Cytokine hypothesis of overtraining: a physiological adaptation to excessive stress? : Medicine & Science in Sports & Exercise

Journal Logo


Cytokine hypothesis of overtraining: a physiological adaptation to excessive stress?


Author Information
Medicine & Science in Sports & Exercise 32(2):p 317, February 2000.
  • Free


The purpose of this paper is to integrate available information pertaining to the overtraining syndrome (OTS) into one paradigm, which will be referred to as the cytokine hypothesis of overtraining. The following hypothesis is not presented as complete but is advanced in an attempt to focus future research efforts. For brevity, references are generally limited to review articles. The predominant focus of this paper will be on the systemic immune/inflammatory response. These terms are frequently used interchangeably due to their extensive overlap; for conciseness, the term systemic inflammation will be used.

Athletes train hard to optimize performance. Inherent in all training programs is the application of the progressive overload principle, which implies working beyond a comfortable level in order to maximize athletic ability (26,27,45,91). Unfortunately, there is a fine line between improved performance and deterioration. When deterioration in performance occurs in association with an arduous training schedule, it is referred to as overtraining, staleness, or burnout (66).

The universal criterion associated with overtraining is a decrease in performance. However, not all aspects of performance are affected simultaneously nor are they impacted to the same degree, making prediction and/or interpretation confusing (66). It is also probable that other signs/symptoms typically associated with overtraining are evident before a deterioration in performance. These might include generalized fatigue, depression, muscle and joint pain, and loss of appetite. However, it is the decline in performance frequently associated with an increased volume or load of training, that captures the attention of the athlete and coach. A large number of symptoms associated with overtraining, have been reported in the literature. Fry et. al. (27) have categorized these according to physiological performance, psychological/information processing, immunological, and biochemical parameters (see Table 1). However, there is no universally agreed upon cluster of symptoms, and no cluster that would conveniently describe overtraining associated with a particular sport, or a particular type of training (such as aerobic versus anaerobic). For the most part, multiple symptoms may be present in a variety of combinations, and it is this cluster that is referred to as OTS.

Table 1:
The major symptoms of overtraining as indicated by their prevalence in the literature (Reprinted from Fry, Morton, and Keast, 1991)
Table 1A:

In contrast to overtraining, overreaching is a term used to imply a temporary deterioration in performance, reflecting the time period between the application of a exacting stimulus, and subsequent recovery and adaptation (26,27,45,48,91). In many training cycles, athletes experience this short-term overreaching as they increase intensity and/or volume but recover rapidly and improve or maintain performance. However, if the athlete continues to show a decrement in performance, even with an appropriate rest/regeneration period, this is most likely OTS.

Since there is a continual risk of imbalance between training, competition, and recovery, OTS is a common problem (48). Sixty percent of distance runners, 21% Australian swimmers, and more than 50% of soccer players, have been classified as overtrained. Presently the only known treatment is a decrease in training volume or in some instances complete rest. “Once the athlete has developed the full-blown overtraining syndrome, he or she must rest completely for anything between 6 to 12 weeks…” (64). OTS is most likely also prevalent amongst recreational athletes, but has not received the same attention, for obvious reasons.

Existing Theories of OTS

A variety of hypotheses have been proposed to account for OTS. A number of these hypotheses remain viable, whereas others have gained minimal support. It will be suggested that many of these hypotheses represent pertinent aspects of the syndrome (45,47,89). For more extensive information, the reader is referred to excellent reviews (24,26,27,91).

Several investigators have focused on the role of the hypothalamus, which results in activation of the autonomic nervous system (47), and the hypothalamic-pituitary-adrenal axis (HPA), as well as involvement of the hypothalamic-pituitary-gonadal axis (HPG); this results in alterations of blood catecholamine, glucocorticoid, and testosterone levels (37). Undoubtedly, there is involvement of these systems in OTS, since heavy training represents an extreme stress, both physically and psychologically. However, it will be proposed that activation of these pathways may be a consequence, and not necessarily a primary initiator.

There is substantial evidence demonstrating reductions in blood levels of the amino acid, glutamine, in OTS (36). Newsholme’s glutamine theory (62) proposes that reduced blood glutamine is responsible for the frequently observed impaired immune response and associated increased rate of infection seen in OTS, since glutamine is a primary fuel utilized by lymphocyte cells (69).

Several investigators (44,62) have focused on the reduction of circulating levels of the amino acid tryptophan (TRY). Reduced blood levels of TRY have been interpreted to reflect a greater uptake of this amino acid by the brain. Tryptophan is the precursor for synthesis of the brain neurotransmitter serotonin. Increased brain levels of serotonin are believed to result in mood and behavioral changes, such as inducing sleep and reducing appetite, both behaviors evident in OTS (44).

The glycogen hypothesis of overtraining (14) has suggested that in response to dramatic increases in training load, certain athletes are unable to maintain sufficient intake of calories, in particular carbohydrate, and that this would result in reduced muscle glycogen, and could account in part, for feelings of fatigue and reduced performance. Although this phenomenon has been frequently observed in OTS, this theory has not been substantiated (89).

Foster and Lehman (24) have suggested that the lack of day to day variation in training, could induce the OTS; this is referred to as the monotony theory of overtraining. Inherent in this theory is the assumption that the psychological monotony can impact on physiological performance. An alternate interpretation for the involvement of monotony in OTS is that the daily “sameness” of intense training will impose excessive stress on the musculo-skeletal-joint system, thus making the athlete more prone to injury.

At present, there is no all encompassing hypothesis for OTS. The view presented in this paper will attempt to integrate the above information into a unifying hypothesis. To be acceptable, it must account for the diverse physical, physiological, behavioral, and psychological changes associated with OTS. It must also explain how OTS, where similarities are more striking than differences, occurs in response to a wide array of training regimens and athletic events.

Muscle Trauma and Systemic Inflammation

The present hypothesis proposes that trauma to the muscular, skeletal, and/or joint system, is frequently the initiator of OTS. However, before presenting this argument, it seems appropriate to discuss the presence of “naturally” occurring, exercise-related, tissue trauma. It is now widely accepted that training and competing results in degrees of microtrauma to muscle, connective tissue, and/or bones and joints (87). This type of “injury” will be referred to as adaptive microtrauma (AMT) and may be regarded as an initial phase along an “injury continuum.” Contending with this AMT may require nothing more than an appropriate training program that includes rest days, and/or hard and easy work days, and or cross-training, to allow for recovery.

It is proposed that AMT may be induced via several mechanisms. It is well documented that the eccentric component of a movement will induce tissue trauma (86). Additionally, it is suggested that exercise requiring elevated local metabolic demands, such as high-intensity cycling, may induce “pockets” of ischemia, resulting in ischemic/reperfusion injury (1,12). Finally, it is also proposed that joint structures involved in high volume repetitions, would induce AMT in these structures (see Fig. 1). The reason for referring to this microinjury as “adaptive” is that it is widely believed that AMT results in a mild inflammatory response, with the final purpose of “healing” (13,50,86). The healing process may result in an “overshoot” phenomenon and be associated with an adaptation (13) of muscle, bone, and/or connective tissue.

Figure 1:
Schematic diagram of proposed manner by which various musculoskeletal actions may result in tissue trauma/injury.

Musculo-skeletal-joint trauma/injury, proposed as the underlying cause of OTS, may be induced by a variety of circumstances. Conceivably, this injury may be due to a progression from the initial benign AMT-stage, to a subclinical injury in the athlete who is training too hard and too frequently (2,71,82). Another possibility is a circumstance involving continued training, before recovery from an acute injury, which may exacerbate the initial injury (39,81,91). Kibler and Chandler (39) suggest that relative to overtraining “the types of injuries identified, range from the overt, that are obvious injuries and will usually prevent performance for some period of time, to the subclinical, that decrease performance but may be seldom recognized.”

As stated previously, the universally accepted sign of OTS is a decrease in performance (6,27,91). Injury would undoubtedly compromise performance. A large body of research demonstrates that even minor muscle trauma, as is seen after an unaccustomed bout of eccentrics, interferes with performance (13). Injury impacts locally on factors such as strength and range of motion, which affects overall performance. Due to injury “the athlete may modify participation, and at times may cause an injury in a distant part of the kinetic chain, likely due to abnormal biomechanical movement patterns” (39).

It has been stated that musculoskeletal overuse injuries represent a “…musculoskeletal manifestation of the overtraining syndrome” (39). This implies first the development of OTS and then the inception of injury. However, it is proposed here, that the injury may be both the initiating and perpetuating cause of OTS. Many reports suggest the presence of injury in an overtrained athlete. Such reports include muscle and joint soreness and tenderness, persistent muscle soreness that increases with each session, and elevated serum creatine kinase (25,64). More direct evidence has recently been made available by the work of Seene and colleagues (78), who reported extensive muscle damage in biopsies of overtrained athletes.

The cytokine hypothesis of overtraining will propose that repetitive trauma to the musculoskeletal system, due to high intensity/volume training, associated with insufficient rest/recovery time, is the predominant cause of overtraining. It will be suggested that many of the physiological, behavioral, and psychological signs and symptoms associated with OTS could emerge from the presence of an injury. Additionally, the cytokine hypothesis will attempt to accommodate alternate stressors that may be causal or may contribute in an additive sense, such as psychological stress (61) or an acute viral infection (36,75).

