Male Reproductive Anatomy: Where Fertility Actually Fails

At four separate points, and which one has gone wrong decides everything that follows. Sperm are made in the testes, mature and are stored in the epididymis, travel out through the vas deferens, and leave mixed with fluid from the seminal vesicles and prostate — which supplies most of the volume.

Four steps, four distinct ways to fail, and tests that exist specifically to tell them apart.

• Testes — make sperm and testosterone. Production failure shows up as azoospermia or a low count, classically alongside a raised FSH and small, soft testes.
• Epididymis — where sperm mature and are stored. A blockage here gives azoospermia with a normal FSH and normal testicular volume: the factory is running, the road is shut.
• Vas deferens — transport. It can be absent from birth, or divided by vasectomy. Either way production continues and nothing arrives.
• Seminal vesicles and prostate — most of the fluid. Obstruction or absence at this level tends to present as a low-volume ejaculate rather than a low count.
• Pituitary and hypothalamus — the FSH and LH signal that drives the testes. Failure here is hypogonadotropic hypogonadism, and it is the one on this list most often correctable with medication.

If one thing on this page is worth understanding, it is this. Azoospermia — no sperm in the ejaculate — means two completely different things depending on whether the testes are producing and something is in the way, or the testes are not producing at all. The word on the report is the same. The outlook is not.

A blockage is the better news of the two. Production is intact, so sperm can usually be retrieved directly from the testis or the epididymis and used with ICSI. A production failure means there may be little or nothing to retrieve, and the procedure that looks for it can come back empty.

Being handed a one-word diagnosis in a short conversation and looking it up afterwards is how most people arrive at a page like this. This is the part that is most often left unexplained, and it is the part that determines what happens next.

Two findings separate them, and neither is on a semen report:

• FSH. A raised FSH means the pituitary is shouting at testes that are not responding — a signal of impaired production. A normal FSH alongside azoospermia points downstream, at a blockage.
• Testicular volume on examination. Small, soft testes go with production failure; normal-sized testes with azoospermia point at obstruction.

This is why the AUA/ASRM 2021 guideline pairs the semen result with hormones and a physical examination rather than reading it alone. Statement 10 calls for FSH and testosterone in infertile males with impaired libido, erectile dysfunction, oligozoospermia or azoospermia, atrophic testes, or evidence of hormonal abnormality on examination.

A third pattern sits further downstream again. A reduced-volume ejaculate points at the delivery end rather than the production end, because most of the volume is accessory gland fluid rather than sperm — and statement 53 directs a post-orgasmic urinalysis in males with aspermia after retroperitoneal lymph node dissection and reduced volume ejaculate who are interested in fertility. Our article on azoospermia covers that workup in more detail.

Where recent fathers happened to sit — not a line between fertile and not. WHO's 2021 fifth-centile values come from a cohort of 3,589 men in 13 countries whose partners conceived within 12 months:

• Total sperm number — 39 million per ejaculate
• Total motility — 42% in that cohort, and progressive motility 30%
• Normal forms — 4% in that cohort

The sixth edition says explicitly that reference ranges and fifth centiles are insufficient to diagnose infertility. A result just below one of them is not a diagnosis, and a result just above is not a clean bill of health. What makes a semen result meaningful is the pattern it forms with the FSH and the examination.

Timing matters too. A cycle of sperm production takes around 74 days — an estimate a 2008 review notes rests on a 1960s data set and may be out by roughly six days — so a repeat test at one month is largely measuring the previous batch. How to read a semen analysis report goes through the columns one by one.

Three different answers, and conflating them is where counselling goes wrong. Only one of the three treats the problem; the other two get around it.

• Signal failure — treated. Hypogonadotropic hypogonadism is evaluated for its cause and treated on that diagnosis (statement 40). This is the one route where medication addresses the problem rather than bypassing it.
• Obstruction — bypassed, reliably. Production is intact, so sperm can be taken from the testis or the epididymis (statement 31) and used with ICSI, fresh or cryopreserved (statement 30). After a vasectomy, reconstruction, retrieval, or both together are all described as viable options (statement 35).
• Production failure — bypassed, uncertainly. Where azoospermia is non-obstructive, retrieval is by microdissection testicular sperm extraction (statement 29), and it may find nothing. That possibility should be part of the conversation before the procedure, not after it.

Where retrieval is the route, the cycle is committed to ICSI, which changes the shape of IVF for the couple — how male factor changes an IVF cycle covers what that involves.

Repair it only if it can be felt, and expect an honest conversation about how uncertain the benefit is. The guideline is specific on both halves: statement 26 says to consider surgical varicocelectomy in males with a palpable varicocele, infertility and abnormal semen parameters, excluding azoospermic males; statement 27 says not to operate on varicoceles found only on imaging.

