Sperm DNA Fragmentation Testing: What It Measures, and Whether It Should Change Anything
By Mr Giangiacomo Ollandini, Consultant Urological Surgeon & Andrologist. Published 31 July 2026.
If you have found your way to this page, there is a reasonable chance you have already read a great deal about sperm DNA fragmentation, quite possibly late at night, and come away more unsettled than when you started. Some of what you read will have told you this test explains everything a standard semen analysis missed. Some of it will have told you the test is worthless and NICE has said so. Both of those things are being said in public, by people who are not fools, and neither is quite the whole picture.
Before anything else, since it is probably why you are here: an elevated result is not a verdict, and it is not a statement that you cannot father a child. Here is what I want to do on this page: set out what the test measures, why the laboratories disagree with one another, what the evidence does and does not support, and why two serious guideline bodies have landed in different places. Then you can decide, with your clinician, whether the result would actually change anything for you — which is the only question that really matters.
A declaration first. I have no commercial interest in sperm DNA fragmentation testing. I do not sell it, I do not have an arrangement with a laboratory that performs it, and I earn nothing whether you have it or you do not. I should also say plainly that I am a urological surgeon: I perform varicocele repair, and I am involved in surgical sperm retrieval. Both are mentioned on this page. On a question this contested you are entitled to know whether the person explaining it stands to gain from your answer — and to notice that the argument I go on to make is an argument against acting on the number.
What the test actually measures
A standard semen analysis counts sperm, watches how they swim, and looks at their shape. It says nothing about the genetic material inside the head of the sperm. Sperm DNA fragmentation testing attempts to fill that gap: it estimates what proportion of sperm in a sample carry breaks — single-stranded or double-stranded — in their DNA.
The rationale is intuitive. A sperm that swims perfectly well may still be carrying damaged genetic cargo. Damage can arise during sperm production, during maturation and transport, or from oxidative stress — a chemical imbalance in which reactive molecules outpace the body's ability to neutralise them, and which is influenced by things like smoking, obesity, infection, heat and age. The result is usually expressed as a DNA Fragmentation Index, a percentage of affected sperm.
That is where the intuitive part stops.
Why the laboratories disagree with each other
There is no single sperm DNA fragmentation test. There are at least four in common use — SCSA, TUNEL, Comet and SCD — and they do not measure the same thing. Some detect breaks directly; others infer damage from how the DNA behaves under chemical or physical stress. They also do not agree with one another in any straightforward way, and the clearest evidence of that is what happens to the thresholds.
When one group ran five assays across a single group of men, the cut-off that best separated fertile donors from infertile patients came out as 20.1% by TUNEL, 18.9% by SCSA, 22.8% by SCD and 45.4% by alkaline Comet (Ribas-Maynou and colleagues, Andrology, 2013). Same men, same laboratory, same day — four different numbers, one of them more than twice another. There is no agreed universal cut-off, and each laboratory sets its own reference range.
Two things about those figures are worth knowing, and both cut in the same direction. That was a single-centre study of a few hundred men, so the numbers themselves are estimates rather than constants. And the cut-offs were chosen to separate fertile donors from infertile patients — a classification — not to separate men who go on to father a child from men who do not. They are a line drawn through a population, not a prediction about you.
There is also the question of whether one test is one measurement. A fragmentation result moves with how long it has been since you last ejaculated, with a recent fever or infection, and between two samples given by the same man a few weeks apart. If you take one message from this page, let it be that a single number, from a single sample, run by a single assay, is a thinner thing than it appears on the report.
This is why you should be sceptical of any source — including a clinic's own literature — that quotes a single number as though it were a fact of nature. A result of 25% means something quite different depending on which assay produced it, and possibly nothing at all if the laboratory's reference range was derived from a different population than the one you belong to.
Which test will my laboratory use?
What it does. The sample is treated with acid and a dye. Sperm whose DNA is already damaged come apart more readily under that treatment, and the dye reports the difference. So SCSA infers damage from how the DNA behaves under stress, rather than counting breaks one by one.
Its threshold. In the population in which all four tests were compared directly, the SCSA cut-off separating “normal” from “elevated” came out at 18.9%.
What that means for your report. A number is only interpretable against the reference range your own laboratory uses, which it will have derived from its own population. Ask for it.
What it does. TUNEL labels the loose ends where DNA strands have broken, so it detects breaks fairly directly rather than inferring them.
