Testosterone, an anabolic-androgen hormone, plays a pivotal role in the healthy aging of men. Serum testosterone concentration in men tends to decline with age, and by the age of 45 years, roughly one-third of U.S. men meet the hormonal criterion for hypogonadism (total testosterone concentration <300 ng·dL−1) (5). Low serum testosterone concentration in men is associated with all-cause mortality (2), cardiovascular disease (41), type 2 diabetes (42), sarcopenia (31), depression, and other cognitive impairments (44). Low serum testosterone concentration in men is also associated with physical changes (e.g., reduced muscle mass, reduced muscle function (13,31), increased fat deposition (25)), which may negatively affect physical function. It was estimated that the United States spent $190–525 billion in direct costs over a 20-year period treating patients with health conditions attributed to hypogonadism in men (30), with costs of testosterone replacement therapy approximately $5 billion per year (6).
Findings from randomized controlled trials (RCTs) indicate testosterone replacement therapy—administering exogenous testosterone (e.g., testosterone enanthate) to return resting serum testosterone to within the physiologic range (i.e., 303–852 ng/dl) in hypogonadal men—has positive effects on libido, erectile function, sexual activity, mood (i.e., decreased negative and increased positive aspects), body composition (i.e., decreased fat mass and increased fat-free mass), muscle strength, and bone mineral density and strength (5). Possible side effects of testosterone replacement include infertility, polycythemia (8), gynecomastia (34), sleep apnea, liver toxicity, and prostate hypertrophy (3). In addition, there is apprehension about the effect of testosterone replacement on cancer and cardiovascular disease risk (6,37); however, the longitudinal evidence needed to evaluate these relationships is currently insufficient (5). Interestingly, it seems that men aged 18–45 years make up the fastest growing segment (i.e., 4-fold increase from 2003 to 2013) of the U.S. population receiving testosterone replacement therapy (33). This is particularly concerning because these men will likely accumulate increased exposure to the therapy because of long-term use (33). It is possible that side effects associated with testosterone replacement may, in part, be due to altered patterns of blood testosterone concentrations resulting from alterations in pulsatile secretion and usual diurnal variation. Compared with pharmacological intervention, maintaining or achieving a physiologic resting serum testosterone concentration through lifestyle changes (e.g., diet and exercise) is preferable because it would preserve natural hormone kinetics, decrease patient risk, and reduce treatment costs.
To date, most of the literature investigating the effectiveness of therapeutic lifestyle changes on resting serum testosterone concentration has focused on weight loss, although experimental trials using exercise training interventions are starting to accumulate. In a meta-analysis of dietary restriction and bariatric surgery trials, Corona et al. (10) reported small-to-moderate benefits of weight loss on resting serum testosterone concentration. Although studies have reported increases in serum testosterone concentration immediately after moderate-intensity or high-intensity exercise in men (11), Hayes and Elliot (18) found small, albeit inconsistent, increases in resting serum testosterone concentration after chronic exercise training in a recent meta-analysis of older men. It is not known, however, whether the results from Hayes and Elliot (18) are generalizable to younger and middle-age men (i.e., <60 years). Furthermore, most studies included in the Hayes and Elliot (18) meta-analysis were uncontrolled trials, with resting serum total testosterone, free testosterone, and bioavailable testosterone included in the analyses, which cloud the interpretation of the effects of exercise on testosterone. The clinical guidelines set forth by the Endocrine Society (5) recommend measurement of serum total testosterone and free testosterone, depending on certain patient characteristics (e.g., borderline initial total testosterone measurement, obesity, aging, and diabetes), for the diagnosis of hypogonadism. However, few RCTs evaluating the effects of exercise training report serum free testosterone and large interassay/intermethod variation complicates its interpretation (5). Randomized controlled trials provide the best evidence for causality because they are less prone to bias and confounding, and meta-analyses of RCTs best guide recommendations for practitioners. Therefore, the aim of this systematic review and meta-analysis was to evaluate the effect of exercise training on resting serum total testosterone concentration in insufficiently active, apparently healthy men when including only RCTs. The secondary aims of this review were to determine whether the effects of exercise training on resting serum total testosterone concentration differed by training mode, age, or body mass status and to assess the effect of exercise training on resting serum free testosterone concentration.