Injury, Inflammation, and Cytokines

The proposed connection between injury and OTS is as follows. Subacute exercise-induced musculoskeletal trauma will result in the release of local inflammatory factors, cytokines. With continued high-volume, high-intensity training and limited rest, typically associated with OTS, local acute inflammation becomes chronic, and the cytokines released in this process activate circulating monocytes (46,71). Activated monocytes produce large quantities of proinflammatory cytokines, resulting in systemic inflammation. Systemic inflammation is proposed as the central underpinning of OTS (see Fig. 2).

Figure 2:
Schematic diagram of proposed exercise-related events leading to the development of a systemic immune/inflammatory response.

Inflammation is the generalized response of the body to tissue injury, irrespective of the damaging stimulus. The primary focus of acute inflammation is healing, a process crucial to survival. Overt signs and symptoms of inflammation include swelling, redness, heat, pain, and reduction in the function of the injured area. However, not all clinical manifestations are consistently detectable. There are undoubtedly variations in the nature and the magnitude of the inflammatory response (9), dependent upon such factors as the extent of the injury, the tissue type, and nutritional status. The present discussion will focus predominantly on inflammation occurring in response to exercise-induced muscular-skeletal injury (87).

In response to tissue injury, the body mounts an elaborate, synchronized response, with extensive amplification at each step. The overall response is characterized by movement of fluid, plasma protein, and leukocytes, from the circulation into injured tissue. Many of the initial events, manifested within a few hours after injury, are directed toward local recruitment of specific white blood cells. Neutrophils represent the first wave of infiltrating cells and play a vital role in the “clean-up” process. Neutrophils predominate during the initial phase of acute inflammation but by 24 h are no longer active (86).

Monocytes form the next line of defense. When these cells move from the circulation into the tissue, they are transformed into macrophages. When activated, either as a circulating monocyte, or as a tissue macrophage, this “complex, powerful and mobile cell,” is capable of secreting over 100 different chemicals and is central to the local and systemic inflammatory process. In the present paper, focus will be on activated, circulating monocytes, representative of a systemic inflammatory response.

Although neutrophils and monocytes are regarded as primary players in an inflammatory response, coordination of these cells, as well as amplification of numerous aspects of inflammation, are accomplished by a group of molecules collectively known as cytokines (84). In recent years, there has been great interest in this group of inflammatory mediators. Cytokines may be defined as soluble hormone-like proteins. However, in contrast to hormones, which are synthesized by specific endocrine tissues, cytokines are produced by a variety of cells such as immune cells, endothelial cells, and fat-storing cells. Furthermore, their synthesis is activated by a large array of stimuli including free radicals, tissue injury, and infectious agents (8,10,80).

Besides involvement in local inflammatory events, cytokines integrate systemic inflammatory events (84). A wide variety of cells, such as lymphocytes, and organs, such as the liver and the brain, are capable of responding to a number of different cytokines (30). Cytokines have the capacity to stimulate surrounding cells (paracrine), or themselves (autocrine), which may lead to further cytokine production and amplification of a particular response. Thus, the cellular source and biologic target of cytokines are not restricted to one cell-type or organ as is often the case with hormones. Cytokines can be broadly grouped according to their structure or function, into interlukins (IL), interferons (INF), tumor necrosis factor (TNF), growth factors, and chemokines (84). Cytokines are generally regarded as pro- or anti-inflammatory. Proinflammatory cytokines include interleukin-1β (IL-1β), IL-6, IL-8, and tumor necrosis factor (TNF)-α. There are also a number of anti-inflammatory cytokines whose sole purpose is to regulate this inflammatory network. Some anti-inflammatory cytokines include IL-4, IL-10, and IL-13, as well as IL-1 receptor antagonist (IL-1ra).

The cytokines central to the proposed theory of overtraining are the proinflammatory IL-1β, and TNF-α. IL-1β and TNF-α are secreted at the onset of an inflammatory cascade and act locally at the site of injury/infection; they are pleiotropic and share many overlapping actions (19). One of their numerous local functions is activation of endothelial cells of local blood vessels, which are stimulated to produce diverse cytokines. Systemically, these proinflammatory cytokines may act on the liver to regulate the synthesis of acute phase proteins, and may also act at the level of the hypothalamus, to initiate the change in the body temperature set-point and thus assist in the control of fever. There are additional multiple areas in the higher brain centers, which contain specific receptor sites for these cytokines (30). Concerning exercise and the production of IL-1 and TNF-α, there are a number of excellent reviews (5,68,82).

The other cytokine believed to be involved in OTS is IL-6. IL-6 is generally synthesized after the initial synthesis of IL-1β and TNF-α. It has been regarded as a proinflammatory cytokine, but more recently, focus has been on its anti-inflammatory effects, as it appears to play a role in the dampening of the inflammatory response (8,19). IL-6 is inducible in nearly every human cell and tissue type (8). Numerous factors are capable of stimulating IL-6 expression, including IL-1 and TNF-α. IL-6 appears to modulate both local and systemic inflammation and immunity. The magnitude of elevation of IL-6 is related to the degree of tissue injury (8). IL-6 involvement in anti-inflammatory/immune responses includes synthesis of glucocorticoids, and certain acute phase proteins that serve as potent antiproteases. It also directly inhibits expression of the pro-inflammatory cytokines IL-1β and TNF-α. In addition, it stimulates macrophage expression of IL-1ra, and soluble TNF receptor, which binds with IL-1 and TNF, truncating the response of these two pro-inflammatory cytokines (8).

IL-6 elevations have consistently been reported after intense exercise or exercise-induced muscle injury (73,81). It appears that muscle cells like myoblasts, satellite cells, and in vivo regenerating myofibers may produce IL-6 when activated in response to muscle injury (70,82).

There is minimal data concerning cytokines and OTS (36,60,75). An attempt was made by this author to induce a state of overtraining and measure blood cytokine levels (99). The exercise protocol failed to induce OTS. However, in a recent study in our laboratory, with the prime focus being on changes in the blood cytokine levels in response to exercise-induced muscle damage, preexercise cytokine values were determined for eight healthy untrained college males, and the mean values compared with one subject, inadvertently found to be suffering from chronic plantar fasciatus. IL-1β, IL-6, and TNF-α (pg·ml−1, mean ± SEM) for the eight healthy subjects were: 1.3 ± .1, 1.8 ± .14, and 1.5 ± .03, respectively. Equivalent values for the chronically injured individual were 6.4, 3.6, and 2.4 pg·mL−1, respectively, displaying cytokine levels several-fold greater than age and activity-matched controls (72).

In addition, two competitive cyclists self-reported as performing well below anticipated levels. They agreed to blood sampling and to a clinical psychological diagnostic interview. Both participants completed the Beck Depression Inventory-2 (BDI-II) (7), a widely used assessment for depression. Participant 1 scored 9, indicating extremely mild symptoms of depression. His blood cytokine levels were: IL-1β = 0.09 pg·mL−1, TNF-α = 1.5 pg·mL−1, and IL-6 = 0.7 mL·kg−1, all within the normal range of the preexercise values for healthy males. Participant 2, on the other hand, scored a 23 on the BDI-II, indicating moderate depression. Interestingly, his cytokine levels were as follows: IL-6 = 0.57 pg·mL−1 was somewhat lower than the mean for age-matched controls; IL-1β was 6.6 pg·mL−1, approximately 5 times the level of matched controls, and TNF was 4.5 pg·mL−1, approximately 3 times the normal level. These preliminary data suggest a possible interaction between psychological mood state and circulating cytokine levels, an issue that will be addressed in the following section. Pitfalls associated with interpreting data from a single subject are acknowledged.

In summary, although certain cytokines may normally be present in the circulation in small amounts, there are a variety of “emergency” circumstances, during which the pro-inflammatory, as well as additional cytokines are produced in large quantities. Local production of cytokines, for example, in injured muscle assists with the development of a local inflammatory response, subsequent healing, and termination of inflammation. At times, due to varying circumstances, increased levels of circulating cytokines will be evident. They may play a primary role in coordinating systemic inflammation, engaging the liver, and the central nervous system. It is suggested that the various signs and symptoms associated with OTS are a consequence of this systemic inflammation.

Mood, Behavior, and Cognitive Changes Associated with OTS

A consistent finding associated with the overtrained athlete is a profound change in global mood/behavior/cognition (61). This pattern varies considerably from athlete to athlete and may reflect individual heterogeneity or may, in fact, be related to the type of training (26). For example, “anaerobic” athletes may tend to experience a greater degree of anxiety/agitation, whereas endurance athletes may experience a greater degree of depression (personal correspondence, Dr. Michael Stone).

Although a reduction in performance is generally considered an initial sign of OTS, several researchers have suggested that this may be accompanied by, or even preceded, by mood, behavioral, and cognitive changes (27,64). Descriptions of these changes reflect a similar theme: a fatigued athlete, discouraged and disinterested in training, in competition, and in life in general. Although there appears to be consensus regarding psychobehavioral changes accompanying overtraining (27,64,91), it is unclear whether these changes are a consequence of intense training or precipitate overtraining (29). Morgan et al. (61) have suggested that the symptoms seen in an overtrained athlete, are remarkably similar to clinical depression (Table 1).

