Journal of Men's Health. 2026; 22(7): 10-36. doi: 10.22514/jomh.2026.056
Systematic Review

Exercise improves sperm quality and pregnancy outcome by attenuating oxidative stress: a systematic review and meta-analysis

Roland Eghoghosoa Akhigbe1,2,*,, Blessing Monica Akindele1,2, Cecilia Adedeji Adegbola1,2, Ayoola Abimbola Oladipo1,2, Olaoluwa Ayoade Afolabi2,3, Precious Jesutofunmi Ashonibare1,2, Deborah Tolulope Mosadomi1,2, Victory Jesutoyosi Ashonibare2,4, Adetomiwa Ezekiel Adetomiwa2,5, Lois Bisola Osuntola1,2, Feyikemi Funmilayo Araba1,2, Tunmise Maryanne Akhigbe2,6

1Department of Physiology, Ladoke Akintola University of Technology, 210214 Ogbomoso, Oyo State, Nigeria

2Reproductive Biology and Toxicology Research Laboratory, Oasis of Grace Hospital, 230212 Osogbo, Osun State, Nigeria

3Department of Obstetrics and Gynecology, Bowen University Teaching Hospital, 212102 Ogbomoso, Oyo State, Nigeria

4Cardiovascular Regenerative Medicine & Tissue Engineering 3D Lab, Department of Cardiovascular Surgery and Research Group for Experimental Surgery, Medical Faculty, Heinrich Heine University, 40225 Düsseldorf, NRW, Germany

5Department of Physiology, Babcock University, 121103 Ilishan Remo, Ogun State, Nigeria

6Department of Agronomy, Osun State University, Ejigbo Campus, 232116 Ejigbo, Osun State, Nigeria

*Corresponding Author(s):reakhigbe@lautech.edu.ng (Roland Eghoghosoa Akhigbe)

History Submitted: 06 December 2025 | Accepted: 09 March 2026 | Published: 30 July 2026
Copyright:  ©2026 The Author(s). Published by MRE Press.
This is an open access article under the CC BY 4.0 license (https://creativecommons.org/licenses/by/4.0/).

Collapse table of contents

Abstract

Background: Male fertility contributes to the life quality of couples. Although exercise has been reported to mitigate the declining trend in sperm quality, the available reports are conflicting, and the precise influence on male fertility indices warrants further investigation. Hence, this study evaluated the effect of exercise on sperm quality, pregnancy outcomes, sex hormones, and markers of oxidative stress and inflammation in men. Methods: This is a systematic review and meta-analysis that evaluated exercise interventions, including aerobic, resistance, and combined modalities, on conventional and advanced semen parameters, and pregnancy and live birth rates. Results: Exercise considerably improved ejaculate volume, sperm count, motility, morphology, and sperm DNA integrity. Also, exercise enhanced seminal total antioxidant capacity (TAC) and decreased seminal reactive oxygen species, malondialdehyde, and pro-inflammatory cytokines levels. More so, pregnancy and live birth rates increased in the exercise group when compared with the control. Nonetheless, exercise did not show any considerable effect on testosterone, luteinizing hormone (LH), and follicle-stimulating hormone (FSH) levels. The observed effects varied by exercise type and duration, with moderate aerobic exercise showing beneficial effects. Moderate aerobic exercise enhances sperm quality and improves pregnancy outcomes by attenuating oxidative stress and inflammation. Conclusions: These findings show that moderate aerobic exercise is a promising non-pharmacological strategy in male infertility management. However, more clinical trials are required to optimize exercise protocols for reproductive benefits. The PROSPERO Registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD42024622166, CRD42024622166.

Keywords:ExerciseInfertilityLive birthMale infertilityPregnancyTestosterone
PDF(6.01 MB)|EndNote (RIS)|BibTeX|RefMan|RefWorks

Cite this article

Roland Eghoghosoa Akhigbe, Blessing Monica Akindele, Cecilia Adedeji Adegbola, Ayoola Abimbola Oladipo, Olaoluwa Ayoade Afolabi, Precious Jesutofunmi Ashonibare, Deborah Tolulope Mosadomi, Victory Jesutoyosi Ashonibare, Adetomiwa Ezekiel Adetomiwa, Lois Bisola Osuntola, Feyikemi Funmilayo Araba, Tunmise Maryanne Akhigbe. Exercise improves sperm quality and pregnancy outcome by attenuating oxidative stress: a systematic review and meta-analysis. Journal of Men's Health. 2026; 22(7): 10-36. doi: 10.22514/jomh.2026.056

1. Introduction

Infertility is the failure of a couple to get pregnant spontaneously despite a year of regular, unprotected intercourse [1]. About one in six people of reproductive age have infertility issues globally [2]. Although it is wrongly perceived as a condition caused by female pathology, male factors account for about 50% of the cases, either alone or in combination with female factors [3, 4]. While the prevalence of male infertility varies across the globe, it constitutes a major public health burden across high-, middle-, and low-income countries alike [5]. Previous studies have reported a declining trend in sperm quality across the globe, suggesting an increase in the prevalence of male infertility [6, 7].

Causes of male infertility include varicocele [8], cryptorchidism [9], testicular torsion/detorsion [10, 11], infections of the reproductive tract [12], hormonal imbalances, problems with sperm production, obstructions of ejaculatory duct and vas deferens, genetic disorders, and exposure to environmental toxicants (like heavy metal, pesticides, and irradiation) [13, 14, 15]. Additionally, lifestyle choices have a significant effect on male fertility. Smoking [16], excessive alcohol intake [17], use of illicit drugs [18, 19], psychological stress [20], and sedentary lifestyle [21] may impair sperm quality, contributing to the global burden of infertility.

There is increasing evidence linking physical activity/exercise with sperm quality. Physical activity maintains a healthy body weight, which is vital as obesity is linked to low testosterone levels and impaired sperm production [22, 23]. Also, exercise enhances natural cellular antioxidant levels, thus protecting sperm cells from damage [22]. However, there are conflicting reports. Some studies revealed that moderate-to-vigorous activity results in increased total sperm count, concentration [24, 25], semen volume, and sperm normal morphology [26], while some studies show that physical activity was not associated with semen quality [27, 28]. On the contrary, long-distance runners were observed to have reduced semen volume and sperm motility and morphology during training [29].

Despite the conflicting reports, there are scanty meta-analyses that have pooled the available data in the literature for a robust evaluation. Although the available meta-analyses on exercise and semen quality reveal the potential benefits of exercise on sperm quality [30, 31], these studies did not explore the likely impact on pregnancy outcomes and the associated mechanisms. Therefore, the present study investigated the impact of exercise on sperm quality and pregnancy outcomes. The impact on male sex hormones, oxidative stress, and pro-inflammatory cytokines was also explored.

2. Materials and methods

2.1 Study protocol and measured outcomes

This systematic review and meta-analysis was prospectively registered on PROSPERO (CRD42024622166) and synthesized data from prior publications examining the effects of physical activity or exercise on semen quality parameters, male sex hormones, pregnancy outcomes, and oxidative stress markers. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [32]. The PRISMA check list is provided as a Supplementary material 1.

Primary outcomes included exercise-induced changes in semen volume, sperm count, concentration, total and progressive motility, normal morphology, DNA fragmentation, and pregnancy and live birth rate. Secondary outcomes included levels of total and free testosterone, follicle-stimulating hormone (FSH), luteinizing hormone (LH), sex hormone-binding globulin (SHBG), cortisol, reactive oxygen species (ROS), malondialdehyde (MDA), total antioxidant capacity (TAC), interleukin (IL)-8, IL-6, IL-1β, and tumor necrosis factor (TNF)-α.

