A Global Survey of Reproductive Specialists to Determine the Clinical Utility of Oxidative Stress Testing and Antioxidant Use in Male Infertility

1. Ethical statement

This global online survey on male infertility, OS testing and AOX use was approved by the Institutional Review Board (IRB) of Cleveland Clinic (IRB No. 20-855).

2. Survey design and participants

This cross-sectional observational study used an internet survey as a data collection tool. The survey was designed to investigate the use of OS testing, AOX utilization and practice patterns of reproductive specialists involved in the care of patients with male infertility. A panel was formed consisting of 58 experts actively involved in the management of male infertility patients from 28 different countries, which included clinicians, researchers and scientists involved in the field of OS, AOX, and male infertility. The survey questions were carefully drafted, reviewed, and approved by these experts. SelectSurvey (https://www.classapps.com/product_ssv5.aspx), a secured tool approved by the Cleveland Clinic's Information Technology Department, was used to populate the survey questions online. The survey was kept open for 3 months (August 10th, 2020 to November 9th, 2020) for the participants to provide their response. During this time period, the online survey link was shared with the members of various professional societies and committees (as listed in the Acknowledgement section). The targeted participants of the survey were medical professionals working in the field of reproductive medicine and those involved in the treatment of infertile couples. Incomplete or duplicate responses were excluded from the analysis.

3. Questionnaire design

The online survey was written in English and composed of 29 questions that were structured into four different sections (Supplement File). The first section (4 questions) was related to the demographics of the participants; the second section (6 questions) was regarding the profession and clinical practice of the respondents; the third section (4 questions) was on OS testing in clinical practice; and the final section (15 questions) was related to the use of AOX in clinical practice.

4. Statistical analysis of data

Survey responses were downloaded in a comma-separated values (CSV) file format from the SelectSurvey application into a secured storage device approved by the Cleveland Clinic. Descriptive statistics was conducted using MedCalc Software (version 19.0.5; MedCalc Software, Ostend, Belgium). Data was reported as the number of participants and percentage of the total who answered each survey section or relevant question. However, participant numbers might be slightly different for some optional questions. For other questions, participants could mark more than one option, hence the results were reported as the percentage calculated based on the total number of participants who answered. Results were analysed using chi-square test to examine the international trends of OS testing and AOX use by participants.

1. Survey participants

After excluding duplicate or incomplete responses (those who did not continue with questions listed in section 2, n=150) from the analysis, a total of 1,327 participants from 88 countries completed the online survey (Table 1). Not all the questions in each section were completed by each participant, as some were optional. The number of responses for each subsection is provided in Table 1. Geographical distribution of the participants included Asia (46.8%), South America (19.7%), Europe (15.7%), Africa (11.3%), North America (4.3%) and Australia (2.2%). The majority of participants were aged 35–64 years old (75.0%), whereas the remaining were aged 25–34 years old (15.1%) or more than 65 years old (9.9%).

Table 1
Classification of survey participants according to demographics and professional details

• Click for larger image
• Click for full table
• Download as Excel file

A total of 1,281 (96.5%) participants responded to the self-reported nature of employment. Here, participants were identified primarily as attending physicians (59.6%) and advanced practice providers such as physician assistants and nurse practitioners (16.0%), along-side resident (6.3%) or fellows (5.7%) and academic staff (3.5%). A total of 1,323 (99.7%) participants responded to the primary work setting, which was most prominently reported as private practice (35.7%) or academic hospitals/clinics (26.9%). Out of 1,325 (99.8%) participants who responded to the question about their profession, nearly half (49.9%) reported themselves to be involved in more than one of the registered professions listed, while the most frequent self-reported single registered profession was in the field of urology (20.2%), andrology (8.8%) and gynecology (6.7%). The full cohort (n=1,327) responded to the questions about years of experience and specific training in male infertility. Here, most of these participants reported clinical practice experience of more than 10 years (61.3%), with special training in the field of male infertility either as part of their residency (47.5%) or fellowship (29.2%) training.

2. Oxidative stress testing

A total of 1,305 (98.3%) participants responded to the use of OS testing as part of male fertility evaluation, where 34.3% reported the use of OS testing in their clinical practice (Table 2). A sub-analysis was conducted to investigate the most commonly used assays to assess OS. Among the tests presented in Fig. 1, SDF (53.4%) was the most commonly utilized, followed by ORP (6.3%). Moreover, 31.3% reported using more than one assay to evaluate OS in infertile men (Fig. 1).