Although the underlying initiator(s) of these psychobehavioral changes are not known, several researchers (44,62) have implicated an increased uptake of tryptophan (TRY) by the brain, resulting in increased brain serotonin levels. Serotonin is regarded as a major contributor to mood/behavior changes. However, it is suggested here that reduced circulating levels of TRY may represent part of OTS; a more global model will now be proposed, based predominantly on a psychoneuroimmunological (PNI) model (51,57,88).

To understand how physiological changes produced by high volume training will impact on the psyche, one needs to focus on the body-mind interaction. Until recently, the field of PNI, as well as the field of exercise science, has focused on two major outflow pathways from the CNS, both activated within the hypothalamus (see Fig. 3) (57). One is the autonomic nervous system, more specifically the sympathetic nervous system, which results in elevated blood levels of catecholamines (Fig. 3, Loop A). The other pathway, the hypothalamic-pituitary-adrenal axis (HPA axis), leads to the release of cortisol by the adrenal cortex glands (Fig. 3, Loop B). What has not been stressed until recently, is the manner in which information is conveyed from the periphery into the CNS. It seems clear that the brain and peripheral immune/inflammatory cells form a bidirectional communication network (Fig. 3, Loop C). In particular, products of the immune system that are external to the CNS, communicate with the brain (30,57). Cytokines appear to be the major messenger molecules, in particular the pro-inflammatory IL-1β, IL-6, and TNF-α (8).

Figure 3:
Loop A and loop B represent the outflow of “information” from the central nervous system (CNS) to the periphery. Loop C represents the manner by which cytokines convey “information” from the periphery to the CNS (inflow) (57). Copyright © 1998 by the American Psychological Association. Reprinted with permission from: Maier, S. F., and L. R. Watkins. Cytokines for psychologists: implications for bidirectional immune-to-brain communication for understanding behavior, mood, and cognition. Psychol. Rev. 105:83–107, 1998. Schematic representation of brain-immune system connections. CRH, corticotropin releasing hormone; ACTH, adrenocortico-tropic hormone; CORT, corticosterone; NE, norepinephrine; E, epinephrine; Enk, enkephlin; SP, substance P; NPY, neuropeptide Y; GH, growth hormone; Mo, macrophages; IL1, interleukin-1; TNF, tumor necrosis factor; IL6, interleukin-6.

Activation of the CNS by these peripheral inflammatory molecules results in a constellation of behaviors referred to as “sickness,” “vegetative,” or “recuperative” (18,32,38,57). This constellation of behaviors generally includes reduced appetite, weight loss, reduced thirst, reduced libido, depression, loss of interest, fear, and sleep disturbances. These behaviors may be initiated by a wide variety of systemic immune/inflammatory conditions, such as rheumatoid arthritis, chronic fatigue syndrome, as well as in response to surgery, or to a common cold.

This constellation of sickness behaviors is believed to have been conserved throughout evolution (32,41) and is most likely a generalized adaptation to infection and injury, and may be regarded as an evolved strategy, aimed at combating infection and injury. These behaviors are therefore not regarded merely as reflexive reactions to “illness”, but rather represent a central motivational state, that assists the organism in recovery. It has been proposed that the changes that occur, may function to reduce the energy cost of behavior so that all available physiological stores can be directed to more imminent aspects of survival, such as the production of fever, the reduction of heat loss, and the activation of the immune/inflammatory systems (57). Certain behaviors, such as reduced activity, exploration, social interaction, sexual behavior, and mood, are apparent in this context. Other behaviors such as reduced feeding, do not fit as obviously, but might be secondary, for example, to the conservation of energy, since searching for food and water in more primitive settings may deplete limited energy reserves (32).

In addition to research that focuses on the development of sickness behavior, there now exists an extensive body of evidence demonstrating a relationship between systemic cytokines and psychological depression, “numerous intriguing findings…are consistent with the argument that nonspecific immune activation and cytokines are involved in the etiology or symptomology of depression.” (57). There is extensive evidence of elevated cytokines in depressed patients who exhibit significantly higher levels of IL-1β and IL-6 in culture supernatant of mitogen-stimulated monocytes, when compared with nondepressed controls (52,54,57); administration of cytokines in the absence of infection, produces a full syndrome of responses (19); and when exogenous cytokines are administered to humans, they often develop a distressed mood state (20,57). Furthermore, there appears to be a dose-dependent relationship between level of cytokines and severity of depression (51). Maes (51) refers to this as the interleukin hypothesis of depression (see Fig. 4). Both IL-1 (57) and/or IL-6 (8) appear to be involved in this cyclic process. In addition, there is evidence implicating TNF-α in depression and mood/behavioral changes (57). Thus, it appears that depressed individuals exhibit a systemic inflammatory-like condition, with elevated serum cytokines, and conversely, an injured or infected individual exhibits sickness/depressive-like behavior.

Figure 4:
The interleukin hypothesis of major depression. A schematic diagram of the association between psychological depression and the development of systemic immune/inflammation, as proposed by Maes (51). Reprinted from Prog. Neuro-Psychopharmacol. & Biol. Psychiatry 19, M. Maes. Evidence for an immune response in major depression: a review and hypothesis, pp. 11–38, 1995, with permission from Elsivier Science.

Cytokines access the CNS via several routes. They may directly access brain structures, either using a transport system to cross the blood brain-barrier, or acting at the level of circumventricular organs, where this barrier does not exist (38,77). They may also inform the CNS indirectly via activation of afferent neurons (57) of the vagus nerve; neural afferents may activate transcription and translation of cytokines within the central nervous system (18). In the brain, there are specific receptors for IL-1, IL-6, and TNF that have a discrete distribution (30). Blocking IL-1 receptors in the brain can prevent some of the sickness responses to peripheral administration of peripheral cytokines (74). Furthermore, administration of certain cytokines directly into the brain produces many or all of the sickness responses (74).

IL-1 and IL-6 receptors in the brain are abundant in the area of the hypothalamus (30,57). The binding of cytokines in the hypothalamus results in activation of the hypothalamic-pituitary-adrenal axis (HPA-axis) and sympathetic nuclei, resulting in increased levels of circulating catecholamines, and cortisol, the traditional stress hormones (21,51). Increased levels of these stress hormones have been consistently associated with mood changes (79) and with OTS (91). IL-1β and IL-6 may also result in increased activation of several discrete hypothalamic nuclei, which may account for many of the sickness-related behavioral changes, including hunger, thirst, sleep, reduced libido, and body core temperature (30,57).

Interleukin receptors, especially IL-1 receptors, are also abundant in the hippocampal area of the brain (17). The hippocampus (49) is implicated in learning, memory, and cognition (16). Thus systemic infection/inflammation may interfere with cognitive processes, such as loss of attention, and with certain types of memory (57). Alterations in cognition have been observed in overtrained individuals (27,64,66). These include: reports of a loss of coordination, the reappearance of mistakes previously corrected, an inability to concentrate at work, impaired academic ability, and changes in learning retention. Unlike the other “sickness” behaviors, this does not appear to be adaptive. Maier and Watkins (57) suggest that the hippocampus is a large structure that participates in many different functions; during illness or injury, certain neurons, usually involved in learning and memory, are diverted to other more pressing functions.

The reader has hopefully discerned overwhelming similarities between these physiological, biochemical, cognitive, and psychological/behavioral signs and symptoms experienced by clinically depressed individuals, by individuals experiencing “sickness behavior” in response to illness/injury, and by many overtrained athletes. Although at present little evidence is available to verify elevated levels of IL-1, IL-6, and/or TNF-α in OTS, results were presented in a previous section, suggesting an association between clinical depression and pro-inflammatory cytokine levels, in an athlete displaying signs and symptoms of overtraining. Research is needed to explore this postulate. If confirmed, the adoption of such an hypothesis would provide an organic, physical cause, as the basis for mood, behavioral, and cognitive changes associated with OTS (88). This approach would be consistent with new approaches used by psychoneuroimmunologists, examining the mind-body connection. “The division of disease into mental and physical could be a fundamental flaw in (the) approach…to dealing with mental illness” (88). Such an inappropriate division may have been propagated in the field of exercise physiology.

Glutamine, Hypercatabolism, and OTS

It has been proposed that intense/long-duration training may cause a marked decrease in blood levels of the amino acid glutamine (62,97). Foster and Lehman (24) reported a decrease in glutamine in overtrained runners, which persisted well into the recovery period, even after performance has begun to normalize; by comparison, there was an increase in blood glutamine in non-overtrained runners. Rowbottom et al. (75), using 10 athletes from a variety of sports, suffering from OTS, surveyed a large range of biochemical, physiological, and immunological parameters and found that glutamine was the only parameter consistently reduced. Keast (36) suggests it is unlikely that reduced levels of glutamine are the prime cause of OTS but that changes in blood glutamine levels may be indicative of some critical aspect of metabolism that is at fault and that glutamine deficit may be an excellent indicator of OTS.