2.2 Eligibility criteria

This study addresses the research question: “Does exercise improve semen quality and pregnancy outcomes?”. It was structured according to the Population, Intervention, Comparator, Outcome, and Study designs (PICO) framework.

2.2.1 Inclusion criteria

Studies were included if they met all of the following PICO-based criteria:

ⅰ. Population: Males of reproductive age.

ⅱ. Intervention: Effects of exercise on semen quality and pregnancy outcomes.

ⅲ. Comparator: Age-matched controls with comparable sociodemographic characteristics, but no exercise involvement.

ⅳ. Outcomes: Reported sperm parameters (semen volume, sperm count, concentration, total motility, progressive motility, normal morphology, DNA fragmentation), pregnancy outcomes (pregnancy rate, live birth rate), male sex hormones (free and total testosterone, FSH, LH, SHBG), cortisol, oxidative stress markers (ROS, MDA, TAC), and inflammatory cytokines (IL-8, IL-6, IL-1β, TNF-α) as means with standard deviations (SDs) or data allowing SD derivation.

ⅴ. Study design: Controlled cross-sectional or longitudinal studies addressing the effect of exercise on semen quality, reproductive hormones, markers of oxidative stress and inflammation, and pregnancy outcomes.

2.2.2 Exclusion criteria

Studies were excluded for any of the following reasons:

ⅰ. Population: No participant age reported.

ⅱ. Intervention: Combined exercise with unspecified lifestyle changes, drugs, or in vitro/animal studies.

ⅲ. Comparator: Lacked non-exercising control group.

ⅳ. Outcomes: No quantifiable data (means/SDs or derivable) for variables of interest; self-reported reproductive outcomes.

ⅴ. Study design: Case studies, reviews, commentaries, letters, or editorials.

Additional exclusions included theses, conference abstracts, retracted papers, and preprints. No language restrictions were applied.

2.3 Search strategies

All authors performed a comprehensive search of Cochrane, Embase, PubMed, Scopus, and Web of Science databases up to 30 March 2025. Search terms combined Medical Subject Headings (MeSH) and Boolean operators: (“sperm” OR “semen” OR “sperm count” OR “sperm concentration” OR “sperm motility” OR “semen volume” OR “sperm DNA” OR “DNA fragmentation index (DFI)”) AND (“pregnancy” OR “live birth”) AND (“testosterone” OR “luteinizing hormone” OR “LH” OR “follicle-stimulating hormone” OR “FSH” OR “sex hormone-binding globulin” OR “SHBG” OR “cortisol”) AND (“oxidative stress” OR “ROS” OR “reactive oxygen species” OR “MDA” OR “malondialdehyde” OR “TAC” OR “total antioxidant capacity”) AND (“IL-1β” OR “IL-6” OR “IL-8” OR “TNF-α” OR “cytokine”).

Eligible studies were compiled, and citation chasing was used to identify additional relevant papers. Five investigators (CAA, PJA, VJA, AEA, TMA) screened abstracts and full texts for eligibility; disagreements were arbitrated by REA.

2.4 Quality assessment and data extraction

Quality of evidence (QoE), risk of bias (RoB), and certainty of evidence (CoE) were evaluated using the Cochrane RoB tool and Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group guidelines, as detailed in prior work [14, 32]. Assessments were performed by five investigators (BMA, AAO, OAA, CAA, DTM), with REA resolving disputes. Relevant data were extracted by BMA, AAO, OAA, CAA, and DTM.

2.5 Statistical analysis

CAA and REA conducted meta-analyses in Review Manager (RevMan) version 5.4.1 (Cochrane Informatics and Technology (IT) Services, London, UK). Random-effects models were applied for significant heterogeneity (p < 0.1 or I2 > 50%); fixed-effects models for low heterogeneity (p ≥ 0.1 or I2 ≤ 50%). Publication bias was inspected via funnel plots. Subgroup analyses targeted longitudinal studies; sensitivity analyses excluded the highest-weighted study, those with high RoB in ≥1 domain, or low CoE. Statistical significance was set at p < 0.05. Results are presented as standardized mean difference (SMD) or weighted mean difference (WMD) with 95% confidence interval (CI) and p values.

3. Results

3.1 Study identification and inclusion

A total of 37,472 articles were collected. After screening out the duplicates and those that were not eligible, only 16 articles [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48] were included (Fig. 1).

PRISMA flowchart for the strategic identification, screening, and inclusion of eligible studies.

Fig. 1.PRISMA flowchart for the strategic identification, screening, and inclusion of eligible studies.

3.2 Study characteristics

All included studies were controlled studies; six were cross-sectional, seven were non-randomized controlled longitudinal studies, and three were randomized controlled trials (RCTs). Four of the studies were from the USA, two from Australia, one each from Egypt and Colombia, two from Italy, and three each from Iran and Spain. Other details of the included studies are in Supplementary Table 1 (Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]).

3.3 Assessment of the quality of the studies

All the included studies had good QoE (≥5/10) (Table 1, Ref, [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). Three of the included studies [33, 34, 36] had unclear RoB on random sequence generation, allocation concealment, and incomplete outcome data domains, one had unclear risk on allocation concealment [35] and random sequence generation [39] only, and one on allocation concealment and incomplete outcome data [37]. Three of the studies [42, 44, 48] had high RoB on at least one domain, while others [40, 41, 43, 45, 46, 47] had low RoB on all domains (Fig. 2A, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). Overall, there was <75% low RoB on the random sequence generation and allocation concealment domain each, 75% on the incomplete outcome data domain, <100% but >75% on the selective reporting domain, and 100% on the blinding of participants/personnel, blinding of outcome assessment, and other bias domains each (Fig. 2B).

Table 1.Assessment of the quality of evidence of the eligible studies.
StudyStudy designStudy sizeMethod of measuring exposureMethod of measuring outcomeAnalysis with adjustmentTotal
Arce et al. [33], 1993202205/10
Bagatell and Bremner [34], 1990202206/10
De Souza et al. [36], 1994202206/10
Denham et al. [35], 2015202228/10
Fahrner and Hackney [37], 1997202206/10
Ismail et al. [38], 2023202206/10
Lalinde et al. [39], 2017202206/10
Lovell et al. [40], 2012202206/10
Luigi et al. [41], 2001212207/10
Maleki and Tartibian [42]222208/10
Maleki et al. [44], 2017222208/10
Maleki et al. [43], 2013222208/10
Montano et al. [45], 2022222208/10
Rosety et al. [46], 2017212207/10
Rosety-Rodriguez et al. [47], 2014212207/10
Vaamonde et al. [48], 2012202206/10
The risk of bias (RoB) of the included studies [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]; RoB summary showing each risk of bias item for each included study (A) and each risk of bias item presented as percentages across all included studies (B). +: Low RoB; −: High RoB.

Fig. 2.The risk of bias (RoB) of the included studies [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]; RoB summary showing each risk of bias item for each included study (A) and each risk of bias item presented as percentages across all included studies (B). +: Low RoB; −: High RoB.

In addition, there was high CoE in all eligible studies except in one [37] that had low CoE (Table 2, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]).

