Does SARS-CoV-2 Affect Human Semen? A Systematic Review and Meta-Analysis

Contradictory results have been reported regarding effects of the SARS-CoV-2 upon human semen. A timely and up-to-date systematic review with meta-analysis appears necessary. This study aimed to deliver pooled prevalence (PP) of SARS-CoV-2 in semen and pooled semen parameters as compared with the uninfected. The relevant databases were scanned by two authors for observational studies reporting analysis of semen in COVID-19 patients. The SARS-CoV-2-infected were assigned to group A (exposed arm), whereas the uninfected to group B (unexposed arm). Newcastle–Ottawa Scale was used to address the risk of bias. PRISMA guidelines were adopted. In case of homogenous studies, fixed-effects model was followed, whereas for heterogenous studies random-effects model was used. Of 990 studies, 24 were eligible involving 1589 subjects (947 in group A and 642 in group B). The “comparability” domain was biased the most. SARS-CoV-2 RNA was detected in three studies among 8 individuals producing the PP of 1.76% (95% CI 0.72–3.21). Sperm concentration was reduced significantly (WMD = –16.23 [95% CI –25.56 to –6.89], as well as total sperm in ejaculate (WMD = –34.84 [95% CI − 43.51 to –26.17]) and sperm volume (WMD =  − 0.48 [95% CI − 0.59 to − 0.36] in group A as compared with controls. There was a non-significant effect upon progressive motility and leukocyte presence in semen. SARS-CoV-2 RNA in semen among the infected individuals is detected infrequently. By this token, sexual transmission through semen is of low probability and little concern for public health. However, significant decrease in sperm volume, sperm concentration, and total sperm in ejaculate has been noted. The current data, though, are limited, and more studies with longer follow-up are needed to evaluate the further impact. Supplementary Information The online version contains supplementary material available at 10.1007/s10508-022-02520-3.


Introduction
An unprecedented scale of the global pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has propelled researchers throughout the world to identify routes of its transmission. As the virus predominantly causes respiratory tract disease officially called coronavirus disease 2019 , it chiefly spreads through respiratory droplets during face-to-face contact with an infected subject. The disease, which until September 4, 2022, was declared in over 600 million cases with almost 6.5 million casualties worldwide (World Health Organization, https:// covid 19. who. int/), has also been of concern in terms of other organs and systems. By means of real-time reverse transcription polymerase chain reaction (rRT-PCR), viral nucleic acid was detected in nasal, nasopharyngeal, and pharyngeal smears, samples of saliva, blood, stool, urine, and tears (Karia et al., 2020;. Regarding male reproductive tract, studies have been emerging focusing on possible presence of the viral RNA in semen and on the impact of COVID-19 upon semen characteristics (Cipriano et al., 2020;Holtmann et al., 2020;Temiz et al., 2021). Cell entry is facilitated by the viral spike protein and cellular angiotensinconverting enzyme 2 (ACE2) interaction (W. Li et al., 2003). Abundance of this enzyme in testes could possibly account for the viral appearance in semen and potentially another route of transmission. As new research has been conducted on the topic, an updated systematic review with meta-analysis is desired. Hence, the aim of this study is to perform the most comprehensive systematic review with statistical approach 1 3 of the observational studies analyzing SARS-CoV-2 RNA in semen and sperm characteristics (sperm concentration, total sperm in ejaculate, volume, sperm motility, and presence of leukocytes). The following null hypotheses were stated: (1) 95% confidence intervals (CI) of weighted mean differences (WMD) of continuous variables encompasses 0 and 95% CI of risk ratio (RR) of binary variables includes 1 indicating SARS-CoV-2 and COVID-19 do not affect semen quality.
(2) SARS-CoV-2 RNA in human semen is undetectable or its prevalence is negligible.

Study Design and Search Strategy
It is a systematic review and meta-analysis of observational studies reporting analysis of semen in subjects diagnosed with

Data Extraction and Curation
An attempt was made to extract the following data: (1) study design, (2) number of participants, (3) mean age, (4) mean/ median follow-up, (5) number of controls, (6) geographical origin of the study, (7) number of patients with SARS-CoV-2 in semen, (8) laboratory method of SARS-CoV-2 confirmation, (9) interval between COVID-19 onset and semen collection, (10) semen parameters (sperm concentration, motility, volume, leukocyte count, bacterial presence, total sperm in ejaculate). Data were disaggregated into its components and then reaggregated for the statistical purpose. Missing means or standard deviations were calculated using the disaggregated data. If a study reported relevant parameters only as medians with interquartile ranges and there were strong hints of the overall distribution not deviating much from the normal distribution, then assumption would be made equaling median and mean as well as standard deviation (SD) = IQR/1.35. Microsoft Excel 2016 (Redmond, USA) was utilized for data curation. Participants were divided into two groups according to the infection status-group A consisting of SARS-CoV-2-positive (active or recovered) patients and group B comprising the uninfected.

Risk of Bias and Quality
Risk of bias was addressed qualitatively and quantitatively. Qualitative assessment was grounded upon the Newcastle-Ottawa Scale (NOS), whereas for quantitative interpretation a funnel plot was schemed. For the NOS, the exposition was SARS-CoV-2 infection. The relevant endpoints were defined as presence of the viral RNA in semen and/or semen parameters as compared to World Health Organization reference values (5th centiles were adopted as the lower cut-off value) (Cooper et al., 2009). Eligible articles were assessed for three spectra: selection, comparability, and endpoints. Four stars could have been assigned for selection, two for comparability, and three for outcome, for "comparability" two stars, and for "outcome" three stars. Lack of any stars in a given domain meant the risk of bias was high. At least one star but shy of a maximum meant the risk of bias of moderate. A maximum number of stars meant the risk of bias was low. If there are no controls at all in a given study, then for the comparability domain as well as for 'Selection of the non-exposed cohort' part, it is assigned null. The most important control factor was age. Additional control factor was mean body mass index (BMI). As it was indicated that SARS-CoV-2 might affect semen even in the acute stage (D. Li et al., 2020aLi et al., , 2020b, adequacy of minimal follow-up of cohorts (time from diagnosis to semen collection) was set at 0 days (the day of diagnosis).