Glutamine is the most abundant amino acid in human plasma and in the muscle free amino acid pool (97). Branched chain amino acids and glutamate are taken up by the muscle, and their carbon skeletons are used for de novo synthesis of glutamine, with muscle being the most abundant glutamine producing tissue. This high rate of glutamine synthesis is probably related to the fact that glutamine plays an important role in human metabolism in many organs. It is also essential for lymphocyte proliferation and macrophage function (69). Because of this latter role, it has been proposed that decreased circulating levels of glutamine is a primary factor causing a decline in immune function, frequently associated with OTS (36).

There is undoubtedly an increased need for glutamine with activation of immune/inflammatory cells. However, a number of additional associated events place increased demands on blood glutamine levels. The presence of systemic inflammation, is associated with a catabolic state (11,43,92), the degree depending on the severity and duration of the trauma/stress, driven, in part, by several cytokines and glucocorticoids (11,92). This catabolic state is adaptive and serves a variety of functions (92). Since tissue trauma is often associated with a reduced food intake, the body is now required to maintain blood glucose levels for specific organs such as the brain. The body achieves appropriate blood glucose levels by up-regulating liver gluconeogenesis. Glutamine and alanine are the primary amino acids released from the muscle, and are the most important precursors for gluconeogenesis and the preservation of blood glucose levels (97).

An additional amino-acid dependent function during systemic inflammation, is de novo synthesis of large quantities of inflammatory-related proteins by the liver, the acute phase proteins, such as C-reactive protein and haptoglobin (59). Synthesis of these proteins represents a crucial aspect of an immune/inflammatory response, helping to contain the potentially lethal amplification of inflammation. Glutamine is a primary precursor for many of these protein molecules (59) (see Fig. 5).

Figure 5:
The movement of amino acids in sepsis and trauma may also reflect what occurs during overtraining. In sepsis and traumatic injury, glutamine and other amino acids are released from skeletal muscle for uptake by tissues involved in the immune response and tissue repair, such as macrophages, lymphocytes, fibroblasts, and the liver. Nitrogen excretion as urea and NH4 + results in negative nitrogen balance. Adapted from: Marks, D. B., A. D. Marks, and C. M. Smith. Intertissue Relationships in the Metabolism of Amino Acids. In:Basic Medical Biochemistry (1st Ed.). Baltimore: Williams and Wilkins, 1996, pp. 647.

Thus, the provision of adequate amounts of amino acid to support biosynthetic pathways in the liver is crucial, with transport of amino acids into hepatocytes being a key regulatory event (59). An interplay between numerous cytokines and the classic stress hormones, redirects the flow of amino acids to the liver. Fischer and Hasselgren (23) reported that IL-6 and TNF-α work with glucocorticoids to stimulate amino acid uptake in human hepatocytes. In human hepatocytes, both alanine and glutamine transport were increased significantly by IL-6 and TNF-α treatment, compared with control.

This increased requirement for amino acids during hypermetabolism is partly satisfied by an augmentation of muscle proteolysis, the major storage pool of amino acids, and by a concomitant reduction in muscle anabolism. Accelerated muscle protein degradation would contribute to a negative nitrogen balance, and this would contribute to the loss of lean body mass (43,92). The necessary excretion of urinary nitrogen by the kidneys requires an increased urine output. This, in turn, would stimulate thirst mechanism (59). All these factors have been associated with illness/trauma and also with OTS (27,64,91).

Associated with hypercatabolism and injury/infection is a shift in fuel usage from a typically mixed glucose-fat substrate to the predominant use of fats (92). This adjustment would support the up-regulation of gluconeogenesis and the need to preserve blood glucose for specific organs. Stoner (92) suggests that if an animal survives a serious injury it may “be condemned to a period of inactivity when it is unable to forage for food…it would make sense to reduce the utilization of carbohydrate and use more fat as fuel since there is much more of it available.” This shift may also explain the finding of an increased reliance on fat metabolism during submaximal running in OTS (40), as well as account for excessive fat loss reported in some athletes (91).

In summary, it appears that low blood glutamine and other OTS-related symptoms could be explained in terms of a catabolic state related to systemic inflammation. These symptoms include elevated basal metabolic rate, negative nitrogen balance, decrease in lean body mass and fat mass, increased uric acid production, increased urination, increased thirst, and fluid intake (64).

Tryptophan and OTS

The central fatigue hypothesis of overtraining proposes an increased uptake of tryptophan (TRY) by the brain, resulting in increased brain serotonin levels (44,62). The rationale for suggesting an increased uptake of TRY is based on two assumptions. First, there is a decrease in circulating levels of TRY, suggesting an increased uptake by the CNS. Second, there is a decrease in circulating levels of branched chain amino acids (BCAA), leucine, isoleucine, and valine, which normally compete with TRY for the same amino acid carrier into the brain (85); thus, a decrease in BCAA favors the entry of TRY into the brain. In the brain, TRY is converted into the neurotransmitter serotonin. In specific areas of the brain, serotonin induces sleep, depresses motor neuron excitability and appetite, and alters autonomic and endocrine function. Since many of these behavioral changes have been seen in OTS, as well as changes in serum TRY:BCAA ratio, Newsholme et al. (62) and Kreider (44) have suggested that this may be germane to mood and behavioral changes. However, evidence of increased uptake of TRY and increased levels of serotonin, although consistently observed in animal research, is inconclusive in human research, possibly due to nonstandardized methodology (28). It also appears that many of the studies investigating BCAA and TRP in humans deal more with the acute response to intense exercise and not OTS (44,96).

The influx of TRY into the brain is certainly dependent on the TRP-BCAA ratio, but it is also dependent on additional factors, such as the free and bound plasma concentration. Normally, tryptophan circulates in the blood with a major fraction (70–90%) loosely bound to serum albumin (Alb). At the blood brain barrier transport site, Alb is stripped off and TRP passes though the brain capillaries. The availability of Alb as a carrier will influence the rate of influx of TRP into the brain. Since serum albumin concentrations are reduced during systemic inflammation (56), this will most likely reduce the availability of TRY to the CNS.

During systemic inflammation (56), an additional drain on available TRY may be due to the fact that TRY is used for leukocyte activity and synthesis of specific inflammatory-related liver proteins. Furthermore, there may be an associated induction of a major TRP-catabolizing enzyme, indoleamine 2,3 dioxygenase. Thus, reduced circulating TRY levels seen during systemic inflammation could be accounted for by a variety of events (55).

A widely held view in the psychology literature (55) is that there is correlation between circulating levels of TRY and brain levels, with low circulating levels reflecting low availability of TRY in the brain. Reduced brain TRY levels are consistently associated with depressive symptoms (55). When comparing normal volunteers with individuals experiencing major-depression, Maes et al. (56) reported a significant group difference in: 1) serum TRY levels, with levels being lower for depressed subjects, and 2) TRY:BCAA ratio, with the ratio being lower for depressed subjects, implying that both TRY and BCAA were reduced. They concluded that lowered TRY levels are related to systemic inflammatory events, evident in clinical depression (51,55,57). It is proposed here that if circulating TRY is reduced in OTS, this would reflect a scenario similar to that seen in clinical depression.

In summary, if serum TRY is reduced in OTS, and OTS does reflect systemic inflammation, then low serum TRY levels could be due to reduced availability of the TRY transporter, Alb, as well as increased usage by leukocytes, increased uptake by the liver for synthesis of liver proteins, and increased degradation. Serum TRY may prove to be a useful marker of immune/inflammatory activation in OTS, since it correlates well with certain aspects of immune changes, as well as with the presence of specific acute phase proteins (55,56).

Acute Phase Proteins, Trace Metals, and OTS

Several researchers have noted changes in various blood proteins and trace metals in OTS (27,64). These alterations could be explained by a series of events known collectively as the acute phase response (APR), which represents a crucial aspect of systemic inflammation (9,27,64,98).

Tissue trauma induces local inflammation at the site of injury, involving factors such as dilation and leakage of blood vessels, aggregation of platelets and clot formation, and accumulation of WBCs in the damaged tissue. This local response is frequently accompanied by a systemic APR. The overall purpose of the APR is to coordinate various physiological systems that will assist in dealing with inflammation; these include the development of fever, recruitment of white blood cells from various sources including bone marrow, as well as increases in systemic levels of cytokines. An integral component of the APR is de novo synthesis of specific proteins by liver hepatocytes (9), the acute phase proteins (APP). IL-1β, IL-6, and TNF-α, are primarily responsible for biosynthesis of these liver proteins, with glucocorticoids acting to enhance their action (8). The liver proteins that increase in concentration are referred to as positive APP (94).

Catabolic enzymes and reactive oxygen species released by phagocytic cells, clear disrupted host tissue in advance of repair. However, they do not discriminate between healthy and damaged cells and so aspects of inflammation can lead to destruction of healthy tissue if uncontrolled. The positive APPs represent the primary mechanism for regulating the inflammatory process (8,9,94). C-reactive protein (CRP) is a primary APP, which may increase 100–1000 fold (43). Associated with the increase in the positive APP, is a concomitant decrease in negative APP, such as albumin (27). With regard to OTS, several studies have reported increases in certain positive APP (64,93,98) and decreases in negative APP (27).

