COVID-19 Pandemic May Have Reversed Declining Sperm Counts in Shanghai Men

For decades, the trajectory of human semen quality has pointed in one troubling direction: downward. Since the landmark 1992 report suggesting that semen volume and sperm concentration had fallen by roughly 50 percent between 1930 and 1991, study after study has documented gradual declines in male reproductive parameters, with some evidence indicating the decline has accelerated in the twenty-first century. Now, a large retrospective study from a reproductive center in Shanghai, China, has produced a result that breaks sharply with that pattern. Analyzing 21,932 semen samples collected between 2021 and 2025, a period encompassing the widespread COVID-19 pandemic, researchers found that semen volume, sperm count, and sperm concentration increased significantly year by year, reversing the declining trends the same laboratory had documented at the same center from 2015 to 2021. The findings, published in Frontiers in Endocrinology, suggest that the COVID-19 pandemic may have fundamentally altered the long-term trajectory of semen quality parameters.

The research team, led by Shiyi Zhu, Jin Dong, and senior authors Chao Feng, Chunhua Tang, and Yuchuan Zhou at the International Peace Maternity and Child Health Hospital of Shanghai Jiao Tong University, extended their previous surveillance work at the same facility. Their earlier report, based on computer-assisted sperm analysis of 49,189 men during 2015 to 2021, had shown conventional semen parameters declining while sperm kinematic parameters rose. The new study covers the years in which the COVID-19 pandemic swept through China, including the nationwide relaxation of restrictions at the end of 2022, offering a rare dataset for exploring whether a global viral pandemic and its associated societal upheaval could leave measurable fingerprints on male reproductive health.

The study population consisted of men attending fertility-related clinical services at the Reproductive Center of the International Peace Maternity and Child Health Hospital between January 2021 and July 2025. From an initial database of 29,246 semen samples, the researchers applied predefined eligibility criteria, restricting analysis to men aged 18 to 60 years with reported sexual abstinence durations of 2 to 7 days, and using only the first sample when an individual had contributed more than one. After this data cleaning, 21,932 samples were retained. The cohort primarily represented men seeking infertility evaluation, assessment of oligoasthenospermia, or sperm freezing for assisted reproductive technology, a clinical context the authors acknowledge as a potential selection bias that likely over-represents men with subfertility compared with a random population sample.

Semen samples were collected by masturbation and allowed to liquefy at 37 degrees Celsius for at least 30 minutes before analysis. Semen volume was measured directly using a graduated collection vessel, while sperm concentration, sperm count, sperm motility, progressive motility, and kinematic parameters were assessed using a computer-assisted sperm analysis system manufactured by Hamilton-Thorne. The instrument was configured with a temperature of 35 degrees Celsius, a chamber depth of 20 micrometers, and a frame rate of 60 Hz, with thirty frames acquired per analysis. For each sample, a 5 microliter aliquot was loaded into a standardized counting chamber and examined immediately to minimize post-liquefaction variability. At least 200 spermatozoa were evaluated across ten microscopic fields per chamber, and all diagnostic assessments were conducted by two experienced technicians and confirmed by a laboratory director with andrology expertise. Crucially, the same equipment, protocols, and trained personnel were used throughout the entire 2021 to 2025 period, reducing the risk that methodological drift could explain the observed trends.

The recorded parameters included total sperm count in millions, sperm concentration in millions per milliliter, total motility in percent, progressive motility in percent, and four CASA-derived kinematic measures: curvilinear velocity (VCL), straight-line velocity (VSL), average path velocity (VAP), and beat-cross frequency (BCF). Spermatozoa with a path velocity exceeding 5 micrometers per second were classified as motile, while those with a path velocity greater than 25 micrometers per second and linearity greater than 80 percent were further categorized as progressively motile. To characterize temporal changes across different baseline semen-quality strata, participants were classified according to the WHO 2010 reference criteria into an above-reference-limit group (ARL; n = 10,504, 47.89 percent) meeting all thresholds, and a below-reference-limit group (BRL; n = 11,428, 52.11 percent) failing at least one. The reference thresholds were semen volume of at least 1.5 milliliters, total sperm count of at least 39 million, sperm concentration of at least 15 million per milliliter, total motility of at least 40 percent, and progressive motility of at least 32 percent.

In the unadjusted annual comparisons across the total cohort, sperm count and sperm concentration tended to increase during the later years of the observation period, while semen volume, total motility, and progressive motility showed no consistent annual changes in the descriptive analyses. The kinematic parameters told a different story: although VCL, VSL, VAP, and BCF fluctuated across individual years, all four showed overall upward trajectories from the beginning to the end of the study period. In the total sample, median VCL rose from 107.1 micrometers per second in 2021 to 113.5 in 2025, median VSL from 61.73 to 65.08, median VAP from 71.6 to 75.7, and median BCF from 27.66 to 28.51 Hz. After adjustment for age and abstinence duration, all four kinematic parameters remained significantly higher in the ARL group than in the BRL group, indicating that sperm kinematic indices provide additional quantitative information on motility performance and may complement conventional motility-based semen assessment.

