While there is no clear consensus, acute and chronic exercise is associated with suppressed endocrine functions at the hypothalamic and testicular levels; specifically, suppression of both gonadotropin-releasing hormone (GnRH) and serum testosterone.17 Taking testosterone (T) as a major regulator of the hypothalamic-pituitary-testicular (HPT) axis, available studies can be included into three categories: those that do not show changes in either T (free [fT] or total [tT] T) or any of the other HPT hormones,18–20 those that show no changes in testosterone (fT or tT) but show changes for the other HPT hormones,7,21,22 and those that show changes in fT or tT but there could or could not be changes, for the other HPT hormones.2,3,10,23–28

Some studies may reflect training stimuli that were not sufficient enough to alter the axis (low or moderate performance level, or low volume used in the study), which could be plausible explanations for studies not detecting changes in T but detecting changes for other hormones (Follicle Stimulating Hormone (FSH) and Luteinizing Hormone (LH)). Among works detecting changes in T the most habitual finding is decrease in T without changes in FSH or LH.2,3,10,22,27 Research has focused primarily on distance runners, but studies have been conducted with other endurance sports. For example cyclists, were findings are controversial. One study reported no differences in hormonal profiles of cyclists.16 Other studies have revealed decreased T values in professional cyclists, during one specific race (Tour of Spain),29,30 researcher found cyclists coming into competition with the greatest training volume had lower T levels than those of other comparative cyclists. This finding is supported by other studies.5,31 Likewise, when trained and untrained subjects performing four hours of road bicycling at the highest possible level were compared, it was observed that FSH levels were higher in the trained subjects, suggestive of a compensatory hypogonadism as result of intensive chronic training or Sertoli cell dysfunction.32

Hypogonadal states have been observed as a result of chronic ultra-endurance-training when running or cycling 10–20 or more hours per week.33 Triathletes, normally undergoing high volume in the three sports modalities composing this discipline, showed, when compared to cyclists and recreationally active men, significantly lower levels of estradiol and T.12 However, not all previous studies have observed such differences.16

Rowing has also assessed the effect of training on HPT. Heavy training seems to result in decreased free Testosterone Cortisol ratio (T/C) when compared to basal levels; though decreases greater than 30% have been reported, but never drop below the threshold value indicating overstrain of 0.35×10−3.34 Urhausen et al.35 observed that T, T/sex hormone – binding globulin (SHBG), and T/C decreased during a 7-week period of training, stopping with recovery of only one week and with two rowers showing normalized values within several weeks of training discontinuation. A 2-week period of high-training volume significantly decreased T 30min after finishing a long-distance rowing test; such observed changes indicate decreased adaptive ability.36

Such decreased adaptivity was also observed in another study as evidenced by reduced basal fT and fT/C ratio as well as an altered T response to an exercise challenge after a heavy training period of three weeks.37 Similarly Vinther et al.38 found that values, albeit within normal range, were in the lower part of the normal spectrum. Therefore, as reported by some, T and cortisol (C) are altered as a result of changes in training volume, and the values of such hormones appear related to weekly training volume.39

Not only does the volume of training induce hormonal alterations, but in fact, the intensity of the exercise may also exert an influence on hormonal profile alterations. In this regard, Safarinejad and co-workers,10 compared moderate-intensity exercise (60% Maximal Oxygen Consumption (VO2max)) to high-intensity exercise (80% VO2max) for a period of 12 weeks and reported alterations in hormones (returning to normal during recovery) with a decrease in T and blunted response to a Gonadotropin Releasing Hormone (GnRH) stimulation challenge in both groups but alterations being more marked in the high intensity group.

Furthermore, our research group has shown that maximal intensity exercise imposed on physically active subjects during a short period of time results in alterations in FSH, LH, prolactin (PRL), dehydroepiandrosterone and C values but no significant changes in T, progesterone and estradiol; such alterations returned to pre-training values after three days during the recovery period.7

In soccer, Grandi and Celani40 have observed that the LH response to a GnRH-Thyrotropin Releasing Hormone (TRH) challenge test was less in the professional players as compared to non-professional players; while no significant hormone changes were observed during the training season, PRL basal levels were increased after a strenuous soccer game. In basketball, an initial transient increase in T was observed to later on decrease, along with a decrease in T/C, from mid-season on.41

The evidence discussed above points to the hormonal response to exercise being affected by both by intensity and volume of training. Although the complexity in hormone feedback mechanisms and the disparity in results in the available literature demands that a consensus in study design and protocols needs to be reached. The heterogeneity observed in different studies with regards to, but not limited, the type of athlete (sedentary vs. amateur vs. professional), the type of intervention (observational vs. interventional) should be minimized.

Though studies analyzing long term response are scarce, imposed training loads seem to elicit an acute response with an initial increase in androgenic hormones, but when programs are sustained, often surpassing the habitual training levels of the athletes, there may be a decrease in basal levels of these hormones.42–45 The loaded used, time of application and type of athletes are key factors for the long-term effects of training. Frequently, exercising subjects present with higher T basal levels that sedentary subjects.46,47 Intervention with moderate intensity aerobic exercise resulted in increased serum dihydrotestosterone (DHT),48 in fT49 and SHBG48,49 in sedentary subjects. Also, Vaamonde and colleagues50 have recently reported improved values for FSH, LH, T and the T/C ratio when comparing sedentary subjects to physically active subjects.

While an increase in the resting levels of T is common feature of chronic response to strength training,5 when resistance training becomes excessive, with the athlete coming close to overtraining or with persistent fatigue, T levels may even decrease from baseline.54

Yet not fully elucidated such response may involve amino acid deficit and central nervous system alterations,2 alterations in neurotransmitters or increase in T-inhibiting hormones, and aging processes.54

For male athletic performance, it may be beneficial to alter the levels of circulating anabolic hormones and the anabolic-catabolic ratio as testosterone, among others, is directly involved in muscle adaptation to exercise.