Its threshold. In that same comparison population, the TUNEL cut-off was 20.1%.
What that means for your report. TUNEL and SCSA thresholds sit close together here, but that is a fact about one population, not a rule. Neither number transfers to a laboratory that has not validated it.
What it does. Sperm are treated so that undamaged DNA fans out into a visible halo around the head, while damaged DNA does not. Damage is read off the absence of a halo — again an inference, not a direct count.
Its threshold. In the same population, 22.8%.
What that means for your report. SCD is widely offered because it is comparatively simple to run. That is a practical virtue, not evidence that its answer is the right one.
What it does. Individual sperm are made to migrate through a gel under an electric field. Broken DNA fragments travel further, trailing behind the head like a comet's tail — hence the name.
Its threshold. In the same population, 45.4% — more than double the SCSA figure.
What that means for your report. Comet returns systematically higher values than the other assays on the very same sample, which is exactly why its bar is set so much higher. That is the clearest illustration of why a bare percentage is meaningless. Comet also expresses its result differently from the other three, so its percentage is not counting quite the same thing — ask your laboratory what its figure represents. The assays are not interchangeable, so the same percentage does not mean the same thing on each, and a number from one cannot be read against a threshold from another.
They are not measuring the same thing. Some detect breaks directly; others infer damage from how the DNA behaves under chemical or physical stress. When the same samples are run through different assays, the agreement between methods is poor — which in plain terms means two tests can look at the same man and disagree substantially about him.
There is no agreed universal cut-off. Each laboratory sets its own reference range. This is why you should be sceptical of any source — including a clinic's own literature — that quotes a single number as though it were a fact of nature.
It is also one of the four reasons NICE gives for recommending against the test: it is unclear which test and which threshold define “elevated”.
What the evidence shows, and what it does not
How strongly fragmentation is linked to reproductive outcome is itself contested, and I am going to be careful with the word "linked". The andrology societies read the evidence as showing the association is clearest in recurrent pregnancy loss and in natural conception, and weaker and inconsistent in IVF and ICSI — ICSI being the technique in which a single sperm is injected directly into an egg. NICE reads the same literature and states that the link between elevated fragmentation and subfertility has not been established at all. I am not going to resolve that disagreement for you in a paragraph; I am telling you it exists, because a page that hands you one of those two readings as settled fact is a page that has taken a side without telling you.
The decisive point is a different one, and it deserves to be stated bluntly: no randomised trial has ever tested whether choosing treatment on the basis of a fragmentation result improves live birth. I can find no published trial in which men were randomly allocated to fragmentation-guided treatment against ordinary treatment and followed to a birth. Sperm-selection techniques themselves have been put through randomised trials — but the men in those trials were not selected by their fragmentation result, which is the question that matters here.
The most informative evidence we do have is a secondary analysis of 953 couples across seven centres, which asked whether fragmentation identified who would do better with ICSI than with conventional IVF. It did not. That was the single most clinically attractive use of the test, and it did not hold up. But note what kind of evidence that is: the couples in it were randomised to ICSI or IVF, not according to their fragmentation result, so this is a very well-powered observation rather than proof of a negative. They also had non-severe male infertility, which means a man with a severe problem is not represented in it.
The treatments proposed for high fragmentation have fared no better. Antioxidant supplements were tested in the MOXI randomised trial, which found no improvement in live birth and none in fragmentation — though it was stopped early by its safety committee and randomised only 171 men, so it is a trial that failed to find a benefit rather than one that established there is none. The Cochrane review does report a pooled signal towards more live births, but its own authors graded that evidence very low certainty, and it disappears once the studies at high risk of bias are removed. Repairing a varicocele — enlarged veins in the scrotum, rather like varicose veins in the leg — does reduce fragmentation by about seven percentage points (95% confidence interval 5.4 to 8.4) across a meta-analysis of 29 studies in men with a varicocele that can be felt on examination. That is a consistent finding with a narrow margin of error, and I do not want to undersell it. But it compares men with themselves before and after surgery, with no untreated group alongside them — and it is a laboratory marker moving. It is not a demonstrated live-birth benefit, and the distinction matters more than it is usually allowed to.