Intimately associated with the APR and synthesis of APP are changes in blood levels of trace metals (9). During an infection, plasma iron and zinc concentrations fall, whereas plasma copper levels are elevated (9). Low plasma iron and zinc have been reported in OTS (9,27,36).

In summary, it appears that the overtrained athlete shows changes in certain blood proteins and metals. These changes mimic an acute phase response, which occurs during a systemic inflammatory event, with many of these changes induced by IL-1β, IL-6, and TNF-α. Changes in positive APP and negative APP and trace metals in OTS would support the notion of systemic inflammation.

Muscle Glycogen, Blood Lactate, Insulin Resistance, and OTS

A number of researchers have reported reduced muscle glycogen levels in overtrained athletes (89). In a classic study, Costill et al. (14) had 12 swimmers more than double their training intensity for 10 d. Eight of the 12 athletes appeared to cope, whereas four developed signs of OTS; they had difficulty completing training loads, had significantly reduced muscle glycogen levels, consumed 1000 fewer kcal than were needed to match the increased energy expenditure, and failed to maintain the required carbohydrate intake. Based on these observations, Costill proposed the glycogen theory of overtraining (14), suggesting that reduced muscle glycogen would cause fatigue and result in a decrement in performance. Furthermore, the low muscle glycogen levels would result in increased uptake and oxidation of circulating branched chain amino acids (BCAA) by the muscle. This would reduce the availability of amino acids for synthesis of central neurotransmitters, resulting in changes in the nervous system, such as fatigue, which has been consistently associated with OTS.

The glycogen theory, however, has not been substantiated. Snyder (89) had cyclists increase their training load for 2 wk, to meet the criteria for short-term overtraining, but also increase carbohydrate intake sufficient to maintain muscle glycogen levels. Although subjects met the criteria for short-term overtraining, muscle glycogen levels were normal. They concluded that a mechanism or combination of mechanisms other than reduced muscle glycogen must be responsible for the development of overtraining.

Although reduced muscle glycogen might not be causal, it is frequently observed in overtrained athletes and warrants attention. It is suggested that excessive stress, including muscle-related trauma, may result in systemic inflammation, with elevated pro-inflammatory cytokines (81), manifesting the adaptive behavioral mood pattern, “sickness” behavior (18,32,38,57), discussed previously. A prominent aspect of this cluster of behaviors is anorexia (18,32,38,57,77). It is thus proposed that reduced muscle glycogen levels in OTS may be a consequence of reduced food intake, mediated by cytokine-induced anorexia.

Directly or indirectly, pro-inflammatory cytokines are clearly implicated in food intake. Cytokines may act directly on specific nuclei in the “hunger centers” of the hypothalamus to suppress food intake in a dose-dependent fashion (51,53,57). Alternatively, certain interleukins may stimulate increases in hypothalamic corticotropin releasing factor (CRF) (19,51,57,77), which suppresses appetite. There is also mounting evidence that energy and weight dysregulation may be related to IL-1β- and TNF-α-activation of the ob gene product, leptin, in white adipose tissue (76). A preliminary study, implicates leptin in overtrained distance runners (34). In addition to the putative role of cytokines on food intake, the reduced carbohydrate intake seen in overtrained swimmers (14) may be a response to conditioned taste aversions associated with IL-1 and “sickness” behavior (18,32,57).

Aside from the role of cytokines in appetite suppression, local, subacute muscle injury could interfere with transport of glucose into the muscle cell and, consequently, muscle glycogen synthesis. In response to eccentrically induced muscle damage, postexercise glycogen synthesis is impaired (15,40). Asp and colleagues (4) found a significant reduction in the glucose transporter protein, GLUT-4, 1 and 2 d after eccentrically induced muscle damage. O’Reilly et al. (65) showed that muscle glycogen stores were markedly reduced for up to 10 d after eccentric exercise. They suggested that a decreased muscle concentration of GLUT-4 protein, possibly due to down-regulation of mRNA by TNF-α (11), would result in decreased transport of glucose into the muscle, and this in turn would sustain low glycogen concentrations seen after muscle damaging eccentric exercise (4). Thus local muscle injury, per se, could contribute to reduced muscle glycogen levels associated with OTS.

In addition to the reduction in GLUT-4 protein and reduced glycogen at the level of the muscle, several investigators (3,40) have reported whole-body insulin resistance associated with muscle injury, most likely mediated by TNF-α (3). Insulin resistance has frequently been reported as part of the metabolic response to tissue trauma and systemic infection (92). Insulin resistance, to date, does not appear to have been tested in the overtrained athlete.

In summary, it is suggested that large volumes of training, systemic inflammation, and elevated levels of pro-inflammatory cytokines, directly and/or indirectly, induce anorexia, resulting in a reduced caloric intake. In addition, local muscle membrane injury and reduced availability of GLUT-4 glucose transporters in muscle cell membrane, attenuates movement of glucose into the cell for glycogen resynthesis. Both factors may contribute to reduced muscle glycogen synthesis in OTS. Although highly speculative, if overtrained athletes experience whole-body insulin resistance, this too could contribute to reduced glycogen stores. Finally, it is conjectured that reduced muscle glycogen could in turn account for the “heavy legs” (64) experienced by many overtrained athletes, as well as the reduced blood lactate levels during both submaximal and maximal exercise.

Hypothalamic-Related Hormones and OTS

The hypothalamus is a major coordinating center for neuroendocrine function (79), controlling blood levels of the stress hormones cortisol, epinephrine, and norepinephrine, as well as gonadal hormones, such as testosterone and estradiol. Generally, with an appropriate training stimulus, the hypothalamic-pituitary axes are stabilized. However, excessive physiological as well as psychological stress may lead to an altered hormonal balance; such an imbalance has been associated with OTS (6,26,27,91), although there is not complete agreement on this issue (47,95).

Cortisol is generally viewed as a catabolic hormone, whereas testosterone is anabolic (26,91). Intense, prolonged physical activity frequently leads to increased blood cortisol levels and decreased free testosterone. An alteration in the typical cortisol:testosterone ratio may be associated with the reported catabolic state in OTS (26,91). Could systemic inflammation direct these events?

During systemic inflammation, pro-inflammatory cytokines are potent activators of the hypothalamic-pituitary-adrenal axis (HPA) (77). The effects of IL-1 (35) and IL-6 (67) on the HPA axis have been studied extensively. These cytokines appear to interact with specific hypothalamic receptors, resulting in release of corticotropin releasing hormone (CRH) (35,67,77). CRH stimulates release of pituitary adrenocorticotropin releasing hormone (ACTH), with subsequent release of cortisol from the adrenal cortex. In addition to the action of cytokines at the level of the hypothalamus, IL-6 may control the release of steroid hormones by direct action on adrenal cells, and regulate adrenal synthesis of mineralocorticoids, glucocorticoids, and androgens, in a time and dose dependent fashion (67). Thus, systemic inflammation and elevated cytokines could account for elevated cortisol levels in OTS (26,27,91).

Reported decreases in testosterone and suppressed reproductive function in OTS (48,90,91) implicate the hypothalamic-pituitary-gonadal (HPG) axis. The controlling hormone in this instance is luteinizing-hormone releasing hormone (LHRH). LHRH controls the pulsatile release of the pituitary gonadal hormones, luteinizing hormone (LH), and follicle stimulating hormone (FSH), which in turn induce the release of ovarian estradiol, and testicular testosterone (58). In reference to cytokines and reproductive function, these inflammatory mediators suppress reproductive function via inhibition of LHRH (58,77,83).

In summary, hypothalamic-related hormonal systems appear to be altered in OTS, although a clear pattern has not emerged. However, there is extensive information demonstrating an interaction between systemic cytokines and the HPA and HPG axes. Thus, inflammatory cytokines may account for alterations in reproductive hormones in OTS.

Immune System and OTS

Although not universally accepted (33), anecdotal evidence suggests an increased incidence of illness associated with OTS (27,36,60,64,68). These include an increased susceptibility to, and severity of colds, and allergies, flu-like illness, slow healing of minor scratches, swelling of lymph glands, reactivation of herpes viral infections, headaches, and gastrointestinal disturbances (see Table 1).

Reasons for the high incidence of illness in OTS are unclear (27,36,60,64,68,81). Intuitively, these conditions are most likely related to impairment of the immune system. Although immune function appears to be enhanced in response to moderate exercise, intense exercise, even one bout, such as a marathon, might result in immune suppression (63). Since overtraining is associated with repetitive bouts of high intensity/volume training, and competing, often in the absence of adequate rest, it is not unreasonable to assume a compromised immune system, although much remains to be learned concerning the influence of overtraining on the immune system (36,60).