Generally speaking, albeit discrepancies exist, including theories that changes in hormones are not mechanistic, but a result of hemoconcentration or decreased hepatic clearance, adequately applied resistance training provokes an increase in tT and fT levels after one training session,51,55 with magnitude of changes depending on different factors such as load (intensity,56 volume,6 recovery,57 amount of muscle mass involved, athlete's age,58 experience and performance level.58 The magnitude of the load used will determine the time that T levels are elevated after training.6

Though controversy, as in hormonal studies, there is evidence that long-term exhaustive exercise seems to negatively affect sperm quality and reproductive potential.11

One study reports that even up to 10% of the assessed runners exhibit severe oligospermia, yet other authors report non-existing or merely subclinical alterations due to exercise training.2,19,23,59 Increase in “round cells” has also been reported2 indicating a possible infectious and/or inflammatory environment. Arce and colleagues2 were able to, retrospectively establish a volume threshold of 100km/wk for semen alterations to occur, as they found alterations in sperm density, motility, morphology and in vitro sperm penetration of standard cervical mucus in endurance-trained runners when compared to resistance athletes or sedentary subjects. Similarly, Safarinejad et al.10 observed a negative effect of training on sperm parameters, though reverting during recovery, in high-intensity training athletes when compared to moderate-intensity ones. Conversely, Hall and co-workers60 could not find an influence on sperm count, morphology, or motility after a period of six weeks of gradually intensified training followed by two weeks of detraining.

High-level athletes have been typically training for many years, making it difficult to establish a potential harmful training threshold (volume and/or intensity) as they normally start training at pre- or peri-pubertal years. Nevertheless, high volume cycle training seems to correlate with sperm morphology anomalies.61 In fact, in triathletes, a cycling volume of 300km/wk correlates with serious degree of morphology abnormalities and possible leading to fertility impairment. Also, ultra-endurance athletes when compared to other athletes show significantly different values for the various sperm parameters with morphology showing the greatest difference, with clinical relevance (<5% normal forms).11

In animal models decreased number of cells from the spermatogenic lineage at different stages of development has been observed with training,14,15 some of these studies also revealed decreased levels of serum hormones, enzymes and antioxidant agents.14,15 Vaamonde et al.62 have reported exercise-related alterations in sperm which may be prevented with antioxidant agents. Vaamonde et al.63 also recently reported that, similarly to sperm morphology, cycling volume positively correlates to sperm deoxycribonucleic acid (DNA) fragmentation, also observing high correlation between training volume, sperm DNA fragmentation, percentage of morphological abnormalities and TUNEL(+) cells (unpublished data Vaamonde).

Regarding reproductive tract abnormalities, it has been reported that a soccer player exhibited testicular mal-development;64 as the athlete had cryptorchidy it is hard to establish a relationship to sports practice, however, with greater training load all other symptoms worsened. In this regard, continuous strenuous exercise has also been linked to erectile dysfunction or impotence65 as tiredness and fatigue may lead to reduced libido, especially in bicycling exercise64,66 due to microtrauma and compression on the genital area. In rats, decreased testicular, epididymal, prostatic, and seminal vesicles somatic indices have been reported as a result of exercise.15

Far less studied is the positive effect of exercise or physical activity on seminal parameters. Vaamonde et al.50 reported improved semen parameters in physically active men when compared to sedentary people, such findings being supported by hormonal differences. Other studies with endurance-trained men report that average semen parameters values are usually within normal limits16 or show subclinical alterations.2,3,19 In this regard, significant alterations, even if subclinical, have been reported to require a certain “training volume-threshold” (∼100km/wk) before occurring.3 Also in animals similar results have been reported, not only on sperm parameters but also on testicular health.67,68

Training load affects oxidative stress; as such, superoxide dismutase (SOD) capacity has been reported to be increased in elite athletes69 which can affect the testicular environment. Physically active men, in comparison to higher level athletes, have been shown to have higher levels of seminal antioxidant compounds and lower levels of seminal reactive oxygen species (ROS), oxidative stress and sperm DNA fragmentation.70 Erectile dysfunction has also been reported to be less common in physically active subjects (non-athletes) than in sedentary; moreover, this condition has been seen to improve as a result of unhealthy lifestyle modification like becoming physically active (minimum energy expenditure of 200kcal/day).71

Also in animal models it has been observed that exercise reduces markers of oxidative stress and increases antioxidant-related enzymes,68 decreasing age-related damage, especially if exercise practice begins earlier in life,72 and also decreasing DNA damage.73 Moreover, testicular atrophy, inflammation and degeneration, related to aging, is decreased in animals submitted to exercise.68

Resistance training is many times intimately relate to androgenic anabolic steroid (AAS) use and regrettably little is known, if any on sperm quality as a result of resistance training without steroid use. It has been observed that the HPT axis and its hormones are altered by supraphysiologic levels of exogenous AAS and that these changes may lead to semen alterations like azoospermia, oligospermia, teratozoospermia (mainly head and mid-piece), and oftentimes testicular atrophy74,75 as evidenced by alterations in germinal epithelium, Leydig cells, among others.76

No clear evidence exists of a positive effect of resistance training on semen parameters. Arce et al.2 has observed, when comparing endurance and resistance athletes, that the latter did not manifest negative alterations as seen in endurance athlete's semen characteristics so this could be indicative that, at least, this type of training is not substantially detrimental to semen quality.