have incorporated surprisingly intense exercise protocols in patients with RA without provocation of symptoms. While exercise may not stop the progression of the disease, it can slow its course, provide for improvements in overall health, and with it, promote better quality of life and function.

As with any disease, precautions should be taken. Signs of inflammation, such as redness of joints, increased temperature, and pain should be monitored closely. In a clinical setting, RA-specific markers, such as ACPA levels and blood leukocyte counts should be monitored to detect changes in disease activity, if available. In the event of inflammation, exercise should be altered. Resistance exercises and aerobic exercise involving weight bearing and joint impact should be reduced. However, inactivity should not be encouraged. Joint damage takes place quickly in arthritic joints, and inactivity and immobilization serve to promote this effect. Low strain, passive range of motion and joint mobilization serve to decrease inflammation and pain, and should be utilized in the event of acute inflammatory event in RA patients.

Frequency: Some activity all days of the week.

Rationale: Activity promotes optimal oxidative homeostasis and mitochondrial health, provides an anti-inflammatory benefit from the release of IL-6, and prevents the occurrence of joint degradation found with immobilization. Joint damage in arthritic joints can occur within 24 hours of immobilization.

Duration: At least 30 minutes, with up to 90 minutes being optimal.

Rationale: Decreases in inflammatory markers and pain provoking enzymes are seen after 10 minutes of low grade joint deformation. This effect has been shown to last up to 90 minutes. At least 30 minutes to is needed to optimize estrogen metabolism.


Resistance exercise: 3 times per week, 3 sets of 10 reps 60-80 percent of 1 rep max.

Rationale: Resistance exercises reduce TNF-α levels, increase osteoblast and bone formation by increasing IGF-1, offsets the effects of rheumatoid cachexia and provides mechanical stress to joint tissues to optimize the effect of strain on inflammation.

Aerobic exercise: 3 times per week, up to 70-85 percent of max heart rate

Rationale: Aerobic exercise on a regular basis, optimizes estrogen metabolism, improves mitochondrial homeostasis, and reduces levels of pro-inflammatory cytokines.

Stretching and range of motions exercises: Daily, passive range of motion to end range. Low grade, long duration stretching and movement of joint tissue

Rationale: Stretching and range of motion exercises reduce pro-inflammatory cytokines through compression and distraction of joint tissue, prevents cartilage damage from the effects of inflammation and immobilization, and reduces levels of pain-provoking enzymes.

Mode: Anything that allows for increased motion with decreased impact on joints, such as stationary cycling and aquatics.

Rationale: Provides for increased intensity to maximize the benefits of exercise while reducing joint impact.

Precautions in patients with acute inflammation:

During periods of acute inflammation, activities involving joint impact, such as resistance exercises and aerobic activity involving weight bearing should be reduced and low impact, non-weight bearing exercise should be performed instead. Inactivity should be discouraged. Long duration (between 10 and 90 minutes) of passive range of motion and low grade-joint mobilizations should be utilized to help reduce inflammation and pain.

Rationale: Large magnitudes of joint strain serves to promote inflammation, while low magnitude strains are inhibitory. Cartilage damage in acutely inflamed joints occurs quickly (within 24 hours), and long duration passive motion serves to protect from this effect.


Exercise is an integral component in the treatment of patients with RA. Activity, including a regular program of moderate aerobic exercise and resistance training, should be encouraged to promote overall health and quality of life. Stretching and joint mobilization serve to protect cartilage from the damaging effects of inflammation, and provide relief from pain. While precaution should be taken to prevent and address acute inflammatory episodes, inactivity should not be encouraged as a pathway to restored health. Through careful monitoring of patients and selective alteration of exercise protocols, exercise can serve to alleviate the symptoms and delay the progression of Rheumatoid Arthritis.