Table 2.Assessment of certainty of evidence of the eligible studies.
StudyInitial RatingDowngrading?Upgrading?Confidence in body of evidence
Arce et al. [33], 1993High1 (imprecision)1 (consistency)High
Bagatell and Bremner [34], 1990High1 (imprecision)1 (consistency)High
De Souza et al. [36], 1994High1 (imprecision)2 (consistency, dose-response)High
Denham et al. [35], 2015High--High
Fahrner and Hackney [37], 1997High2 (unexplained inconsistency, indirectness)-Low
Ismail et al. [38], 2023High--High
Lalinde et al. [39], 2017High-1 (rare outcome)High
Lovell et al. [40], 2012High--High
Luigi et al. [41], 2001High--High
Maleki and Tartibian [42]High1 (risk of bias)1 (rare outcome)High
Maleki et al. [44], 2017High1 (risk of bias)1 (rare outcome)High
Maleki et al. [43], 2013High--High
Montano et al. [45], 2022High--High
Rosety et al. [46], 2017High--High
Rosety-Rodriguez et al. [47], 2014High--High
Vaamonde et al. [48], 2012High--High

3.4 Quantitative analysis

3.4.1 Semen volume

Exercise significantly increased semen volume when compared with the control (−0.48 [−0.82, −0.15] 0.004), but there was inter-study heterogeneity (I2 = 91%; χ2p < 0.00001) (Fig. 3, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). Publication bias was observed (Supplementary Fig. 1).

Effect of exercise on semen volume [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 3.Effect of exercise on semen volume [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

The subgroup analysis of longitudinal studies also showed a significant increase in semen volume in subjects who exercised when compared with the control counterpart (−0.13 [0.23, −0.04] 0.007), and there was no inter-study heterogeneity (I2 = 29%; χ2p = 0.19). Subgroup analysis of the moderate aerobic exercise studies revealed that exercise significantly improved semen volume (−0.59 [−1.06, −0.12] 0.01), but there was also an inter-study heterogeneity (I2 = 93%; χ2p < 0.00001). In addition, the sensitivity analysis revealed that exercise involvement significantly increased semen volume in comparison with the control (−0.66 [−1.22, −0.11] 0.02), but there was inter-study heterogeneity (I2 = 91%; χ2p < 0.00001) (Fig. 3).

3.4.2 Sperm count

Fourteen studies were examined for the influence of exercise on sperm count, consisting of 438 controls and 430 subjects in the exercise group. Sperm count significantly increased in the exercise group in comparison with the control group (−8.22 [−13.50, −2.94] 0.002). However, there was inter-study heterogeneity (I2 = 73%; χ2p < 0.00001) (Fig. 4, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 2).

Effect of exercise on sperm count [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 4.Effect of exercise on sperm count [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Subgroup analysis of longitudinal studies of sperm count among subjects who exercised showed an insignificant increase in comparison with the control group (−4.69 [−9.41, 0.02] 0.05) and there was inter-study heterogeneity (I2 = 79%; χ2p < 0.0001). Subgroup analysis of the moderate aerobic exercise studies showed that exercise significantly improved sperm count (−6.80 [−12.54, −1.07] 0.02), but there was also an inter-study heterogeneity (I2 = 80%; χ2p < 0.00001). Also, the sensitivity analysis revealed that exercise significantly increased sperm concentration (−8.30 [−15.80, −0.81] 0.03), but there was inter-study heterogeneity (I2 = 56%; χ2p = 0.01) (Fig. 4).

3.4.3 Sperm concentration

The effect of exercise on sperm concentration was assessed in fifteen studies with 593 and 595 subjects in the control and exercise group, respectively. There were no significant changes in sperm concentration (−0.24 [−5.77, 5.30] 0.93), but there was inter-study heterogeneity (I2 = 78%; χ2p < 0.00001) (Fig. 5, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). Publication bias was observed (Supplementary Fig. 3).

Effect of exercise on sperm concentration [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 5.Effect of exercise on sperm concentration [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

However, subgroup analysis of longitudinal studies revealed that there was a significant increase in sperm concentration in exercising subjects compared with the control counterpart (−2.80 [−4.77, −0.82] 0.005) and there was no inter-study heterogeneity (I2 = 9%; χ2p = 0.36), while subgroup analysis of the moderate aerobic exercise studies revealed that exercise did not significantly alter sperm concentration (2.37 [−7.25, 11.99] 0.63), but there was also inter-study heterogeneity (I2 = 77%; χ2p < 0.00001). The sensitivity analysis revealed that sperm concentration was not significantly altered (8.42 [−4.06, 20.90] 0.19), but there was inter-study heterogeneity (I2 = 81%; χ2p < 0.00001) (Fig. 5).

3.4.4 Sperm total motility

Sperm total motility was assessed in six studies to evaluate the effect of exercise in 304 control subjects and 316 subjects who engaged in exercise. Exercise showed no significant impact on sperm total motility (−5.64 [−15.53, 4.24] 0.26), but inter-study heterogeneity was significant (I2 = 96%; χ2p < 0.00001) (Fig. 6, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). Publication bias was also noted (Supplementary Fig. 4).

Effect of exercise on sperm total motility [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Fig. 6.Effect of exercise on sperm total motility [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

The subgroup analysis of longitudinal studies showed an insignificant increase in sperm total motility among the exercise-group subjects when compared with the control group (−6.07 [−12.22, 0.09] 0.05), and there was inter-study heterogeneity (I2 = 84%; χ2p = 0.0002). Subgroup analysis of studies on the moderate aerobic exercise demonstrated that exercise considerably increased sperm total motility (−10.20 [−18.47, −1.93] 0.02), but there was also inter-study heterogeneity (I2 = 93%; χ2p < 0.00001). The sensitivity analysis showed that exercise had no significant impact on sperm total motility (−5.38 [−19.19, 8.43] 0.45), but there was inter-study heterogeneity (I2 = 96%; χ2p < 0.00001) (Fig. 6).

3.4.5 Sperm progressive motility

Twelve studies were evaluated for the outcome of exercise on sperm progressive motility, comprising 480 control subjects and subjects who exercised. The global analysis showed no significant changes, reflecting that exercise does not influence sperm progressive motility (−0.32 [−5.24, 4.59] 0.90). Nonetheless, inter-study heterogeneity was significant (I2 = 89%; χ2p < 0.00001) (Fig. 7, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). Publication bias was also significant (Supplementary Fig. 5).

Effect of exercise on sperm progressive motility [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 7.Effect of exercise on sperm progressive motility [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

On the contrary, the subgroup analysis of longitudinal studies demonstrated a substantial increase in sperm progressive motility in subjects who participated in exercise when compared with subjects in the control group (−4.85 [−9.46, −0.25] 0.04), and there was inter-study heterogeneity (I2 = 87%; χ2p < 0.00001). Subgroup analysis of the moderate aerobic exercise studies revealed that exercise did not considerably alter sperm progressive motility (−2.22 [−7.89, 3.45] 0.44), but there was also an inter-study heterogeneity (I2 = 90%; χ2p < 0.00001). Then again, the sensitivity analysis revealed that exercise involvement did not alter the sperm progressive motility (2.33 [−4.53, 9.19] 0.51), but there was an inter-study heterogeneity (I2 = 88%; χ2p < 0.00001) (Fig. 7).

3.4.6 Sperm normal morphology

Since sperm morphology was evaluated by the eligible studies using different criteria (as stated in Supplementary Table 1), SMD was used. The effect of exercise on sperm normal morphology was assessed in the intervention group, which revealed that there was a marginal increase in comparison with the control subjects (−0.24 [−0.50, 0.02] 0.07). Likewise, there was inter-study heterogeneity (I2 = 79%; χ2p < 0.00001) (Fig. 8, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). Also, a significant publication bias is shown (Supplementary Fig. 6).

Effect of exercise on sperm normal morphology [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 8.Effect of exercise on sperm normal morphology [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Furthermore, the subgroup analysis of longitudinal studies proved that sperm normal morphology was significantly increased in subjects who exercised when compared with the control counterpart (−0.35 [−0.48, −0.22] <0.00001), inter-study heterogeneity was significant (I2 = 44%; χ2p = 0.08). Subgroup analysis of the moderate aerobic exercise studies showed that exercise significantly improved sperm normal morphology (−0.41 [−0.64, −0.17] 0.0006), but there was also an inter-study heterogeneity (I2 = 58%; χ2p = 0.006). In contrast, the sensitivity analysis demonstrated that there was no considerable impact of exercise on sperm normal morphology (−0.14 [−0.57, 0.29] 0.52) but with an inter-study heterogeneity (I2 = 83%; χ2p < 0.00001) (Fig. 8).