Fig. 1
Tests of oxidative stress that are commonly ordered by participants in the evaluation of male infertility.

• Click for larger image
• Download as PowerPoint slide

Table 2
Participants (n=1,305) response to the use of oxidative stress-related biomarkers in the clinical evaluation of male infertility

• Click for larger image
• Click for full table
• Download as Excel file

A total of 416/448 (92.9%) participants responded to the clinical conditions for OS testing. Survey results revealed that 59% (n=249) of clinicians recommended OS testing for patients with abnormal semen parameters and 55.8% of clinicians ordered OS testing for patients with UMI (n=232) (Fig. 2A). Survey participants reported that patients with risk factors such as smoking, alcohol consumption or drug abuse/misuse (n=308, 74.0%) and advanced paternal age (n=275, 66.1%) were most indicated for lifestyle-related OS testing (Fig. 2B).

Fig. 2
Indication for oxidative stress testing based on (A) clinical conditions and (B) lifestyle risk factors. SDF: sperm DNA fragmentation, ART: assisted reproduction techniques, IUI: intrauterine insemination, IVF: in vitro fertilization, ICSI: intracytoplasmic sperm injection.

• Click for larger image
• Download as PowerPoint slide

3. Indication for antioxidant therapy

A total of 1,260 (94.9%) responded to the question on the prescription of AOX for the treatment of male infertility. Here, a high percentage (n=1,078, 85.6%) of participants recommend AOX as a therapeutic option in the management of male infertility, either routinely (n=550, 43.7%) or for specific groups of patients (n=528, 41.9%) (Table 3: a). Of those participants who did not use OS testing as a part of male infertility evaluation (n=857), a high percentage (n=679, 79.2%) reported prescribing AOX treatment.

Table 3
Indications for AOX therapy

• Click for larger image
• Click for full table
• Download as Excel file

A total of 1,039 out of 1,078 (96.4%) responded to the multiple option question on the clinical conditions for which AOX-based therapy was prescribed. Here, the most common clinical conditions for which AOX-based therapy were prescribed were risk factors for OS (such as obesity, age, smoking) (n=707, 68.0%), idiopathic oligoasthenoteratozoospermia (n=659, 63.4%), UMI (n=651, 62.6%), isolated asthenozoospermia (n=629, 60.5%) and teratozoospermia (n=501, 48.2%) (Table 3: b).

A total of 1,039 participants responded to the question on the duration of AOX treatment of male infertility. Here, the majority of the participants recommend AOX treatment for a duration of 3 months (n=454, 43.7%) or up to six months (n=401, 38.6%) (Table 3: c).

Furthermore, the survey results revealed that 584 participants out of 1,039 (56.2%) who responded to this question treat 25%–75% of their patients with AOX (Fig. 3). A total of 1,039 participants responded to the question on the preference of AOX used in clinical practice. Here, the most commonly prescribed individual AOX were zinc (n=753, 70.9%), vitamin E (n=717, n=69.0%), L-carnitine (n=706, 67.9%) and Co-enzyme Q10 (n=675, 65.0%) (Table 4).

Fig. 3
Percentage of patients recommended for AOX treatment by the participants to the survey. AOX: antioxidant.

• Click for larger image
• Download as PowerPoint slide

Table 4
Most commonly prescribed individual antioxidants for the treatment of male infertility

• Click for larger image
• Click for full table
• Download as Excel file

A total of 708 participants were familiar with dietary AOX supplements, while 688 (97.2%) responded to the question on the top 10 most common commercial dietary AOX supplements (Table 5). While 177 participants did not recommend the AOX treatment, 1,039 participants recommended it mostly based on the scientific evidence (n=718, 69.1%) and personal experience (n=616, 59.3%) (Table 6).

