Within-subject variability of the percentage of morphologically abnormal spermatozoa among fertile men: biological and measurement components.

Two groups of men were studied to determine how much an individual's spermatozoal morphology varies and which part of the variability is due to biological and to measurement variation. A first group of ten subjects each provided five ejaculates; the other group of four subjects each provided one ejaculate, from each of which five smears were prepared, and evaluated twice. The slides were examined in random order and blindly by the same experienced technician. The total within-subject variability of the percentage of abnormal forms was large (SD = 8.8%), as was the component of this variability due to biologic fluctuation (SD = 5.8%). The distribution of the various abnormalities also varied greatly from one ejaculate to another. The results indicate that more work is needed to determine why biological variation exists, and that several ejaculates (three if possible, two at minimum) are needed for an accurate evaluation of spermatozoal morphology, whereas the examination of greater than 100 spermatozoa/ejaculate is not advantageous.