Urhausen, A., Gabriel, H., & Kindermann, W. (1995). Blood hormones as markers of training stress and overtraining. Sports Medicine, 20, 251-276.

This excellent review covers most features analyzed in blood when attempts are made to associate biochemical factors and overtraining.

A distinction is made between overreaching, short-term accumulated fatigue that can be erased by longer than normal rest periods, and overtraining, a result of continued exposure to excessive training stimuli without adequate recovery. Overtraining is generally characterized by a decrease in performance when training load is maintained or increased, enhanced fatiguability, disruptions to sleep, rest, and social behaviors, and complaints of poor well-being.

Performance Testing

Using performance testing to diagnose overtraining is difficult because of the problem with standardizing procedures, the rarity of valid sport-specific tests, and the lack of generalization between laboratory test and field test performances. Laboratory performance factors associated with overtraining are a reduction in short-term endurance and maximal anaerobic lactic capacity. These can only be tested when performance tests are maximal, something overtrained athletes are hesitant to perform. The level of exertion that should be sought in performance tests is at least 110% of anaerobic threshold, or 10% above maximum lactate steady state.

Hormonal Testing

Periodic hormonal assessments are difficult because of the potential to confuse readings with possible causes. Generally, most changes in blood parameters are caused by incidental exercise fatigue, such as that generated by a hard training session, and by psychological stress. Altered hormonal levels recover with adequate rest or stress removal and/or coping. It is incorrect to assume that these transient changes will become more permanent with overtraining.

Catecholamines

Plasma levels of free catecholamines indicate sympatho-adrenergic activation due to exercise. Catecholamines modulate metabolic and circulatory reactions and adaptations to physical and psychological stresses.

Catecholamine levels, commonly epinephrine and norepinephrine, are associated with the duration and intensity of exercise. However, a relationship with overtraining is not clear. It appears that the type of overtraining (sympathetic or parasympathetic dominant) will alter the response and habitual status of catecholamines. Because of this confusion, it is unlikely that this could be a universal field-sensitive measure.

Catecholamines and Psychological Stress

Exercise forms with high nervous stress components (e.g., sprints, high lactate activities) are probably more likely to induce overtraining syndrome, especially if adequate periods of recovery are neglected. The hypothesis that adrenaline reflects mental stress and noradrenaline reflects physical stress is not supported by modern research.

Implication. The measurement of free plasma catecholamines in the diagnosis of overtraining is unclear and usually unreliable because of measurement problems due to the lack of agreement between different assays. Essentially, this means that there is no one reliable measure of this index and that it has yet to be shown to be a marker of overtraining in any consistent manner.

Testosterone and Cortisol

Depending on the intensity and duration of physical work, hormones with anabolic or catabolic properties, such as testosterone and cortisol, show changes signaling a catabolic state which reverses with rest. This has led to this being considered a potential marker for recognizing overtraining.

Testosterone. Several relationships have been investigated.

1. Findings with strength athletes. An increase in strength is not necessarily associated with a simultaneous increase in the testosterone/cortisol ratio.

2. Findings with endurance athletes. Only a few studies are available relating these factors. At best, one could tentatively assert that testosterone increases with adaptive training and decreases with overtraining. However, this is partly conjecture since no studies have been conducted with overtrained athletes.

3. Findings with female athletes. There is no consistent relationship between these hormones and extended physical activity in females, primarily because insufficient research work has been completed.

4. Overtraining studies. Resting levels of testosterone, cortisol, and sex-hormone-binding globulin (SHBG) are not altered with overtraining nor are they correlated with performance in overtrained states. It is probable that the behavior of free testosterone and cortisol is a physiological indicator of training load stress rather than overtraining. The testosterone/cortisol ratio is occasionally overinterpreted.

   In most studies, SHBG showed no changes during overtraining. Total testosterone mainly parallels changes of free testosterone.

5. During and after training. Serum testosterone shows a two-phase behavior in exercise: after short-duration stimulation it increases in relation to intensity, volume, and muscle mass. On the other hand, after extended duration exercise (e.g., three hours) it decreases.

Testosterone levels are decreased by endurance training or work during overtraining. To what extent decreased testosterone levels influence energy-supply cannot be definitely answered at this time.

Cortisol. The behavior of cortisol during overtraining has been inconclusive in the literature. There is no consistent finding. However, the type of training (aerobic or anaerobic) leads to different hormonal adaptations. An exercise-induced cortisol increase depends upon the duration and intensity of exercise. Short intense activity causes the change. After exercise, it decreases rapidly and will be "normal" after a few hours.

Summary. Although testosterone and cortisol are the most frequently investigated hormones as a measure of overtraining, hormonal monitoring of overtraining does not yet seem to be justified.