Statistical Analysis
Meta-analysis was conducted by means of MetaXL 5.3, Epi-Gear International Pty Ltd. (Brisbane, Australia) and Comprehensive Meta-Analysis V3 (Englewood, New Jersey, USA). Relative risk (risk ratio; RR) and weighted mean differences (WMD) were calculated with corresponding confidence intervals of 95%. Heterogeneity was evaluated based on I 2 and chi 2 . Interpretation of I 2 value was made as broadly accepted: 0-40%-homogenous; 40-60%-moderate heterogeneity; 60-80%-substantial heterogeneity; and 80-100%-considerable heterogeneity. Level of significance for Cochrane Q p value was arbitrarily set to < 10% (< 0.10), whereas level of significance for p value of comparative tests was universally set to < 5% (< 0.05). As in good research practice, in case of I 2 ≥ 40% (heterogeneity), random-effects model would be adopted, whereas for I 2 < 40% (homogeneity), fixed-effects model would be employed.

Risk of Bias Assessment
Tabular display of Newcastle-Ottawa Scale is shown in  Table 3 for the summary (Fig. 2).

Sensitivity Analysis
In order to evaluate sperm quality specifically before and after COVID-19 in the same subjects, a separate analysis was conducted for studies that included longitudinal semen assessment. 7 of 24 papers included longitudinal cohorts (Best et al., 2021;Erbay et al., 2021;Hamarat et

Spermatogenesis Deregulation
The following semen characteristics were significantly affected in those who had been infected with SARS-CoV-2: total sperm per ejaculate, sperm concentration, and sperm volume. Although these parameters were reduced, when juxtaposed with norms of World Health Organization they were still above 5th centiles (Cooper et al., 2009). On one hand, febrile course of any disease might lead to transient spermatogenesis deregulation (Carlsen et al., 2003). Thus, there have been speculations whether fever could be the only culprit of the impaired spermatogenesis in COVID-19 (Bendayan & Boitrelle, 2020). On the other hand though, studies have shown that staying afebrile throughout the entire SARS-CoV-2 infection does not necessarily protect from deregulation of spermatogenesis (Donders et al., 2022;Holtmann et al., 2020). Disaggregation of the authors' data to perform a subgroup meta-analysis of febrile versus afebrile was not feasible, thereby this matter remains unresolved. Interestingly, after having performed sensitivity analysis with studies of longitudinal semen evaluation, WMD of sperm progressive motility between two groups changed to statistically significant. This may reflect more heterogeneous nature of cross-sectional cohorts and more homogeneity of longitudinal cohorts.

Impact of Other Viral Infections upon Semen
SARS-CoV-2 is not unique in having an impact on sperm quality. It has been also documented for human immunodeficiency virus (HIV) which may diminish ejaculate volume and progressive motility, although this might potentially be a side-effect of antiretroviral treatment (Goulart et al., 2020). The sperm parameters in AIDS patients could be correlated with CD4 leukocyte count: in a group with count greater than 350 cells/µl sperm vitality, sperm penetration, and sperm motility were significantly better than in those with CD4 count less than 350 cells/µl (D. Wang et al., 2014). In terms of other viruses, L1 capsid protein of human papillomavirus (HPV) binds to syndecan-1 in spermatozoa, which was shown to reduce progressive motility and increase a rate of antisperm antibodies (Garolla et al., 2013). Herpes simplex (HSV) infection usually does not affect sperm motility but was proven to reduce sperm count per ejaculate and seminal volume (Kurscheidt et al., 2018;Monavari et al., 2013). Furthermore, there might be an association between Epstein-Barr virus (EBV)-positive semen samples and leukocytospermia, without an impact upon sperm count or motility (Kaspersen & Höllsberg, 2013).

Limitations
Although comprehensive, this meta-analysis is not without limitations. While some recent studies were of class II (Hamarat et al., 2021;Rafiee & Tabei, 2021;M. Wang et al., 2022), still majority of the literature on the effect of COVID-19 upon semen present evidence no stronger than class III. As authors are not unanimous in their expression of the semen parameters, not all of the characteristics could have been pooled (e.g., prevalence of bacteria in semen or febrile versus afebrile subgroup analysis). As identified through qualitative and quantitative risk of bias assessment, most studies might be biased, particularly in terms of comparability.

Conclusions
Although SARS-CoV-2 has been detected in semen by several independent researchers, the pooled data suggest its prevalence is very low. Thus, semen as a serious transmission route is unlikely. Even though the viral RNA is usually undetectable in semen, COVID-19 alters its characteristics: sperm concentration, total sperm in ejaculate, and sperm volume are significantly reduced as compared to the uninfected. Progressive motility and prevalence of leukocytes in semen might remain unaffected. As more and more research is being conducted on the topic, this updated systematic review with meta-analysis pools more data and provides narrower confidence intervals and stronger evidence.

Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.

Ethics Approval
For this kind of retrospective study, no ethics approval was required.

Consent to Participate
As this was a retrospective analysis of the studies that had been published in the past, no consent to participate was necessary.
Consent for Publication (include appropriate statements) Authors of this study grant the Publisher the sole and exclusive license of the full copyright.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.