On whether surgery produces more pregnancies, the Cochrane review behind that caution pooled 48 studies and 5,384 participants. Treatment may improve pregnancy rates against delayed or no treatment (RR 1.55, 95% CI 1.06 to 2.26, across 13 randomised trials and 1,193 men) — rated low certainty, and moving a baseline chance of pregnancy of 21% to somewhere between 22% and 48%. That range is wide because the evidence is weak, not because the surgery is unpredictable.

On whether it produces more babies, the answer is that nobody knows. The live birth estimate is RR 2.27 with a 95% confidence interval running from 0.19 to 26.93, across 2 randomised trials and 204 men, rated very low certainty. An interval that wide is not a weak positive result; it is a statement that the question is unanswered.

If surgery is going ahead, the technique has better evidence behind it than the decision does. Microscopic subinguinal repair probably improves pregnancy slightly over other surgical approaches (RR 1.18, 95% CI 1.02 to 1.36, across 12 randomised trials and 1,473 men, moderate certainty). That is worth asking about by name.

At the severe end of the range, and the guideline draws the line numerically rather than by clinical impression. Statement 12 recommends karyotype testing at azoospermia or a sperm concentration below 5 million/mL where FSH is elevated, the testes are atrophic, or sperm production is known to be impaired. Statement 13 recommends Y-chromosome microdeletion analysis at azoospermia or 1 million/mL or below, with the same accompanying findings.

Absence of the vas deferens is the genetic failure with an anatomical signature, and it is the one where the result changes something for the couple rather than only explaining the diagnosis. Its reported prevalence is approximately 0.1% of men — probably an underestimate, since unilateral absence in fertile men is rarely diagnosed. The majority of those affected carry at least one cystic fibrosis-causing mutation, which is why CFTR testing and genetic counselling before conception are warranted: the question becomes what the couple's child could inherit, not only how to conceive.

The genetic picture is incomplete, and it is worth knowing that before a negative result is over-read. Around 2% of cases carry a loss-of-function ADGRG2 variant, and the prevalence data leave 10 to 20% of bilateral cases with no genetic explanation at all. Our article on the genetic causes of male infertility covers what each test actually returns.

The workup is mostly cheap, quick and outpatient. It is the conclusions drawn from it that are expensive, which is a good reason to make sure the conclusions are properly supported.

1. A semen analysis, usually more than one. Results vary between samples from the same man, and a single abnormal result is a reason to repeat rather than a diagnosis.
2. A physical examination. Testicular size and consistency, whether the vas deferens can be felt on each side, and whether a varicocele is palpable. None of these is on a report and all three change the route.
3. Blood tests — FSH and testosterone at minimum, per statement 10, because the pairing with the semen result is what separates a blockage from a production failure.
4. Targeted tests only if the numbers cross the guideline's thresholds: karyotype, Y-microdeletion, CFTR, scrotal or transrectal ultrasound, post-orgasmic urinalysis.
5. A named diagnosis, or an honest statement that the cause has not been identified. Both are acceptable answers. "Male factor" on its own is not one.

Five questions worth asking before leaving: which of the four steps do you think has failed, and what points you there; is this obstructive or non-obstructive, and how confident are you; what would change your mind; if retrieval is proposed, what is the chance of finding no sperm; and what is the next test rather than the next treatment.

On cost, the diagnostic side is far lighter than the treatment side. In an ICMR-led study across five Indian tertiary facilities, couples seeking care for male infertility reported a median annual out-of-pocket spend of Rs 13,211, with an interquartile range of Rs 6,654 to Rs 21,521, covering diagnosis, management and IUI. The median across all infertility diagnoses was Rs 11,317, higher in private facilities (Rs 14,217) than public ones (Rs 8,355). Those figures exclude IVF — see fertility treatment cost in India for that.

Four honest limits, because each of them is routinely stated with more confidence than the evidence carries.

That varicocele repair produces more babies. On live birth the Cochrane estimate is RR 2.27 with a 95% confidence interval from 0.19 to 26.93, across 2 randomised trials and 204 men, rated very low certainty. That is not a small effect; it is an unmeasured one.

That a semen analysis inside the reference range means the tract is working. The WHO's sixth edition says plainly that reference ranges and fifth centiles are insufficient to diagnose infertility.

That a repeat test at one month means anything. A cycle of sperm production takes around 74 days, and the 2008 review of that estimate notes it rests on a 1960s data set and may be out by roughly six days either way.

And that a negative genetic result excludes a genetic cause. The prevalence data leave 10 to 20% of bilateral absence of the vas deferens with no identified genetic explanation, so a clear CFTR result does not rule the condition out.