Why NICE and the andrology societies differ
NICE's 2026 guideline (NG257) is unambiguous. It recommends: do not carry out testing for sperm DNA integrity (fragmentation), and separately, do not offer supplements, antioxidants or medical treatments to improve sperm DNA integrity. Its reasoning is stated plainly, in four parts: treating fragmentation showed no convincing benefit; the link between elevated fragmentation and subfertility has not been established; it is unclear which test and which threshold define "elevated"; and these are expensive assays that can take weeks to come back. That last reason is the one usually left out when this guidance is quoted, and for a couple watching a calendar it may be the one that matters most. That is the UK position, and any page that buries it is not being straight with you.
In practice this also means the test is not something you should expect to be offered on the NHS. If you have it, you will almost certainly be paying for it privately — which is one more reason to be clear, before you do, about what you intend to do with the answer.
Against that, the European Association of Urology strongly recommends testing after recurrent pregnancy loss. The Global Andrology Forum's 2025 clinical guidelines — 52 experts, GRADE methodology with a modified Delphi process, 24 statements of which 21 are strong — hold that testing may be considered, or should be discussed, in unexplained infertility, recurrent pregnancy loss, clinical varicocele, recurrent failure of assisted reproduction, and in men over 40 or with risk factors for oxidative stress. Their verbs are worth noticing: "may be considered" is not "should have". And "21 of 24 statements are strong" describes how firmly those 52 experts agreed with each other, not how strong the underlying evidence is — the two are easy to run together and they are not the same thing. Notably, the same group explicitly does not recommend testing in subclinical varicocele, the kind detectable only on a scan. These are not people waving the test through.
So how can both be right? Because they are answering different questions. NICE asks a national question: should a health system commission this for everyone, on outcome-grade evidence — meaning randomised proof that doing it produces more babies? On that standard, the answer today is no, and the answer is defensible. An andrologist asks a narrower question about the individual man sitting across the desk, often one who has already had three miscarriages and two failed cycles, where the alternative to imperfect information is no information at all. Neither is being foolish. They are optimising different things.
There is a precedent worth knowing, not because it settles anything but because it teaches you how to read guidance. For over a decade NICE advised against treating varicocele to improve fertility. In 2026 it changed, and now says repair should be considered for men whose varicocele is detectable on clinical examination, who have reduced semen parameters, and who are trying to conceive without assistance. Those qualifiers are the point: not everybody, and not nobody. That is not an embarrassment. It is an institution revisiting a position on the evidence, in public — which is the system working as intended. It is emphatically not a reason to assume NICE will change its mind about fragmentation. It is a reason to read "do not" as a statement about the strength of today's evidence, rather than a permanent verdict on the biology.
What this means for you
The useful question is not "is my fragmentation high?" It is: would the result change what we do next?
If the answer is no — if you would proceed with the same plan whatever the number said — then the test is not information. It is anxiety with a price attached, and quite possibly a nudge towards supplements that a randomised trial has already found do nothing.
If the answer is yes, the calculation shifts. In recurrent pregnancy loss, and in repeated failure of assisted reproduction after other explanations have been looked for, a result plausibly can alter the conversation — about lifestyle and oxidative stress, about whether a varicocele that can be felt deserves attention, about how sperm are selected in the laboratory for the next cycle, and in some cases about whether sperm are retrieved from the testis rather than the ejaculate. Two of those are operations. Varicocele repair and surgical sperm retrieval each carry their own risks, and neither is a decision to be taken on the strength of a laboratory number alone — they are worth discussing on their own merits, with someone who can examine you. Those last two are real options an andrologist may raise, and they are why "would it change anything" is a fairer question than it first sounds. That is where a serious clinician might reasonably order it, while being honest that the evidence supporting the decision that follows is weaker than anybody would like.
Where I have landed, and it has moved
I should tell you where I stand, because a page that reviews everyone else's position and hides its author's is doing half a job.
Over the past few years I have come to include sperm DNA fragmentation among the tests I consider useful — in very selected situations, and I mean that literally rather than as a hedge. I am not alone in this; a good many andrologists have moved the same way over the same period, which is neither an argument that we are right nor a reason to dismiss it.
The situation where I find it genuinely informative is this one: a man who is not conceiving, whose conventional semen analysis is good, and who has a varicocele that can be felt on examination. That combination is a puzzle. The standard test says the sperm are fine. The couple are not conceiving. There is a plausible mechanism sitting right there in the scrotum, and nothing in the routine work-up is measuring what it might be doing. Fragmentation measures something the semen analysis cannot see, and in that specific configuration a result changes what I would advise — which is the test of whether a test is worth doing.
