Numerous studies have been published reporting male infertility and a drop in sperm quality (concentration, motility, and morphology) or dismissing the same (Carlsen et al., 1992; Auger et al., 1995; Fisch et al., 1996; Marimuthu et al., 2003; Lackner et al., 2005; Kumar and Singh, 2015). Retrospective data analysis indicates that overall sperm parameters have declined in some parts of the world with geographical variations in semen quality (Auger and Jouannet, 1997; Jørgensen et al., 2001; Swan, 2006; Kumar and Singh, 2015). These geographical variations in semen characteristics could be due to environmental, nutritional, socioeconomic, or other unknown causes (Fisch and Goluboff, 1996) and this coincides with ever-increasing male genital tract abnormalities including testicular cancer and cryptorchidism in various countries (Giwercman and Skakkebaek, 1992; Bussen et al., 2004; Kumar and Singh, 2015).

Inability to achieve clinical conception despite 1 year of unprotected intercourse with the same partner is categorized as human infertility by the World Health Organization (WHO) (Zhao et al., 2016; Buhling et al., 2019; Cao et al., 2019; Hosseini et al., 2019; Wang et al., 2020). A total of 15% (48.5 million) couples worldwide are affected by any kind of infertility, and male infertility dominates the trend with 70% of all cases. Higher male infertility rates are seen in the population of Central Europe, Eastern Europe, and Africa (Zhao et al., 2016; Buhling et al., 2019; Cao et al., 2019; Hosseini et al., 2019; Wang et al., 2020).

The etiology of male factor infertility is multifactorial, primarily based on reduced and poor sperm quality. Concentration, morphology, and motility are three primary endpoints to assess the sperm quality (Buhling et al., 2019). Other factors include genetics (chromosomal aberrations, gene mutations, congenital anomalies, and polymorphisms), morphology of the reproductive system (testicular cancer, aplasia of the germinal cells, alteration in reproductive hormone, varicocele, defects in the transport of sperm, and bilateral sperm ducts), addictive disorders (alcoholism, smoking, and drug addiction), environmental factors (cigarette, smoking, exposure to certain chemicals, and nutritional deficiency), and alteration in spermatogenesis due to pathophysiological conditions (infectious diseases, pregnancy-related infections, urogenital infections, genitourinary dysplasia, immune system-related factor, abnormal levels of biochemical components of seminal plasma, and presence of antiserum antibodies (ASAs)) (Zhao et al., 2016; Buhling et al., 2019; Cao et al., 2019; Hosseini et al., 2019; Wang et al., 2020). These factors can lead to numerous abnormalities of the reproductive system and capacity for fertilization. Moreover, infertility can have severe damaging impacts on social, psychological, and economic well-being and health of the couples (Zhao et al., 2016; Buhling et al., 2019; Cao et al., 2019; Hosseini et al., 2019; Wang et al., 2020). Hormones, selective estrogen receptor modulator, antioxidants, vitamins, and enzymes are some of the treatment protocols for male infertility these days. Although several micronutrient supplementation products are available in the market that promises to improve the spermiogram, there are only a few existing evidence based data that address male factor infertility and micronutrient treatment options (Hamada et al., 2012; Cannarella et al., 2019; Cardoso et al., 2019; Hosseini et al., 2019; Wang et al., 2020).

To date, many systematic reviews and meta-analyses revealed the impact of different interventions in male infertility (Kamischke and Nieschlag, 1999; Boivin, 2003; Lafuente et al., 2013; Giahi et al., 2016; Hosseini et al., 2019; Smits et al., 2019; Wang et al., 2019; Garolla et al., 2020). Though systematic reviews and pairwise meta-analysis are vital tools for decision makers for contriving guidelines and clinical protocols, they produce partial information, because amongst several accessible interventions merely few are examined in head-to-head comparisons (Greco et al., 2015; Tonin et al., 2017). To overcome these issues, WHO introduced network meta-analysis (NMA). One of the several advantages of NMA is its ability to compare interventions quantitatively which otherwise are not directly comparable (Greco et al., 2015). This NMA will facilitate clinicians to mold interventions according to their choice to achieve desired therapeutic outcomes in male infertility, with maximum utilization of resources available.

This study compares effectiveness of multiple pharmacological groups including hormones (follicle stimulating hormone, testosterone undecanoate), selective estrogen receptor modulators (SERMs) (tamoxifen citrate and clomiphene citrate), supplements (coenzyme Q10, carnitine, L-Carnitine, zinc sulfate, fish oil, and Profertil), vitamins (vitamins A, D, and E and folic acid), and enzymes (kallikrein) alone and in combination through NMA. We choose to use sperm concentration, sperm motility, and sperm morphology as primary outcomes for male infertility related complications (Kumar and Singh, 2015; Buhling et al., 2019). Other secondary outcomes include serum total testosterone and serum follicle stimulating hormones (FSH).