The most robust findings emerged from multiple linear regression models fitted within the ARL group, adjusting for calendar year, age, and abstinence duration. After adjustment, semen volume increased by 0.08759 milliliters per year, sperm count by 4.815 million per year, and sperm concentration by 0.6929 million per milliliter per year, all statistically significant. In contrast, total motility declined significantly over the study period by 0.711 percentage points per year, while progressive motility showed a nonsignificant downward tendency of 0.1056 percentage points per year. The kinematic parameters displayed significant annual increases as well: VCL by 0.9423 micrometers per second, VSL by 0.3553, VAP by 0.515, and BCF by 0.1172 Hz, all with p values below 0.0001. The coexistence of increasing sperm quantity-related parameters and improving kinematic velocity indices alongside declining sperm motility suggests that different dimensions of semen quality may not change in parallel, highlighting the importance of integrating conventional semen parameters with CASA-derived kinematic measures when evaluating temporal changes in male reproductive health.

The dissociated pattern of declining motility but rising kinematic parameters may be partly explained by the distinct definitions of these measures. The motility parameter reflects the proportion of spermatozoa exhibiting any movement, whereas VCL, VSL, VAP, and BCF quantify movement characteristics, including velocity patterns and flagellar beating, within the motile sperm subpopulation. Slower or poorly motile spermatozoa may become classified as immotile or be underrepresented in kinematic calculations, resulting in higher average velocities among the remaining motile cells. Beyond this methodological explanation, the authors propose that improved kinematic values among motile spermatozoa may represent a compensatory response that partially offsets the decline in the proportion of motile cells, although aberrant or premature sperm hyperactivation cannot be excluded. The mechanisms underlying these dissociated changes remain insufficiently understood, and future longitudinal studies incorporating advanced sperm functional assays are needed.

Several potential reasons may account for the significant shift in semen quality parameters at the same reproductive center after the COVID-19 outbreak. The detection of SARS-CoV-2 in the testes and semen of infected individuals suggests the possibility of direct local effects on sperm cells, and SARS-CoV-2 infection has been associated with oxidative stress in the seminal environment, reflected by increased levels of reactive oxygen species. Such oxidative stress may be particularly detrimental to human spermatozoa, which are rich in polyunsaturated fatty acids and have limited antioxidant capacity. Meanwhile, recent studies have generally shown that COVID-19 vaccination does not adversely affect sperm quality, and some have reported increases in total sperm count, sperm concentration, and semen volume after vaccination, with sperm motility-related kinematic parameters also reported to differ before and after vaccination. In addition, lifestyle changes during lockdowns, including psychological stress, altered daily routines, increased sedentary behavior, and reduced outdoor activity, may have exerted bidirectional effects. On one hand, reduced physical activity and altered dietary patterns may increase the risk of weight gain and metabolic deterioration, impairing spermatogenesis. On the other hand, survey-based studies have reported improvements in dietary habits after the COVID-19 outbreak, including increased consumption of home-prepared and nutrient-rich foods, and higher intake of fruits and vegetables has been associated with improved semen parameters.

The study has important limitations that temper interpretation. The generalizability of the results is limited because the data were obtained from a single reproductive center and included both apparently healthy men and men seeking fertility evaluation, so the observed trends may not fully represent semen-parameter trends in the general population. The number of observations also decreases substantially over the study period, from 9,569 samples in 2021 to 940 in 2025, mainly due to the opening of a new branch hospital and the renovation of the old hospital area, which led to the referral of participants. Several important factors that may influence semen quality, including fever, COVID-19 severity, lifestyle, smoking, obesity, environmental exposures, medications, and changes in the patient population, were not captured, and the investigation lacks individual-level COVID-19 and vaccination information, meaning the relationship between the changing trend of semen quality and the COVID-19 pandemic is only indirectly inferred. Residual confounding by unmeasured factors cannot be excluded, and the clinical significance of the observed changes in kinematic parameters remains uncertain. Nevertheless, the authors conclude that the increasing trends in semen volume, sperm count, and sperm concentration from 2021 to 2025 are opposite to the decreasing trends observed at the same center before and during the COVID-19 outbreak, and that to their knowledge this is the first study to investigate the potential long-term association between the COVID-19 pandemic and semen quality. Whether the reversal reflects viral exposure, vaccination, pandemic-mediated lifestyle changes, or a complex interplay among these factors remains an open question that only continued surveillance and mechanistic studies can answer.

Subject of Research: Reproductive Medicine

Article Title: Temporal trends in semen quality at a single reproductive center in Shanghai, China, during 2021–2025 amid the COVID-19 pandemic

Article References: Zhu, S., Dong, J., Wu, Z., Wang, Z., Yang, J., Feng, C., Tang, C., & Zhou, Y. (2026). Temporal trends in semen quality at a single reproductive center in Shanghai, China, during 2021–2025 amid the COVID-19 pandemic. Frontiers in Endocrinology, 17, Article 1980263. https://doi.org/10.3389/fendo.2026.1980263

Image Credits: AI Generated

DOI: 10.3389/fendo.2026.1980263

Keywords: semen quality, COVID-19, SARS-CoV-2, sperm motility, computer-assisted sperm analysis, male fertility, Shanghai, Frontiers in Endocrinology, Temporal, trends, semen, quality