Table 5
Top 10 most common commercial dietary AOX supplements recommended for treatment of male infertility

• Click for larger image
• Click for full table
• Download as Excel file

Table 6
Reasons selected by clinicians to recommend (n=1,039) or not recommend (n=177) the use of AOX in treatment of male infertility

• Click for larger image
• Click for full table
• Download as Excel file

Participants testing for seminal OS and prescribing AOX, respectively, were further analyzed based on their age, geographical origin, experience and the training in the field of male infertility. The participants using the seminal OS markers during evaluation of male infertility were not influenced by their age or years of experience in the field (Table 7: a). Conversely, the geographic origin significantly influenced this decision (p<0.001), with 53.2% of respondents from Asia opting for no testing (Fig. 4). Participants who had received a specific training in male infertility were more likely to order OS testing (n=378 out of 448), although interestingly the majority of participants chose no testing (n=626 out of 857) (p<0.001) (Fig. 4). With respect to prescribing AOX, the percentage of participants who recommended AOX was significantly related to the participant's age (p<0.001), geographic origin (p<0.004), experience or training in male infertility (p<0.001) (Table 7: b, Fig. 5).

Fig. 4
Oxidative stress testing based on the geographic origin of the participants and training in male infertility.

• Click for larger image
• Download as PowerPoint slide

Fig. 5
Antioxidant prescription based on the age, geographic origin, experience of the participants and their training in male infertility.

• Click for larger image
• Download as PowerPoint slide

Table 7
Patterns of oxidative stress testing and prescription of AOX according to the participants' age, geographic origin, experience (in years) and training in male infertility

• Click for larger image
• Click for full table
• Download as Excel file

The perception of AOX treatment in clinical practice from the participants point of view

A total of 1,152 participants out of 1,327 (86.8%) responded to the question on the strength of the evidence for AOX use in male infertility. Here, more than half of the participants considered the evidence supporting AOX use in clinical practice as modest (52.3%), that is a score of four in a scale where 5 was strong, while the remaining practitioners were divided between “the evidence supporting its use is strong” (19.7%) and “no good evidence supporting its use” (22.2%) (Table 8).

Table 8
Perception of AOX treatment in clinical practice from the point of view of the participants

• Click for larger image
• Click for full table
• Download as Excel file

A total of 1,172 out of 1,327 (88.3%) participants responded to the multiple response question on the limitations of AOX use. Here, the most important reasons for limiting its use were cost (n=611, 52.1%), duration of treatment which could postpone other treatments (such as assisted reproduction techniques [ART]) (n=528, 45.0%), and low adherence of patients to the therapy (n=323, 27.6%).

A total of 1,172 out of 1,327 (88.3%) participants responded to the multiple response question on the most appropriate outcomes to be evaluated after AOX treatment. Here, survey results indicated standard semen parameters (n=648, 55.3%), live birth rate (n=644, 54.9%), SDF (n=576, 49.1%) and clinical/ongoing pregnancy rate (n=555, 47.3%) as possible endpoints for the evaluation of AOX efficacy.

A total of 1,172 out of 1,327 (88.3%) participants responded to the multiple response question on the development of clinical guidelines. This was favoured by most of the participants (96.3%), who considered it either as much needed (n=607, 51.8%) and/or helpful (n=521, 44.5%) (Table 8).

1. Oxidative stress testing in clinical practice

The survey results show that there is a relatively low proportion of participants conducting OS testing. This may be due to lack of consensus on appropriate tests and their clinically relevant cut-off values, and an absence of standardization of laboratory techniques [29]. Although sensitivity and specificity of various tests have been published, they remain variable, nonstandardized and without general diagnostic recommendations [30, 31]. Furthermore, many tests for OS are expensive, may be time-consuming, often require specialized training, and are therefore not readily available in clinical and diagnostic laboratories [29]. This may be particularly apparent in developing countries, where empirical prescription of AOX in suspected cases of OS may represent a less expensive alternative to advanced testing. Interestingly, though 85.6% (Table 3: a) of participants reported the use of AOX, either routinely (43.7%) or in specific patient groups (41.9%), the majority (65.7%) (Table 2) did not use OS testing in the assessment of their patients. This may be due to the fact that the prescription of AOX is likely cheaper and more convenient compared to the assessment of OS. However, this is a potential clinical concern, as without identifying men with OS and monitoring ROS levels during therapy, there exists the risk of inducing reductive stress due to excessive AOX therapy. Reductive stress can be as damaging as OS [25, 32] and may itself be a cause of male infertility [33].