3.4.7 Sperm DNA fragmentation

The influence of exercise on sperm DNA fragmentation was assessed in five studies incorporating 239 control and 230 subjects in the exercise group. The global analysis showed that sperm DNA fragmentation significantly increased in the subjects of the control group when compared with the subjects who exercised (0.60 [0.28, 0.93] 0.0003), but there was inter-study heterogeneity (I2 = 62%; χ2p = 0.03) (Fig. 9, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 7).

Effect of exercise on sperm DNA fragmentation [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 9.Effect of exercise on sperm DNA fragmentation [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Additionally, the subgroup analysis of longitudinal studies revealed that there was a substantial increase in sperm DNA fragmentation in the control group compared to those who exercised (0.71 [0.51, 0.90] <0.00001) and there was no inter-study heterogeneity (I2 = 13%; χ2p = 0.33). Subgroup analysis of the moderate aerobic exercise studies did not show any significant changes in sperm DNA fragmentation (0.50 [−0.15, 1.15] 0.13), but there was also an inter-study heterogeneity (I2 = 75%; χ2p = 0.02). The sensitivity analysis revealed that there were no significant changes in sperm DNA fragmentation in both groups (0.38 [−0.92, 1.68] 0.57), but there was inter-study heterogeneity (I2 = 86%; χ2p = 0.007) (Fig. 9).

3.4.8 Total testosterone

Twelve reports were studied for the effect of exercise on total testosterone. There were 194 subjects in the control group and 191 subjects who exercised. Exercise had no significant impact on total testosterone when compared with the control (−0.27 [−0.95, 0.42] 0.45), but inter-study heterogeneity was significant (I2 = 88%; χ2p < 0.00001) (Fig. 10, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 8).

Effect of exercise on total testosterone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Fig. 10.Effect of exercise on total testosterone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

No significant increase was seen in total testosterone in the subgroup analysis of longitudinal studies in subjects who exercised when compared with the control counterpart (−0.51 [−1.16, 0.15] 0.13), and there was substantial inter-study heterogeneity (I2 = 83%; χ2p < 0.0001). Subgroup analysis of the moderate aerobic exercise studies revealed that exercise did not significantly affect total testosterone level (−0.70 [−1.76, 0.36] 0.20), but there was also an inter-study heterogeneity (I2 = 91%; χ2p < 0.00001). In addition, the sensitivity analysis demonstrated that exercise had no considerable influence on total testosterone (−0.17 [−0.86, 0.51] 0.62), but also with inter-study heterogeneity (I2 = 85%; χ2p < 0.00001) (Fig. 10).

3.4.9 Free testosterone

A control and intervention group consisting of 114 and 110 subjects, respectively, from nine studies, were assessed for the effect of exercise on free testosterone. Exercise had no significant impact on free testosterone when both groups were compared (0.00 [−0.61, 0.61] 1.00), but there was inter-study heterogeneity (I2 = 78%; χ2p < 0.0001) (Fig. 11, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was no significant publication bias (Supplementary Fig. 9).

Effect of exercise on free testosterone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Fig. 11.Effect of exercise on free testosterone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

The subgroup analysis of longitudinal studies also showed no noteworthy increase in free testosterone in subjects who exercised when compared with the control group (−0.00 [−0.36, 0.35] 0.98) and there was no inter-study heterogeneity (I2 = 0%; χ2p = 0.42). Subgroup analysis of the moderate aerobic exercise studies also revealed that exercise did not considerably affect free testosterone level (−0.32 [−1.10, 0.46] 0.42), but there was also inter-study heterogeneity (I2 = 81%; χ2p < 0.0001). More so, the sensitivity analysis revealed exercise had no considerable impact on free testosterone (0.24 [−0.35, 0.82] 0.42); however, inter-study heterogeneity was significant (I2 = 66%; χ2p = 0.008) (Fig. 11).

3.4.10 Follicle-stimulating hormone

The effect of exercise on follicle-stimulating hormone was assessed in nine studies, including 152 controls and 150 subjects in the exercise group. Exercise had no substantial influence on FSH when both groups were compared (−0.21 [−0.60, 0.19] 0.31), but there was inter-study heterogeneity (I2 = 62%; χ2p = 0.007) (Fig. 12, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 10).

Effect of exercise on follicle-stimulating hormone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Fig. 12.Effect of exercise on follicle-stimulating hormone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Conversely, the subgroup analysis of longitudinal studies showed a significant increase in FSH in the control group than those who exercised (0.37 [0.07, 0.67] 0.02), and there was no inter-study heterogeneity (I2 = 4%; χ2p = 0.35). In contrast, subgroup analysis of the moderate aerobic exercise studies revealed that exercise did not significantly alter FSH (−0.17 [−0.64, 0.31] 0.49), but there was an inter-study heterogeneity (I2 = 69%; χ2p = 0.003). The sensitivity analysis also demonstrated that the effect of exercise on FSH was comparable across both groups (−0.30 [−0.81, 0.21] 0.25), but with inter-study heterogeneity (I2 = 65%; χ2p = 0.009) (Fig. 12).

3.4.11 Luteinizing hormone

Nine studies were evaluated for the influence of exercise on LH, consisting of 152 controls and 150 subjects in the exercise group. The global analysis revealed no significant impact of exercise on LH among those who exercised when compared with the control group (−0.05 [−0.28, 0.18] 0.66), and there was no inter-study heterogeneity (I2 = 15%; χ2p = 0.31) (Fig. 13, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was no significant publication bias (Supplementary Fig. 11).

Effect of exercise on luteinizing hormone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Fig. 13.Effect of exercise on luteinizing hormone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

The subgroup analysis of longitudinal studies also showed no significant effect of exercise on LH in subjects who exercised when compared with the control counterpart (0.15 [−0.15, 0.45] 0.33), and there was no inter-study heterogeneity (I2 = 0%; χ2p = 0.65). Subgroup analysis of the moderate aerobic exercise studies showed that exercise did not significantly alter LH (−0.06 [−0.30, 0.19] 0.66) and there was no inter-study heterogeneity (I2 = 27%; χ2p = 0.66). In addition, the sensitivity analysis revealed that exercise involvement did not significantly alter LH in comparison with the control (−0.21 [−0.49, 0.08] 0.16), and there was no inter-study heterogeneity (I2 = 2%; χ2p = 0.41) (Fig. 13).

3.4.12 Prolactin and estrogen

Five studies were assessed for the effect of exercise on prolactin, including 70 controls and 68 subjects in the exercise group. For the effect of exercise on estrogen, four studies were assessed, including 95 controls and 93 subjects in the exercise group. Exercise did not affect prolactin levels when compared with the control (−0.25 [−0.91, 0.41] 0.45), but there was inter-study heterogeneity (I2 = 69%; χ2p = 0.01) (Fig. 14A, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 12). Subgroup analysis of the moderate aerobic exercise studies showed that prolactin was not significantly affected by exercise (−0.07 [−0.95, 0.81] 0.88), but there was also an inter-study heterogeneity (I2 = 75%; χ2p = 0.02). Also, the sensitivity analysis revealed that exercise involvement significantly increased prolactin levels in comparison with the control (−0.51 [−0.97, −0.05] 0.03), but there was no inter-study heterogeneity (I2 = 0%; χ2p = 0.41) (Fig. 14A).

Exercise also did not have any effect on estrogen levels when compared with the control (0.23 [−0.06, 0.52] 0.12), and there was no inter-study heterogeneity (I2 = 11%; χ2p = 0.34) (Fig. 14B). There was a significant publication bias (Supplementary Fig. 13). In addition, the subgroup analysis of longitudinal studies for estrogen did not reveal any significant alteration in subjects who exercised when compared with the control counterpart (0.13 [−0.38, 0.65] 0.61), but there was inter-study heterogeneity (I2 = 60%; χ2p = 0.12). Subgroup analysis of the moderate aerobic exercise studies revealed that exercise did not significantly alter oestrogen level (0.19 [−0.11, 0.49] 0.22), and there was also no inter-study heterogeneity (I2 = 25%; χ2p = 0.26). However, the sensitivity analysis for estrogen levels revealed that exercise involvement did not affect levels of estrogen in comparison with the control (0.09 [−0.31, 0.49] 0.65) and there was no inter-study heterogeneity (I2 = 19%; χ2p = 0.29) (Fig. 14B).

Effect of exercise on (A) prolactin and (B) oestrogen [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Fig. 14.Effect of exercise on (A) prolactin and (B) oestrogen [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

3.4.13 Sex hormone-binding hormone

Four studies were assessed for the effect of exercise on sex hormone-binding globulin, including 44 controls and 42 subjects in the exercise group. Exercise did not significantly increase SHBG when compared with the control (−0.02 [−0.44, 0.41] 0.94), and there was no inter-study heterogeneity (I2 = 0%; χ2p = 0.97) (Fig. 15, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 14).

Effect of exercise on sex hormone-binding hormone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Fig. 15.Effect of exercise on sex hormone-binding hormone [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

The subgroup analysis of longitudinal studies also showed that exercise did no significantly alter SHBG (0.04 [−0.44, 0.52] 0.87) and there was no inter-study heterogeneity (I2 = 0%; χ2p = 0.99). Subgroup analysis of the moderate aerobic exercise studies revealed that exercise did not considerably alter SHBG (−0.05 [−0.53, 0.44] 0.85), and there was also no inter-study heterogeneity (I2 = 0%; χ2p = 0.91). In addition, the sensitivity analysis revealed that exercise involvement did not significantly increase SHBG in comparison with the control counterpart (0.06 [−0.54, 0.66] 0.84) and there was no inter-study heterogeneity (I2 = 0%; χ2p = 0.96) (Fig. 15).

3.4.14 Seminal ROS

Five studies were assessed for the effect of exercise on seminal fluid reactive oxygen species (ROS), including 310 control and 297 subjects in the exercise group. There was no significant impact of exercise on seminal ROS (72.96 [−165.65, 311.58] 0.55), but there was inter-study heterogeneity (I2 = 98%; χ2p < 0.00001) (Fig. 16, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 15).

Effect of exercise on seminal fluid reactive oxygen species (ROS) [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 16.Effect of exercise on seminal fluid reactive oxygen species (ROS) [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

There was a substantial reduction in the seminal ROS in subjects who exercised when compared with the control counterpart in the subgroup analysis of longitudinal studies (202.58 [5.09, 400.07] 0.04), and there was inter-study heterogeneity (I2 = 95%; χ2p < 0.00001). Subgroup analysis of the moderate aerobic exercise studies revealed that exercise significantly reduced seminal ROS (137.79 [80.67, 194.90] <0.00001), and there was also no inter-study heterogeneity (I2 = 0%; χ2p = 0.56). On the contrary, the sensitivity analysis revealed that exercise involvement did not significantly increase seminal ROS in comparison with the control (−123.15 [−670.96, 424.66] 0.66), but there was an inter-study heterogeneity (I2 = 99%; χ2p < 0.00001) (Fig. 16).

3.4.15 Seminal MDA

An assessment of five studies was carried out on the effect of exercise on seminal MDA concentration, including 310 control and 301 subjects in the exercise group. Exercise did not significantly affect seminal MDA concentration when compared with the control (0.02 [−0.17, 0.22] 0.81), but there was inter-study heterogeneity (I2 = 98%; χ2p < 0.00001) (Fig. 17, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 16).

Effect of exercise on seminal fluid malondialdehyde (MDA) concentration [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 17.Effect of exercise on seminal fluid malondialdehyde (MDA) concentration [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

However, the subgroup analysis of longitudinal studies showed a considerably higher seminal MDA concentration in subjects who were in the control group in comparison with those who exercised (0.17 [0.08, 0.26] 0.0003), and there was also inter-study heterogeneity (I2 = 86%; χ2p = 0.0007). Subgroup analysis of the moderate aerobic exercise studies also revealed that exercise markedly reduced seminal MDA (0.18 [0.03, 0.32] 0.02), but there was also an inter-study heterogeneity (I2 = 92%; χ2p = 0.0003). The sensitivity analysis revealed that exercise involvement did not significantly increase seminal MDA concentration in comparison with the control (−0.20 [−0.79, 0.39] 0.51), but there was an inter-study heterogeneity (I2 = 99%; χ2p < 0.00001) (Fig. 17).

3.4.16 Seminal TAC

Six studies were assessed for the effect of exercise on seminal TAC, including 436 controls and 438 subjects in the exercise group. Seminal TAC was significantly higher in the subjects who exercised compared with those in the control group (−1.71 [−2.85, −0.58] 0.003), and there was inter-study heterogeneity (I2 = 98%; χ2p < 0.00001) (Fig. 18, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was a significant publication bias (Supplementary Fig. 17).

Effect of exercise on seminal fluid total antioxidant capacity (TAC) [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

Fig. 18.Effect of exercise on seminal fluid total antioxidant capacity (TAC) [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval; HICT: High-intensity continuous exercise; HIIT: High-intensity interval exercise; MICT: Moderate-intensity continuous exercise.

The subgroup analysis of longitudinal studies showed that exercise had a marginal positive impact on seminal TAC in subjects who exercised when compared with the control counterpart (−0.29 [−0.59, 0.02] 0.06); inter-study heterogeneity was significant (I2 = 73%; χ2p = 0.01). Subgroup analysis of the moderate aerobic exercise studies demonstrated that exercise significantly improved seminal TAC (−5.86 [−8.43, −3.29] <0.00001), but there was also inter-study heterogeneity (I2 = 99%; χ2p < 0.00001). In contrast, the sensitivity analysis revealed that exercise involvement did not significantly improve seminal TAC in comparison with the control (−10.69 [−34.32, 12.95] 0.38), but there was an inter-study heterogeneity (I2 = 100%; χ2p < 0.00001) (Fig. 18).

3.4.17 Cortisol

A total of five studies were assessed for the effect of exercise on serum cortisol, consisting of 57 controls and 57 subjects in the exercise group. Exercise did not significantly impact serum cortisol when compared with the control (−0.32 [−1.02, 0.38] 0.37), but there was inter-study heterogeneity (I2 = 69%; χ2p = 0.01) (Fig. 19, Ref. [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]). There was no significant publication bias (Supplementary Fig. 18).

Effect of exercise on serum cortisol [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Fig. 19.Effect of exercise on serum cortisol [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. SD: Standard deviation; IV: Inverse variance; CI: Confidence interval.

Subgroup analysis of the moderate aerobic exercise studies revealed that cortisol level was not significantly altered by exercise (−0.33 [−1.23, 0.57] 0.47), but there was also an inter-study heterogeneity (I2 = 77%; χ2p = 0.005). In addition, the sensitivity analysis revealed that exercise involvement did not significantly increase serum cortisol in comparison with the control (−0.30 [−1.09, 0.50] 0.46), but there was an inter-study heterogeneity (I2 = 59%; χ2p = 0.09) (Fig. 19).