Experimental Approach to the Problem
The review protocol was prospectively registered with the International Prospective Register of Systematic Review (PROSPERO; CRD42020131461). This review followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement (28).
Studies were eligible for inclusion if they reported on the effects of exercise training (e.g., aerobic or resistance) on resting total testosterone concentration in insufficiently active (i.e., less than 150 minutes of moderate or 75 minutes of vigorous physical activity [or an equivalent combination thereof] per week), apparently healthy men (i.e., free from known disease or signs and symptoms of disease; aged ≥18 years). Studies were not excluded on the basis of recruiting men with low testosterone concentration or exhibiting signs and symptoms of hypogonadism. Only RCTs lasting a minimum of 4 weeks and reporting preintervention and postintervention resting total testosterone concentration assessed from serum, or changes in such concentration, were included. It was a requirement that the control group maintained habitual physical activity levels throughout the study period. Studies were excluded if the training intervention was below moderate intensity (4 metabolic equivalent of task) (14) or was conducted in special interest groups (e.g., elite athletes). Only peer-reviewed journal articles were considered, with all other publication types (e.g., abstracts, reviews, theses, books, and letters) excluded.
Studies were identified by searching electronic databases, reference lists, and topical systematic reviews (28). A systematic search of the literature was performed on January 27, 2019, in MEDLINE, Cumulative Nursing and Allied Health Literature (CINAHL), Scopus, and SPORTDiscus. Secondary and tertiary searches were performed on September 19, 2019, with date restrictions from January 1 to September 19, 2019, and on October 22, 2020, with date restrictions from September 1, 2019, to October 22. All sources were limited to the English language. An academic librarian experienced in systematic reviews was consulted during the search strategy development, with the search strategy piloted by 2 researchers to identify the combination of search terms most likely to yield relevant articles. This study was not evaluated by the University of North Dakota Institutional review board because it only uses descriptive data previously published in randomized control trials that have undergone institutional review. The descriptive data was obtained from the published articles and no raw data was reviewed or used in this study. Thus, this systematic review does not require institutional review as per University of North Dakota IRB guidelines.
The electronic database search included a keyword search of the full text and the citation record. Search terms that were included in a group were combined by the Boolean OR and entered before being combined by the Boolean AND. For some search terms, proximity operators were used to search for the root word. Selection of search terms was informed by the PICO (population, intervention, comparison, outcome) method. The first group of search terms identified exercise training intervention (exercise training OR aerobic training OR resistance training OR high-intensity interval training OR sprint training), the second group identified the outcome (testosterone OR hormone* OR androgen*), and the third group identified the subjects (men OR males). Randomized controlled trials were then screened at the full-text level. Supplement 1 shows the full search strategies for each database (see text, Supplemental Digital Content 1, https://links.lww.com/JSCR/A280).
One researcher executed the electronic database search and imported all results into RefWorks (ProQuest, Ann Arbor, MI). At the first level, 2 researchers independently screened all studies by title and abstract. At the second level, 2 researchers independently screened all studies at the full-text level. Consensus (at both levels) was required between both researchers for final inclusion; if not, then a third researcher resolved any discrepancies. Two researchers who published on the effects of exercise training on resting testosterone concentration were also contacted to locate additional studies, and referred studies were evaluated against inclusion criteria.
Descriptive data from included studies were extracted and imported into a study-specific template created in Excel (Microsoft Corp., Redmond, WA). Data were extracted independently by 2 reviewers and examined for errors. Corresponding authors were contacted through email if additional data were needed.
The following descriptive data were extracted: study title, training exposure (exercise modality, training volume, sessional intensity, frequency, and duration), age, height, body mass, body mass index (BMI), serum testosterone assay, pretest, posttest and change measures for testosterone concentration (sample sizes, means, and SDs), and effect size data. If necessary, means and SDs were estimated for testosterone concentration data using equations described in the study by Wan et al. (43) or using the RevMan calculator (12). Serum free testosterone concentration data were extracted, if available.