Seven electronic databases (PubMed, Scopus, Cochrane library, Embase, EBSCOhost, Ovid, and Google scholar) were searched for data sources and strategies from inception till 31 April 2020. Medical subject headings [MeSH] and text terms were included for search terms in this review. The strategic search terms were as follows: “infertility” or “subfertility” or “subfertile” and “azoospermia” or “oligospermia” or “oligozoospermia” or “oligoasthenoteratozoospermia” or “genital disease” or “genitalia” or “genital” or “low sperm count” or “semen.” Details of search strategies used for each database are provided in the Supplementary Material.

The inclusion criteria were set according to the PICOT framework (population, intervention, comparison, outcomes, and time), as follows (Table 1). Studies which were included comprise of the following:

1) Randomized control trials or cluster-randomized controlled trials.

Studies conducted using the observational study designs mentioned were excluded:

1) Observational studies.

Two reviewers MNS and TMK screened titles and abstracts extracted from various databases using the well-defined selection criteria. Appropriate articles were then screened individually by the reviewers to access their inclusion eligibility. Resolution of disagreement was primarily through discussion.

MNS extracted the data from the studies included using standardized procedure. TMK independently reviewed the data for proper extraction. Details about title, publication year, authors, design of the study, country and study settings, sample size, age of the patients, gender of the patients, follow-up duration, interventions given, criteria for study inclusion and exclusion, and outcome of the study were extracted from each included study. In this review, changes from baseline to end of the intervention were summarized for both intervention and control groups as a result for the outcome measures.

MNS and TMK evaluated the risk of bias (ROB) of the included studies using the Cochrane ROB tool. For RCTs, each ROB item was ranked as “low risk” if it was suspected that a bias would seriously alter the result; it would be “unclear” if it was expected that a bias would raise some uncertainty about the results; or it would be “high risk” if it was prospective that a bias would completely alter the result. Discussion was used to resolve disagreements among reviewers.

Meta-analysis (MA) and NMA were executed by using Review Manager 5.3 and STATA 14. Comparative efficacies for all interventions were estimated through calculation of mean difference using the random effect model. The assessed significance of results p-values were set to be < 0.05 with 95% confidence interval (CI).

Subgroup analyses were executed for primary and secondary clinical outcomes for different interventions to clarify the heterogeneity among the studies. Robustness of the study results was analyzed through subgroup analysis of the baseline values for sperm concentration, motility and morphology levels, intervention duration, influence of studies on primary clinical outcomes, and geographical areas of the performed studies. Moreover, forest plot was generated to study pairwise comparison for the study of treatment effect in NMA. In addition, treatment effect and mean difference were used to generate league tables to evaluate all the effects (direct and indirect) among interventions.

The protocol was registered at PROSPERO (CRD42020152891).

In total, 302,869 articles were extracted from the electronic database searches; after confiscating duplications (n = 193,268), the final count was reduced to 109,601. Evaluation of title and abstract excluded 108,994 studies not fulfilling inclusion criteria. 607 studies remained and full-text assessment was done from which 80 studies (n = 9529 participants) were finally included for qualitative synthesis, whereas quantitative analysis (MA) was conducted for 29 studies, details of which are presented in PRISMA flow diagram (Figure 1). Reasons for omission after full-text assessment are presented in Supplementary Table 1.

