Azoospermia: Obstructive or Not Decides Everything

Not on this evidence, and not yet. A finding of no sperm in one ejaculate is the beginning of an investigation, not its conclusion. Some men with this result have sperm being produced normally and simply blocked on the way out, which is often fixable or bypassable. Others have impaired production, where sperm can still frequently be found inside the testicle even when none reach the ejaculate.

What the result does mean is that you need a urologist or andrologist rather than reassurance, and that the next steps are specific and worth knowing. The single question that organises everything that follows is whether the azoospermia is obstructive or non-obstructive.

Check this before absorbing anything else. Hubbard and colleagues (Asian J Androl 2025;27:298-306) define azoospermia as the absence of spermatozoa in the sediment of a properly centrifuged ejaculate. If the lab looked at a drop of uncentrifuged semen under a microscope and reported nil, the diagnosis is not established — spinning the sample down and examining the pellet is what distinguishes true azoospermia from very few sperm.

The distinction is not academic. Hubbard's review cites a series in which cytocentrifugation and staining found sperm in 60 of 251 men already diagnosed as azoospermic. Finding even rare sperm changes the plan, because sperm from the ejaculate can be used for ICSI without an operation. A confirmed diagnosis normally rests on more than one properly processed sample.

Mostly by two cheap things: an FSH blood level and a physical examination of the testicles. In obstructive azoospermia the testicles are working, so they are normal in size and the pituitary is not straining to drive them, which keeps FSH normal. In non-obstructive azoospermia the testicles are usually smaller and FSH is usually raised.

The separation is better than it sounds. In the cohort Hubbard and colleagues cite, over 96% of men with obstructive azoospermia had an FSH below 7.6 mIU/mL together with a testicular length above 4.6 cm. In that same cohort a testicular long axis under 4.6 cm predicted spermatogenic failure roughly 90% of the time, and a total testicular volume above 15 mL tends to indicate obstruction.

Two honest caveats. A normal FSH does not always mean obstruction — Hubbard's review names spermatogenic arrest and Sertoli cell-only syndrome as non-obstructive conditions that can present with a normal FSH. And ejaculate volume, contrary to what is often assumed, is seldom a reliable indicator, because any obstruction above the ejaculatory ducts leaves the volume normal.

The sequence below is what Hubbard and colleagues describe as the contemporary workup. Not all of it applies to every man, and which half applies depends on the obstructive question above.

• A repeat semen analysis with proper centrifugation and examination of the pellet, plus pH and fructose where the volume is low.
• Hormones: FSH, LH, prolactin and total testosterone in every man with azoospermia. Inhibin B is optional.
• Examination and measurement of testicular volume, and palpation for the vas deferens, epididymal nodules and a varicocele.
• Scrotal ultrasound, and transrectal ultrasound where obstruction low in the tract is suspected.
• In non-obstructive azoospermia: karyotype, and Y-chromosome microdeletion testing for the AZFa, AZFb and AZFc regions. Hubbard's review calls these mandatory in non-obstructive azoospermia and says they should not be used in confirmed isolated obstruction.
• Where the vas deferens is absent or underdeveloped: CFTR testing. Up to 78% of men with congenital bilateral or unilateral absence of the vas carry at least one CFTR mutation, on the prevalence figures in that review.

One thing that has changed: international guidance now recommends against a testicular biopsy done purely for diagnosis, and advises combining it with a therapeutic sperm retrieval so a man is not put through two operations. If a biopsy is proposed on its own, that is a reasonable thing to question. See also genetic counselling for male infertility.

In obstructive azoospermia, usually — sperm are being made, and the problem is access. In non-obstructive azoospermia it is roughly a coin toss, and no honest figure is better than that.

Major and colleagues pooled 29 studies in a systematic review and meta-analysis (Asian J Androl 2023;25:38-42). Across 20 studies and 4,760 men with non-obstructive azoospermia, microdissection TESE retrieved sperm in 51.9% (95% CI 46.1-57.7), versus 40.1% (95% CI 34.8-45.5) across 9 studies and 1,227 men undergoing conventional TESE. The difference in favour of micro-TESE was about 11.8 percentage points.

Genetics shifts the odds in both directions, and this is the part worth knowing before surgery. Jiao and colleagues systematically reviewed men with an AZFc microdeletion (BJU Int 2024;134:6-12): sperm were obtained in 275 of 441 men, an overall rate of 62.4%, though individual studies ranged from 25.0% to 85.7%. But complete deletions of AZFa, AZFb, AZFb+c or AZFa+b+c are different in kind — Hubbard's review states these carry no prospect of retrieval at micro-TESE. That is precisely why the genetic test comes before the operation rather than after it.

Retrieval is not assured in non-obstructive azoospermia, and a man should be counselled on what happens if nothing is found, including when donor sperm becomes the question. For the procedures themselves, see what TESE is and when it is used.

Yes, and it is the cause most often missing from material written outside India. Kumar, in the Indian Journal of Urology (2008;24:392-5), describes how genital tuberculosis obstructs the reproductive tract through granulomas and scarring, and notes that infertility may be the first and only presentation — with the man recalling no illness. In one series Kumar cites, the epididymis was involved in over 78% of 69 patients with genital tract tuberculosis.

