ICSI is used well beyond male infertility. A review finds no live-birth benefit
Injecting one sperm into each egg was meant for severe male-factor cases, yet clinics now use it widely. Pooling six randomised trials found no live-birth advantage over standard IVF when sperm is not the problem.
| Group | Value (%) |
|---|---|
| ICSI | 32.8 |
| Conventional IVF | 34.5 |
Intracytoplasmic sperm injection — ICSI, in which an embryologist injects a single sperm directly into each egg — was developed for couples whose infertility is driven by poor sperm. In many clinics it is now the default for almost everyone, used for unexplained infertility, mild sperm problems, or no male factor at all. A 2026 systematic review and meta-analysis of randomised trials asked whether that expansion is justified, and found it is not: among couples without severe male-factor infertility, ICSI did not improve the chance of a live birth over conventional IVF [s1].
What the review did
The authors searched ten databases from inception to 31 May 2025, without language restrictions, for randomised controlled trials comparing ICSI with conventional IVF (cIVF) per couple [s1]. They excluded non-randomised designs, studies comparing effects per egg rather than per couple, and trials that did not meet predefined trustworthiness criteria, assessing the remainder with the Cochrane Risk of Bias 2 tool and the TRACT integrity checklist, and grading the overall certainty of evidence with GRADE [s1]. Six randomised trials in couples without severe male-factor infertility met the bar [s1].
What it found
For the outcome that matters most to patients — a live birth — there was no difference. Pooling four trials with 1,438 couples, the live birth rate was 32.8% with ICSI against 34.5% with conventional IVF, a pooled risk ratio of 0.96 (95% CI, 0.85 to 1.09; I² = 37%), rated high-quality evidence [s1]. The cumulative live birth rate, which counts births from frozen as well as fresh transfers, told the same story: 43.2% versus 47.4% across three trials and 1,911 couples, a risk ratio of 0.92 (95% CI, 0.84 to 1.01; I² = 41%), again high-quality evidence [s1].
If anything, the signal tilted slightly against ICSI. The preterm birth rate was lower with conventional IVF's comparator — 4.6% with ICSI versus 6.0% without, a risk ratio of 0.77 (95% CI, 0.59 to 1.00; P = 0.0447) across three trials and 222 births — though this rests on small numbers [s1]. No significant differences emerged for the other fertility and pregnancy outcomes measured [s1].
Why clinics do it anyway
ICSI spread because it feels like more control. It bypasses the step where sperm must bind and penetrate the egg on their own, which can reassure a clinic worried about total fertilisation failure. But the trial evidence says that for couples whose sperm is adequate, that extra manipulation does not translate into more babies — it adds cost, laboratory time, and a procedure performed on every mature egg retrieved. The review's framing is pointed: ICSI is effective for severe male-factor infertility, but its use for other reasons has run ahead of the evidence [s1].
This is the kind of result that is easy to lose in a clinic consultation, where add-ons and intensive techniques can be presented as doing more, and therefore as better. Health Newspapers has written before about how IVF add-ons often fail to deliver when tested in randomised trials; the choice between ICSI and standard insemination belongs in that same conversation.
How much to trust it
The strengths here are the randomised design of the included studies and the high GRADE rating for the live-birth outcomes, which means the authors judged further research unlikely to overturn the finding [s1]. The limits are familiar for any meta-analysis: only six trials qualified, some outcomes rested on a few hundred births, and the trials varied in setting and patient mix, reflected in the moderate statistical heterogeneity (I² around 37–41%) for the main outcomes [s1].
A funding note belongs here too. The work was supported by Chinese government research programmes — including the National Key Research and Development Program and the National Natural Science Foundation — and by Peking University Third Hospital, an academic fertility centre rather than a manufacturer of laboratory equipment [s1]. That is an independent, non-commercial funding base, which matters when the question is whether a widely sold procedure earns its place.
What it means for a reader
If you are facing IVF and have no significant sperm problem, it is reasonable to ask your clinic why ICSI is being recommended rather than conventional insemination, and what evidence supports it for your situation. The honest answer, on current data, is that it is unlikely to raise your chance of taking home a baby [s1]. This article is informational and is not medical advice.
What to watch
Whether professional bodies and clinics narrow their use of ICSI to the indications the evidence supports, and whether registry data start to show the overuse receding. For now, a high-quality synthesis has put a number on a long-standing suspicion: for couples without a sperm problem, the fancier technique is not the better one [s1].
Sources
Sources
More on
Testosterone gel before IVF did not lift pregnancy rates in women with low ovarian reserve
A European trial stopped early for futility: clinical pregnancy occurred in 15.7% of women given transdermal testosterone before IVF and 14.9% given placebo. The trial was funded by Ferring.
How soon can you get pregnant after stopping birth control?
Two systematic reviews point the same way: for the pill, IUDs, implants and most methods, about 8 in 10 women conceive within a year of stopping — similar to using no method at all.
What egg freezing actually delivers: age at freezing and egg count decide the odds
In two large clinics, the live birth rate per woman who returned to use her eggs was around 39-50% under 35 or 38, but far lower for older women — and most women who froze never came back to use them.
For frozen embryo transfer, a natural cycle matched hormones, with fewer complications
In a 4,376-woman trial across 24 Chinese centres, natural-cycle embryo transfer gave the same healthy-live-birth rate as a programmed hormone regimen, with less pre-eclampsia and postpartum haemorrhage.