Among all 80 selective articles, 57 RCTs were blinded while the remaining 23 were not blinded. Among the studies included in this review, 13 were conducted in Iran (Safarinejad, 2009; Safarinejad and Safarinejad, 2009; Moradi et al., 2010; Safarinejad, 2011a; Safarinejad, 2011b; Nadjarzadeh et al., 2011; Safarinejad et al., 2011; Safarinejad et al., 2012; Mehni et al., 2014; Nadjarzadeh et al., 2014; Haghighian et al., 2015; Hosseini et al., 2016; Modarresi et al., 2019), 13 in Italy (Izzo et al., 1984; Foresta et al., 2002; Caroppo et al., 2003; Lenzi et al., 2003; Lenzi et al., 2004; Balercia et al., 2005; Foresta et al., 2005; Paradisi et al., 2006; Selice et al., 2011; Colacurci et al., 2012; Paradisi et al., 2014; Farrag et al., 2015; Maretti and Cavallini, 2017), 6 in the United States of America (Haas and Manganiello, 1987; Clark and Sherins, 1989; Gerris et al., 1991; Sigman et al., 2006; Wiehle et al., 2014; Helo et al., 2015), 6 in England (Willis et al., 1977; Pryor et al., 1978; Inton et al., 1979; Badenoch et al., 1988; Scott et al., 1998; Williams et al., 2020), and 4 from Austria (Pusch, 1988; Maier and Hienert, 1990; Imhof et al., 2012; Lipovac et al., 2016) and Germany (Knuth et al., 1987; Krause et al., 1992; Keck et al., 1994; Kamischke et al., 1998) while 3 were from India (Kumar et al., 2011; Ambiye et al., 2013; Gopinath et al., 2013), Iraq (Hadwan et al., 2012; Haje and Naoom, 2015; Alsalman et al., 2018), Japan (Yamamoto et al., 1995a; Yamamoto et al., 1995b; Matsumiya et al., 1998), and Greece (Adamopoulos et al., 1997; Adamopoulos et al., 2003; Farmakiotis et al., 2007). Two were from Egypt (Ghanem et al., 2010; ElSheikh et al., 2015), Denmark (Fedder et al., 2014; Jensen et al., 2020), China (Ding et al., 2015; Guo et al., 2015), and Switzerland (Comhaire et al., 1986; Crottaz et al., 1992) and n = 1 from Turkey (Çakan et al., 2009), Australia (Baker et al., 1984), Malaysia (Ismail et al., 2014), Nigeria (Mbah et al., 2012), France (Almeida et al., 1985), Saudi Arabia (Suleiman et al., 1996), South Africa (Wong et al., 2002), Finland (Park et al., 2016), Yugoslavia (Mićić and Dotlić, 1985), Canada (AinMelk et al., 1987), Israel (Glezerman et al., 1993), Mexico (Merino et al., 1997), and Scotland (Hargreave et al., 1984). Moreover, one multiple center study was conducted in Australia, Germany, Italy, Poland, Spain, and United Kingdom (Nieschlag et al., 2017).

Eighty studies (n = 9529 patients) fulfilled the selection criteria of this review. The participants claimed infertility issue (WHO criteria) for at least 12 months or tested to have sperm count less than 20 million per ml. Sample size of the included studies ranged from n = 9 to n = 1,679; details are presented in Table 2

Figures 2 and 3 present ROB for all the included RCTs. Cochrane ROB tool was used to generate the graphs. More than 20% of the studies were categorized as free of attrition bias, reporting bias and other sources of bias. Performance bias and selection bias were granted in only 40 and 35% of the studies, respectively.

Sperm concentration (10⁶/ml), sperm motility (%), and sperm morphology (%) were the three primary clinical outcomes discussed in this study. A total of n = 29 studies were included for quantitative synthesis (Mićić and Dotlić, 1985; AinMelk et al., 1987; Pusch, 1988; Krause et al., 1992; Keck et al., 1994; Kamischke et al., 1998; Foresta et al., 2002; Wong et al., 2002; Adamopoulos et al., 2003; Caroppo et al., 2003; Lenzi et al., 2004; Paradisi et al., 2006; Safarinejad, 2009; Çakan et al., 2009; Ghanem et al., 2010; Safarinejad, 2011b; Selice et al., 2011; Colacurci et al., 2012; Hadwan et al., 2012; Imhof et al., 2012; Safarinejad et al., 2012; Nadjarzadeh et al., 2014; Paradisi et al., 2014; Ding et al., 2015; Farrag et al., 2015; Haje and Naoom, 2015; Lipovac et al., 2016; Alsalman et al., 2018; Jensen et al., 2020). All the studies were randomized control trials and encompass data on the stated outcomes involving interventions grouped under categories like hormones, selective estrogen receptor modulators, supplements, vitamins, and enzymes (Supplementary Figures 1–3).

Pairwise MA results were in the favor of the SERMs (Mićić and Dotlić, 1985; AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009; Ghanem et al., 2010; Haje and Naoom, 2015) where the levels of sperm concentration increased to 6.00 million per mL [95% CI 5.27, 6.72; p = 0.43] followed by supplements [5.99; 95% CI 2.83, 9.15; p < 0.00001; I² = 91%] (Wong et al., 2002; Lenzi et al., 2004; Safarinejad, 2009; Nadjarzadeh et al., 2011; Hadwan et al., 2012; Imhof et al., 2012; Safarinejad et al., 2012; Nadjarzadeh et al., 2014; Haje and Naoom, 2015; Lipovac et al., 2016; Alsalman et al., 2018; Jensen et al., 2020) (Supplementary Figure 4).

Subgroup analysis of studies examining the effect of SERMs revealed (Mićić and Dotlić, 1985; AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009; Ghanem et al., 2010; Haje and Naoom, 2015) maximum effect with clomiphene citrate [6.91; 95% CI 5.62, 8.20; p = 0.95; I² = 0%] (Mićić and Dotlić, 1985; Ghanem et al., 2010), followed by tamoxifen citrate [5.61; 95% CI 4.75, 6.46; p = 0.50; I² = 0%] (AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009; Haje and Naoom, 2015) (Supplementary Figure 5).