That is also where the guidance repays careful reading.
The European Association of Urology says, strongly, that varicocele should not be treated in infertile men whose semen analysis is normal. It also says, weakly, that varicocelectomy may be considered in men with raised DNA fragmentation whose infertility is otherwise unexplained, or who have had assisted reproduction fail.
Read quickly, those look like they contradict each other. Read properly, the second is the carve-out from the first — and it has to be, because otherwise it would have no field of application at all: it would advise considering surgery in precisely the men the first sentence forbids operating on. Guidelines are not written that way. The specific qualifies the general.
And the clinical logic points the same way. Fragmentation is a test on the semen. If it comes back high, the semen assessment is not normal — what is normal is only the part the standard report happens to measure. The man I described does not have a normal semen analysis. He has a normal conventional semen analysis and an abnormal one on the measure nobody ordered.
I should be straight about the status of what I have just said: that is a reading, not a sentence you will find in the guideline. The EAU does not spell out the relationship between its two recommendations, and someone could argue the other way — that "semen analysis" there means the conventional parameters and nothing else. I think that reading makes the guideline incoherent, so I do not hold it. But you are entitled to know that the reconciliation is mine and not theirs, and that the recommendation I am relying on is a weak one.
You should also know what improves. What has been shown to fall after varicocele repair in these men is the DNA-damage marker. Not a demonstrated pregnancy, and not a demonstrated live birth.
I want to be exact about the limits of this. It is a position about whom to test, not a claim that the evidence has changed. Everything earlier on this page still stands: no trial has tested whether choosing treatment on a fragmentation result produces more babies, the assays still disagree, and NICE still advises against the test. I hold this view knowing all of that, as a judgement about an individual man in front of me rather than a recommendation for a health system — which is exactly the distinction I drew above between the two kinds of question, and I am not going to exempt myself from it.
If you have read this far and want to know what would change my mind: a properly designed trial in this population, randomising on the fragmentation result and following couples to a birth. It has not been done. Until it is, this is a considered judgement and I will keep calling it one.
Should I have this test?
“Unexplained” means the standard investigations have come back normal and no cause has been found.
Where the guidance sits. The Global Andrology Forum's 2025 clinical guidelines hold that testing may be considered here. NICE recommends against it, on the basis that the link between elevated fragmentation and subfertility is not established.
What a result would change. Honestly, often very little. If the plan would be the same whatever the number said, the test is not information — it is anxiety with a price attached, and quite possibly a nudge towards supplements that a randomised trial has already found do nothing.
Where the guidance sits. This is the situation with the strongest support for testing. The European Association of Urology strongly recommends it after recurrent pregnancy loss, and the Global Andrology Forum supports it here too. NICE's blanket “do not carry out” still stands as the national position.
What a result would change. The association between fragmentation and outcome is read by the andrology societies as clearest here and in natural conception — while NICE reads the same literature and says the link with subfertility has not been established. A result can plausibly alter the conversation — about lifestyle and oxidative stress, about whether a varicocele you can feel deserves attention, about how a fertility unit approaches the next cycle.
What it would not change. It would not tell you that acting on the number will bring a live birth. No randomised trial has ever tested whether choosing treatment on the basis of a fragmentation result improves live birth.
A varicocele is a group of enlarged veins in the scrotum, rather like varicose veins in the leg. “Clinical” means large enough to be found on examination, as opposed to “subclinical”, visible only on a scan.
Where the guidance sits. The Global Andrology Forum holds that testing may be considered in clinical varicocele — and explicitly does not recommend it in subclinical varicocele. That distinction is worth noticing: this is not a group waving the test through.
What a result would change. Repairing a varicocele does reduce fragmentation in men whose varicocele can be felt on examination, by roughly seven percentage points across a meta-analysis of 29 studies — though those studies compare men with themselves before and after surgery, with no untreated group alongside them. But that is a laboratory marker moving. It is not a demonstrated live-birth benefit, and the distinction matters more than it is usually allowed to. Repair is surgery, with its own risks, and is a decision to be taken on its own merits rather than on a laboratory number. Separately, NICE changed its position on varicocele repair in 2026: repair should now be considered for men whose varicocele is detectable on clinical examination, who have reduced semen parameters, and who are trying to conceive without assistance. All three conditions are part of the recommendation — a decision that does not depend on a fragmentation result.
Where the guidance sits. The Global Andrology Forum holds that testing may be considered after recurrent failure of assisted reproduction. NICE does not.
What a result would change. The most clinically attractive use of the test was to decide between conventional IVF and ICSI — the technique in which a single sperm is injected directly into an egg. A secondary analysis of 953 couples in a randomised trial found that fragmentation did not identify which couples would do better with ICSI. That is the use that did not hold up.
Where it may still help. After other explanations have been looked for, a result can reasonably inform how a unit approaches the next cycle — while being honest that the evidence supporting the decision that follows is weaker than anybody would like.
Oxidative stress is a chemical imbalance in which reactive molecules outpace the body's ability to neutralise them. It is influenced by smoking, obesity, infection, heat and age.
Where the guidance sits. The Global Andrology Forum holds that testing may be considered in men over 40 or with risk factors for oxidative stress. NICE recommends against the test, and separately recommends against offering supplements, antioxidants or medical treatments to improve sperm DNA integrity.
What a result would change. If it prompts a conversation about smoking, weight and heat exposure, those conversations were available without the test. If it prompts antioxidant supplements, note that these were tested in the MOXI randomised trial, which found no benefit — though it was stopped early with 171 men, so it failed to show a benefit rather than proving there is none. The pooled Cochrane signal was graded very low certainty by its own authors, and does not survive the removal of studies at high risk of bias.
Then both NICE and the andrology societies point the same way, and the answer is straightforward: there is no case for the test.
It is also worth knowing that you should not expect to be offered it on the NHS. If you have it, you will almost certainly be paying for it privately — which is one more reason to be clear, before you do, about what you intend to do with the answer.
A note on the numbers on this page
Percentage points versus per cent. When we say varicocele repair lowers fragmentation by "about seven percentage points", we mean a man whose result was 30% would typically read about 23% afterwards. That is not the same as "a 7% reduction", which would mean a fall from 30% to about 27.9%. The two are often confused, and the difference is large.
A secondary analysis is when researchers take a trial designed to answer one question and use its data to ask a second one. The people in it were never randomly allocated according to that second question, so the answer — however large the study — is a strong observation rather than proof. Better than most evidence, weaker than a trial built for the purpose.
A laboratory marker is not an outcome. A treatment can reliably move a number on a report without changing whether you take a baby home. Both things can be true at once, and when anyone tells you a number improved, the fair next question is always: and did anything else?
Where no range is given, treat the figure as approximate. Every number here comes from a study of particular men in a particular place, and would come out somewhat differently elsewhere.
What to ask
If someone offers you this test, four questions will tell you most of what you need to know. Which assay will be used, and what is the laboratory's own reference range? What decision will change if the result comes back elevated — and what changes if it comes back normal? What is the evidence that that particular decision improves the chance of a live birth, as opposed to improving the number on the report? And is anybody in this conversation paid more if I say yes?
Before you agree to it
Four questions to ask whoever offers you this test
Which assay, and what is the range?
Which of the four tests will the laboratory use, and what is that laboratory's own reference range? A number without an assay and a reference range behind it is not a result.
What changes if it is elevated?
And — the question people forget — what changes if it comes back normal? If the plan is the same either way, the test is not informing a decision.
Where is the live-birth evidence?
What is the evidence that that particular decision improves the chance of a live birth, as opposed to improving the number on the report? These are not the same claim.
Who is paid more if I say yes?
Is anybody in this conversation paid more if you agree to the test? You are entitled to ask, and to hear a straight answer.
If the answers are vague, that is itself an answer.
Declaration of interest: Mr Ollandini has no commercial interest in sperm DNA fragmentation testing and does not sell it.
This page is general information about a contested area of andrology. It is not medical advice and cannot account for your individual circumstances. Please discuss your own situation with your GP, your fertility unit, or a urologist or andrologist who can review your full history.
Frequently asked questions
No. It is not a verdict. How strongly fragmentation is linked to reproductive outcome is itself contested. The andrology societies read the association as clearest in recurrent pregnancy loss and in natural conception, and weaker and inconsistent in IVF and ICSI. NICE reads the same literature and says the link with subfertility has not been established. And no randomised trial has ever tested whether choosing treatment on the basis of a fragmentation result improves live birth. Sperm-selection techniques have themselves been put through randomised trials — but the men in them were not selected by their fragmentation result, which is the question that matters here.
It depends entirely on which test produced it, and there is no way round that. In one group of men where all four assays were compared directly, the cut-off for “elevated” came out as 18.9% by SCSA, 20.1% by TUNEL, 22.8% by SCD and 45.4% by alkaline Comet. Those are four different bars, not four readings of one man — and the assays are not interchangeable, so your 25% cannot simply be held up against all four. A Comet result of 25% and an SCSA result of 25% are not the same finding. Ask which assay was used and what reference range that laboratory applies. If nobody can tell you, the number cannot be interpreted.
I have not been able to source a reliable UK price range from a primary source, and I would rather say so than quote a figure I cannot stand behind. Ask the laboratory or clinic directly, before the sample is given, what the test costs and which assay that price buys. Cost is one of the four reasons NICE gives for advising against the test, so it is a fair thing to ask about plainly.
You should not expect to be offered it. NICE's 2026 fertility guideline (NG257) recommends: do not carry out testing for sperm DNA integrity (fragmentation). If you have it, you will almost certainly be paying for it privately.
Because NICE and an individual andrologist are answering different questions. NICE asks a national one: should a health system commission this for everyone, on randomised proof that doing it produces more babies? On that standard the answer today is no, and it is a defensible answer. An andrologist asks a narrower question about the individual man in front of them — often one who has already had several miscarriages and failed cycles, where the alternative to imperfect information is no information at all. The European Association of Urology strongly recommends testing after recurrent pregnancy loss, and the Global Andrology Forum holds that it may be considered in five defined situations. Neither position is foolish. They are optimising different things.
The evidence does not support it, and NICE recommends against offering supplements, antioxidants or medical treatments to improve sperm DNA integrity. Antioxidants were tested in the MOXI randomised trial, which found no benefit — though it was stopped early with 171 men, so it failed to show a benefit rather than proving there is none. The pooled Cochrane signal was graded very low certainty by its own authors, and does not survive the removal of studies at high risk of bias. Please discuss any supplement with your own clinician rather than starting one on the strength of a test result.
It does move the number, in men whose varicocele can be felt on examination. Across a meta-analysis of 29 studies, varicocele repair reduced fragmentation by roughly seven percentage points — in men whose varicocele can be felt on examination, and comparing men with themselves before and after surgery, with no untreated group for comparison. What has not been shown is that repairing a varicocele on the strength of a fragmentation result will bring a live birth. No randomised trial has ever tested whether choosing treatment on the basis of a fragmentation result improves live birth. NICE changed its position in 2026: repair should now be considered for men whose varicocele is detectable on clinical examination, who have reduced semen parameters, and who are trying to conceive without assistance.
The primary sources for the claims on this page.
- NICE NG257, Fertility problems: assessment and treatment, 2026 — the two “do not” recommendations on sperm DNA integrity testing and on supplements, and NICE's stated reasoning. https://www.nice.org.uk/guidance/ng257
- Ribas-Maynou J et al., Andrology 2013;1(5):715–722 — the five-assay comparison from which the four thresholds are taken. PMID 23843251.
- Global Andrology Forum, Clinical Guidelines on the Relevance of Sperm DNA Fragmentation in Reproductive Medicine, World Journal of Men's Health, 2025 — 52 experts, GRADE with modified Delphi, 24 statements. PMID 40263962.
- EAU Guidelines on Male Sexual and Reproductive Health, 2025 update on male infertility — the recommendations on testing after recurrent pregnancy loss and on varicocelectomy with raised fragmentation.
- Human Reproduction Open 2026;2026(2):hoag011 — the secondary analysis of 953 couples on whether fragmentation identifies who benefits from ICSI over conventional IVF.
- Steiner AZ et al. (MOXI trial), Fertility and Sterility 2020;113(3):552–560 — antioxidants and male infertility. PMID 32111479.
- de Ligny W et al., Cochrane Database of Systematic Reviews 2022, CD007411.pub5 — antioxidants for male subfertility. PMID 35506389.
- Cannarella R et al., World Journal of Men's Health 2024;42(2):321–337 — meta-analysis of 29 studies on fragmentation after varicocele repair.