A model that may be relevant to understanding a compromised immune system in the overtrained athlete is the model adopted to explain the high susceptibility to infection, postsurgery/injury (8,22). Immediately postsurgery/injury, inflammation is dramatically up-regulated so as to mobilize cellular and humoral immune mechanisms (8). Frequently, this early inflammation is hyperinflammatory. As stated earlier, considerable anti-inflammatory factors are associated with the up-regulation of inflammation. Anti-inflammatory factors are expressed in various “forms” and have a variety of targets. Interleukin-1 receptor antagonist (IL-1ra) acts specifically to block the action of IL-1 (20); a variety of soluble serum receptors, such as TNF-receptors, bind and thus limit cytokine activity (31); hormones, specifically cortisol, have profound anti-inflammatory action (22,23); and finally, augmented expression of liver acute phase proteins, such as C-reactive protein, serve as potent anti-inflammatory agents (42). Although these anti-inflammatory effects are necessary to counteract the pro-inflammatory effects, the ultimate result of prolonged, intense, counterregulation is immunosuppression (8). In reference to trauma patients, Biffl and colleagues (8) have suggested that it is paradoxical that “the hyper-inflammatory response may predispose the individual to the subsequent development of immuno-suppression” (see Fig. 6) (8).

Figure 6:
A model of immunosuppression, proposed by Biffle et al. (8). The physiologic response to injury involves an early hyper-inflammatory response, which is accompanied by a degree of compensatory anti-inflammatory effect. If the early inflammatory response is excessive, an appropriate persistence of anti-inflammatory compensation will result in later immunosuppression. Reprinted with permission from: Biffl, W. L., E. E. Moore, F. A. Moore, and V. M. Peterson. Interleukin-6 in the injured patient. Marker of injury or mediator of inflammation? Ann. Surg. 224:647–664, 1996.

A model of early hyperinflammation followed by late immunosuppression may be applicable to understanding the immune response of the overtrained athlete (5). Possibly by the time true overtraining has manifested itself, the athlete has been exposed to pro-inflammatory cytokines, with associated counterregulatory anti-inflammatory factors, for an extended period. Thus, immunosuppression may reflect the body’s highly developed attempt to contain inflammation through the production of endogenous anti-inflammatory molecules (19).

Theoretical Implications

The stress theory, developed by Selye (79), was based on the observation that a wide variety of diseases manifest themselves in a similar physiological fashion, with extensive involvement of the hypothalamic-pituitary-adrenal axis. The disease process may progress through three stages, each stage being characterized by a cluster of associated symptoms. Selye refers to this as the General Adaptation Syndrome (GAS), with the three stages being the alarm, resistance, and exhaustion stage. The initial two phases are considered adaptive and are implicated in adjustments to a wide variety of psychological and physiological stresses. However, Selye suggests that the final phase of exhaustion represents a breakdown of the adaptive capacity; he reasons that the organism possesses a limited amount of adaptive energy and stage 3 represents depletion of these reserves.

Several researchers have suggested that OTS is a manifestation of the exhaustion stage of the GAS (24,37,91), most likely due to excessive physical/physiological stress related to intense training, with psychological stressors being additive. It is proposed here that this third stage represents a generalized response to excessive stress (GRES) and that this final stage also be viewed as adaptive, but in a more profound sense, with the focus not being on “improvement” but more specifically on recovery/survival per se, with the aim being to regain the homeostatic condition of “wellness.”

It is further proposed that GRES be viewed as a self perpetuating cycle, consisting of a primary stimulus (muscle-related trauma, psychological stress, and/or a viral infection) that would result in the activation of circulating monocytes and the biosynthesis of pro-inflammatory cytokines. These cytokines would in turn coordinate the whole body response, including the CNS, the liver, and the immune system, attempting to negate the effects of the stressor. If the stressor persists, then the cyclic, mind-body response will continue (see Fig. 7). Withdrawal of the egregious stimulus would probably be the most appropriate means for terminating this cycle. Withdrawal in this instance implies rest. It is ironic that despite the wonders of modern medicine, rest may be the most potent healing agent, universally recommended by coaches and exercise physiologists. If, however, OTS proves to be a form of systemic inflammation, drug therapy such as nonsteroidal anti-inflammatories (22,38), anti-depressants, and anti-cytokine drugs (19), as well as dietary factors (88), may prove useful adjuncts.

Figure 7:
Cytokine theory of overtraining: proposed events leading to, and sustaining the overtraining syndrome (Schematic diagram under “IMMUNE CELLS” is reprinted with permission from: Biffl, W. L., E. E. Moore, F. A. Moore, and V. M. Peterson. Interleukin-6 in the injured patient. Marker of injury or mediator of inflammation? Ann. Surg. 224:647–664, 1996).

Finally, if this hypothesis proves viable, it will be important to guard against overdiagnosing OTS. Selye (79) has emphasized that the different stages of the general adaptation syndrome are represented by clusters of several symptoms and is not represented by one or two manifestations. It might be important to apply similar thinking to OTS. If an athlete develops an upper respiratory tract infection (URTI), as is frequently the case after a marathon (63), this single event should not be interpreted as overtraining. If however, an URTI occurs in association with a array of symptoms, such as changes in sleep patterns, reduced appetite, lethargy, and depression, this might then be suggestive of OTS. It is hoped that diagnostic criteria will eventually be established.


It is suggested that the overtraining syndrome is a response to excessive musculoskeletal stress, associated with insufficient rest and recovery, which may induce a local acute inflammatory response that may evolve into chronic inflammation and produce systemic inflammation. Part of systemic inflammation involves activation of circulating monocytes, which may synthesize large quantities of pro-inflammatory cytokines, IL-1β, IL-6, and TNF-α. The cytokines act on the CNS and induce a cluster of motivated behaviors, commonly referred to as “sickness” behavior (reduced appetite, depression, etc.), which is conducive to healing/recuperation. The cytokines also activate the sympathetic nervous system and hypothalamic-pituitary-adrenal axis, while suppressing activity of hypothalamic-pituitary-gonadal axis, thus accounting for changes in blood levels of catecholamines, glucocorticoids, and gonadal hormones. Pro-inflammatory cytokines also up-regulate liver function, to maintain blood glucose levels (gluconeogenesis), and to synthesize inflammatory-related acute phase proteins. Immune-related changes may be related to an immuno-suppression, possibly due to anti-inflammatory factors that accompany a pro-inflammatory response, that occurs in response to tissue trauma.

Thus, if OTS is viewed under the rubric of systemic inflammation, it is possible to reconcile a variety of previously proposed mechanisms. It is hoped that future research pertaining to OTS, will examine the role of systemic inflammatory markers to test this hypothesis.

This manuscript was supported by a grant from The Procter & Gamble Company.

Thanks to Dr. Joseph Houmard, East Carolina University, for editorial suggestions. Thank you to Alta Bender, Appalachian State University, for assistance in developing the graphics. Thank you to Denise Martz-Ludwig, Ph.D. (Psychology), Appalachian State University, for administering psychological assessments. I would like to thank the following graduate students from Appalachian State University, for assistance in preparation of this manuscript: Max Shute, Mark Lehmkeul, and Elizabeth Hogen.