The Physiotherapy Evidence Database (PEDro) tool for RCTs was used to assess the quality of included studies. The PEDro tool assesses a study's methodological quality and completeness in reporting using an 11-criteria scale. Scores range from 0 to 10 points as the first criterion is not scored. Scores of 0–3, 4–5, and 6–10 were interpreted as poor, moderate, and high quality, respectively. The PEDro tool exhibits good reliability (27). Quality assessment was independently conducted by 2 reviewers. The intraclass correlation coefficient (ICC) determined the test-retest agreement between raters.
Meta-analyses were performed in RevMan 5 (v5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014). The primary analysis was conducted to determine the overall effect of exercise training on resting serum total testosterone concentration in all subjects, with secondary subgroup analyses conducted to determine whether the overall training effect was influenced by training mode, age, or body mass status, and to assess the effect of exercise training on resting serum free testosterone concentration. Standardized mean differences (SMDs) weighted by the inverse of the pooled variance, and corresponding 95% confidence intervals, were calculated for each intervention group using a random effects model. Experimental effects were calculated relative to the control group using the weighted mean difference method. Positive mean differences indicated favorable experimental effects, and negative mean differences indicated unfavorable experimental effects. Standardized mean differences were interpreted using Cohen's (9) thresholds of 0.2, 0.5, and 0.8 for small, medium, and large, respectively, with SMDs < 0.2 considered to be negligible. Statistical heterogeneity was quantified using the I2 statistic, with values of 25, 50, and 75% used as thresholds for small, medium, and large, respectively (19). Funnel plot asymmetry analysis was used to assess the risk of publication bias for analyses with 10 or more group comparisons because the power of this test to distinguish chance from real asymmetry is too low when there are fewer than 10 comparison groups (19).
A total of 3,570 unique articles were located through the online database search, with 71 studies retained after the first level of screening (title and abstract) and 8 studies retained after the second level of screening (full-text). Three additional studies were located from reference list searching (n = 1) and consultation with researchers in the field (n = 2), resulting in 11 included studies (1,16,17,20,22,26,29,35,38–40). Figure 1 shows the identification of included studies.
All RCTs were published in English between 2004 and 2020, collectively represented 421 insufficiently active, apparently healthy men with a mean ± SD age range from 18.8 ± 1.5 to 75.2 ± 3.0 years. Overall, there were 16 intervention groups using 3 modes of exercise training (resistance [k = 11], aerobic [k = 3], and a combination of resistance and aerobic [k = 2]) lasting between 6 and 52 weeks in duration. Study characteristics are presented in Table 1.
Table 1 -
Description of training for included studies.*†
||E resting testosterone change
|C resting testosterone change
size E (n)
size C (n)
||Training duration (wk)
||Frequency (per wk)
|Ahtiainen et al. (1)
|E: 28 (5)
C: 25 (3)
E: 61 (5)
C: 64 (8)
||3 leg exercises and 4–5 upper body, 3 7-wk phases. Phase 1 (multiple sets, 10–20 reps, 40–60% 1 RM); phase 2 (multiple sets 8–12 reps, 60–80% 1RM); phase 3 (multiple sets 5–8 reps, 70–90% 1RM)
|Harber et al. (16)
||E: 23.6 (1.8)
C: 20.5 (1.0)
||10 exercises 1–3 sets 12–20 reps at 40–60% 1RM
|Hawkins et al. (17)
||E: 56.2 (6.7)
C: 56.6 (7.6)
||60-min sessions at 60–85% HRmax (e.g., treadmill, rowers, and stationary bike)
|Hiruntrakul et al. (20)
||E: 21 (2)
C: 20 (1)
||5-min warm-up followed by 50 min of biking at 60% V̇o
2max and 5-min cool down
|Katznelson et al. (22)
||E: 72 (5.4)
C: 72 (5.2)
||11 extremity and trunk resistance strengthening exercises using therabands
|Lovell et al. (26)
||RT: 17.3 (100.4)
AT: 31.7 (133.8)
||RT: 74.1 (2.7)
AT: 75.2 (3.0)
C: 73.5 (3.3)
|Incline squat 3 sets of 8 at 50% 1RM, increased to 6–10 reps at 70–90% 1RM.