Moreover, the subgroup analysis for supplements (Wong et al., 2002; Lenzi et al., 2004; Safarinejad, 2009; Nadjarzadeh et al., 2011; Hadwan et al., 2012; Imhof et al., 2012; Safarinejad et al., 2012; Nadjarzadeh et al., 2014; Haje and Naoom, 2015; Lipovac et al., 2016; Alsalman et al., 2018; Jensen et al., 2020) revealed that studies involving zinc sulfate (Wong et al., 2002; Hadwan et al., 2012; Alsalman et al., 2018) had large effect [11.49; 95% CI -6.42, 29.39; p = 0.001; I² = 81%] followed by Profertil [10.90; 95% CI 7.98, 13.82] (Imhof et al., 2012) and CoQ10 [6.53; 95% CI 1.88, 11.17; p < 0.00001; I² = 94%] (Safarinejad, 2009; Nadjarzadeh et al., 2011; Safarinejad et al., 2012; Nadjarzadeh et al., 2014), respectively. Statistically significant results were seen in CoQ10 (Supplementary Figure 6).

Pairwise MA results were in the favor of the SERMs (Mićić and Dotlić, 1985; AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009; Ghanem et al., 2010; Haje and Naoom, 2015) where the percentage of sperm motility was increased [6.62; 95% CI 3.69, 9.54; p = 0.005; I² = 63%] followed by supplements (Wong et al., 2002; Lenzi et al., 2004; Safarinejad, 2009; Nadjarzadeh et al., 2011; Hadwan et al., 2012; Imhof et al., 2012; Safarinejad et al., 2012; Nadjarzadeh et al., 2014; Haje and Naoom, 2015; Lipovac et al., 2016; Alsalman et al., 2018; Jensen et al., 2020) with a standard mean difference of 6.51 [95% CI 3.15, 9.86; p < 0.00001; I² = 96%] (Supplementary Figure 7).

Subgroup analysis revealed that among SERMs (Mićić and Dotlić, 1985; AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009; Ghanem et al., 2010; Haje and Naoom, 2015) large effects were made by studies which involved clomiphene citrate [8.17; 95% CI 425, 12.10; p = 0.79; I² = 0%] (Mićić and Dotlić, 1985; Ghanem et al., 2010) followed by tamoxifen citrate [6.26; 95% CI 2.62, 9.90; p = 0.002; I² = 71%] (AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009; Haje and Naoom, 2015) (Supplementary Figure 8).

Moreover, the subgroup analysis for supplements (Wong et al., 2002; Lenzi et al., 2004; Safarinejad, 2009; Nadjarzadeh et al., 2011; Hadwan et al., 2012; Imhof et al., 2012; Safarinejad et al., 2012; Nadjarzadeh et al., 2014; Haje and Naoom, 2015; Lipovac et al., 2016; Alsalman et al., 2018; Jensen et al., 2020) revealed that studies involving zinc sulfate (Wong et al., 2002; Hadwan et al., 2012; Alsalman et al., 2018) had large effect [16.78; 95% CI 14.27, 19.29; p = 0.76; I² = 0%] followed by Profertil [7.00; 95% CI -1.50, 15.50] (Imhof et al., 2012) and CoQ10 [6.97; 95% CI 1.94, 12.01; p < 0.00001; I² = 98%] (Safarinejad, 2009; Nadjarzadeh et al., 2011; Safarinejad et al., 2012; Nadjarzadeh et al., 2014), respectively (Supplementary Figure 9).

Pairwise MA results were in the favor of the hormones (Pusch, 1988; Kamischke et al., 1998; Foresta et al., 2002; Adamopoulos et al., 2003; Caroppo et al., 2003; Paradisi et al., 2006; Selice et al., 2011; Colacurci et al., 2012; Paradisi et al., 2014; Ding et al., 2015; Farrag et al., 2015) where the percentage increase in sperm morphology was maximum [3.68; 95% CI 0.97, 6.39; p < 0.00001; I² = 83%], followed by supplements [1.93; 95% CI 0.43, 3.43; p < 0.00001; I² = 89%] (Wong et al., 2002; Lenzi et al., 2004; Safarinejad, 2009; Nadjarzadeh et al., 2011; Hadwan et al., 2012; Imhof et al., 2012; Safarinejad et al., 2012; Nadjarzadeh et al., 2014; Haje and Naoom, 2015; Lipovac et al., 2016; Alsalman et al., 2018; Jensen et al., 2020) (Supplementary Figure 10).