1. Ala, Y., O. Palluy, and J. Favero. Hypoxia/reoxygenation stimulates endothelial cells to promote interleukin-1 and interleukin-6 production: effects of free radical scavengers. Agents Actions 37:134–139, 1992.
2. Almeida, S. A., K. M. Williams, R. A. Shaffer, and S. K. Brodine. Epidemiological patterns of musculoskeletal injuries and physical training. Med. Sci. Sports Exerc. 31:1176–1182, 1999.
3. Asp, S., J. R. Daugaard, S. Kristiansen, B. Kiens, and E. A. Richter. Eccentric exercise decreases maximal insulin action in humans: muscle and systemic effects. J. Physiol. 494:891–898, 1996.
4. Asp, S., J. R. Daugaard, and E. A. Richter. Eccentric exercise decreases glucose transporter GLUT4 protein in human skeletal muscle. J. Physiol. 482:705–712, 1995.
5. Bagby, G., J., C. D. Larry, and R. E. Shepherd. Exercise and cytokines: spontaneous and elicited resosnes. In:Exercise and Immune Function, L. Hoffman-Goetz (Ed.). Boca Rota, FL: CRC Press, 1996, pp. 55–79.
6. Barron, J. L., T. D. Noakes, W. Levy, C. Smith, and R. P. Millar. Hypothalamic dysfunction in overtrained athletes. J. Endocrinol. Metab. 60:803–806, 1985.
7. Beck, A. T., R. A. Steer, and G. K. Brown. BDI-II Manual. San Antonio: Harcourt Brace & Company, 1996, pp. 38.
8. Biffl, W. L., E. E. Moore, F. A. Moore, and V. M. Peterson. Interleukin-6 in the injured patient: marker of injury or mediator of inflammation? Ann. Surg. 224:647–664, 1996.
9. Cannon, J. G. Exercise and the acute phase response. In: Exercise and Immune Function, L. Hoffman-Goetz (Ed.). Boca Raton, FL: CRC Press, 1996, pp. 39–55.
10. Cavaillon, J. M. Cytokines and macrophages. Biomed. Pharmacother. 48:445–453, 1994.
11. Chang, H. R., and B. Bistrain. The role of cytokines in the catabolic consequences of infection and injury. J. Parent. Ent. Nutr. 22:156–166, 1998.
12. Child, R. B., D. M. Wilkinson, J. L. Fallowfield, and A. E. Donnelly. Elevated serum antioxidant capacity and plasma malondialdehyde concentration in response to a simulated half-marathon run. Med. Sci. Sport Exerc. 30:1603–1607, 1998.
13. Clarkson, P. M., and I. Tremblay. Exercise-induced muscle damage, repair, and adaptations in humans. J. Appl. Physiol. 65:1–6, 1988.
14. Costill, D. L., M. G. Flynn, J. P. Kirwan, et al. Effects of repeated days of intensified training on muscle glycogen and swimming performance. Med. Sci. Sports Exerc. 20:249–254, 1988.
15. Costill, D. L., D. R. Pearson, and W. J. Fink. Impaired muscle glycogen resynthesis after eccentric exercise. J. Appl. Physiol. 69:46–50, 1990.
16. Cunningham, A. J., C. A. Murray, L. A. J. O’Neill, M. A. Lynch, and J. J. O’Connor. Interleukin-1b (IL-1b) and tumor necrosis factor (TNF) inhibit long-term potentiation in the rat dentate gyrus in vitro. Neurosci. Lett. 203:17–20, 1996.
17. Cunningham, E. T., and E. B. de Souza. Interleukin-1 receptors in the brain and endocrine tissue. Immunol. Today 14:171–176, 1993.
18. Dantzer, R., R. Bluthe, S. Kent, and G. Goodall. Behavioral effects of cytokines: an insight in mechanisms of sickness behavior. In: Neurobiology of Cytokines, E. G. DeSouza (Ed.). San Diego: Academic Press, 1993, pp. 130–151.
19. Dinarello, C. Role of pro- and anti-inflammatory cytokines during inflammation: experimental and clinical findings. J. Biol. Regul. Homeost. Agents 11:91–103, 1997.
20. Dinarello, C. A., and R. C. Thompson. Blocking IL-1: interleukin-1 receptor antagonist in vivo and in vitro. Immunol. Today 12:404–410, 1991.
21. Dunn, A. J. Interactions between the nervous system and the immune system: implications for psychopharmacology. In: Psychopharmacology: The Fourth Generation of Progress, F. E. Bloom and D. J. Kupfer (Eds.). New York: Raven Press, 1995, pp. 719–733.
22. Faist, E., C. Schinkel, and S. Zimmer. Update on the mechanisms of immune suppression of injury and immune modulation. World J. Surg. 20:454–459, 1996.
23. Fischer, J., and P.-O. Hasselgren. Cytokines and glucocorticoids in the regulation of the “hepato-skeletal muscle axis” in sepsis. Am. J. Surg. 161:266–271, 1991.
24. Foster, C., and M. Lehman. Overtraining syndrome. In: Running Injuries, G. N. Guten (Ed.). Philadelphia: W.B. Saunders Company, 1997, pp. 173–188.
25. Fry, A. C. The role of training intensity in resistance exercise-overtraining and overreaching. In: Overtraining in Sport, R. B. Kreider, A. C. Fry, and M. L. O’Toole (Eds.). Champaign, IL: Human Kinetics, 1998, pp. 107–130.
26. Fry, A. C., and W. J. Kraemer. Resistance exercise overtraining and overreaching. Sports Med. 23:106–129, 1997.
27. Fry, R. W., A. R. Norton, and D. Keast. Overtraining in athletes: an update. Sports Med. 12:32–65, 1991.
28. Gastmann, U. A. L., and M. J. Lehmann. Overtraining and the BCAA hypothesis. Med. Sci. Sports Exerc. 30:1173–1178, 1998.
29. Gross, J. D. Hardiness and mood disturbances in swimmers while overtraining. J. Sport Exerc. Psychol. 16:135–149, 1994.
30. Haas, H. S. and K. Schauenstein. Neuroimmumodulation via limbic structures: the neuroanatomy of psychoimmunology. Prog. Neurobiol. 51:195–222, 1996.
31. Hamblin, A. S. Cytokines and Cytokine Receptors. In: In Focus, D. Rickwood and D. Male (Eds.). New York: Oxford University Press Inc., 1993, pp. 1–19.
32. Hart, B. L. Biological basis of the behavior of sick animals. Neurosci. Biobehav. Rev. 12:123–137, 1988.
33. Holger, H., W. Gabriel, A. Urhausen, G. Valet, U. Heidelbach, and W. Kinderman. Overtraining and immune system: a prospective longitudinal study in endurance athletes. Med. Sci. Sports Exerc. 30:1151–1157, 1998.
34. Ishigaki, T., K. Koyama, M. Takemura, et al. Why does leptin increase? Med. Sci. Sports Exerc. 30:S185, 1998.
35. Kalra, P. S., A. Sahu, and S. P. Kalra. Interleukin-1 inhibits the ovarian steroid-induced luteinizing hormone surge and release of hypothalamic luteinizing hormone-releasing hormone in rats. Endocrinology 126:2145–2152, 1990.
36. Keast, D. Immune responses to overtraining and fatigue. In: Exercise and Immune Function, L. Hoffman-Goetz (Ed.). Boca Rotan, FL: CRC Press, 1996, pp. 121–141.
37. Keizer, H. A. Neuroendocrine aspects of overtraining. In: Overtraining in Sport, R. B. Kreider, A. C. Fry, and M. L. O’Toole (Eds.). Champaign, IL: Human Kinetics, 1998, pp. 145–168.
38. Kent, S., R. Bluthe, K. W. Kelley, and R. Dantzer. Sickness behavior as a new target for drug development. TiPS 13:24–28, 1992.
39. Kibler, W. B., and T. J. Chandler. Musculoskeletal and orthopedic considerations. In: Overtraining in Sport, R. B. Kreider, A. C. Fry, and M. L. O’Toole (Eds.). Champaign, IL: Human Kinetics, 1998, pp. 169–192.
40. Kirwan, J. P., R. C. Hickner, K. E. Yarasheski, W. M. Kohrt, B. V. Wiethop, and J. O. Holloszy. Eccentric exercise induces transient insulin resistance in healthy individuals. J. Appl. Physiol. 72:2197–2202, 1992.
41. Kluger, M. J. The evolution and adaptive value of fever. Am. Sci. 66:38–43, 1978.
42. Koj, A. Metabolic studies of acute phase proteins. In: Pathophysiology of Plasma Protein Metabolism, G. Mariani (Ed.). London: McMillan, 1983, pp. 221–248.
43. Kotani, G., M. Usami, H. Kasahara, and Y. Saitoh. The relationship of IL-6 to hormonal mediators, fuel utilization, and systemic hypermetabolism after surgical 1 trauma. Kobe J. Med. Sci. 47:187–205, 1996.
44. Kreider, R. B. Central fatigue hypothesis and overtraining. In: Overtraining in Sport, R. B. Kreider, A. C. Fry, and M. L. O’Toole (Eds.). Champaign, IL: Human Kinetics, 1998, pp. 309–334.
45. Kuipers, H. Training and overtraining: an introduction. Med. Sci. Sports Exerc. 30:1137–1139, 1998.
46. Kunkel, S. L. Th1- and Th2-type cytokines regulate chemokine expression. Biol. Signals 5:197–202, 1996.
47. Lehmann, M., C. Foster, H. H. Dickhuth, and U. Gastman. Autonomic imbalance hypothesis and overtraining syndrome. Med. Sci. Sports Exerc. 30:1140–1145, 1998.
48. Lehmann, M., C. Foster, N. Netzer, et al. Physiological response to short- and long-term overtraining in endurance athletes. In: Overtraining in Sport, R. B. Kreider, A. C. Fry, and M. L. O’Toole (Eds.). Champaign, IL: Human Kinetics, 1998, pp. 19–46.
49. Linthorst, A. C., C. Flachskamm, F. Holsboer, and J. M. H. M. Reul. Local administration of recombinant human interleukin-1b in the rat hippocampus increases serotonergic neurotransmission, hypothalamic-pituitary-adrenocorticol axis activity, and body temperature. Endocrinology 135:520–532, 1994.
