Effect of Antioxidants on Sperm Quality Parameters in Subfertile Men: A Systematic Review and Network Meta-Analysis of Randomized Controlled Trials

ABSTRACT

Antioxidant supplementation has been identified as an important intervention for subfertile men. However, the effectiveness of different antioxidants in improving sperm quality remains unclear. In this study, a network meta-analysis (NMA) was designed to evaluate the effects of different antioxidants on sperm quality parameters in subfertile men. Published randomized controlled trials (RCTs) of antioxidants in subfertile men were searched in the PubMed, Embase, and Cochrane Library databases from inception to 31 January, 2021. Eight antioxidants (folic acid, zinc, vitamin E, carnitine, selenium, coenzyme q10 [CoQ10], N-acetylcysteine, and vitamin C) and a placebo (control) were included in our study. A Bayesian NMA with random effects was performed for each outcome (sperm concentration, sperm motility, and sperm morphology); the surface under the cumulative ranking curves (SUCRAs) for the effectiveness of each intervention was applied to identify the optimal intervention. Eighteen studies with 1790 subfertile men were included in the study. CoQ10 elicited a significant increase in sperm concentration (mean difference [MD] = 5.95; 95% CI: 0.05, 10.79) compared with the placebo; it achieved the highest rank in efficacy among all the antioxidants (SUCRA: 79.4%). With regard to sperm motility, carnitine (MD = 12.43; 95% CI: 4.07, 20.26) and CoQ10 (MD = 7.33; 95% CI: 0.35, 14.17) showed significant beneficial effects compared with the placebo; the efficacy of carnitine was the highest among all the antioxidants (SUCRA: 88.7%). With regard to sperm morphology, the efficacy of vitamin C tended to be the highest (SUCRA: 93.6%), although it did not show a significant beneficial effect (MD = 7.73; 95% CI: –0.94, 16.33) compared with the placebo. Overall, for subfertile men, CoQ10 and carnitine interventions showed better effectiveness in increasing sperm concentration and sperm motility, respectively.

Statement of Significance: The effectiveness of different antioxidants in improving sperm quality in subfertile men remains unclear. To the best of our knowledge, this is the first study to compare the effectiveness of different antioxidants on sperm quality parameters in subfertile men.

Introduction

Subfertility generally describes any form or grade of reduced fertility among couples unsuccessfully trying to conceive (1). Around the world, ∼15% of all couples are affected by this problem; the male factor is estimated to account for half of the cases (2). Currently, semen analysis is the most common method of assessing male fertility. Abnormal semen quality is determined using quantitative (azoospermia, oligospermia) and qualitative (asthenozoospermia, teratozoospermia) analyses (3). A previous global meta-analysis reported that human semen quality has significantly declined in the last 40 y (4). Therefore, the search for treatments to improve semen quality in subfertile men has been a public health concern.

Spermatogenesis is a complex series of molecular and biochemical procedures. Thus, the explicit pathophysiology of subfertile men remains elusive (5). Oxidative stress triggered by reactive oxygen species (ROS), a common pathology in subfertile men, is associated with 30–80% of cases (6). ROS are unstable and highly reactive oxygen-derived molecules produced by cellular metabolism. They are important signaling mediators in biological systems and play a key role in sperm fertilization (7). However, ROS concentrations become elevated with abnormal production and excessive accumulation, thereby overwhelming the cellular antioxidant defense system. As a result, oxidative stress arises and affects sperm development, function, and DNA structure (8). In addition, previous studies have observed that spermatozoa of oligospermic men remain within the epididymis for a longer period, consequently increasing the risk of ROS-mediated damage (9). Considering the enormous economic cost generated by assisted reproductive techniques, antioxidant supplementation has been an important and widely used alternative treatment to improve sperm quality by reducing oxidative stress in subfertile men. Furthermore, the wide availability of oral antioxidants is supported by their excellent safety and bioavailability (10).

Over the past decades, many studies have demonstrated the importance of antioxidants (e.g. carnitine, vitamin E, folate, and zinc) for sperm protection (7, 10–12). Antioxidant supplementation has been reported to improve semen parameters in subfertile men and further increase the probability of spontaneous pregnancy, while simultaneously controlling the cost of pregnancy (13). Antioxidant therapies for subfertile men were summarized in a Cochrane meta-analysis and subsequently updated in 2019. The Cochrane meta-analysis concluded that antioxidant therapy may improve semen parameters as well as increase the clinical pregnancy rate (14). This conclusion provides evidence for the effective treatment of subfertile men with antioxidants; however, there are various types of antioxidants. The Cochrane meta-analysis lacked conclusions on the effects of different antioxidants, as studies comparing different antioxidant agents remained scarce. Thus, a network meta-analysis (NMA) is deemed to be an attractive approach for novel evidence synthesis. When direct evidence is insufficient, NMA is a useful tool that can synthesize indirect and direct comparisons as well as evaluate the effect of interventions in 1 statistical model (15). Therefore, in this study, the effects of different antioxidants on sperm quality parameters in subfertile men were evaluated using a systematic review and NMA.

Methods

This NMA was registered in the International Prospective Register of Systematic Reviews as CRD42021234266. The results are reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Extension for reporting NMAs. (16).

Search strategy

The PubMed, Embase, and Cochrane Library databases were searched from inception to 31 January, 2021. According to a Cochrane meta-analysis, antioxidants for male subfertility (14) and 9 substances with direct antioxidant action (arginine, folic acid, zinc, vitamin E, carnitine, selenium, coenzyme q10 [CoQ10], N-acetylcysteine, and vitamin C) were searched. No language restrictions were applied in the systematic database searches. The details of the literature search strategy are presented in Supplemental Appendix 1. All eligible studies were assessed independently by 2 reviewers (LS and JZ). Disagreements were resolved by another reviewer (Y-zJ).

Selection of studies

The inclusion criteria for this study were as follows: 1) randomized controlled trials (RCTs) with parallel or crossover design; however, crossover trials were only used for the first-phase data; 2) subfertile men with abnormal semen parameters (oligozoospermia, asthenozoospermia, or teratozoospermia); 3) interventions for single-type antioxidant supplementation versus placebo or other types of antioxidants; and 4) outcomes including sperm quality parameters (sperm concentration, sperm motility, and sperm morphology). Furthermore, the exclusion criterion was: cointervention with other fertility-enhancing drugs being inapplicable to all intervention arms.

Data extraction

Two reviewers (LS and JZ) independently extracted the following data from each included study: name of first author, publication year, country, sample size, mean age, mean BMI, interventions (the type of antioxidants or placebo), duration of treatment, sperm concentration, sperm motility, and sperm morphology. The outcome data included postintervention values with corresponding SDs.

Risk of bias assessment

Two reviewers (LS and JZ) independently evaluated the quality of each included study using the Cochrane Collaboration Recommendations assessment tools (17). We assessed the following sources of bias: random sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other bias. Studies were considered as follows: 1) high risk of bias (if ≥1 item was rated with a high risk); 2) low risk of bias (if ≥3 out of a maximum of 7 items were rated as low risk, and no item was rated with a high risk); and 3) moderate/unclear risk of bias (all other studies) (18).

Data synthesis and analysis

For all included studies, direct comparisons between different interventions were represented using a network plot for each outcome (19). The size of the node is proportional to the number of studies involving each specific intervention; the thickness of the line is proportional to the number of comparisons included in the network.

We performed a direct pairwise meta-analysis with a random-effects model using Stata15 software (STATA Corporation), in view of the random-effects model providing more conservative results (20). For each outcome, the pooled results were presented as mean differences (MDs) with 95% CIs. When the same type of antioxidant was divided into subgroups in the original studies, we calculated the pooled mean and SD, as suggested by the Cochrane Handbook (21). When only medians and IQRs were provided, the means and SDs were calculated using the methods proposed by Luo et al. (22) and Wan et al. (23), respectively.