Subgroup analysis suggested that among hormones favourable effect was made by testosterone [12.24; 95% CI 1.00, 23.49; p = 0.01; I² = 83%] (Pusch, 1988; Adamopoulos et al., 2003), followed by FSH at a dose range of ≥200–300 IU [6.53; 95% CI -0.54, 5.59; p = 0.99; I² = 0%] (Paradisi et al., 2006; Paradisi et al., 2014; Ding et al., 2015) (Supplementary Figure 11).

Moreover, the subgroup analysis for supplements revealed that Profertil has large effect [14.50; 95% CI 7.31, 21.69] (Imhof et al., 2012) followed by zinc sulfate [4.23; 95% CI -2.39, 10.84; p < 0.0001; I² = 86% ] (Wong et al., 2002; Hadwan et al., 2012; Alsalman et al., 2018) (Supplementary Figure 12).

Serum total testosterone (ng/ml) and serum FSH (mIU/ml) were two secondary clinical outcomes discussed in this study. Total n = 12 studies were included (AinMelk et al., 1987; Pusch, 1988; Krause et al., 1992; Kamischke et al., 1998; Paradisi et al., 2006; Safarinejad, 2009; Çakan et al., 2009; Selice et al., 2011; Safarinejad et al., 2012; Paradisi et al., 2014; Helo et al., 2015; Jensen et al., 2020). All the studies were randomized control trials involving interventions grouped under categories like hormones, selective estrogen receptor modulators, supplements, vitamins, and enzymes (Supplementary Figures 13 and 14).

Pairwise MA results were in the favor of the supplements (Safarinejad and Safarinejad, 2009; Safarinejad, 2011a; Jensen et al., 2020) where the total serum testosterone concentration was increased [2.74; 95% CI 1.81, 3.68; p = 0.78; I² = 0%] followed by SERMs [1.50; 95% CI 1.20, 1.79; p < 0.00001; I² = 92%] (AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009) (Supplementary Figure 15).

Subgroup analysis revealed that among supplements (Safarinejad and Safarinejad, 2009; Safarinejad, 2011a; Jensen et al., 2020) coenzyme Q10 ^34,39 showed better results in increasing total serum testosterone concentration [2.77; 95% CI 1.83, 3.71; p = 0.76; I² = 0%]. None of remaining supplements showed substantial effects (Supplementary Figure 16).

Moreover, the subgroup analysis for SERMs (AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009) showed that all the total serum testosterone concentration effect was due to tamoxifen citrate [1.50; 95% CI 1.20, 1.79; p < 0.00001; I² = 92%] (AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009); none of the studies reported clomiphene citrate (Supplementary Figure 17).

Pairwise MA results were in the favor of the SERMs (AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009) where the total serum FSH concentration was increased [3.66; 95% CI 1.27, 6.05; p < 0.00001; I² = 90%] followed by hormones [1.24; 95% CI -0.29, 2.77; p < 0.00001; I² = 89%] (Supplementary Figure 18).

Subgroup analysis revealed that among SERMS studies only tamoxifen citrate 1.05 [95% CI 0.40, 1.71; p < 0.0001; I² = 84%] (AinMelk et al., 1987; Krause et al., 1992; Çakan et al., 2009) effect was reported. None of the studies reported clomiphene citrate (Supplementary Figure 19).

Moreover, the subgroup analysis for hormones (Pusch, 1988; Kamischke et al., 1998; Paradisi et al., 2006; Selice et al., 2011; Paradisi et al., 2014; Helo et al., 2015) revealed that studies involving FSH in dose <200–50 IU) (Kamischke et al., 1998; Selice et al., 2011) showed better results in increasing total serum FSH concentration [2.74; 95% CI -0.00, 5.48; p = 0.0009; I² = 91%] followed by studies involving FSH in dose ≥200–300IU [0.94; 95% CI -0.82, 2.70; p = 0.03; I² = 80%] (Paradisi et al., 2006; Paradisi et al., 2014) (Supplementary Figure 20).