30-min cycling at 50% V̇o
2max, increasing to 70% V̇o
2max for 45 min
||E: 26.5 (2.8)
C: 27.4 (2.9)
||10 exercises at 3 sets of 8–12 reps at 60–80% 1RM. 20–40 min per session
|Roberts et al. (35)
||8 upper-body and lower-body exercises. Phase 1 (2 wk, 2 sets 12–15 reps at 100% effort); phase 2 (5 wk, 3 sets 8–12 reps at 100% effort); phase 3 (5 wk, 2 sets 6–8 reps at 100% effort)
|Sedliak et al. (38)
||M: −66 (276)
A: 28 (91)
||ME: 23 (2)
AE: 24 (4)
C: 24 (3)
||3 leg exercises and 5 core and upper-body exercises. First 5 wk at 40–60% 1RM 10–15 reps for 3 sets; Second 6 wk increased to 50–80% 1RM 8–12 reps in 4 sets
|Sharifi et al. (39)
||TS: 16 (123)
OS: 25 (98)
||RT2: 20.1 (2.2)
RT1: 19.8 (1.5)
C: 22.7 (0.7)
||3 leg exercises and 3 upper-body exercises. First 2 wk at 70% 1RM 10 reps for 3 sets; Second 2 wk at 75% 1RM 10 reps for 3 sets; Third 2 wk at 80% 1RM 10 reps for 3 sets
|Sheikholeslami-Vatani et al. (40)
||CRE: 30 (120)
CER: 10 (160)
||Running at 70–75% HRmax for 10 min, increasing to 80% for 21.5 min, then 5 resistance exercises 3 sets of 8 at 80% 1RM
*PEDro = Physiotherapy Evidence Database; RT = resistance training; AT = aerobic training; E = exercise group; C = control group; M = morning group; ME = morning exercise group; A = afternoon group; AE = afternoon exercise group; TS = 2 sessions per day group; OS = one session per day group; CRE = concurrent resistance-endurance group; CER = concurrent endurance-resistance group; RM = repetition max; V̇o2max = maximal oxygen uptake; HRmax = maximal heart rate.
†Change mean data in ng·dl−1.
There was perfect agreement (ICC = 1.00) between the 2 reviewers who assessed methodological quality using the PEDro scale. Included studies were moderate to high in quality, with scores ranging from 4 of 10 (k = 3) to 7 of 10 (k = 1) and most scoring 6 of 10 (k = 6). None of the studies scored positive for blinding of subjects and therapists. One study did have blinding of assessors. All studies scored positively for similar baseline, random allocation to groups, point measures, and comparison between groups.
Synthesis of Results
Resting serum total testosterone concentration change data were collected. The mean resting serum total testosterone concentration for the exercise training group was 517 ± 90 ng·dl−1 at the start of the study. Exercise training had a negligible effect on resting serum total testosterone concentration (mean SMD [95% CI]: 0.00 [–0.20 to 0.20]; Figure 2). The heterogeneity for exercise training on resting serum total testosterone concentration was negligible (I2 = 0%), and the funnel plot showed no evidence of asymmetry (see Figure, Supplemental Digital Content 2, https://links.lww.com/JSCR/A280).