Go to Rheumatoid Arthritis Cell Biology Page

1. Goebel MU, Mills PJ, Irwin MR, Ziegler. Interleukin-6 and tumor necrosis factor-a production after acute psychological stress, exercise, and infused isoproterenol: differential effects and pathways. Psychosom Med. 2000; 62:591-598.
2. Moldoveanu AI, Shephard RJ, Shek PN. Exercise elevates plasma levels but not gene expression of IL-1β, IL-6, and TNF-α in blood mononuclear cells. J Appl Physiol. 2000; 89:1499-1504.
3. Scharhag J, Meyer T, Gabriel HW, et al. Does prolonged cycling of moderate intensity affect immune cell function? Br J Sports Med. 2005; 39:171-177.
4. Pool, AJ. and Axford, JS. The effects of exercise on hormonal and immune systems in rheumatoid arthritis. Rheumatology. 2001. 40: 610-614.
5. Pool, AJ., Whipp, BJ., Skasick, AJ., Alavi, A., Bland, JM., and Axford, JS. Serum cortisol reduction and abnormal prolactin and CD4+/CD8+ T cell response as a result of controlled exercise in patients with rheumatoid arthritis, systemic lupus erythematosus despite unaltered muscle energetics. Rheumatology. 2004. 43(1): 43-48.
6. Petersen AW and Pedersen BK. The anti-inflammatory effect of exercise. J Appl Physiol. 2005; 98:1154-1162.
7. Walsh NP, Gleeson M, Shephard RJ, et al. Position statement part one: immune function and exercise. Exerc Immunol Rev. 2011; 17:6-63.
8. Cooney JK, Law RJ, Matschke V, et al. Benefits of exercise in rheumatoid arthritis. J Aging Res. 2011; 2011: 681640.
9. Choi Y, Arron JR, Townsend MJ. Promising bone-related targets for rheumatoid arthritis. Nat Rev Rheumatol. 2009; 5: 543-8.
10. Rubin J, Rubin C, Jacobs CR. Molecular pathways mediating mechanical signaling in bone. Gene. 2006; 367: 1-16.
11. Karatay S and Yildirim K. Effects of dynamic exercise on circulating IGF-1 and IGFBP-3 levels in patients with rheumatoid arthritis or ankylosing spondylitis. Clin Rheumatol. 2007; 26: 1635-9.
12. Lemmey AB, Marcora SM, Chester K, Wilson S, Casanova F, Maddison PJ. Effects of high-intensity resistance training in patients with rheumatoid arthritis: a randomized controlled trial. Arthritis Rheum. 2009; 61(12): 1726-34.
13. Engvall IL, Tengstrand B, Brismar K, Hafstrom I. Infliximab therapy increases body fat mass in early rheumatoid arthritis independently of changes in disease activity and levels of leptin and adiptonectin: a randomized study over 21 months. Arthritis Res Ther. 2010; 12(5): R197.
14. Hakkinen A, Sokka T, Kautiainen H, Kotaniemi A, Hannonen P. Sustained maintenance of exercise induced muscle strength gains and normal bone mineral density in patients with early rheumatoid arthritis: a 5 year follow up. Ann Rheum Dis. 2004; 63(8): 910-6.
15. Czarkowska-Paczek B, Wesolowska K, Przybylski J. Physical exercise prevents osteoporosis. Przegl Lek. 2011; 68(2): 103-6
16. Martin RB. Fatigue microdamage as an essential element of bone mechanics and biology. Calcif Tissue Int. 2003; 73: 101-7.
17. Ferretti M, Srinivasan A, Deschner J, Gassner R, Baliko F, Piesco N, Salter R, Agarwal S. Anti-inflammatory effects of continuous passive motion on meniscal fibrocartilage.
J Orthop Res. 2005;23(5):1165-71.
18. Ferretti M, Gassner R, Wang Z, Perera P, Deschner J, Sowa G, Salter R, Agarwal S. Biomechanical signals suppress proinflammatory responses in cartilage: early events in experimental antigen-induced arthritis. J Immunol. 2006;177(12):8757-66.
19. Walsmith J and Roubenoff R. Cachexia in rheumatoid arthritis. Int J Cardiol. 2002; 85(1): 89-99.