To assess the effect of intervention on primary and secondary parameters, twenty-eight studies (Mićić and Dotlić, 1985; AinMelk et al., 1987; Pusch, 1988; Krause et al., 1992; Kamischke et al., 1998; Foresta et al., 2002; Wong et al., 2002; Adamopoulos et al., 2003; Caroppo et al., 2003; Lenzi et al., 2003; Paradisi et al., 2006; Safarinejad, 2009; Çakan et al., 2009; Ghanem et al., 2010; Nadjarzadeh et al., 2011; Selice et al., 2011; Colacurci et al., 2012; Hadwan et al., 2012; Imhof et al., 2012; Safarinejad et al., 2012; Nadjarzadeh et al., 2014; Paradisi et al., 2014; Ding et al., 2015; Farrag et al., 2015; Haje and Naoom, 2015; Lipovac et al., 2016; Alsalman et al., 2018; Jensen et al., 2020) were included for primary outcomes and seventeen studies (AinMelk et al., 1987; Pusch, 1988; Krause et al., 1992; Kamischke et al., 1998; Paradisi et al., 2006; Safarinejad, 2009; Çakan et al., 2009; Selice et al., 2011; Safarinejad et al., 2012; Paradisi et al., 2014; Helo et al., 2015; Jensen et al., 2020) were included for secondary outcomes. NMA was executed to compare the effects of intervention on primary and secondary parameters. The network plots exhibit the association of all available evidences. The thickness of the lines denotes the number of trials and the size of node represents the sample size (Figure 4).

With respect to primary outcomes (sperm concentration, sperm morphology, and sperm motility) when the reference arm was set as a placebo in the analysis, the sperm concentration was found to be considerably higher for supplement 6.26 [CI 95% 3.32, 9.21; p = 0.00] followed by SERMs 4.97 [CI 95% 1.61, 8.32; p = 0.004] and hormone 4.14 [CI 95% 1.83, 6.45; p = 0.00] groups versus the placebo group. On the other hand, the sperm morphology was significantly increased in hormone 3.71 [CI 95% 1.34, 6.07; p = 0.002] followed by supplement 2.22 [CI 95% −0.12, 4.55; p = 0.63] and SERMs 2.21 [CI 95% -0.78, 5.20; p = 0.15] groups versus the placebo group. Moreover, increasing trends in sperm motility were observed in SERMs 6.69 [CI 95% 2.38, 10.99; p = 0.002] trailed by supplement 6.46 [CI 95% 2.86, 10.06; p = 0.00] and hormone 3.47 [CI 95% 0.40, 6.54; p = 0.027] groups versus the placebo group (Table 3).

For secondary outcomes (serum total testosterone and serum FSH) by setting the reference arm as placebo in the analysis serum total testosterone was significantly increased in supplement 2.70 [CI 95% 1.33, 4.07; p = 0.00] followed by SERMS 1.82 [CI 95% 1.15, 2.49; p = 0.00] and hormone 0.40 [CI 95% -0.48, 1.28; p = 0.377] whereas serum FSH was significantly increased in SERMs 3.63 [CI 95% 1.48, 5.78; p = 0.001] followed by hormones 1.28 [CI 95% −0.78, 3.36; p = 0.223] and vitamins 0.033 [CI 95% −2.69, 2.76; p = −2.692] groups versus the placebo group (Table 3).

League table was created by using NMA to elaborate all probable pairwise comparisons between any two of the four interventions and customary pairwise meta-analysis (Table 3). It was obvious from the NMA that all of four interventions show analogous efficacy in increasing primary outcomes (sperm concentration, sperm morphology, and sperm motility) and secondary outcomes (serum total testosterone and serum FSH) (Table 4).

For primary outcomes, supplements were found statistically significant in increasing the sperm concentration 6.26 [CI 95%, 3.32, 9.21; p = 0.00] followed by SERM 4.97 [CI 95%, 1.61, 8.32; p = 0.004] and hormones 4.14 [CI 95%, 1.83, 6.46; p = 0.00]. Sperm motility was significantly increased by SERM 6.69 [CI 95%, 2.38, 10.99; p = 0.002] followed by supplements 6.46 [CI 95%, 2.87, 10.06; p = 0.00] and hormones 3.47 [CI 95%, 0.40, 6.54; p = 0.027]. Moreover, hormones proved to be statistically significant in improving the sperm morphology 3.71 [CI 95%, 1.34, 6.07; p = 0.002], followed by supplements 2.22 [CI 95%, 0.12, 4.55; p = 0.063] and SERMS 2.21 [CI 95%, −0.78, 5.20; p = 0.15] (Figure 5).

For secondary outcomes, supplements were found statistically significant in increasing serum total testosterone concentration 2.70 [CI 95%, 1.34, 4.07; p = 0.00] followed by SERMS 1.83 [CI 95%, 1.16, 2.50; p = 0.00] and hormones 0.40 [CI 5%, −0.49, 1.29; p = 0.37]. Moreover, SERMS proved to be statistically significant in increasing the serum FSH concentration 3.63 [CI 95%, 1.48, 5.79; p = 0.001], followed by hormones 1.29 [CI 95%, −0.79, 3.36; p = 0.223] and vitamins 0.03 [CI 95%, −2.69, 2.76; p = −2.692] (Figure 5).

Numerous curbs are associated with this study. Due to lack of resources, non-English studies were not reviewed as it was difficult to translate them to other languages. Combination of data from non-English literature might alter the significance of the current analysis of various male infertility interventions. Secondly, there were few studies with different subgroups making it difficult to get a perfect picture of the overall comparison. In addition, heterogeneity among the one‐on‐one and pairwise comparison was in excess of 70%. However, subgroup analysis and removal of poor quality studies from NMA resolved this issue to some extent but it is still a limitation in this study. Also, population of interest, intervention, comparators, and outcomes were the same across all the studies included in the NMA. Therefore, the chance of clinical heterogeneity is at very minimal to negligible level ruling out statistical heterogeneity. Lastly, due to diversified types of the male infertility along with interventions, all such interventions were classified into five categories, i.e., supplements, hormones, SERMs, vitamins, and enzymes. Along with all the limitations, our NMA is the first study estimating and establishing comparison among all available interventions regarding male infertility and this comprises a very significant aspect of this work.

This review is the first of its kind to present NMA on the comparative effect of numerous interventions from different pharmacological groups in managing males with infertility, conducted worldwide, in diverse health care settings under different experimental practices.

The studies included in this review encompass different RCTs conducted using moieties from different pharmacological groups including hormones (FSH, testosterone, and anastrozole), selective estrogen receptor modulators (SERMs) (clomiphene citrate and tamoxifen citrate), supplements (zinc sulfate, CoQ10, carnitine, L-Carnitine, fish oil, and Profertil), vitamins (folic acid, vitamin C, vitamin D, and vitamin E), and enzymes (kallikrein). It is evident from the studies that concomitant administration of supplements, hormones, and SERMS in a patient with male infertility can enhance the production of healthy motile sperms through maintaining adequate serum FSH and testosterone levels. The average duration of clinical outcome was reported to be six months. However, it should be noted that no major side effects were reported from any of the included studies. Common side effects were mild GIT disorders, occasional rashes, nervousness, and drowsiness and in few cases there were hot flashes with increased appetite.

The overall analysis on primary outcomes revealed that supplements, SERMs, and hormones increased sperm concentration 6.26 [95% CI 3.32, 9.21], sperm motility 6.69 [95% CI 2.38, 10.99], and sperm morphology, respectively, superior to other included interventions and placebo. Similar findings were reported by Manish Kuchakulla et al. and Rossella Cannarella et al. that supplements improve fertility, sperm concentration and sperm motility but do not effects sperm morphology, which was also proved through multiple RCTs (Fallah et al., 2018; Cannarella et al., 2020; Kuchakulla et al., 2020). Specifically, zinc sulfate proved to be the best in increasing sperm concentration 11.49 [95% CI −6.42, 29.39] and sperm motility 16.78 [95% CI 14.27, 19.29] among all supplement interventions. This effect of zinc sulfate was due to its ability of increasing low and high molecular weight ligands in the semen (Ahmadi et al., 2016). Among all SERMS interventions, clomiphene citrate was the best in increasing sperm concentration 6.91 [95% CI 5.62, 8.20] and sperm motility 8.17 [95% CI 4.25, 12.10]. Clomiphene citrate exerts its effect through raising the endogenous serum FSH, LH and testosterone levels and initiating gametogenesis (Patankar et al., 2007). Meanwhile, testosterone was the best in increasing sperm morphology among all hormones 12.24 [95% CI 1.00, 23.49]. Studies suggested that increased serum testosterone levels lead to better morphology due to its role in pathogenesis of teratozoospermia (Tang et al., 2012).

In addition to primary outcomes, the results of NMA showed statistically noteworthy effect of supplements, hormones, SERMs, and vitamins on secondary outcomes as well. Total serum testosterone levels were significantly enhanced by supplements 2.70 [95% CI 1.34, 4.07] followed by SERMs 1.83 [95% CI 1.16, 2.50], hormones 0.40 [95% CI −0.49, 1.29], and vitamins −0.70 [95% CI -6.71, 5.31] in comparison to placebo. These are in agreement with previous reports (Thakur et al., 2015; Lo et al., 2018). CoQ10 2.77 [95% CI 1.83, 3.71] and tamoxifen citrate 1.50 [95% CI 1.20, 1.79] showed the best results in improving serum total testosterone levels among all supplements and SERMs, respectively (AinMelk et al., 1987; Krause et al., 1992; Safarinejad, 2009; Çakan et al., 2009; Safarinejad et al., 2012). CoQ10 supplementation was found to ameliorate the reduction in testosterone induced by chemicals mainly by neutralizing the generated free radicals (Banihani, 2018). However, tamoxifen citrate stimulated release of LH and FSH through negative feedback of estrogen at the hypothalamus and pituitary which in turn increases the testosterone biosynthesis and stimulates spermatogenesis (Rambhatla et al., 2016).

