This retrospective study assessed the biological intra-individual variability of the percentage of sperm with DNA damage (SDF) observed in subsequent ejaculates of the same individual. Variation in SDF was analyzed using the Mean Signed Difference (MSD) statistic based on 131 individuals, comprising 333 ejaculates. Either two, three or four ejaculates were collected from each individual. With this cohort of individuals two main questions were addressed; (1) does the number of ejaculates analyzed influence the variability in the level of SDF associated with each individual? and (2) is the variability observed in SDF similar when individuals are ranked according to their level of SDF? Results showed that the variation observed in mean SDF was not different when 2, 3 or 4 ejaculates were analyzed; consequently, we suggest that the assessment of SDF based on two ejaculates is likely to be representative of the mean SDF expected for the individual. In parallel, it was determined that the variation in SDF increased as SDF increased; in individuals presenting with an SDF value of lower than 30% (potentially fertile), only 5% possessed levels of MSD that could be considered as variable as that presented by individuals presenting with a recurrent high SDF. Finally, we showed that a single assessment of SDF in individuals with medium SDF (20-30%) was less likely to be predictive of the SDF value in the next ejaculate, and therefore, less informative of the patient's SDF status.
Keywords: Sperm DNA fragmentation; clinical andrology; ejaculate variability; human reproduction; male factor.