After stratifying for either exercise training mode (resistance [k = 11] vs. aerobic [k = 3]) or age (younger men [<60 years; k = 12] vs. older men [≥60 years; k = 4]), exercise training had a negligible effect on resting serum total testosterone concentration for all subgroups (mean SMD [95% CI] range: −0.05 to 0.01 [–0.40 to 0.42]). Exercise training had a small, nonsignificant effect on resting serum total testosterone concentration in studies recruiting only obese men (BMI ≥ 30 [k = 3]; mean SMD [95% CI]: 0.43 [–0.08 to 0.94]) and a negligible effect in studies lacking obesity as an inclusion criterion (mean SMD [95% CI]: −0.07 [–0.29 to 0.14]). Exercise training had a small, nonsignificant effect on resting serum free testosterone concentration (k = 5; mean SMD [95% CI]: −0.22 [–0.49 to 0.06]) and a negligible effect after removing a study (20) that used an assay that is no longer recommended by the Endocrine Society (k = 4; mean SMD [95% CI]: −0.15 [–0.45 to 0.15]) (17,26,35) (see Figures, Supplemental Digital Content 3–4, https://links.lww.com/JSCR/A280). Heterogeneity for all subgroup analyses was negligible (I2 = 0%), with funnel plots for analyses with 10 or more group comparisons showing no evidence of asymmetry (see Figures, Supplemental Digital Content 5–7, https://links.lww.com/JSCR/A280).
The primary aim of this systematic review and meta-analysis was to evaluate the effect of exercise training on resting serum total testosterone concentration in insufficiently active, apparently healthy men. Overall, the effect of exercise training on resting serum total testosterone concentration was negligible and did not differ by training mode, age, or body mass status. In addition, exercise training did not have a statistically significant effect on resting free testosterone concentration.
In a recent meta-analysis, Hayes and Elliot (18) found small effects of aerobic and interval, but not resistance, training on resting testosterone concentration in older men. Consistent with Hayes and Elliot (18), most of the studies identified in our systematic review evaluated the effects of resistance training and found negligible effects of resistance training on resting serum total testosterone concentration. In addition, although a recent meta-analysis (11) found that acute physical exercise was associated with a moderate-to-large increase (SMD = 0.71) in serum testosterone, the effects were limited to within 30 minutes postexercise and were negligible thereafter. Collectively, these data indicate a lack of effectiveness of short-term (median duration = 12 weeks) resistance training to increase resting total testosterone concentration in younger and older eugonadal men. In contrast to Hayes and Elliot (18), we found aerobic training to have a negligible effect on resting serum total testosterone concentration in men. This discrepancy may be due to differences in exclusion criteria between studies (as our study was limited to RCTs in insufficiently active, apparently healthy men) and outcomes included in analyses (Hayes and Elliot (18) analyzed multiple measures of resting testosterone [total, free, and bioavailable testosterone]). In subgroup analyses, we did not find evidence that the effects of exercise training on resting serum total testosterone concentration differed by age (younger vs. older men) nor did we find an effect of exercise training on resting free testosterone concentration. In addition, baseline resting testosterone concentration before exercise training does not seem to explain differences between meta-analyses because the studies recruiting subjects with lower testosterone concentration were included in the resistance training analysis in Hayes and Elliot (18). However, we found a trend toward a difference when analyses were stratified by body mass status, with studies recruiting only obese subjects yielding small, positive, albeit nonsignificant, effects. A plausible explanation for this trend is that exercise training trended toward greater weight loss in obese men (−3.0 ± 1.4 kg or 3.1% decrease) when compared with nonobese men (0.0 ± 1.8 kg or 0.1% increase) in the included studies. Reduced fat mass may decrease aromatization of testosterone to estradiol and reduce leptin resistance, leading to increased gonadotropin secretion and a subsequent small increase in resting serum testosterone concentration (7). The effect of weight loss on both free and total resting testosterone concentration has been reported in a meta-analysis by Corona et al. (10). On average, they found a 9.8% weight loss during dietary interventions resulted in small increases in testosterone concentration, which is slightly greater than the weight loss observed in the studies used for the meta-analysis in our study.