A Bayesian meta-analysis with random effects and noninformative priors was performed to incorporate the estimates of direct and indirect comparisons using the aggregate data drug information system (ADDIS) 1.16.5 (Drug Information Systems, Groningen, The Netherlands) (24). A random-effects model was used with a Markov chain Monte Carlo method. For the preset model parameters, 4 chains were utilized for simulation analysis, with an initial value of 2.5, a step size of 10, tuning iterations of 20,000, and simulation iterations of 50,000. The model convergence was assessed using the Brooks–Gelman–Rubin method, in which the major reference index was the potential scale reduction factor (PSRF). When the PSRF approaches 1, convergence is better. Additionally, we calculated the ranking probabilities for the effectiveness of each intervention using ADDIS 1.16.5 and estimated the different interventions for each outcome employing surface under the cumulative ranking curves (SUCRAs). Network transitivity was assumed by comparing the distribution of potential effect modifiers in terms of age and duration of treatment. To evaluate the presence of statistical inconsistency in the network, node-splitting analysis was used to assess the agreement between direct and indirect evidence (25). Sensitivity analysis was conducted by excluding articles previously employing imputation methods as well as studies rated as having a high risk of bias. To evaluate the small study effects and publication bias, we created a comparison-adjusted funnel plot to assess the magnitude of asymmetry for each outcome (19).

Results

A total of 1992 publications were identified by the literature searches, of which 48 were reviewed in full, and 18 studies (26–43) involving 1790 subfertile men were included in the systematic review and NMA (Figure 1, Supplemental Appendix 2). The included 18 RCTs investigated 8 antioxidants (folic acid, zinc, vitamin E, carnitine, selenium, CoQ10, N-acetylcysteine, and vitamin C); however, arginine was not included due to the lack of associated trials. Only 5 RCTs (26, 31, 32, 39, 42) directly compared the effects of different types of antioxidants.

FIGURE 1.

Flow diagram of the study selection process.

The study and participant characteristics of the RCTs included in this NMA are presented in Table 1. These trials have been conducted in various countries. Eight trials were conducted in Iran (30–34, 36, 38, 43); 3 in Italy (35, 40, 41); and 1 each in The Netherlands (26), Saudi Arabia (27), USA (28), United Kingdom (29), Brazil (37), China (39), and Iraq (42). The duration of treatment ranged from 12 to 26 wk. The mean age of the participants included in these trials was between 25.54 and 36.24 y. All participants had fertility problems with abnormal semen parameters. Sperm concentration was investigated in 13 RCTs (26, 30–33, 35–37, 39–43); sperm motility was investigated in 16 RCTs (26–36, 39–43); and sperm morphology was investigated in 10 RCTs (26, 30, 31, 33, 34, 36–38, 42, 43).

TABLE 1.

