This study was designed to determine whether COVID-19 affects sperm quality in men and to identify which sperm parameters were affected. Impact of COVID-19 on the semen parameters were demonstrated from different perspectives (with different study designs), including a cross sectional study of 515 patients who contracted COVID-19 (exposure group) and 89 controls who were free from the disease, and a longitudinal study (n = 140) with comparison of parameters before and after the onset of the infection in the same individuals. In addition, we included a COVID-19 negative natural control group (n = 149) with two consecutive samples over the same period of time to rule out the possibility of natural parameter change (worsening) over time due to other potential confounding factors such as adverse climate or environmental changes rather than the COVID-19 infection. As a result, we were able to generate several interesting findings and tentative conclusions: (1) The total sperm count and the motility including progressive motility and total motility were all significantly reduced while the sperm DFI was increased after the onset of COVID-19 in both the cross sectional and longitudinal observations, which strongly indicated that COVID-19 was associated with impaired sperm quality; (2) The finding that sperm quality of the latter samples were better than the former samples in the COVID-19 negative natural control group indicated that patients attending the fertility center generally benefited from the consultation or treatment by the physicians; (3) Comparison of samples between the COVID-19 positive and negative groups in the longitudinal study revealed that after the onset of COVID-19, sperm quality was worsened, further implicating COVID-19 as the causative factor; (4) The finding that in comparison with the COVID-19 negative group, a larger proportion of COVID-19 positive patients showed decreased sperm count and motility and increased DFI further indicated the adverse impact of COVID-19 on sperm quality from another perspective.

Total sperm count reflects the capacity of the testis to produce sperms, and is considered one of the most important conventional sperm quality parameters. COVID-19 was generally believed to be able to impair sperm production. However, there are still controversies as some studies did not reach the same conclusions¹². Our findings in this study supported the conclusion that sperm count was adversely affected by COVID-19, because the COVID-19 exposure group showed much fewer total sperm count than the controls, and in the longitudinal study, over half (57.86%, 81/140) of the cases showed reduction in the total sperm count, with scale of the reduction being 11.8% (a actual reduction of 20.49 × 10⁶ sperms in number) after COVID-19. Our study results contrasted with some previous studies, but were in line with a recent meta-analysis which showed that the infection was associated with a reduction of SMD = − 0.30 in total sperm count¹³. We speculate that the causes of the discrepancies might be due to differences in the study design or smaller sample size in the other studies^9,14.

Sperm motility is another important conventional semen parameters found affected by COVID-19 in this study. In this study, we considered total motility as well as progressive sperm motility and grade A motility, because progressive sperm motility was known related to pregnancy rates^15,16, and grade A motility was useful for predicting outcomes of the assisted reproductive technology^17,18,19. Our study showed that in comparison with the controls, the exposure group had much lower sperm motility, and the longitudinal study more specifically revealed that after COVID-19, most (69.29%, 97/140) of the cases showed reduction in the total motility, and the scale of reduction is 17.4% (an actual reduction of -5.62%). It is interesting as well as important to note that among all the categories of the motility, the most affected is grade A sperm motility. As shown in the longitudinal study, after COVID-19, as high as 71.43% (100/140) of the cases showed reduction in grade A sperm motility, and scale of the reduction was as high as 37.41% (an actual reduction of -3.18%). Similarly, although not as pronounced as with grade A motility, the exposure group also showed much lower progressive motility, and the longitudinal study showed that after COVID-19, 65.0% (91/140) of the cases showed reduction in the progressive motility and the scale of reduction was 20.38% (an actual reduction of -5.07%). Similar findings have been reported in previous studies by other researchers, though their sample size was smaller^20,21 or the study design was different²². Again, there are some discrepancies. For example, one study found there were a decrease only in total motility, not in the progressive motility²¹. We speculate that the slight differences in finding might be due to the smaller sample size or the long interval between onset of the disease and the semen collection (e.g. median = 111.5 days in one study ) in the previous studies.

Another important finding of this study was about the effect of COVID-19 on sperm DNA integrity: Both the cross sectional and the longitudinal study demonstrated that COVID-19 was associated with increased sperm DFI. Specifically, in the cross sectional study, the exposure group had much higher sperm DFI than the controls; in the longitudinal study, 75% (21/28) of the COVID-19 positive patients showed increased sperm DFI after onset of the disease, and the scale of increase was 21.18% (an actual increase of + 5.61%); whereas in the COVID-19 negative patients, sperm DFI was actually decreased by 12.68% (an actual decrease of -4.80%) over the same period of time. This finding is important in that it would be useful for clinicians in the field of reproduction to consult or manage patients, because it had been known that elevated sperm DNA fragmentation (SDF) affected outcomes of assisted reproductive techniques (ART) such as pregnancy rates and live birth rates, and high DFI was also associated with miscarriage²³. This finding is important also because research on the effect of COVID-19 on sperm DNA integrity has been sparse, and the limited literature had inconsistent opinions. For example, one previous cohort study (n = 120) by Donders GGG and colleagues²⁴ showed that DNA damage was most pronounced in the early stage following the onset of COVID-19, and they found abnormal DFI (defined as DFI > 25%) in 29% of the early samples as compared with 11-15% of the more later (1–2 months) samples after the onset. Similarly, Dipankar and colleagues²⁵ reported impact of COVID-19 on sperm DNA integrity in 30 patients suffering from mild COVID-19; they found that DFI was high (74.25%) at the first semen sampling after patients’ recovery from COVID-19 and became lower (66.94%) at the second semen sampling 74 days after the first sampling. Finally, Falahieh FM and colleagues²⁶ reported their study on 20 patients who had previously suffered moderate COVID-19, and found that DFI of patients at Day 120 after the diagnosis was lower than that at Day 14 after the diagnosis which fell within the normal range (defined as DFI < 30%). One of the issues with the above three studies was that they lacked baseline value of DFI before COVID-19. In this regard, it is interesting to note that Moryousef J²⁷ recently published a case report in which pre-COVID-19 values of DFI were included, and the patient showed substantially elevated sperm DFI one month following the onset of COVID-19 as compared with the pre-COVID-19 baseline values, and the DFI returned to normal level 4 months later.These aforementioned study results indicated that the COVID-19 could affect sperm DNA integrity. In contrast, however, a more recent study done by Paoli and colleagues⁸ reported that sperm DFI of the recovered COVID-19 patients was within the normal range, and DFI was not related to severity or presence/absence of COVID-19, concluding that COVID-19 had no effects on DFI. Our findings supported the conclusion that sperm DNA integrity was adversely affected by COVID-19. We think that discrepancy or uncertainty over the issue probably occurred due to the fact that many previous studies lacked controls with COVID-19 negative patients and lacked the controls with baseline values of semen parameters before the onset of COVID-19, and their sample sizes were usually small. In addition, with respect to Paoli’s study⁸, the fact that their COVID-19 patients were those already 3 months recovered from COVID-19 could be the reason for the null finding. All in all, we guess that the study designs in previous studies could make it difficult to reach a firm conclusion on the impact of COVID-19. In this sense, our finding with DFI might be important in that it helped in clarifying the issue.