The results from our systematic review and meta-analysis do not support exercise training as an effective therapeutic lifestyle intervention for increasing resting serum total testosterone concentration in insufficiently active, apparently healthy men. It should be noted that the mean resting serum total testosterone concentration for all studies analyzed in this review was above the threshold for hypogonadism, and it is possible that having low resting serum total testosterone concentration before exercise training would modify the response to exercise training. Future RCTs should consider recruiting age-related and obesity-related hypogonadal men and evaluate the effect of exercise training on resting serum testosterone concentrations using effect sizes and clinical end points (e.g., crossing threshold for hypogonadism). Furthermore, future research could evaluate the effect of exercise-induced weight loss on total and free testosterone concentration in obese men. However, if present, the effects of exercise training are likely small and may not be clinically meaningful when the testosterone concentration is well below the physiologic range (7,10) and adherence to lifestyle interventions is low (36).
Of greater interest to practitioners may be the effect of exercise training on signs and symptoms of hypogonadism (e.g., reduced sexual function and desire, decreased mental health, infertility, increased body fat, decreased muscle mass and strength, and decreased bone mineral density), which are the primary diagnostic criterion in clinical settings given that the evaluation of testosterone concentration is not indicated until a patient reports 1 or more signs or symptoms (5). Evidence from RCTs indicates that exercise training may improve many of the signs and symptoms of hypogonadism, for example, sexual dysfunction (21), infertility (15), poor mental health (32), increased body fat (23), decreased muscle mass, low strength (4), and low bone mineral density (24). Future research should evaluate the effects of lifestyle change interventions on less studied outcomes to elucidate the effects of such interventions to allow for a better understanding of the practical utility of lifestyle interventions for the prevention and treatment of hypogonadism.
This study is not without limitations. Only 11 RCTs met our inclusion criteria, with exercise training programs varying greatly across studies (e.g., mode, intensity, volume, and frequency). Statistical power to detect potentially meaningful effects was limited in subgroup analyses. Most of the studies used an exercise training frequency of 3 days per week or less, and exercise session duration was difficult to ascertain for some studies but seemed to vary substantially from approximately 20 to 60 minutes, presumably resulting in considerable variability in training volume across studies. It is possible that greater exercise training volumes are required to increase resting serum total testosterone concentration in insufficiently active men. Most men in the studies analyzed were eugonadal, which limits generalizability to clinical populations. In addition, information relevant to signs and symptoms of hypogonadism was not tracked (or was not reported) nor were both free and total resting serum testosterone concentration assessed in most studies. Although most studies measured testosterone concentration in the morning after a period of fasting, a few studies did not, or did not provide this information. Studies failed to report on population demographic descriptive data (i.e., race and ethnicity), which limits our ability to comment on this aspect. The studies identified in this review were primarily of moderate quality as assessed using the PEDro scale. Future studies should use more rigorous study designs.
In conclusion, this systematic review and meta-analysis found exercise training to have a negligible effect on resting serum total testosterone concentration in insufficiently active, eugonadal men, which did not significantly differ by training mode, age, or body mass status. Future research should examine the effect of exercise training on testosterone concentrations in obese, hypogonadal men to determine whether exercise training has a favorable effect on resting testosterone concentration. The role of exercise training in addressing signs and symptoms of hypogonadism needs further evaluation.
Health practitioners and exercise professionals should not expect improvements in resting total testosterone concentrations with short-duration exercise training in previously insufficiently active, eugonadal men. The results of our meta-analysis have limited generalizability to hypogonadal men because this population is underrepresented in the current exercise training literature. Practitioners, researchers, and exercise professionals should also consider exercise training for the primary prevention and treatment of certain signs and symptoms of hypogonadism (e.g., increased body fat, decreased muscle mass and strength, decreased bone mineral density, reduced sexual function and desire, decreased mental health, and infertility) because testosterone replacement therapy is not indicated in the absence of such signs and symptoms. Reducing the need for testosterone replacement therapy has the potential to decrease patient risk and reduce treatment costs, especially among younger men (18–45 years) who are more likely to be subject to long-term use.
The authors have no external funding sources or conflicts of interest to disclose. The results of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association. The contents of this publication do not necessarily reflect the views or policies of the U.S. Department of Agriculture or the Agricultural Research Service nor does mention of trade names, commercial products, or organizations imply endorsement from the U.S. government. USDA is an equal opportunity provider and employer.