3.5 Qualitative analysis of inflammatory markers and pregnancy outcomes

It was observed that exercise attenuated pro-inflammatory cytokines (IL-1β, IL-6, IL-8, and TNF-α) when compared with the control group [44]. Also, exercise resulted in increased pregnancy rate as well as live birth rate when compared with the control [42].

4. Discussion

Findings from this study demonstrate that exercise improves sperm quality, especially semen volume, sperm count, concentration, progressive motility, and normal morphology. In addition, exercise reduced sperm DNA damage and increased pregnancy rate and live birth rate. These are associated with increased TAC and reduced ROS, MDA, and pro-inflammatory cytokines.

Conventional semen analysis assesses fundamental physiological sperm variables, such as semen volume, sperm count, concentration, motility, and morphology, to screen for male infertility, while advanced analyses, like DNA fragmentation and oxidative stress markers, give deeper insights into sperm function and sperm fertilization capacity. Sperm DNA fragmentation promotes sperm damage, and elevated sperm DNA fragmentation is associated with reduced pregnancy and increased adverse fertility outcomes [49, 50]. This study shows that exercise significantly improves basic sperm parameters and sperm DNA. This agrees with a previous study that reported reduced sperm DNA fragmentation in men who are on an exercise regimen and antioxidants [51]. Also, it agrees with previous studies that demonstrated improved sperm DNA in men on a weight loss program [52, 53]. However, this does not agree with the study of Håkonsen and his colleagues [54], who did not observe any significant improvement in sperm chromatin structure in men undergoing weight control and exercise. Although studies reporting the impact of exercise on pregnancy rate and live birth rate in couples whose husbands were on an exercise regimen are scanty, the present findings that exercise improved pregnancy rate and live birth rates may be attributed to the improvement of sperm quality in men who are engaged in exercise since sperm DNA integrity is positively correlated with improved pregnancy rate and live birth rate [49, 50]. This agrees with an earlier study that reported increased pregnancy rate and live birth rates [52], but not with another that showed no association between exercise and clinical pregnancy or live birth rates [24]. The observed variation may be due to the type, intensity, and duration of the exercise regimen.

In an attempt to explore the role of male sex hormones and stress in the link between exercise and sperm quality, the impact of exercise on testosterone, LH, FSH, SHBG, oestrogen, prolactin, and cortisol levels was investigated. SHBG regulates the bioavailability of testosterone [55], which, at an optimal level, maintains spermatogenesis and sperm protein expression, thus enhancing sperm quality [56]. LH stimulates the Leydig cells to produce testosterone, while FSH promotes Sertoli cell function that is essential for sperm maturation [57]. Elevated oestrogen disrupts hormonal balance, impairing spermatogenesis and lowering sperm quality [58], while elevated prolactin suppresses gonadotropin-releasing hormone (GnRH), LH, FSH, and testosterone, leading to hypogonadism and poor semen quality [59]. Overall, physiological levels of these hormones are essential for spermatogenesis, sperm quality, and overall fertility potential. Remarkably, the present study shows that exercise does not significantly alter male sex hormones and cortisol levels. Although exercise was found to increase prolactin level following a global analysis of the included studies, subgroup analysis of the moderate aerobic exercise revealed no significant changes in the prolactin level; this suggests that the increased prolactin level may be due to the type and intensity of the exercise engaged in.

Sperm DNA fragmentation is mediated by seminal oxidative stress [60, 61]. In addition, seminal ROS level and MDA generation are positively correlated with adverse pregnancy outcomes [62]. Cellular antioxidant levels and TAC influence incident oxidative stress [63], thus modulating sperm DNA fragmentation and pregnancy outcomes. The present study showed that exercise elevated seminal TAC and reduced seminal ROS and MDA. This infers that exercise, especially moderate aerobic exercise, improves sperm quality and pregnancy outcomes by downregulating seminal ROS and MDA generation and enhancing TAC. Oxidative stress is a cause and consequence of inflammation [64]; thus, seminal oxidative stress may promote seminal inflammation and vice versa. The present findings that exercise reduces pro-inflammatory cytokines in association with improved sperm quality and pregnancy outcomes, and attenuated oxidative stress, suggest that exercise protects sperm cells by downregulating cytokine accumulation via suppression of seminal oxidative stress.

Despite the substantial facts presented in the present study, it has some limitations. First, there are scanty RCTs that were eligible for inclusion. This possibly affected the quality of the included studies and the study outcome. Also, most of the analyses showed publication bias. This may be due to the varying degree of exercise intensity and duration, study design, and different methods of determining the endpoints. Nonetheless, this study seems to be the first meta-analysis on the effect of exercise on semen quality and pregnancy outcomes with an exploration of the possible role of inflammation and oxidative stress.

5. Conclusions

In conclusion, exercise improves sperm quality and pregnancy outcomes by downregulating seminal inflammation and oxidative stress. This may be independent of testosterone and cortisol. However, well-designed RCTs should be conducted to investigate the role of graded aerobic and resistance exercises with varying durations on sperm quality. This will offer a useful tool in the non-pharmacological and non-surgical management of male infertility.

Availability of data and materials

The data used to support the findings of the present study are available from the corresponding author upon request.

Author contributions

REA—Conceptualized and designed the study, supervised the study, validated the data, participated in fund acquisition, developed the methodology, participated in the investigation, participated in project administration, wrote the original draft, reviewed, edited and approved the final draft. BMA and CAA—developed the methodology, participated in the investigation, participated in project administration, wrote the original draft, reviewed, edited and approved the final draft. AAO, OAA, PJA, DTM, VJA, AEA, LBO, FFA—participated in the investigation, participated in project administration, reviewed, edited and approved the final draft. TMA—participated in fund acquisition, developed the methodology, participated in the investigation, participated in project administration, wrote the original draft, reviewed, edited and approved the final draft. All authors read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Acknowledgment

Authors are grateful to the members of Reproductive Biology and Toxicology Research Laboratory (ReBTReL), Oasis of Grace Hospital, Nigeria and Global for the support provided in the course of the study and manuscript preparation.

Funding

This research received no external funding.

Conflict of interest

The authors have no conflicts of interest. The authors declare no conflict of interest. Roland Eghoghosoa Akhigbe is serving as one of the Editorial Board members of this journal. We declare that Roland Eghoghosoa Akhigbe had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to MS.

Supplementary material

Supplementary material associated with this article can be found, in the online version, at https://oss.jomh.org/files/article/2082723457680719872/attachment/Supplementary%20material.zip.

References

Akhigbe RE, Hamed MA, Dutta S, Sengupta P. Influence of ejaculatory abstinence period on semen quality of 5165 normozoospermic and oligozoospermic Nigerian men: a retrospective study. Health Science Reports. 2022; 5: e722.

[Google Scholar]

Liang Y, Huang J, Zhao Q, Mo H, Su Z, Feng S, et al. Global, regional, and national prevalence and trends of infertility among individuals of reproductive age (15–49 years) from 1990 to 2021, with projections to 2040. Human Reproduction. 2025; 40: 529–544.

[Google Scholar]

Agarwal A, Majzoub A, Baskaran S, Panner Selvam MK, Cho CL, Henkel R, et al. Sperm DNA fragmentation: a new guideline for clinicians. The World Journal of Men’s Health. 2020; 38: 412–471.

[Google Scholar]

Ajayi AF, Akhigbe RE. The physiology of male reproduction: impact of drugs and their abuse on male fertility. Andrologia. 2020; 52: e13672.

[Google Scholar]

Barratt CLR, Björndahl L, De Jonge CJ, Lamb DJ, Osorio Martini F, McLachlan R, et al. The diagnosis of male infertility: an analysis of the evidence to support the development of global WHO guidance—challenges and future research opportunities. Human Reproduction Update. 2017; 23: 660–680.