Baseline characteristics of the included randomized controlled trials

Author, year, reference	Study origin (country)	Population	Sample size (n)	Mean age1 (y)	Mean BMI1 (kg/m2)	Intervention (per day)	Duration
Wong et al. 2002 (26)	The Netherlands	Subfertile men	70	34.27	NR	Placebo vs. folic acid 5 mg vs. zinc sulphate 66 mg	26 wk
Suleiman et al. 1996 (27)	Saudi Arabia	Asthenospermic men	87	22–52 (range)	NR	Placebo vs. vitamin E 300 mg	6 mo
Sigman et al. 2006 (28)	USA	Idiopathic asthenospermia	21	35.81	NR	Placebo vs. LC 2 g + LAC 1 g	24 wk
Scott et al. 1998 (29)	United Kingdom	Low sperm motility	34	32.76	NR	Placebo vs. selenium 100 ug	3 mo
Safarinejad et al. 2012 (30)	Iran	Oligoasthenozoospermia	225	31.5	NR	Placebo vs. CoQ10 200 mg	26 wk
Safarinejad et al. 2009 (31)	Iran	Idiopathicoligo-asthenoteratospermia	316	31.33	26.33	Placebo vs. selenium 200 ug vs. NAC 600 mg	26 wk
Raigani et al. 2014 (32)	Iran	Oligoasthenoteratozoospermia	62	NR	NR	Placebo vs. folic acid 5 mg vs. zinc sulphate 220 mg	16 wk
Nadjarzadeh et al. 2011 (33)	Iran	Oligoasthenoteratozoospermia	47	34.42	23.1	Placebo vs. CoQ10 200 mg	3 mo
Mehni et al. 2014 (34)	Iran	Idiopathic oligoasthenoteratozoospermia	110	30	23.39	Placebo vs. LC 1 g	3 mo
Lenzi et al. 2004 (35)	Italy	Asthenozoospermia	56	20–40 (range)	NR	Placebo vs. LC 2 g + LAC 1 g	6 mo
Eslamian et al. 2020 (36)	Iran	Idiopathic asthenozoospermic	90	32.92	26.42	Placebo vs. vitamin E 600 IU	12 wk
da Silva et al. 2013 (37)	Brazil	Subfertile	49	36.24	NR	Placebo vs. folic acid 5 mg	3 mo
Cyrus et al. 2015 (38)	Iran	Infertile men with varicocele	115	27.6	NR	Placebo vs. vitamin C 500 mg	3 mo
Cheng et al. 2018 (39)	China	Idiopathic oligoasthenozoospermia	125	30.96	18.2–26.1 (range)	LC 2 g vs. CoQ10 60 mg	3 mo
Balercia et al. 2005 (40)	Italy	Idiopathic asthenozoospermia	59	30	NR	Placebo vs. LC 3 g vs. LAC 3 g vs. LC 2 g + LAC 1 g	6 mo
Balercia et al. 2009 (41)	Italy	Idiopathic asthenozoospermia	60	32	NR	Placebo vs. CoQ10 200 mg	6 mo
Alahmar et al. 2020 (42)	Iraq	Oligoasthenoteratospermia	70	25.54	NR	CoQ10 200 mg vs. selenium 200 μg	3 mo
Safarinejad et al. 2009a (43)	Iran	Idiopathic oligoasthenoteratospermia	194	28	28	Placebo vs. CoQ10 300 mg	26 wk

CoQ10, coenzyme Q10; LC, L-carnitine; LAC, L-acetylcarnitine; NAC, N-acetylcysteine; NR, not reported.

¹Mean age or BMI of subjects who completed the study.

Network diagrams

The network diagrams of the outcomes are shown in Figure 2. The most common comparisons occurred between the antioxidants and placebo. Figure 2A presents the direct comparisons of the network diagrams for sperm concentration (n = 13 RCTs), including 7 antioxidants and the placebo. Figure 2B presents the direct comparisons of the network diagrams for sperm motility (n = 16 RCTs), including 7 antioxidants and the placebo. Figure 2C presents the direct comparisons of the network diagrams for sperm morphology (n = 10 RCTs), including 8 antioxidants and the placebo.

FIGURE 2.

The network diagrams for sperm concentration (A), sperm motility (B), and sperm morphology (C). The size of the node is proportional to the number of studies involving each specific intervention; the thickness of the line is proportional to the number of the comparisons included in the network. CoQ10, coenzyme Q10.

Risk of bias assessment

The assessments for risk of bias are shown in Supplemental Figure 1. With respect to the individual risk of bias, 11 trials illustrated the specific methods of random sequence generation, although all trials mentioned this. Nine trials stated allocation concealment, whereas others did not. Fifteen trials had a low risk of bias in the blinding of participants and personnel, whereas 8 trials had a low risk of bias in the blinding of outcome assessment. One trial had a high risk of bias in the blinding of participants and personnel. Furthermore, there were 12 trials with a low risk and 2 trials with a high risk of incomplete outcome data. There were 3 trials with a low risk and 1 trial with a high risk of selective reporting. Overall, among the 18 RCTs included in this NMA, 3 trials were classified as having a high risk of bias (27, 30, 37), 10 were classified as having a low risk of bias (26, 28, 29, 31–33, 36, 38, 40, 43), and 5 were judged to have a moderate/unclear risk of bias (34, 35, 39, 41, 42).

Transitivity analyses showed minor differences among the studies in terms of age (Supplemental Figure 2). In terms of treatment duration, the included studies had acceptable transitivity, as all studies covered ≥1 human spermatogenesis cycle (74 d) (44).

Outcomes

Sperm concentration

In the pairwise meta-analysis, only CoQ10, selenium, and N-acetylcysteine showed statistically significant differences (P <0.05) compared with the placebo (Supplemental Figure 3). In the NMA, CoQ10 elicited a significant increase in sperm concentration (MD = 5.95; 95% CI: 0.05, 10.79) compared with the placebo. However, no statistical differences were observed between folic acid, zinc, vitamin E, carnitine, selenium, N-acetylcysteine, and the placebo (Supplemental Table 1). Among the different antioxidants, CoQ10 had the highest SUCRA (79.4%), followed by carnitine (65.4%) and folic acid (56.3%) (Supplemental Figure 4).

Sperm motility

In the pairwise meta-analysis, only carnitine, CoQ10, and N-acetylcysteine showed statistically significant differences (P <0.05) compared with the placebo (Supplemental Figure 5). In the NMA, carnitine (MD = 12.43; 95% CI: 4.07, 20.26) and CoQ10 (MD = 7.33; 95% CI: 0.35, 14.17) significantly increased sperm motility compared with the placebo. However, no statistical differences were observed between folic acid, zinc, vitamin E, selenium, N-acetylcysteine, and the placebo (Supplemental Table 2). Among the different antioxidants, carnitine had the highest SUCRA (88.7%), followed by selenium (71.1%) and CoQ10 (65.3%) (Supplemental Figure 6).

Sperm morphology

In the pairwise meta-analysis, vitamin C, CoQ10, selenium, and N-acetylcysteine showed statistically significant improvement in sperm morphology (P <0.05) compared with the placebo (Supplemental Figure 7). In the NMA, no statistical differences were observed between folic acid, zinc, vitamin E, carnitine, selenium, CoQ10, N-acetylcysteine, vitamin C, and the placebo (Supplemental Table 3). Among the different antioxidants, vitamin C had the highest SUCRA (93.6%), followed by CoQ10 (76.4%) and N-acetylcysteine (53.3%) (Supplemental Figure 8). Given the lack of statistical differences among the different interventions, the ranking results should be interpreted cautiously.

Inconsistency and convergence analysis

The inconsistency between direct and indirect evidence was evaluated by node-splitting analysis. In this study, no significant inconsistencies in sperm concentration, motility, and morphology were observed (Supplemental Tables 4–6, respectively). This reveals that the consistency model was dependent. Meanwhile, the PSRF reached 1, indicating good convergence.

Sensitivity analyses

We conducted sensitivity analyses based on the exclusion of trials with a high risk of bias (27, 30, 37) as well as those using imputation methods (26, 32). The findings obtained from the sensitivity analyses were consistent with the results of the main outcome analyses. CoQ10 had the highest SUCRA (80.4%) for sperm concentration (Supplemental Figure 9); carnitine had the highest SUCRA (86.2%) for sperm motility (Supplemental Figure 10); and vitamin C had the highest SUCRA (89.3%) for sperm morphology (Supplemental Figure 11). The results of the sensitivity analyses are shown in Supplemental Tables 7–9. We observed no significant inconsistencies in sperm concentration, motility, and morphology (Supplemental Tables 10–12).

Small study effects

Comparison-adjusted funnel plots were used for all outcomes. There was no obvious funnel plot asymmetry for sperm concentration (Supplemental Figure 12), motility (Supplemental Figure 13), and morphology (Supplemental Figure 14).

Discussion

In this study involving 18 RCTs with 1790 participants, we evaluated the effects of different antioxidants on sperm quality parameters in subfertile men. Our data provided the following findings: 1) among 7 different varieties of antioxidants (CoQ10, folic acid, zinc, vitamin E, carnitine, selenium, and N-acetylcysteine), CoQ10 might be the most effective for improving sperm concentration (SUCRA: 79.4%); 2) among 7 various types of antioxidants (CoQ10, folic acid, zinc, vitamin E, carnitine, selenium, and N-acetylcysteine), carnitine might be the most effective for improving sperm motility (SUCRA: 88.7%); and 3) among 8 different kinds of antioxidants (CoQ10, folic acid, zinc, vitamin E, vitamin C, carnitine, selenium, and N-acetylcysteine), vitamin C appeared to be the best for improving sperm morphology (SUCRA: 93.6%). Given the lack of statistical differences among the different interventions, the ranking results should be interpreted cautiously. Additionally, we performed sensitivity analyses by excluding studies utilizing imputation methods and those rated as having a high risk of bias. The findings obtained from the sensitivity analyses were consistent with the results of the main outcome analyses with regard to sperm concentration, motility, and morphology.

Previously published pairwise meta-analyses have assessed the effect of antioxidants on sperm parameters and suggested that antioxidants could beneficially regulate sperm quality as well as affect male fertility (14, 45); a similar conclusion has also been supported by in vitro studies (46). Excessive generation of ROS has been cited as responsible for sperm injury during the freeze-thaw process. A recent meta-analysis showed that antioxidants have a significant positive impact on sperm motility by reducing ROS during the freeze-thaw process (47). However, these meta-analyses did not quantitatively analyze different antioxidants; therefore, the related authors suggested that an NMA might be an attractive method for seeking the optimal antioxidant intervention to improve sperm parameters (47). To the best of our knowledge, our work is the first to adopt an NMA design to compare the direct and indirect effects as well as quantitatively evaluate the action of different antioxidants on sperm quality parameters in subfertile men. Oxidative stress triggered by elevated ROS concentrations is a common pathology in subfertile men; therefore, the underlying mechanisms of antioxidants could explain our results. CoQ10 is a lipid-soluble antioxidant with properties that inhibit the oxidation of cell membranes and lipoproteins. The sperm plasma membrane contains high concentrations of PUFAs and is highly susceptible to damage induced by excessive ROS concentrations (9). Another study reported that CoQ10 supplementation could decrease testicular oxidative stress and lipid peroxidation, restore antioxidant defense mechanisms, and maintain testicular function (48). Moreover, CoQ10 regulates energy metabolism as a component of the respiratory chain (9). In humans, CoQ10 is highly expressed in the mitochondria of the sperm midpiece; a lower concentration of CoQ10 was observed in the seminal plasma and spermatozoa of infertile men with asthenospermia (49). In fact, our NMA results also demonstrate that CoQ10 improves sperm motility and morphology. Furthermore, CoQ10 has other important biological activities, such as being a recycler of vitamin E, an autophagy modulator, and a regulator of the physicochemical properties of cell membranes (50). With spermatogenesis being a complex molecular biochemical process, the effect of CoQ10 on sperm parameters might have been due to its extensive biological role in regulating the metabolic environment. Carnitine is a small, water-soluble, quaternary amine. Its antioxidant activities are mainly elicited by L-carnitine and L-acetylcarnitine. In human tissues, L-carnitine is mainly derived from exogenous ingestion, such as dietary supplements (51). Carnitine can scavenge superoxide anions and hydrogen peroxide radicals, thereby preventing lipid peroxidation. It can also provide energy for sperm by transporting and regulating long-chain fatty acids into the mitochondria (49). The concentration of L-carnitine in epididymal plasma and spermatozoa is ∼2000-fold higher than its circulating concentrations (52). The epididymis is principally the site of sperm maturation; fatty acid oxidation is the main source of energy metabolism in epididymal spermatozoa (52). Therefore, a potential biological relation could be established between carnitine supplementation and increased sperm motility in subfertile men. Vitamin C, as a water-soluble compound, can neutralize hydroxyl, superoxide, and hydrogen peroxide free radicals to suppress endogenous oxidative damage (53). In the seminal plasma, the concentration of vitamin C was 10-fold higher than that in the serum (54). Additionally, its concentrations in seminal plasma correlated positively and negatively with the proportion of normal sperm morphology and with the DNA fragmentation index, respectively (55). Thus, vitamin C ranked first in improving sperm morphology with the highest SUCRA; however, no statistical differences were observed between vitamin C and other antioxidants as well as with the placebo. Therefore, this conclusion needs to be verified by further prospective studies.

Our study compared the effects of different antioxidants as well as placebo on sperm quality parameters in subfertile men; the results are reported based on the PRISMA Extension Statement for conducting NMAs. Ideally, the optimal regimen for oral antioxidant treatment should be regulated according to the redox status of the patient. However, this proved to be a challenge in clinical practice because of the lack of a widely accepted method for measuring redox status as well as determining a precise normal redox concentration or acceptable threshold (56). Our findings provide an important reference for the clinical application of antioxidants. Specifically, we ranked the effects of different antioxidants on sperm quality parameters in subfertile men; different antioxidants were selected based on the results of the sperm analysis. CoQ10 may be a better treatment option for oligozoospermia, whereas carnitine may be a better choice for asthenospermia. Furthermore, considering the safety and convenience of vitamin C as well as its ranking results, daily dietary supplementation may be a preferable option to improve sperm morphology.

Although we strictly followed the PRISMA Extension Statement for reporting NMAs, our study has some notable limitations. First, due to the limited number of trials, outcomes between different antioxidants were mainly derived from the results of indirect comparisons, thereby limiting the evaluation of the overall quality of evidence using the evidence assessment approach. Second, our study did not report the outcome of clinical pregnancy, which is the ultimate goal in the treatment of fertility problems. However, we cannot ignore the fact that embryological and female factors have a great influence on pregnancy and live birth odds; the majority of present studies on antioxidants failed to adjust for these confounding factors (10). Third, the insufficient number of included RCTs has limited our capability to compare the effects of similar antioxidants with varying doses. If there are enough studies in the future, a dose-response analysis may be able to determine the potential optimal dose. This facilitates the evaluation of optimal treatment options. Finally, although our study included common antioxidants used for treating male subfertility according to a Cochrane meta-analysis (14), arginine and carotenoids were not included. Specifically, arginine was not included due to the lack of associated trials. Furthermore, carotenoids exist in different forms, including β-carotene, lycopene, lutein, and zeaxanthin. Thus, the differences among their transformations render them unsuitable for this NMA.

In summary, CoQ10 produces a better clinical outcome for subfertile men in terms of sperm concentration, whereas carnitine produces a better clinical outcome for sperm motility. Further, vitamin C ranked first in improving sperm morphology with the highest SUCRA. Given the lack of statistical differences among the different interventions, this conclusion needs to be verified by further prospective studies. CoQ10 and carnitine were selected based on the results of sperm analysis in subfertile men; at present, both may be considered as better treatment options. These findings provide an important clinical reference for the treatment of subfertile men using antioxidants.

Notes

This work was supported by the National Natural Science Foundation of China (No. 81673987; 82074446); and China Academy of Chinese Medical Sciences Innovation Fund (No. CI2021A02202)

Author disclosures: The authors report no conflicts of interest.

Supplemental Appendices 1–2, Supplemental Figures 1–14, and Supplemental Tables 1–12 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/advances/.

LS and HQ contributed equally to this article.

Abbreviations used: ADDIS, aggregate data drug information system; CoQ10, coenzyme Q10; MD, mean difference; NMA, network meta-analysis; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses; PSRF, potential scale reduction factor; RCT, randomized controlled trial; ROS, reactive oxygen species; SUCRA, surface under the cumulative ranking curve.