50. Macintyre, D. L., W. D. Reid, and D. C. McKenzie. The inflammatory response to muscle injury and its clinical implications. Sports Med. 20:24–40, 1995.
51. Maes, M. Evidence for an immune response in major depression: a review and hypothesis. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 19:11–38, 1995.
52. Maes, M., E. Bosmans, E. Suy, C. Vandervorst, C. Dejonckheere, B. Minner, and J. Raus. Depression-related disturbances in mitogen-induced lymphocyte responses, interleukin-1-beta, and soluble interleukin-2-receptor production. Acta. Psychiatr. Scand. 84:379–386, 1991.
53. Maes, M., M. Y. Meltzer, S. Scharpe, et al. Psychomotor retardation, anorexia, weight loss, sleep disturbances, and loss of energy: psychopathological correlates of hyperhaptoglobinemia during major depression. Psychiatr. Res. 47:229–241, 1993.
54. Maes, M., S. Scharpe, H. Y. B. Meltzer, et al. Relationships between interleukin-6 activity, acute phase proteins and HPA-axis function in severe depression. Psychiatr. Res. 49:11–27, 1993.
55. Maes, M., R. Verkerk, E. Vandoolaeghe, et al. Serotonin-immune interactions in major depression: lower serum tryptophan as a marker of immune-inflammatory response. Eur. Arch. Psychiatry Clin. Neurosci. 247:154–161, 1997.
56. Maes, M., A. Wauters, R. Verkerk, et al. Lower serum L-tryptophan availability in depression as a marker of a more generalized disorder in protein metabolism. Neuropsychopharmacology 15:243–251, 1996.
57. Maier, S. F. and L. R. Watkins. Cytokines for psychologists: implications for bidirectional immune-to-brain communication for understanding behavior, mood, and cognition. Psychol. Rev. 105:83–107, 1998.
58. Marchetti, B., F. Gallo, Z. Farinella, C. Tirolo, N. Testa, C. Romeo, and M. C. Morale. Luteinizing hormone-releasing hormone is a primary signaling molecule in the neuroimmune network. Ann. N. Y. Acad. Sci. 840:205–248, 1998.
59. Marks, D. B., A. D. Marks, and C. M. Smith. Intertissue relationships in the metabolism of amino acids. In:Basic Medical Biochemistry. (1st Ed.). Baltimore: Williams and Wilkins, 1996, pp. 647–666.
60. McKinnon Traeger, L. Effects of overreaching and overtraining on immune function. In: Overtraining in Sport, R. B. Kreider, A. C. Fry, and M. L. O’Toole (Eds.). Champaign, IL: Human Kinetics, 1998, pp. 219–242.
61. Morgan, W. P., D. R. Brown, J. S. Raglin, P. J. O’Connor, and K. A. Ellikson. Psychological monitoring of overtraining and staleness. Br. J. Sports Med. 21:107–114, 1987.
62. Newsholme, E. A., M. Parry-Billings, N. McAndrew, and R. Budgett. A biochemical mechanism to explain some characteristics of overtraining. In: Advances in Nutrition and Sport, F. Brouns (Ed.). Basel: Karger, 1991, pp. 79–83.
63. Nieman, D. C. Exercise, infection, and immunity. Int. J. Sports Med. 15:S131–S141, 1994.
64. Noakes, T. Lore of Running. Champaign, IL: Human Kinetics Publishers, Inc., 1991, pp. 408–425.
65. O’Reilly, K. P., M. J. Warhol, R. A. Fielding, W. R. Frontera, C. N. Meredith, and W. J. Evans. Eccentric exercise-induced muscle damage impairs muscle glycogen repletion. J. Appl. Physiol. 63:252–256, 1987.
66. O’Toole, M. Overreaching and overtraining in endurance athletes. In: Overtraining in Sport. R. B. Kreider, A. C. Fry, and M. L. O’Toole (Eds.). Champaign, IL: Human Kinetics, 1998, pp. 3–18.
67. Path, G., S. R. Bornstein, M. Ehrhart-Bornstein, and W. A. Scherbaum. Interleukin-6 and the interleukin-6 receptor in the human adrenal gland: expression and effects on steroidogenesis. J. Clin. Endocrinol. Metabol. 82:2343–2349, 1997.
68. Pedersen, B. K. Exercise and infection. In: Exercise Immunology, B. K. Pedersen (Ed.). New York: Chapman & Hall, 1997, pp. 113–122.
69. Pedersen, B. K. and T. Rohde. Exercise, glutamine and the immune system. In: Exercise Immunology, B. K. Pedersen (Ed.). New York: Chapman & Hall, 1997, pp. 75–88.
70. Pedersen, B. K., T. Rohde, and H. Bruunsgaard. Exercise and Cytokines. In: Exercise Immunology, B. K. Pedersen (Ed.). New York: Chapman & Hall, 1997, pp. 89–112.
71. Perry, J. D. Exercise, injury and chronic inflammatory lesions. Br. Med. Bull. 48:668–682, 1992.
72. Rananto, C., E. Hogen, K. Person, et al. Elevated serum cytokines associated with plantar fasciitis. Med. Sci. Sport Exerc. 31:S60, 1999.
73. Rohde, T., D. A. MacLean, E. A. Richter, B. Kiens, and B. K. Pedersen. Prolonged submaximal eccentric exercise is associated with increased levels of plasma IL-6. Am. J. Physiol. 273:E85–91, 1997.
74. Rothwell, N. J. and S. J. Hopkins. Cytokines and the nervous system: II. Actions and mechanisms of action. Trends Neurosci. 18:130–136, 1995.
75. Rowbottom, D., D. Keast, C. Goodman, and A. R. Mortan. Glutamine and the overtraining syndrome. Eur. J. Physiol. 70:502, 1995.
76. Sarraf, P., R. C. Frederich, E. M. Turner, et al. Multiple cytokines and acute inflammation raise mouse leptin levels: potential role in inflammatory anorexia. J. Exp. Med. 185:171–175, 1997.
77. Schobitz, B., J. M. H. M. Reul, and F. Holsboer. The role of the hypothalamic-pituitary-adrenocorticol system during inflammatory conditions. Crit. Rev. Neurobiol. 8:263–291, 1994.
78. Seene, T., M. Umnova, and P. Kaasik. The exercise myopathy. In: Overload, Performance Incompetence and Regeneration in Sport, M. Lehmann, C. Foster, U. Gastmann, H. Keizer, and J. Steinacker (Eds.). New York: Kluwer Academic/Plenum Publishers, 1999, pp. 119–130.
79. Selye, H. The Stress of Life. London: Longmans, Green and Co., 1950, pp. 52–67.
80. Sen, C. K. and L. Packer. Antioxidant and redox regulation of gene transcription. FASEB J. 10:709–720, 1996.
81. Shephard, R. J. and P. N. Shek. Acute and chronic over-exertion: do depressed immune responses provide useful markers? Int. J. Sports Med. 19:159–171, 1998.
82. Shepherd, R. J. Physical Activity, Training, and the Immune Response. Carmel, In: Cooper Publishing Group, 1997.
83. Simon, C. and M. L. Polan. Cytokines and reproduction. West J. Med. 160:425–429, 1994.
84. Simpson, K. J., N. W. Lukas, L. Colletti, R. M. Strieter, and S. K. Kunkel. Cytokines and the liver. J. Hepat. 27:1120–1132, 1997.
85. Skeie, B., V. Kvetan, K. M. Gil, et al. Branch-chain amino acids: their metabolism and clinical utility. Crit. Care Med. 18:549–571, 1990.
86. Smith, L. L. Acute inflammation: the underlying mechanism in delayed onset muscle soreness? Med. Sci. Sports Exerc. 23:542–551, 1991.
87. Smith, L. L. and M. Miles. Exercise-induced muscle injury and inflammation. In:Applied Sports Science, W. E. Garret, and D. T. Kirkendall (Eds.). Media, PA: Williams & Wilkins, 1999, in press.
88. Smith, R. S. The macrophage theory of depression. Med. Hypoth. 35:298–306, 1991.
89. Snyder, A. Overtraining and glycogen depletion hypothesis. Med. Sci. Sports Exerc. 30:1146–1150, 1998.
90. Stone, M. H. and A. C. Fry. Increased training volume in strength/power athletes. In: Overtraining in Sport, R. B. Kreider, A. C. Fry, and M. L. O’Toole (Eds.). Champaign, IL: Human Kinetics, 1998, pp. 87–106.
91. Stone, M. H., R. E. Keith, J. T. Kearney, S. J. Fleck, G. D. Wilson, and N. T. Triplett. Overtraining: a review of the signs, symptoms and possible causes. J. Appl. Sport Sci. Res. 5:35–50, 1991.
92. Stoner, H. B. Metabolism after trauma and in sepsis. Circ. Shock 19:75–87, 1986.
93. Strachan, A. F., T. D. Noakes, G. Kotsenberg, A. E. Nel, and F. C. DeBeer. C-reactive protein levels during long-distance running. Br. Med. J. 289:1249–1251, 1984.
94. Tizard, I. R. Immunology. Philadelphia: Saunders, 1997, pp. 446–449.
95. Urhausen, A., H. H. W. Gabriel, and W. Kinderman. Impaired pituitary hormonal response to exhaustive exercise in overtrained endurance athletes. Med. Sci. Sports Exerc. 30:447–448, 1998.
96. Uwe, A., L. Gastman, and M. J. Lehmann. Overtraining and the BCAA hypothesis. Med. Sci. Sports Exerc. 30:1173–1178, 1998.
97. Wagenmakers, J. M. Muscle amino acid metabolism at rest and during exercise: role in human physiology and metabolism. In: Exercise and Sport Sciences Reviews, J. O. Holloszy (Ed.). Baltimore: Williams and Wilkins, 1998, pp. 287–314.
98. Weight, L. M., D. Alexander, and P. Jacobs. Strenuous exercise: analogous to the acute-phase response? Clin. Sci. 81:677–683, 1991.
99. Wilson, J., L. L. Smith, D. Holbert, A. Anwar, and J. A. Houmard. The effect of combined versus eccentric training on markers of overstress. Med. Sci. Sports Exerc. 29:S52, 1997.


©2000The American College of Sports Medicine