An evaluation of changes over time in the semen parameters data used for the World Health Organization semen analysis...

Previous meta-analyses concluded that there is a decline in sperm parameters over time. This conclusion might be incorrect due to inherent biases or focusing only on a single parameter - sperm concentration.

To study trends in sperm parameters over the past 20 years using data from the trials that defined the reference ranges of the World Health Organization manual.

Retrospective evaluation of the data used to define the World Health Organization reference ranges. The data from 11 studies, including 3589 participants between 1996 and 2016, were divided into three period groups based on the decade of study. Differences in semen parameters' distribution were presented in boxplot. p-values were calculated by the Kruskal-Wallis rank-sum test followed by Dunn post hoc test. Analyses were conducted using the R programming language.

A small decrease was noted in mean sperm concentrations (88.1 million/ml, 87.6 million/ml, and 77.2 million/ml for the first, second, and third decades, respectively) (p < 0.01). However, the 5th percentile of sperm concentration for the third decade was higher than the first or second decades (18 million/ml versus 14.9 million/ml and 15 million/ml, respectively). No significant differences were noted in progressive motility over the years (p = 0.32). The percent of morphologically normal sperm decreased between the first (24.2%) and the second (12.6%) periods of the study (p < 0.001) and then increased in the third decade (14.2%) (p < 0.01). Total motile sperm count (TMC) declined between the second and third decades (189 million and 153.9 million, respectively, p < 0.001), at levels unlikely to decrease fertility. However, the 5th percentile of the TMC remained stable at 24.9, 20.8, and 20.6 million, for the first, second, and third decades respectively (p = 0.36).

Trends in sperm parameters over the last three decades do not seem to be clinically significant.