How often is it the answer? Yadav and colleagues, reviewing genital tuberculosis (Transl Androl Urol 2017;6:222-33), report a prevalence of around 4% to 9.1% among men with infertility who carry a clinical diagnosis of obstructive azoospermia — which is the group you are in if you are reading this section, and is the reason anyone screens at all. The same review puts the prevalence of infertility as a presenting complaint among men with genital tuberculosis at about 10%. That review attributes the range to three earlier studies rather than measuring it, so it is a secondary source, and it shares an author with the Kumar review cited above — the two are not independent.

The clinical signs to raise it are a nodule in the epididymis or beading of the vas on examination, or a low-volume, acidic, fructose-negative ejaculate pointing to obstruction at the ejaculatory ducts. Kumar also cites a series of 50 men who had ejaculatory duct surgery, in which 17 had atrophic seminal vesicles on ultrasound and 15 of those 17 had a history of pulmonary tuberculosis.

Where in the tract the scarring sits also changes the report itself, which is useful when the obstructive picture is uncertain. Yadav and colleagues note that where tuberculosis involves only the epididymis or the vas, the ejaculate is of normal volume and fructose-positive, with azoospermia or severe oligospermia and a normal hormone profile. It is involvement further down — the ejaculatory ducts, seminal vesicles and prostate — that produces the small, acidic, fructose-negative sample. A normal volume therefore does not argue against tuberculous obstruction.

Two honest qualifications. Kumar describes infertility as an uncommon manifestation of genitourinary tuberculosis overall, so this is a cause to consider rather than to assume. And testing for it has limits: Gupta and colleagues screened 100 Indian men with idiopathic obstructive azoospermia using semen TB PCR (J Hum Reprod Sci 2015;8:43-7). Seven of 100 tested positive, four of whom had other clinical evidence of tuberculosis; none of those treated had any improvement in semen parameters, and the authors concluded that PCR screening added nothing beyond clinical evaluation. The practical reading is that tuberculosis is worth a doctor examining for and asking about, and not worth buying a PCR test for independently.

If empirical antitubercular therapy is offered to you, the evidence behind it is genuinely divided and you are entitled to know that before starting six months of four-drug treatment. Yadav's review cites reappearance of sperm in the ejaculate in about a third of cases treated empirically with antitubercular therapy plus steroids. Gupta and colleagues saw no improvement in semen parameters in any of the men they treated. Both are cited here and neither settles it; what is not in dispute is that curing the infection does not undo scarring that has already formed.

The hopeful part is real. Kumar reports that where tuberculous obstruction is present, the testis is usually spared and testicular sperm is almost always available, and that assisted reproduction in this group gives results comparable to obstructive azoospermia from other causes.

Diagnosis can be largely settled within a few weeks — a repeat centrifuged semen analysis, a hormone panel, an examination and, where indicated, genetic tests. Anything aimed at sperm production is slower. Heller and Clermont measured human spermatogenesis directly with labelled thymidine and serial biopsies, and put one cycle of the seminiferous epithelium at 16 days and the whole process at about 64 days (Science 1963;140:184-6). Sperm then have to transit the epididymis. That is why any treatment intended to affect production is reassessed at around three months rather than three weeks.

The gaps here are large enough that pretending otherwise would mislead a decision.

• No pre-operative test reliably predicts whether sperm will be found. Major and colleagues found FSH inversely related to retrieval, but Hubbard's review reports that neither FSH nor inhibin B was predictive of retrieval rates. These two sources genuinely disagree, and the honest position is that FSH helps separate obstructive from non-obstructive azoospermia but does not tell an individual man whether his operation will succeed.
• Testicular volume is widely quoted as a predictor of retrieval success. Major and colleagues could not assess it: they state the available records were insufficient to identify statistically significant effects for testicular volume or testosterone. It helps distinguish obstruction from impaired production; it is not an established predictor of retrieval.
• The 51.9% figure is a weighted mean of 29 studies' averages, not an individual probability, and the constituent studies ranged widely. It should not be read as one man's odds.
• The tuberculosis figure is weaker than a single number looks. The 4% to 9.1% prevalence comes from a review citing three earlier studies rather than from one measurement, that review shares an author with the Kumar review cited alongside it, and a 2026 systematic review and meta-analysis of tuberculosis presenting as infertility (Int J Tuberc Lung Dis 2026;30:193-7) pooled 11 studies and 2,197 patients but reported that male cases could not be analysed for lack of detailed studies.
• Whether empirical antitubercular therapy restores sperm to the ejaculate is unsettled. Yadav's review cites reappearance of sperm in about a third of cases treated with antitubercular therapy plus steroids; Gupta and colleagues saw no improvement in any man they treated. Both are cited above and neither settles it.
• Known genetic anomalies explain only around 25% of non-obstructive azoospermia versus roughly three-quarters that remain idiopathic (Cervan-Martin et al., J Clin Med 2020;9:300). A normal karyotype and normal AZF testing do not identify a cause.
• This page names no drug, supplement or hormonal regimen as restoring sperm to the ejaculate in non-obstructive azoospermia, because no source retrieved for it establishes that.

One thing worth saying directly, because azoospermia is often read as a final verdict: the most common outcome of this investigation is not that nothing can be done. It is a specific diagnosis, with a specific route, and for a substantial proportion of men that route leads to their own genetic child. For some it will not, and that possibility deserves to be discussed early rather than discovered late.