[Google Scholar]

Rosa-Villagrán L, Barrera N, Montes J, Riso C, Sapiro R. Decline of semen quality over the last 30 years in Uruguay. Basic and Clinical Andrology. 2021; 31: 8.

[Google Scholar]

Li Y, Lu T, Wu Z, Wang Z, Yu T, Wang H, et al. Trends in sperm quality by computer-assisted sperm analysis of 49,189 men during 2015–2021 in a fertility center from China. Frontiers in Endocrinology. 2023; 14: 1194455.

[Google Scholar]

Çayan S, Pinggera GM, Alipour H, Altay B, Shah R, Giulioni C, et al. The effects of varicocele repair on testicular sperm retrieval, sperm recovery in the ejaculate and clinical pregnancy rates in non-obstructive azoospermic men with clinical varicocele: a systematic review and meta-analysis. The World Journal of Men’s Health. 2025; 43: e44.

[Google Scholar]

Zamani P, Hemati Z, Kelishadi R, Kolahdozan S, Dianatinasab M, Keikha M. Association between anogenital distance as a noninvasive index in the diagnosis and prognosis of reproductive disorder: a systematic review. International Journal of Reproductive Biomedicine. 2023; 21: 599–618.

[Google Scholar]

Hu S, Guo M, Xiao Y, Li Y, Luo Q, Li Z, et al. Mapping trends and hotspot regarding testicular torsion: a bibliometric analysis of global research (2000–2022). Frontiers in Pediatrics. 2023; 11: 1121677.

[Google Scholar]

Akhigbe RE, Odetayo AF, Akhigbe TM, Hamed MA, Ashonibare PJ. Pathophysiology and management of testicular ischemia/reperfusion injury: lessons from animal models. Heliyon. 2024; 10: e27760.

[Google Scholar]

Ashonibare VJ, Ashonibare PJ, Akhigbe TM, Akhigbe RE. SARS-CoV-2 impairs male fertility by targeting semen quality and testosterone level: a systematic review and meta-analysis. PLOS ONE. 2024; 19: e0307396.

[Google Scholar]

Adegbola CA, Akhigbe TM, Adeogun AE, Tvrdá E, Pizent A, Akhigbe RE. A systematic review and meta-analysis of the impact of triclosan exposure on human semen quality. Frontiers in Toxicology. 2024; 6: 1469340.

[Google Scholar]

Akhigbe RE, Akhigbe TM, Adegbola CA, Oyedokun PA, Adesoye OB, Adeogun AE. Toxic impacts of arsenic bioaccumulation on urinary arsenic metabolites and semen quality: a systematic and meta-analysis. Ecotoxicology and Environmental Safety. 2024; 281: 116645.

[Google Scholar]

Akhigbe RE, Oyedokun PA, Akhigbe TM, Adenike S, Oladipo AA, Hughes JR. Does pyrethroid exposure lower human semen quality? A systematic review and meta-analysis. Frontiers in Toxicology. 2024; 6: 1395010.

[Google Scholar]

Osadchuk L, Kleshchev M, Osadchuk A. Effects of cigarette smoking on semen quality, reproductive hormone levels, metabolic profile, zinc and sperm DNA fragmentation in men: results from a population-based study. Frontiers in Endocrinology. 2023; 14: 1255304.

[Google Scholar]

Amor H, Hammadeh ME, Mohd I, Jankowski PM. Impact of heavy alcohol consumption and cigarette smoking on sperm DNA integrity. Andrologia. 2022; 54: e14434.

[Google Scholar]

Akhigbe RE, Afolabi OA, Ajayi AF. L-Arginine abrogates maternal and pre-pubertal codeine exposure-induced impaired spermatogenesis and sperm quality by modulating the levels of mRNA encoding spermatogenic genes. Frontiers in Endocrinology. 2023; 14: 1180085.

[Google Scholar]

Hamed MA, Ekundina VO, Akhigbe RE. Psychoactive drugs and male fertility: impacts and mechanisms. Reproductive Biology and Endocrinology. 2023; 21: 69.

[Google Scholar]

Odetayo AF, Akhigbe RE, Bassey GE, Hamed MA, Olayaki LA. Impact of stress on male fertility: role of gonadotropin inhibitory hormone. Frontiers in Endocrinology. 2024; 14: 1329564.

[Google Scholar]

Zańko A, Martynowicz I, Citko A, Konopka P, Paszko A, Pawłowski M, et al. The influence of lifestyle on male fertility in the context of insulin resistance—identification of factors that influence semen quality. Journal of Clinical Medicine. 2024; 13: 2797.

[Google Scholar]

Adelowo OE, Akindele BM, Adegbola CA, Oyedokun PA, Akhigbe TM, Akhigbe RE. Unraveling the complexity of the impact of physical exercise on male reproductive functions: a review of both sides of a coin. Frontiers in Physiology. 2024; 15: 1492771.

[Google Scholar]

Wei R, Lang ZT, Wang EH. Impact and action mechanisms of obesity on male reproductive function: an update. National Journal of Andrology. 2025; 31: 357–362. (In Chinese)

[Google Scholar]

Gaskins AJ, Afeiche MC, Hauser R, Williams PL, Gillman MW, Tanrikut C, et al. Paternal physical and sedentary activities in relation to semen quality and reproductive outcomes among couples from a fertility center. Human Reproduction. 2014; 29: 2575–2582.

[Google Scholar]

Gaskins AJ, Mendiola J, Afeiche M, Jørgensen N, Swan SH, Chavarro JE. Physical activity and television watching in relation to semen quality in young men. British Journal of Sports Medicine. 2015; 49: 265–270.

[Google Scholar]

Li Y, Li Y, Zhou N, Han X, Ma M, Li L, et al. Socio-psycho-behavioural factors associated with male semen quality in China: results from 1346 healthy men in Chongqing. Journal of Family Planning and Reproductive Health Care. 2013; 39: 102–110.

[Google Scholar]

Jóźków P, Mędraś M, Lwow F, Zagrodna A, Słowińska-Lisowska M. Associations between physical activity and semen quality in young healthy men. Fertility and Sterility. 2017; 107: 373–378.e2.

[Google Scholar]

Mínguez-Alarcón L, Chavarro JE, Mendiola J, Gaskins AJ, Torres-Cantero AM. Physical activity is not related to semen quality in young healthy men. Fertility and Sterility. 2014; 102: 1103–1109.

[Google Scholar]

Jensen CE, Wiswedel K, McLoughlin J, van der Spuy Z. Prospective study of hormonal and semen profiles in marathon runners. Fertility and Sterility. 1995; 64: 1189–1196.

[Google Scholar]

Ibanez-Perez J, Santos-Zorrozua B, Lopez-Lopez E, Matorras R, Garcia-Orad A. An update on the implication of physical activity on semen quality: a systematic review and meta-analysis. Archives of Gynecology and Obstetrics. 2019; 299: 901–921.

[Google Scholar]

Song W, Peng Y, Jiang Z, Quan Z. Effectiveness of exercise interventions on sperm quality: a systematic review and network meta-analysis. Frontiers in Endocrinology. 2025; 16: 1537271.

[Google Scholar]

Akhigbe TM, Fidelis FB, Adekunle AO, Ashonibare VJ, Akorede BA, Shuaibu MS, et al. Does coenzyme Q10 improve semen quality and circulating testosterone level? A systematic review and meta-analysis of randomized controlled trials. Frontiers in Pharmacology. 2025; 15: 1497930.

[Google Scholar]

Arce JC, De Souza MJ, Pescatello LS, Luciano AA. Subclinical alterations in hormone and semen profile in athletes. Fertility and Sterility. 1993; 59: 398–404.

