What does an abnormal semen analysis really mean?
Welcome back Techno Tidbits! Last week, we reviewed the major components of the semen analysis. This week, in keeping with the spirit of Movember let’s explore what it means to have an “abnormal” semen analysis.
Traditionally, normal values were determined by examining sperm from lots of men who had demonstrated fertility (ie, had kids). Average levels were calculated, and levels below the mean were labeled as abnormal. The fact that half of the men with less than average values also had demonstrated fertility was ignored by those who set the early standards.
In 1999, the WHO changed the label from “normal values” to “reference values”, acknowledging that results outside typical ranges do not rule out the potential for in vivo fertility (aka, doing it the old-fashioned way). The threshold parameters are actually of little value in discriminating between fertile and infertile men . In statistical terms, semen analysis results have relatively low positive predictive value for infertility.
The most controversial parameter in terms of influence on fertility is definitely morphology, which refers to the proportion of normally shaped sperm. The concept arose from the observation that only sperm of a particular shape appeared in the cervical mucus, compared to the variable appearance of sperm in the entire ejaculate. Later, as IVF evolved, the observation was extended to the fact that only a certain shape of sperm bound to the zona pellucida (the outer surface of the egg).
Here’s a picture of what sperm bound to the egg look like (from the WHO manual):
The standards of what constitutes a normally shaped sperm, and the percentage below which were thought to signify infertility has changed numerous times over the years. Currently, the strict Tygerberg criteria, more commonly known as Kruger criteria after the scientist that championed the approach, uses a cut-off level of 4% normal sperm. Depending on the study that you read, a low Kruger morphology level can identify males with poor fertility potential.
However, when criteria are so strict, it is much harder to judge what is a normal vs abnormal sperm under the microscope. What appears as normal to one andrologist might look abnormal to another. There is more interobserver variation: beauty is in the eye of the beholder. This is why there is still abundant skepticism about the value of morphological grading even when numerous studies have documented its usefulness. Improvements in computer vision and artificial intelligence promise to improve the objectivity of grading sperm.
Most of the time, semen analysis results are in the gray zone: maybe not perfect, but not absolutely terrible either. Furthermore, the significance of the different sperm parameters is not equally weighted. A higher concentration may compensate for a lower motility, or vice versa. Similarly, an abnormality of just one parameter is less than impactful than two problems, which is not as bad as three.
Numerous researchers have tried to use statistical methods such as ROC (receiver operator curves) and CART (classification and regression tree) to determine the most predictive combination of values. Guzick et al classified results as normal, indeterminant, and abnormal based on the combination of concentration, motility, and morphology . However, there was significant overlap in values between groups, and even the lowest category was labeled subfertile rather than infertile, implying that these results still don’t absolutely rule out the chance of spontaneous conception.
Bottom line: the semen analysis is a critical part of the fertility evaluation, but like most screening test, has significant limitations in predictive value. In rare instances, such as when no sperm are seen (azoospermia), or none are moving (necrospermia), the results will be definitive for a diagnosis of infertility. However, in most instances, while “abnormal” semen analysis results may indicate a reason for infertility, they do not absolutely rule out the ability to conceive naturally. The results need to be interpreted in the big picture.
On the flip side, a perfectly normal semen analysis does not mean that there is no problem on the male side either. There can theoretically be problems that we cannot see in the traditional semen analysis, such as DNA fragmentation, oxidative stress, etc. More on this later…
 S. E. M. Lewis, “REPRODUCTION REVIEW Focus on Determinants of Male Fertility Is sperm evaluation useful in predicting human fertility?,” Reproduction, vol. 134, pp. 31–40, 2007.
 D. S. Guzick et al., “Sperm Morphology, Motility, and Concentration in Fertile and Infertile Men,” N. Engl. J. Med., 2001.