Numerous testing methods have been developed over the past few decades to determine OS or measure ROS in semen [29]. However, these tests currently have limited practical use and are mostly limited to research [34]. This is reflected in the current survey results, with chemiluminescence detection being used by only 2.2% of participants (Fig. 1). Indirect tests determine the impact of OS on lipids, proteins or DNA, or determination of conditions closely associated with OS, and therefore do not include direct measures of ROS or AOX. These include assays for malondialdehyde (MDA), thiobarbituric acid (TBARS) or 4-hydroxynonenal (4-HNE) as mutagenic by-products of lipid peroxidation, 8-hydroxy-2-deoxyguanosine (8-OHdG) as a marker of damage to DNA, and other assessments of SDF [5, 34, 35, 36].

OS is a well-recognized cause of SDF [29, 37]. Recently, Agarwal et al [38] proposed clinical guidelines for SDF testing and showed that there is strong evidence that varicocele, IMI, UMI and a detrimental lifestyle are associated with increased OS, and hence necessitate SDF assessment. This is supported by the survey results, where SDF assessment was most strongly reported for use as a marker of OS by 53.4% of respondents (Fig. 1). SDF has been found to correlate positively with MDA, and negatively with superoxide dismutase and glutathione peroxidase, in semen samples of infertile men [39]. This correlation has also been reported in males with subclinical, normozoospermic, asthenozoospermic and oligozoospermic varicocele [40]. Linear correlations between SDF and seminal OS with increasing paternal age [41] have also been noted. However, it is suggested that both SDF and another measure of OS should be used in the evaluation of infertile men [39]. This is supported by Homa et al [42], who suggests that neither chemiluminescence nor ORP should be used alone to determine OS. This may explain a high proportion (31.3%) of participants reporting the use of more than one assessment for OS. Although there are numerous SDF testing methods for diagnostic and research use, with terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL), sperm chromatin structure analysis (SCSA), and sperm chromatin dispersion (SCD) being most commonly used, these tests remain unstandardized and without clinical consensus [38, 43, 44].

The survey results show that 6.3% of participants used ORP for assessment of OS, third most frequent after SDF and combination testing (Fig. 1). The ORP has been introduced recently to determine seminal OS directly as an electrochemical measurement using the male infertility oxidative system (MiOXSYS; Aytu Bioscience, Englewood, CO, USA) [31, 45]. Increased seminal ORP, indicative of OS, has been reported to correlate positively with SDF when normalized with sperm concentration and motile sperm concentration [46].

Important clinical indications for OS testing includes varicocele, genitourinary tract infections, obesity, diabetes, IMI and UMI [6, 35]. The most commonly reported clinical reason for OS testing in this survey was abnormal semen parameters (59.9% of participants that assess OS, Fig. 2A). This is consistent with the literature, where asthenozoospermia, teratozoospermia, leukocytospermia and increased seminal viscosity have each been suggested as potential surrogate markers of seminal OS [47, 48, 49]. Abnormal semen parameter(s) are also reflected in the diagnosis of IMI, defined as having one or more abnormal semen parameter(s) without identifiable cause [6]. IMI was reported to be an indication for OS testing by a further 44.2% of participants that assess OS (Fig. 2A). The condition of UMI, defined as patients being infertile despite having normal semen parameters, was the second most reported clinical indication for testing (55.8% of participants that assess OS, Fig. 2A). Importantly, OS is reported in 11%–78% of males diagnosed with UMI and 30%–80% of cases of IMI [50, 51]. For the latter, the term MOSI has been introduced for males with IMI and who are positive for seminal OS [6].

Numerous lifestyle factors are known to increase the risk for male infertility and abnormal semen parameters. These include alcohol, tobacco and recreational drug use, increased heat exposure to the genitals, exposure to endocrine disrupting chemicals, and radiation [52, 53, 54]. Detrimental lifestyle factors are also associated with increased OS as a common underlying mediator of male infertility [52]. This reflects in the results of this survey in which there was a relatively even spread across these risk factors as indications for OS testing (Fig. 2B).