20. Sharif S, Thomas JM, Donley DA, et al. Resistance exercise reduces skeletal muscle cachexia and improves muscle function in rheumatoid arthritis. Case Report Med. 2011; 2011: 205691.
21. Favier FB, Benoit H, Freyssenet D. Cellular and molecular events controlling skeletal muscle mass in response to altered use. Pflugers Arch. 2008; 456(3): 587-600.
22. Berdeaux R and Stewart R. cAMP signaling in skeletal muscle adaptation: hypertrophy, metabolism and regeneration. Am J Physiol Endocrinol Metab. 2012.
23. Morgan JE and Partridge TA. Muscle satellite cells. Int J Biochem Cell Biol. 2003; 35(8): 1151-6.
24. Harty LC, Biniecka M, O’Sullivan J, Fox E, Mulhall K, Veale DJ, Fearon U. Mitochondrial mutagenesis correlates with the local inflammatory environment in arthritis. Ann Rheum Dis. 2011; 1-7.
25. Bo H, Zhang Y, Ji LL. Redefining the role of mitochondria in exercise: a dynamic remodeling. Ann. N.Y. Acad. Sci. 2010; 1201: 121-128.
26. Jackson MJ. Reactive oxygen species and redox-regulation of skeletal muscle adaptations to exercise. Phil Trans R Soc B. 2005; 360: 2285-2291.
27. Powers SK, Jackson MJ. Exercise-induced oxidative stress: cellular mechanisms and impact on muscle force production. Physiol Rev. 2008; 88: 1243-1276.
28. Falone S, Mirabilio A, Pennelli A et al. Differential impact of acute bout of exercise on redox- and oxidative damage-related profiles between untrained subjects and amateur runners. Physiol Res. 2010;59:953-961.
29. Reid MB, Moylan JS. Beyond atrophy: redox mechanisms of muscle dysfunction in chronic inflammatory disease. J Physiol. 2011: 2171-2179.
30. Hitchon CA, El-Gabalawy. Oxidative in rheumatoid arthritis. Arthritis Res Ther. 2004; 6: 265-278.
31. James MJ, Cleland LG, Rofe AM, Leslie AL. Intraarticular pressure and the relationship between synovial perfusion and metabolic demand. J Rheumatol. 1990; 17: 521-527.
32. Levick JR. Hypoxia and acidosis in chronic inflammatory arthritis: relation to vascular supply and dynamic effusion pressure. J Rheumatol. 1990; 17: 579-582.
33. Jackson MJ, McArdle A. Age-related changes in skeletal muscle reactive oxygen species generation and adaptive responses to reactive oxygen species. J Physiol. 2011; 589: 2139-2145.
34. Bottinelli R, Westerblad H. Reactive oxygen and nitrogen species in skeletal muscle: acute and long-term effects. J Physiol. 2011; 589.9: 2117-2118.
35. Okamoto H, Cujec T, Yamanaka H, Kamatani N. Molecular aspects of rheumatoid arthritis: role of transcription factors. FEBS J. 2008 ;275(18):4463-70.
36. Bogdan C. Nitric oxide and the immune response. Nat Immunol. 2001;2(10):907-16.
37. Silveira E, Rodrigues M, Krause M, Vianna D, Almeida B, Rossato J, Oliveira L, Curi R, de Bittencourt P. Acute exercise stimulates macrophage function: possible role of NF-kappaB pathways. Cell Biochem Funct. 2007;25(1):63-73.
38. Durham W, Li Y, Gerken E, Farid M, Arbogast S, Wolfe R, Reid M. Fatiguing exercise reduces DNA binding activity of NF-kappaB in skeletal muscle nuclei. J Appl Physiol. 2004;97(5):1740-5.
39. Ho R, Hirshman M, Li Y, Cai D, Farmer J, Aschenbach W, Witczak C, Shoelson S, Goodyear L. Regulation of IkappaB kinase and NF-kappaB in contracting adult rat skeletal muscle. Am J Physiol Cell Physiol. 2005;289(4):C794-801.
40. Dossumbekova A, Anghelina M, Madhavan S, He L, Quan N, Knobloch T, Agarwal S. Biomechanical signals inhibit IKK activity to attenuate NF-kappaB transcription activity in inflamed chondrocytes. Arthritis Rheum. 2007;56(10):3284-96.
41. Madhavan S, Anghelina M, Sjostrom D, Dossumbekova A, Guttridge DC, Agarwal S. Biomechanical signals suppress TAK1 activation to inhibit NF-kappaB transcriptional activation in fibrochondrocytes. J Immunol. 2007;179(9):6246-54.