1. Ahtiainen JP, Hulmi JJ, Kraemer WJ, et al. Heavy resistance exercise training and skeletal muscle androgen receptor expression in younger and older men. Steroids 76: 183–192, 2011.
2. Araujo AB, Dixon JM, Suarez EA, et al. Endogenous testosterone and mortality in men: A systematic review and meta-analysis. J Clin Endocrinol Metab 96: 3007–3019, 2011.
3. Bassil N, Alkaade S, Morley JE. The benefits and risks of testosterone replacement therapy: A review. Ther Clin Risk Manag 5: 427–448, 2009.
4. Beaudart C, Dawson A, Shaw SC, et al. Nutrition and physical activity
in the prevention and treatment of sarcopenia: Systematic review. Osteoporos Int 28: 1817–1833, 2017.
5. Bhasin S, Brito JP, Cunningham GR, et al. Testosterone therapy in men with hypogonadism: An endocrine society clinical practice guideline. J Clin Endocrinol Metab 103: 1715–1744, 2018.
6. Braun SR. Promoting “low T”: A medical writer's perspective. JAMA Intern Med 173: 1458–1460, 2013.
7. Carrageta DF, Oliveira PF, Alves MG, Monteiro MP. Obesity and male hypogonadism: Tales of a vicious cycle. Obes Rev 20: 1148–1158, 2019.
8. Chin-Yee B, Lazo-Langner A, Butler-Foster T, Hsia C, Chin-Yee I. Blood donation and testosterone replacement therapy. Transfusion (Paris) 57: 578–581, 2017.
9. Cohen J. The t Test for Means. In: Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers, 1988. pp. 19–74.
10. Corona G, Rastrelli G, Monami M, et al. Body weight loss reverts obesity-associated hypogonadotropic hypogonadism: A systematic review and meta-analysis. Eur J Endocrinol 168: 829–843, 2013.
11. D'Andrea S, Spaggiari G, Barbonetti A, Santi D. Endogenous transient doping: Physical exercise acutely increases testosterone levels—results from a meta-analysis. J Endocrinol Invest 43: 1349–1371, 2020.
12. Drahota A, Beller E. RevMan Calculator. 2008. Available at: https://training.cochrane.org/resource/revman-calculator
. Accessed March 2, 2020.
13. Emmelot-Vonk MH, Verhaar HJJ, Nakhai Pour HR, et al. Effect of testosterone supplementation on functional mobility, cognition, and other parameters in older men: A randomized controlled trial. JAMA 299: 39–52, 2008.
14. Garber CE, Blissmer B, Deschenes MR, et al. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: Guidance for prescribing exercise. Med Sci Sports Exerc 43: 1334–1359, 2011.
15. Hajizadeh Maleki B, Tartibian B. Combined aerobic and resistance exercise training for improving reproductive function in infertile men: A randomized controlled trial. Appl Physiol Nutr Metab 42: 1293–1306, 2017.
16. Harber MP, Fry AC, Rubin MR, Smith JC, Weiss LW. Skeletal muscle and hormonal adaptations to circuit weight training in untrained men. Scand J Med Sci Sports 14: 176–185, 2004.
17. Hawkins VN, Foster-Schubert K, Chubak J, et al. Effect of exercise on serum sex hormones in men: A 12-month randomized clinical trial. Med Sci Sports Exerc 40: 223–233, 2008.
18. Hayes LD, Elliott BT. Short-term exercise training inconsistently influences basal testosterone in older men: A systematic review and meta-analysis. Front Physiol 9: 1878, 2019.
19. Higgins J, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. The Cochrane Collaboration, 2011. Available at: http://handbook.cochrane.org
. Accessed March 2011.
20. Hiruntrakul A, Nanagara R, Emasithi A, Borer KT. Effect of endurance exercise on resting testosterone levels in sedentary subjects. Cent Eur J Public Health 18: 169–172, 2010.
21. Hoffman BM, Babyak MA, Sherwood A, et al. Effects of aerobic exercise on sexual functioning in depressed adults. Ment Health Phys Act 2: 23–28, 2009.