2. Antioxidant use in clinical practice

This survey confirms the widespread use of AOX to treat men with infertility. Almost half (43.7%) of clinicians routinely propose AOX therapy to all patients, while another 41.9% offer it to men with specific indications (Table 3: a). More than half the respondents (56.2%) stated that 25%–75% of their patients received AOX therapy (Fig. 3). Even the vast majority (80%) of those who did not test for OS or SDF still prescribed AOX therapy. There are numerous possible reasons underlying this practice. Firstly, there is no specific therapy for most men with infertility and even in cases with a correctable male factor the use of AOX therapy may be beneficial. Secondly, there is widespread availability of “over the counter” AOX both individually and/or in combination, with 55.3% of respondents being familiar with dietary AOX supplements (Table 5). Thirdly, AOX have been widely represented as having numerous health benefits with a good safety profile. Thus, both clinicians and patients perceive AOX to be safe. There is also relative ease of use, where taking AOX tablets is preferable to more invasive procedures like varicocele surgery or ART. Lastly, considering patient demographics, 70.1% of the respondents were either from Asia or South America (Fig. 4). In both these regions, there is a long history of use of natural medicines, which possibly makes this patient population more compliant with long-term empirical medical therapy.

On the other hand, the study also identified several potential inconsistencies to the use of AOX. The duration of therapy recommended was generally either for 3 months (reflecting the average length of the spermatogenesis process) or 6 months, while some give it indefinitely until pregnancy (Table 3: c). Despite relative ease of taking tablets daily, patient compliance can be a challenge when the duration of treatment is long, especially since infertile couples usually want a rapid positive result. Furthermore, this may delay the utilization of more effective therapies, such as ART, while the couple awaits the unpredictable outcome of AOX therapy. Although relatively cost-effective, AOX combinations containing compounds like L-carnitine and Co-enzyme Q10 can be expensive, and when used for a long duration, may prove to be costly for economically constrained patients. The survey further showed that there is no consensus on what the outcome measure should be to evaluate the benefits obtained from AOX therapy. While 55.3% of participants suggested improvement in semen parameters, there was an equal number who felt that the end point should be live birth rate (Table 8). Other end-points suggested were reduction in SDF or clinical pregnancy rates (Table 8). This diversity of opinion causes difficulty in conducting good clinical studies on AOX therapy that would gain widespread acceptance. Significant diversity was found in the therapies used by the respondents with 36 different AOX being listed (the top 10 AOX are listed in Table 5). It is in this context that there is a clear need for evidence-based guidelines for AOX treatment in male infertility.

3. Evidence based medicine

In the modern age, clinical care has shifted towards the practice of evidence-based medicine (EBM), defined as “the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients” [55]. EBM incorporates the clinicians' experience and patient values with high quality medical research aiming to optimize clinical decision making. Amongst survey participants, AOX treatment was recommended mostly based on scientific evidence (69.1%) and personal experience (59.3%) (Table 6), reflecting some adherence to EBM guidelines.

Most of the clinical evidence investigating the efficacy of AOX therapy on male fertility comes from three published reviews by the Cochrane library in 2011, 2014, and 2019 [20, 56, 57]. The last update included 61 randomized clinical trials (RCTs) and collectively reported that AOX therapy for the treatment of male infertility is associated with a significant increase in pregnancy and live birth rates [20]. However, the quality of the reported evidence was low, which was attributed to the small study size and extensive heterogeneity across the included studies, with multiple AOX regimens, varied doses and different treatment durations. Moreover, the primary treatment groups were different. Some AOX trials were using it to boost natural or intrauterine insemination (IUI) conception, where others were aimed at augmenting in vitro fertilization-intracytoplasmic sperm injection (IVF-ICSI) outcomes in male factor infertility. As the latter is more reliant on improving sperm DNA quality, rather than motility or count, it is understandable that AOX therapy may be more advantageous in an IVF setting than in natural conception. Finally, many trials did not screen for OS and hence may include subjects who may not have OS and therefore would not benefit from AOX therapy.

Recently Agarwal et al [27] published clinical guidelines on the utility of AOX in the treatment of male infertility that were based on a systematic review and critical analysis of the evidence. The authors included 97 articles in their systematic review that were evaluated using multiple quality scores. They found that the majority of low-quality studies reported a significant improvement in semen parameters (85.7%) and sperm function tests (89.6%) following AOX therapy. Furthermore, 78.6% of low-quality studies showed a positive effect for AOX treatment on reproductive outcomes. While a significant effect was not detected among high-quality studies investigating the impact of AOX therapy on male fertility potential, 65% of low-quality studies investigating semen parameters, 58.3% sperm function and 60% reproductive outcomes, showed a positive effect. Overall, positive effects in semen parameters and sperm function following AOX therapy were reported by both the low- and high-quality studies for men with varicocele, abnormal semen parameters, IMI and UMI. This is supported by proteomic evidence that AOX therapy enhances AOX defence mechanisms at the subcellular level and is associated with overexpression of sperm proteins that are essential for the fertilization process [58].