On the other hand, our study did not find any effect of COVID-19 on sperm morphology, and this finding was in line with the study result by Can Balcı MB who reported a pronounced reduction of sperm concentration in COVID-19 positive patients as compared with COVID-19 negative patients but no difference in sperm morphology was found between the two groups²⁸. We also explored sperm chromatin maturity (immature sperm nucleoprotein) in this study because previous study had indicated that chromatin maturity or condensation was a valuable parameter in assessing male infertility²⁹ and sperm chromatin maturity was related to zygote development in ART programs³⁰. However, we did not identify any change in sperm nucleoprotein in either the cross sectional study or the longitudinal study.

Our study has some strengths. First, to our knowledge, our study had the largest sample size among all the observational studies up to date concerning the impact of COVID-19 on sperm quality. Second, design of the study with both cross sectional and longitudinal observations with COVID-19 negative controls allowed us reaching relatively confident conclusions concerning the effect of COVID-19 on semen parameters. In addition, with this type of design we could measure the scale of the effect, which should be useful for the clinicians to evaluate condition of the patients. Third, this study was done at the time of change in the control strategy for COVID-19 in China when over 80% of the population were infected with the virus in a concentrated period of time (less than two months). By seizing the window period for the virus infection, we were able to recruit large number of participants for the study. More important, because of the concentration of time and number of COVID-19 patients, many of the potential confounding factors for the study such as uncertainties over the diagnosis of COVID-19 infection or possible influences from climate or environment change could be eliminated so that results of the study might be more reliable. Finally, our study provided a relatively more complete description of the effect of COVID-19 on sperm qualities because more sperm parameters including the conventional semen parameters and the more recent parameters such as sperm chromatin maturity and sperm DNA fragmentation were explored.

Nevertheless, some limitations in this study should be noted. First, all the semen samples came from one reproductive center, and a multi-center study involving more cases is still needed. Second, due to the concentration of the outbreak, some baseline information on the sample was not fully collected. Third, the study only reported the semen characteristics shortly after COVID-19 recovery. We can conduct follow-up on these cases and perform semen analysis in later periods to understand if the changes acquire completely reversion. Furthermore, studies on chromosomic aberrations and imprinting changes, capable of being transmitted, could also be a subject for the future research.

Although WHO has declared that “COVID-19 is now an established and ongoing health issue which no longer constitutes a public health emergency of international concern”, the long-term careful management of COVID-19 pandemic is still advocated by WHO³¹. More important, SARS-CoV-2 virus is still currently rampant in many parts of the world, and fertility centers are bound to encounter many patients recovering from or currently suffering COVID-19. Therefore, proper handling of COVID-19 is definitely required in the diagnosis and management of male infertility. Based on our experience with the SARS-CoV-2 infection and the knowledge on its effects on sperm quality, we strongly suggest that screening or diagnosis of COVID-19 should be integrated into the workup of the patients with male infertility. In addition, keeping in mind the fact that SARS-CoV-2 continues to evolve, clinicians should remain vigilant against any potential adverse effects of different SARS-CoV-2 variants. In this sense, our study with SARS-CoV-2 infection might serve as a good example to observe effects of a specific SARS-CoV-2 variant on sperm quality. COVID-19 in our study has been known caused by a newly identified variant of the novel coronavirus, Omicron. Omicron is more infectious than previous variants, but less pathogenic in terms of causing pneumonia^32,33. Present study is very important, because up to date, the effect of Omicron on sperm quality has not been reported and our study provided novel information in this regard. In addition, because Omicron is more contagious than the previous variants, large population will be infected with the variant and the characteristics of the infection should be addressed specifically. In this sense, our study findings may be very meaningful in clinical terms.

In conclusion, our study showed that COVID-19 was associated with poor sperm quality manifested by reduced sperm count and sperm motility, and increased sperm DNA fragmentation. Further research is needed to observe the long-term effect of COVID-19 on sperm quality or reproductive outcomes. In addition, there may be a need to constantly monitor the SARS-CoV-2 infection and integrate screening and diagnosis of the disease in the management of male infertility.