How to Read a Semen Analysis and the WHO Reference Limits

A reference limit is a percentile, not a verdict

The World Health Organization's 2021 sixth edition lower reference limits for semen analysis describe where the bottom 5 percent of recently fertile men fall, not a line that separates fertile from infertile. A result under one of these values means only that fewer than 1 in 20 men who fathered a child within a year scored that low on that measurement. It is a statistical benchmark drawn from a reference population, so a single number below the cutoff rarely settles whether a couple will conceive, and a single sample rarely captures a man's true baseline.

That framing matters because the numbers are easy to misread as a grade. Here is what they are and how to hold them.

Where the numbers come from

The sixth edition set its reference limits from roughly 3,589 fertile men across 12 countries and 5 continents whose partners conceived within 12 months of trying. For each measurement, the manual reports the 5th percentile of that group as the lower reference limit. By construction, 5 percent of demonstrably fertile men sit below it. The Fertility and Sterility overview from the manual's contributors is explicit that these are reference values for interpretation, not thresholds that diagnose infertility. A pooled analysis of the underlying distributions published in Andrology found that the 5th centiles were broadly similar to the 2010 edition, which suggests the values are reasonably stable.

The current lower reference limits are:

• Semen volume: 1.4 mL
• Sperm concentration: 16 million per mL
• Total sperm number: 39 million per ejaculate
• Total motility: 42 percent
• Progressive motility: 30 percent
• Vitality: 54 percent alive
• Normal morphology: 4 percent normal forms

Two of these often surprise people. The 4 percent morphology figure is not a sign that something is deeply wrong; it reflects strict criteria under which even fertile men typically show a low share of textbook forms. And the volume figure describes the fluid, which carries little independent weight on its own.

"Below the limit" is not "infertile"

Because each limit is a 5th percentile, falling under it places a man in the lower tail of the fertile distribution, not outside it. Plenty of men below a cutoff conceive without help, and plenty above every cutoff struggle for reasons a semen analysis never measures, from partner factors to timing. The Reproductive BioMedicine Online analysis of the 2021 values makes the point directly: reference limits describe a population, and they are useful but insufficient to diagnose infertility in an individual. A number just under a limit and a number far under it are different situations, and neither is a diagnosis by itself.

The mirror image error is just as common. A result comfortably above every limit is reassuring about those specific measurements, yet it does not rule out a fertility problem, because conception depends on far more than the handful of things a basic analysis counts.

Why one sample rarely settles it

Semen parameters swing a great deal within the same man from week to week. A fever in the prior two to three months, the length of abstinence before collection, a recent illness, or an incomplete collection can all move the numbers substantially. That is why laboratory and clinical guidance generally treats an abnormal result as a prompt to repeat the test, often after several weeks, rather than as a conclusion. Two or more samples spaced out give a far better read on a man's baseline than any single snapshot, and a trend across samples usually tells you more than one value against a table.

How the sample is produced and handled matters too. The sixth edition placed heavy emphasis on standardized methods and laboratory quality, because a low count can sometimes reflect how a specimen was collected, transported, or read rather than the man who produced it.

What the reference population can and cannot tell you

The reference group behind these limits has real strengths, including broader geographic coverage than earlier editions, and real limits. A critical reappraisal in Human Reproduction found the underlying studies heterogeneous enough that treating them as one clean population is questionable, with 5th percentile estimates for some parameters varying widely from study to study. The practical point is that these are approximate reference points rather than precise dividing lines, closer to guidance than to a hard cutoff. Some andrologists distinguish reference limits, which describe a population, from decision limits, which would be tuned to predict an outcome such as conception or a response to treatment. The WHO values are the former.

Reading your own report

A workable way to read a semen analysis is to notice which measurements are flagged, by how much, and whether the pattern holds on a repeat. One value a hair under the limit, on a single sample, with a short abstinence interval, carries a very different weight than several values well below across two properly collected samples. The result belongs next to the rest of the clinical picture, including the partner's evaluation, rather than standing alone as a score. Interpreting all of that is the job of the clinician who ordered the test.

This article is educational and is not medical advice; decisions about testing and fertility belong with a qualified clinician who knows the specific situation.