Despite the reported data, recent practice statements published by international societies have showcased more negative views towards the use of AOX as a treatment modality for infertile men. The European Association of Urology (EAU) states that “No clear recommendation can be made for treatment of patients with idiopathic infertility using antioxidants, although antioxidant use may improve semen parameters (weak strength rating)” [59]. The American Society for Reproductive Medicine (ASRM) recommended that “clinicians should counsel patients that the benefits of supplements (e.g., antioxidants, vitamins) are of questionable clinical utility in treating male infertility. Existing data are inadequate to provide recommendation for specific agents to use for this purpose. (Conditional Recommendation; Evidence Level: Grade B)” [60]. However, the extensive research that has been published on the study of AOX certainly warrants the formulation of new specific guidelines on the topic that would consider the utility of OS diagnostic testing, AOX type, dose and duration, and the clinical indications for AOX therapy.

4. Limitations of online survey

It is important to highlight some sources of bias which might have affected the collection of results through the online survey [61]. The survey link was shared around the world through the personal and professional society contacts of the experts associated with this study. The official and unofficial channels used might have created a selection bias, leading to an over- or under- representation of some countries in the results. Secondly, participants routinely prescribing AOX may have been more likely to fill out the survey, leading to participation bias [62]. There is also the impossibility to calculate a response rate, as it was not possible to keep track of the number of individuals who were given the opportunity to participate in the survey [63]. The length of the survey as well as the extensive number of questions might also have discouraged participants from completing it, leading to incomplete results.

Several strategies were put in place to reduce the bias of the results. First, the survey was carefully created, with several rounds of revisions by experts in the field. The aim was to collect reliable observations, by reducing the possibility of ambiguous, uncertain data or misinterpretation [64]. The questionnaire was anonymous, to reduce the risk of desirability bias, where participants provide the feedback which is hypothetically-preferred [65]. The English language was also revised to make questions clearer and easily understandable for non-English speaking participants or for participants who spoke English as a second language. Furthermore, the online nature of the survey made it possible to reach a larger audience. Hence, it is reasonable to speculate that our results may represent a reliable and accurate global picture of the use of AOX in clinical practice for male infertility treatment.

5. Experts' opinion based on AOX survey results

This survey reports the opinions of 1,327 healthcare practitioners from around the world regarding the utility of AOX therapy for male infertility in clinical practice. Despite the recently published recommendations from international societies, including ASRM and EAU [60], 85.4% of respondents believe that AOX are a useful treatment option for men with infertility, with 69.1% of them basing their decision on scientific evidence in support for AOX use (Table 6). Indeed, such evidence exists, although the studies supporting the use of AOX are small and heterogeneous. As such, the question remains whether the quality of evidence is sufficient to justify the use of AOX therapy for male infertility, mainly when considering that it has a very low risk of side effects. High-quality evidence is difficult to obtain practically and is not currently available. The possibility of conducting placebo-controlled RCTs of AOX therapy is limited by the fact that patients can buy AOX over the counter at the local pharmacy or supermarket. As such, we may never be in a position to gather Level 1 evidence. In the meantime, clinical practice recommendations based on the currently available best evidence seems the most reasonable approach. Infertility is a unique medical condition which, despite the tremendous progress in its management, still suffers from many unknowns. Conception results from the highly complex interplay between a spermatozoon and an oocyte and is influenced by numerous variables, many of which are yet to be discovered. This is the reason why infertility is still unexplained in up to 30% of couples [50]. Furthermore, male factors causing an impairment in semen quality are unknown in about 50% of cases, and hence the term IMI [6]. AOX are dietary supplements that generally have a good safety profile and are supported by low-quality evidence showing that a positive effect on semen quality and reproductive function can be expected following treatment. Moreover, recent evidence revealed that when AOX are used for specific clinical indications (IMI and UMI), the improvement in semen quality has been shown by a high quality of evidence [27]. For these reasons, we advocate the use of AOX in clinical practice, provided that the following criteria are met: