Abstract
BACKGROUND: Ejaculatory abstinence (EA) is the key to assessing the semen analysis. While its fundamental roles on all sperm and semen parameters have been studied for decades, there are still controversies about whether shortening or lengthening EA would be beneficial. Despite natural variations of human semen, most studies in this field investigate the influence of EA using between-individual approaches that cannot control intra-individual covariates. There is still little evidence on how deviation in EA between two samplings affects variations in semen parameters. This study aimed to revisit the relationship between EA and semen parameters, especially in the within-individual analysis and in terms of two-time EA deviations. METHODS: A cross-sectional study was conducted on 11,297 conventional semen examinations from 9,595 men who presented for reproductive health check-ups between May 2017 and December 2022, aiming to assess between-individual variation. Among them, 1,702 men doing semen analysis twice within 1 month were selected to investigate the role of two-time EA deviation further. RESULTS: EA positively correlated with the semen volume, sperm concentration, and total motile sperm count (TMSC), consistent in both between- and within-individual analyses. However, according to the linear regression model, there were no clear peaks in the above parameters following EA elongation. Sperm concentration and TMSC from the two samplings differed when the two-time EA deviation was no more than 1 day. On the other hand, the proportion of total motility tends to increase with lengthening the EA (β=0.16, P<0.01) in between-individual but not in within-individual analysis. Moreover, this study showed no correlation between the straight-line velocity (VSL) and EA. Variations in semen parameters would be reduced when the EA deviation between two samplings was decreased. CONCLUSIONS: This study reaffirms the importance of EA in sperm quantity. EA should be maintained consistently or deviate by no more than 1 day to minimize variations between the two samples.