42. Agarwal S, Deschner J, Long P, Verma A, Hofman C, Evans C, Piesco N. Role of NF-kappaB transcription factors in antiinflammatory and proinflammatory actions of mechanical signals.
Arthritis Rheum. 2004;50(11):3541-8.
43. Nam J, Aguda B, Rath B, Agarwal S. Biomechanical thresholds regulate inflammation through the NF-kappaB pathway: experiments and modeling. PLoS One. 2009;4(4):e5262.
44. Phaneuf S, Leeuwenburgh C. Apoptosis and exercise. Med Sci Sports Ex. 2001; 293-296.
45. Wang K, Hampson P, Hazeldine J, et al. Cyclin-dependent kinase 9 activity regulates neutrophil spontaneous apoptosis. PLoS One. 2012; 7(1): e30128.
46. Syu GD, Chen H, Jen CJ. Severe exercise and exercise training exert opposite effects on human neutrophil apoptosis via altering the redox status. PLoS One. 2011; 6(9): e24385.
47. Morand, EF., and Leech, M. Hypothalamic-pituitary-adrenal axis regulation of inflammation in rheumatoid arthritis. Immunol Cell Biol. 2001. 79(4):395-399.
48. Becker, KL. Critical Illness and Systemic Inflammation: Principles and Practice of Endocrinology Metabolism. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 2001. 2072-2073.
49. Vega, SR., Hollmann, W., and Struder, HK. Influences of exercise and training on the circulating concentration of prolactin in humans. Journal of Neuroendocrinology. 2011. 24:395-402.
50. Luger, A., Watchinger, B., Deuster, P., Svoboda, T., Clodi, M., and Chrousos, GP. Plasma growth hormone and prolactin responses to graded levels of acute exercise and to lactate infusion. Neuroendocrinology. 1992. 56(1): 112-117.
51. Daly, W. Seegers CA., Rubin, DA., Dobridge, JD., and Hackney, AC. Relationship between stress hormones and testosterone with prolonged endurance exercise. Eur J Appl Physiol. 2005. 93(4): 375-80.
52. Cutolo M, Capellino S, Montagna P, Villaggio B, Sulli A, Seriolo B, and Straub RH. New roles for estrogens in rheumatoid arthritis. Clin Exp Rheumatol. 2003. 21(6):687-690.
53. Cutolo, M., Villaggio, B., Seriolo, B., Montagna, P., Capellino, S., Straub, RH., and Sulli, A. Synovial fluid estrogens in rheumatoid arthritis. Autoimmun Rev. 2004. 3(3):193-198.
54. Capellino, S., Montagna, P., Villaggio, B., and Cutolo, M. Effects of estrogen peripheral metabolism in rheumatoid arthritis. Rheumatismo. 2005. 57(2): 78-82.
55. Schmidt, M., Hartung, R., Capellino, S., Cutolo, M., Pfeifer-Leeg, A., and Straub, RH. Estrone/17beta-estradiol conversion to, and tumor necrosis factor inhibition by, estrogen metabolites in synovial cells of patients with rheumatoid arthritis and patients with osteoarthritis. Arthritis Rheum. 2009. 60(10)2913-2922.
56. Weidler, C., Struharova, S., Schmidt, M., Ugele, B., Scholmerich, J., and Straub, RH. Tumor necrosis factor inhibits conversion of dehydroepiandrosterone sulfate (DHEAS) to DHEA in rheumatoid arthritis synovial cells: a prerequisite for local angrogen deficiency. Arthritis Rheum. 2005. 52(6):1721-1729.
57. Schmitz, KH., Warren, M., Rundle, A., Williams, NI., Gross, MD., and Kurzer, MS. Exercise effect on oxidative stress is independent of change in estrogen metabolism. Cancer Epidemiol Biomarkers Prev. 2008. 17: 220-223.
58. Bertone-Johnson, ER., Tworoger, SS., and Hankinson, SE. Recreational physical activity and steroid hormone levels in post-menopausal women. American Journal of Epidemiology. 2009. 170(9): 1095-1104.


Add a New Comment
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License