Family physicians are often involved in the initial stages of an infertility work-up, and this typically includes a sperm analysis. While various entities have published suggested normal values for sperm count, motility, and morphology, these values have been questioned for their evidence basis and reliability for predicting infertility. Guzick and colleagues report on a comparison of sperm analysis parameters in infertile and fertile men.
Infertile men were defined as those whose female partners had regular menstrual cycles, normal hysterosalpingogram results, laparoscopy and luteal-phase endometrial biopsies, yet had failed to conceive for more than 12 months. Fertile men had partners who had become pregnant within the two years before the study. Semen specimens were collected, on two different occasions at least two weeks apart, from 765 infertile men and 696 fertile age-matched control subjects. Demographic analysis showed that infertile men were more likely to be white, to have lower educational levels, to be current smokers, and to consume alcohol.
There was a wide range of values for sperm count, motility, and percentage of normal morphologic forms for infertile and fertile men, and these values largely overlapped. For example, the average sperm count for infertile men was 52 million per mL, with a standard deviation range of 10 to 94 million per mL. Fertile men averaged 67 million sperm per mL, with a range of 17 to 117 million per mL. The authors used statistical analysis to identify values for each parameter that would have approximately 95 percent specificity for infertility. Men with values above these levels would be fertile in 95 percent of cases. Looking for this high degree of specificity—so that very few men would be incorrectly labeled as infertile—had a tradeoff in lower sensitivity. Depending on the parameter, only 14 to 43 percent of all infertile men were identified as such by the threshold values selected.
The authors conclude that a sperm count of less than 13.5 million per mL, motility less than 32 percent, and less than 9 percent normal morphologic sperm forms predicted infertility (see accompanying table). Fertility was predicted by a sperm count greater than 48 million per mL, motility greater than 63 percent, and greater than 12 percent normal forms. Values between these thresholds were considered indeterminate.