Serum FSH concentration was majorly increased by SERMs 3.63 [95% CI 1.48, 5.79] followed by hormones, vitamins, placebo, and supplements. Hormones also increase the serum FSH levels 1.29 [95% CI −0.79, 3.36] followed by a minor increase by vitamins 0.03 [95% CI −2.69, 2.76]. Supplements failed to increase serum FSH levels but rather decreased the serum FSH level significantly −4.45 [95% CI −7.15, −1.76] as compared to placebo. The reason involves understanding of the role of serum FSH in sperm production. FSH, along with testosterone, is necessary for maintaining normal sperm count and function in males. Normal spermatogenesis yields low levels of FSH whereas compromised spermatogenesis can yield high serum FSH levels. As discussed above, supplements are the best in improving sperm concentration through normal spermatogenesis resulting in decreased levels of serum FSH (Orlowski and Sarao, 2018). Tamoxifen citrate 3.66 [95% CI 1.27, 6.05] and FSH 2.74 [95% CI −0.00, 5.48] were the best in increasing the serum FSH levels among all SERMs and hormones, respectively (AinMelk et al., 1987; Krause et al., 1992; Kamischke et al., 1998; Çakan et al., 2009; Selice et al., 2011).

It is evident from the NMA of this review that concomitant use of supplements, SERMs, and hormones was associated with additional clinical benefits beyond sperm concentration, sperm motility, and sperm morphology and these include improvement in sex life and conception (Ring et al., 2016). Strict adherence to therapy along with healthy diet could be clinically significant in reducing male infertility and increasing the chances of conception among couples. Finding of our NMA shows that concomitant use of supplements, hormones, and SERMs irrespective of their types has shown significant increase in sperm concentration, sperm motility, sperm morphology, serum total testosterone, and serum FSH in males’ infertility. Furthermore, considering the type of treatment, combination of zinc sulfate, clomiphene citrate, and testosterone undecanoate can be used to increase the sperm parameters (sperm concentration, sperm motility, and sperm morphology) and combination of CoQ10, tamoxifen citrate, and FSH can be used to improve the hormonal profile in infertile males. The verdict of this NMA will enable policy makers to formulate or choose the different accessible interventions, keeping in view the anticipated advantageous outcomes and existing healthcare assets.

Among all RCTs included, the selection, detection, and performance bias were observed to be <35%. This could perhaps influence the NMA results; therefore, it is recommended to infer our NMA results with caution. Due to lack of methodical content elaboration and varied nature of interventions, it is very difficult to conclude which type of intervention will be the most effective. This is a common issue exclusive in complex interventions in which the description of methods is insufficient to extract data which contribute to success of the regimen. Publishing a separate protocol of study is recommended to empower the readers and investigators to better recognize and comprehend study element, enabling them to be replicated in future.

For better outcome reporting, additional investigation is desirable to evaluate the male infertility interventions with respect to time, frequency, and contents. In addition, this reading endows imperative insights for future research focusing on a couture intervention and economic cost investigation in delivering such interventions, and, hence, designing cost effective interventions. The outcomes of this read will assist policy makers in assortment of suitable interventions keeping in view the paramount utility of the available resources.

It is observed that supplements appeared to be the best in increasing sperm concentration [6.26, 95% CI 3.32, 9.21; p = 0.00] and serum total testosterone levels [2.70, 95% CI 1.34, 4.07; p = 0.00]. On the other hand, hormones and SERMs intervention groups showed better sperm morphology [3.71, 95% CI 1.34, 6.07; p = 0.002), sperm motility [6.69, 95% CI 2.38, 10.99; p = 0.002], and serum FSH level [3.63, 95% CI 1.48, 5.79; p = 0.001], respectively.

This is perhaps the first paper to compare the one-on-one comparison among the interventions and controls used to improve the sperm morphology and count. In addition, this paper has also compared the groupwise and overall effect of the intervention using network meta-analysis which will serve as an ideal approach to optimize the therapy based on the effect size and might be useful in optimizing the cost of therapy as well. This study is of significant value for healthcare providers and policy makers in selecting perfect blend of interventions for male infertility patients keeping in view of the existing health resources. This review establishes that all interventions had a significantly positive effect on male infertility (sperm count, sperm motility, sperm morphology, serum testosterone, and FSH). Statistically significant increased sperm parameters (sperm concentration, sperm motility, and sperm morphology) were noted in combinations of zinc sulfate (220 mg BID), clomiphene citrate (50 mg BID), and testosterone undecanoate and CoQ10; tamoxifen citrate and FSH were shown to improve the hormonal profile in infertile males. There is a need for the future experimental studies on these interventions with significant effect size so that a better pharmacotherapy can be planned to improve the outcome of therapy.