[Google Scholar]

Bagatell CJ, Bremner WJ. Sperm counts and reproductive hormones in male marathoners and lean controls. Fertility and Sterility. 1990; 53: 688–692.

[Google Scholar]

Denham J, O’Brien BJ, Harvey JT, Charchar FJ. Genome-wide sperm DNA methylation changes after 3 months of exercise training in humans. Epigenomics. 2015; 7: 717–731.

[Google Scholar]

De Souza MJ, Arce JC, Pescatello LS, Scherzer HS, Luciano AA. Gonadal hormones and semen quality in male runners. International Journal of Sports Medicine. 1994; 15: 383–391.

[Google Scholar]

Fahrner CL, Hackney AC. Effects of endurance exercise on free testosterone concentration and the binding affinity of sex hormone binding globulin (SHBG). International Journal of Sports Medicine. 1998; 19: 12–15.

[Google Scholar]

Ismail AM, Abdelghany A, Atef H. Response of testosterone and semen parameters to a 14-week aerobic training in sedentary obese men with hyperglycaemia. Physiotherapy Quarterly. 2023; 31: 28–33.

[Google Scholar]

Lalinde-Acevedo PC, Mayorga-Torres BJ, Agarwal A, Du Plessis SS, Ahmad G, Cadavid ÁP, et al. Physically active men show better semen parameters than their sedentary counterparts. International Journal of Fertility & Sterility. 2017; 11: 156–165.

[Google Scholar]

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. 2012; 77: 413–418.

[Google Scholar]

Luigi L, Gentile V, Pigozzi F, Parisi A, Giannetti D, Romanelli F. Physical activity as a possible aggravating factor for athletes with varicocele: impact on the semen profile. Human Reproduction. 2001; 16: 1180–1184.

[Google Scholar]

Maleki B, Tartibian B. Combined aerobic and resistance exercise training for improving reproductive function in infertile men: a randomized controlled trial. Applied Physiology, Nutrition, and Metabolism. 2017; 42: 1293–1306.

[Google Scholar]

Maleki B, Tartibian B, Eghbali M, Asri‐Rezaei S. Comparison of seminal oxidants and antioxidants in subjects with different levels of physical fitness. Andrology. 2013; 1: 607–614.

[Google Scholar]

Maleki B, Tartibian B, Chehrazi M. The effects of three different exercise modalities on markers of male reproduction in healthy subjects: a randomized controlled trial. Reproduction. 2017; 153: 157–174.

[Google Scholar]

Montano L, Ceretti E, Donato F, Bergamo P, Zani C, Viola GC, et al. Effects of a lifestyle change intervention on semen quality in healthy young men living in highly polluted areas in Italy: the FASt randomized controlled trial. European Urology Focus. 2022; 8: 351–359.

[Google Scholar]

Rosety MÁ, Díaz A, Rosety JM, Pery MT, Brenes-Martín F, Bernardi M, et al. Exercise improved semen quality and reproductive hormone levels in sedentary obese adults. Nutrición Hospitalaria. 2017; 34: 608–612.

[Google Scholar]

Rosety-Rodriguez M, Rosety JM, Fornieles G, Rosety MA, Diaz AJ, Rosety I, et al. Home-based treadmill training improved seminal quality in adults with type 2 diabetes. Actas Urológicas Españolas. 2014; 38: 589–593.

[Google Scholar]

Vaamonde D, Da Silva-Grigoletto ME, García-Manso JM, Barrera N, Vaamonde-Lemos R. Physically active men show better semen parameters and hormone values than sedentary men. European Journal of Applied Physiology. 2012; 112: 3267–3273.

[Google Scholar]

Lewis SE, Aitken RJ, Conner SJ, De Iuliis G, Evenson DP, Henkel R, et al. The impact of sperm DNA damage in assisted conception and beyond: recent advances in diagnosis and treatment. Reproductive Biomedicine Online. 2013; 27: 325–337.

[Google Scholar]

McQueen DB, Zhang J, Robins JC. Sperm DNA fragmentation and recurrent pregnancy loss: a systematic review and meta-analysis. Fertility and Sterility. 2019; 112: 54–60.e3.

[Google Scholar]

Humaidan P, Haahr T, Povlsen BB, Kofod L, Laursen RJ, Alsbjerg B, et al. The combined effect of lifestyle intervention and antioxidant therapy on sperm DNA fragmentation and seminal oxidative stress in IVF patients: a pilot study. International Brazilian Journal of Urology. 2022; 48: 131–156.

[Google Scholar]

Faure C, Dupont C, Baraibar MA, Ladouce R, Cedrin-Durnerin I, Wolf JP, et al. In subfertile couple, abdominal fat loss in men is associated with improvement of sperm quality and pregnancy: a case-series. PLOS ONE. 2014; 9: e86300.

[Google Scholar]

Mir J, Franken D, Andrabi SW, Ashraf M, Rao K. Impact of weight loss on sperm DNA integrity in obese men. Andrologia. 2018; 50: e12957.

[Google Scholar]

Håkonsen LB, Thulstrup AM, Aggerholm AS, Olsen J, Bonde JP, Andersen CY, et al. Does weight loss improve semen quality and reproductive hormones? Results from a cohort of severely obese men. Reproductive Health. 2011; 8: 24.

[Google Scholar]

Winters SJ. SHBG and total testosterone levels in men with adult onset hypogonadism: what are we overlooking? Clinical Diabetes and Endocrinology. 2020; 6: 17.

[Google Scholar]

Grande G, Barrachina F, Soler-Ventura A, Jodar M, Mancini F, Marana R, et al. The role of testosterone in spermatogenesis: lessons from proteome profiling of human spermatozoa in testosterone deficiency. Frontiers in Endocrinology. 2022; 13: 852661.

[Google Scholar]

Oyedokun PA, Akhigbe RE, Ajayi LO, Ajayi AF. Impact of hypoxia on male reproductive functions. Molecular and Cellular Biochemistry. 2023; 478: 875–885.

[Google Scholar]

Rato L, Sousa AC. The impact of endocrine-disrupting chemicals in male fertility: focus on the action of obesogens. Journal of Xenobiotics. 2021; 11: 163–196.

[Google Scholar]

Gangwar PK, Sankhwar SN, Pant S, Krishna A, Singh BP, Mahdi AA, et al. Increased gonadotropins and prolactin are linked to infertility in males. Bioinformation. 2020; 16: 176–182.

[Google Scholar]

Akhigbe RE. Oxidative stress and male fertility. Frontiers in Reproductive Health. 2025; 7: 1657872.

[Google Scholar]

Vaughan DA, Tirado E, Garcia D, Datta V, Sakkas D. DNA fragmentation of sperm: a radical examination of the contribution of oxidative stress and age in 16 945 semen samples. Human Reproduction. 2020; 35: 2188–2196.

[Google Scholar]

Agarwal A, Parekh N, Panner Selvam MK, Henkel R, Shah R, Homa ST, et al. Male oxidative stress infertility (MOSI): proposed terminology and clinical practice guidelines for management of idiopathic male infertility. The World Journal of Men’s Health. 2019; 37: 296–312.

[Google Scholar]

Amirkhizi F, Hamedi-Shahraki S, Rahimlou M. Dietary total antioxidant capacity is associated with lower disease severity and inflammatory and oxidative stress biomarkers in patients with knee osteoarthritis. Journal of Health, Population and Nutrition. 2023; 42: 104.

[Google Scholar]

Akhigbe RE, Dutta S, Hamed MA, Ajayi AF, Sengupta P, Ahmad G. Viral infections and male infertility: a comprehensive review of the role of oxidative stress. Frontiers in Reproductive Health. 2022; 4: 782915.

[Google Scholar]