Pituitary Hormones

Gonadotrophins (Luteinising hormone - LH and follicle-stimulating hormone). It is uncertain whether isolated measurements of hormonal levels in peripheral blood are sensitive enough to investigate these complex mechanisms.

Corticotrophin (ACTH). Few studies relating this hormone to overtraining are available. Corticotrophin secretion depends on the intensity and duration of exercise and is stimulated by different stresses, particularly blood glucose depletion. There is a hint that in an overtrained state, an increase in ACTH during exercise might be impaired.

Somatotrophic hormone (STH). The most common factor in this hormonal group, insulin-like growth factor 1 (IGF-1), and its function in exercise, is not clearly understood and controversially reported.

No pattern of association with overtraining has been established. Conflicting and confounded reports do not provide support for the logical postulation of Beta-Endorphin being associated with responses to exercise or overtraining.

Prolactin (PRL). The response of PRL to exercise reveals large individual differences. There are no grounds for using it to monitor training or diagnose overtraining.

Other Hormones

Insulin. Hardly any well-founded data with respect to athletes in intensive training periods or in a state of overtraining exist for the other hormones. Even with insulin, which theoretically should demonstrate a lowered maximal exercise-induced level independent of blood glucose or exercise duration, results have been conflicting.

Thyroid-stimulating hormones and estradiol/progesterone. No findings that would suggest an applicability in the monitoring of training or diagnosis of overtraining exist at present.

Special Conditions

Diet. After many months of low-fat high-CHO diets, ice-hockey players showed increased free testosterone, cortisol, and LH in comparison to controls. Similar changes are not evidenced with short-term diets even when performances change. An inadequate caloric intake or eating disorders, impair central hormonal regulations including a reduced pulsatile LH secretion. An imbalance in caloric intake decreases the metabolic clearance of steroid hormones.

Hormonal drugs. One of the major reasons for using testosterone and anabolic steroids, growth hormone, clenbuterol, erythropoeitin, and releasing factors during training seems to be the supposed prevention or treatment of overtraining. After withdrawal of sustained steroid use, pituitary function can be affected leading to long-lasting impairment of testicular endocrine function. The concentration of growth hormone and estradiol are reported to increase after the administration of testosterone or aromatisable anabolic-androgenic steroids in men.

Altitude. Hormonal responses at altitude are modified by the degree of hypoxia, climate, duration of exposure and stage of acclimatization, individual experience with altitude training, plasma volume changes as well as reduced absolute exercise intensity. However, if these factors are controlled and relative workloads are established, acute moderate hypoxia does not seem to affect metabolic and hormonal responses to short bouts of exercise.

Pathophysiological Considerations

Energy supply in recovery. The impaired secretion of centrally or peripherally acting hormones during overtraining contrasts with the usually increased hormonal levels induced by training. A decreased exercise-induced rise of pituitary hormones, cortisol and insulin as well as lower resting levels of testosterone possibly affects the resynthesis of protein and glycogen during regeneration after exercise. The lower respiratory quotient repeatedly described in overtraining probably indicates a shift of the energy-supplying processes in favor of an increased fat and decreased CHO utilization.

Immunofunctions. The increased susceptibility for infections of overtrained athletes will likely find an explanation in a fine but complex regulation of both closely interacting hormonal and immune systems.

Research Problems

This topic is particularly difficult to research in the field. It requires careful control, repeated monitoring to set individual levels and response patterns, and standardized testing procedures probably of a level that is not possible in field or applied settings. Some of the confounding difficulties in researching this aspect of the overtraining response are as follows:

• lack of reference values and considerable individual differences;
• limited knowledge about regulation (supra-hypothalamic level) and peripheral modulation (receptor level);
• methodological limitations such as relatively high inter- and intra-assay variability, the need for quick results, the relatively large sample volumes required, and expense;
• circadian and pulsatile rhythms;
• influences of external modifying factors (diet, altitude, etc.);
• different plasma half-lives; and
• in female athletes, physiologically low testosterone levels and menstrual cyclic variations.

When these factors are considered, it is understandable why the "field" use of hormonal measures is extremely unreliable as well as having revealed few worthwhile criteria when related to overtraining.

Implications

• The testosterone/cortisol ratio indicates the physiological strain of training load rather than overtraining syndrome.
• The frequency of competitions or training sessions with higher anaerobic lactic demands, should be carefully limited in order to prevent overtraining syndrome.
• The role of hormones in the recovery phase and their effects on the receptor and intracellular level remain to be better established.
• External and measurement factors influence hormonal blood levels and need to be controlled very well before any test results can be considered for the diagnosis of overtraining.

Return to Table of Contents for this issue.