22. Katznelson L, Robinson MW, Coyle CL, Lee H, Farrell CE. Effects of modest testosterone supplementation and exercise for 12 weeks on body composition and quality of life in elderly men. Eur J Endocrinol 155: 867–875, 2006.
23. Keating SE, Johnson NA, Mielke GI, Coombes JS. A systematic review and meta‐analysis of interval training versus moderate‐intensity continuous training on body adiposity. Obes Rev 18: 943–964, 2017.
24. Kelley GA, Kelley KS, Tran ZV. Exercise and bone mineral density in men: A meta-analysis. J Appl Physiol 88: 1730–1736, 2000.
25. Kelly DM, Jones TH. Testosterone: A metabolic hormone in health and disease. J Endocrinol 217: R25–R45, 2013.
26. Lovell DI, Cuneo R, Wallace J, McLellan C. The hormonal response of older men to sub-maximum aerobic exercise: The effect of training and detraining. Steroids 77: 413–418, 2012.
27. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther 83: 713–721, 2003.
28. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. BMJ 339: b2535, 2009.
29. Moradi F. Changes of serum adiponectin and testosterone concentrations following twelve weeks resistance training in obese young men. Asian J Sports Med 6: 1–7, 2015.
30. Moskovic DJ, Araujo AB, Lipshultz LI, Khera M. The 20‐year public health impact and direct cost of testosterone deficiency in U.S. Men. J Sex Med 10: 562–569, 2013.
31. Ottenbacher KJ, Ottenbacher ME, Ottenbacher AJ, Acha AA, Ostir GV. Androgen treatment and muscle strength in elderly men: A meta-analysis. J Am Geriatr Soc 54: 1666–1673, 2006.
32. Penedo FJ, Dahn JR. Exercise and well-being: A review of mental and physical health benefits associated with physical activity
. Curr Opin Psychiatry 18: 189–193, 2005.
33. Rao PK, Boulet SL, Mehta A, et al. Trends in testosterone replacement therapy use from 2003 to 2013 among reproductive-age men in the United States. J Urol 197: 1121–1126, 2017.
34. Rhoden EL, Morgentaler A. Risks of testosterone-replacement therapy and recommendations for monitoring. N Engl J Med 350: 482–492, 2004.
35. Roberts CK, Croymans DM, Aziz N, Butch AW, Lee CC. Resistance training increases SHBG in overweight/obese, young men. Metabolism 62: 725–733, 2013.
36. Rogers M, Lemstra M, Bird Y, Nwankwo C, Moraros J. Weight-loss intervention adherence and factors promoting adherence: A meta-analysis. Patient Prefer Adherence 10: 1547–1559, 2016.
37. Schwartz LM, Woloshin S. Low “T” as in “template”: How to sell disease. JAMA Intern Med 173: 1460–1462, 2013.
38. Sedliak M, Zeman M, Buzgó G, et al. Morphological, molecular and hormonal adaptations to early morning versus afternoon resistance training. Chronobiol Int 35: 450–464, 2018.
39. Sharifi S, Monazzami A, Nikousefat Z, Heyrani A, Yari K. The acute and chronic effects of resistance training with blood flow restriction on hormonal responses in untrained young men: A comparison of frequency. Cell Mol Biol 66: 1–8, 2020.
40. Sheikholeslami-Vatani D, Siahkouhian M, Hakimi M, Ali-Mohammadi M. The effect of concurrent training order on hormonal responses and body composition in obese men. Sci Sports 30: 335–341, 2015.
41. Traish AM, Saad F, Feeley RJ, Guay A. The dark side of testosterone deficiency: III. Cardiovascular disease. J Androl 30: 477–494, 2009.
42. Traish AM, Saad F, Guay A. The dark side of testosterone deficiency: II. Type 2 diabetes and insulin resistance. J Androl 30: 23–32, 2008.
43. Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol 14: 135, 2014.
44. Zarrouf FA, Artz S, Griffith J, Sirbu C, Kommor M. Testosterone and depression: Systematic review and meta-analysis. J Psychiatr Pract 15: 289–305, 2009.