ICSI vs IVF: which do you actually need

Being told you might need ICSI can leave you trying to work out whether this means something is seriously wrong, or whether you are being steered toward a procedure you may not actually need. It is reasonable to want a clear explanation of why ICSI is being suggested, and whether it is likely to help in your specific situation.

IVF and ICSI are different ways of doing the fertilisation step

In conventional IVF, eggs and sperm are placed together and fertilisation happens without direct sperm injection.[1] In ICSI, a single sperm is injected into an egg.[2]

Both IVF and ICSI are used after egg collection, but they are not interchangeable by default.[3][1] The evidence in this dossier supports using ICSI when the aim is to address specific sperm-related fertilisation problems, rather than using it routinely for everyone.[4][5]

When ICSI is supported by the evidence

Male-factor infertility

HFEA says a doctor may recommend ICSI for male-factor infertility.[4] That fits with the broader evidence base here, which points to ICSI as a way to address sperm-related fertilisation problems rather than to improve outcomes universally.[5][6]

The ESHRE evidence summary for unexplained infertility also highlights that the men in one IVF-versus-ICSI trial had sperm that were normal by WHO 2010 criteria, defined as “total sperm count ≥39 × 10⁶ sperm, progressive motility ≥32%.”[7] In practice, that matters because if semen parameters are normal, the case for ICSI needs a more specific explanation than simply saying it gives “better chances.”[3][7]

Surgically retrieved sperm and azoospermia

A 2025 systematic review protocol states that “Intracytoplasmic sperm injection (ICSI) with surgically retrieved sperm from the testis or epididymis enables biological fatherhood in obstructive (OA) and non-obstructive azoospermia (NOA) patients.”[6] This is one of the clearest evidence-based scenarios in the dossier where ICSI has a defined role.[6]

Some previous fertilisation problems

The dossier supports considering ICSI when there has been a sperm-related fertilisation problem in the past, but it does not support assuming ICSI will solve every case of failed fertilisation.[4][5] HFEA specifically says that if previous trouble fertilising was due to “poor quality or immature eggs,” then “ICSI is unlikely to help.”[4]

When routine ICSI is not clearly helpful

Unexplained infertility

ESHRE gives a strong recommendation that “ICSI is not recommended over conventional IVF in couples with unexplained infertility.”[3] In the same guideline, a subgroup analysis of a large randomised trial in unexplained infertility found “no significant difference in live birth rate between groups (35.5% (65/183) vs. 36.7% (73/199), RR 1.03 (95% CI 0.79-1.35)).”[1]

Another study cited by ESHRE found no difference in fertilisation rate between IVF and ICSI in unexplained infertility: “14/30 (46.7%) vs. 15/30 (50%).”[1] ESHRE also cites an RCT subgroup of 100 couples with unexplained infertility showing no difference in pregnancy rates: “32% vs. 38%, RR 0.83, 95% CI 0.48-1.45.”[1]

Normal sperm count and motility

One of the key trial populations in the ESHRE evidence summary involved couples in whom the male partner had sperm parameters that were normal by WHO 2010 criteria: “total sperm count ≥39 × 10⁶ sperm, progressive motility ≥32%.”[7] That does not mean ICSI is never used when semen parameters are normal, but it does mean the guideline evidence against routine ICSI in unexplained infertility is directly relevant to couples without a documented male-factor problem.[3][7]

Poor-quality or immature eggs

HFEA says that if previous fertilisation problems happened because of poor-quality or immature eggs, ICSI is unlikely to help.[4] That is important because ICSI is not a fix for every reason fertilisation may fail.[4][5]

What ICSI does not mean

It does not mean “better for everyone”

The evidence here does not support ICSI as a universally better version of IVF.[3][1] In unexplained infertility, ESHRE explicitly recommends against using ICSI over conventional IVF, and the live birth results quoted above were similar rather than better with ICSI.[3][1]

It does not guarantee fertilisation

The dossier does not support any claim that ICSI guarantees fertilisation.[5] A centre-specific UCSF source reports an ICSI fertilisation rate of “80 to 85 percent,” which is not 100%, and it is a single-centre figure rather than a universal benchmark.[8]

It does not remove every sperm-related difficulty

A 2025 retrospective cohort study found that “Some severe male infertility factors, i.e. severe oligoasthenozoospermia (OAT-S) and non-obstructive azoospermia (NOA), may be negatively associated with fertilization, embryo development, and cumulative live birth rates, but not with neonatal outcomes.”[5] That means ICSI may still be used in severe male-factor infertility, but severe male-factor infertility can still affect outcomes.[5]

Trade-offs and downsides to ask about

ICSI involves injecting individual mature eggs, so not every egg collected will necessarily be suitable for the procedure.[9] The dossier also supports that ICSI is a technically demanding lab procedure, which is one reason clinics should be able to explain why they are recommending it in your case rather than as routine practice.[10]

There is also a procedural risk to the eggs themselves. A critical review returned by the research tool states: “Studies have reported anywhere from a 5%–19% mechanical damage rate to oocytes during the ICSI procedure.”[9]

For long-term safety questions, the evidence in this dossier is incomplete rather than definitive. In azoospermia cases using surgically retrieved sperm, a 2025 protocol notes that comparative studies remain “fragmented, particularly for neonatal and long-term offspring outcomes.”[6] The male-factor cohort study above found no negative association with neonatal outcomes, but it did find that some severe male infertility factors were negatively associated with fertilization, embryo development, and cumulative live birth rates.[5]

Questions to ask your clinic before agreeing to ICSI

• Ask what specific problem ICSI is meant to solve in your case, because the evidence here supports ICSI for defined indications such as male-factor infertility, surgically retrieved sperm, or sperm-related fertilisation problems, not as a universal upgrade.[4][6][5]

• Ask to review the semen analysis, including whether the sperm count and progressive motility are actually abnormal, because ESHRE cites WHO 2010 normal criteria as “total sperm count ≥39 × 10⁶ sperm, progressive motility ≥32%.”[7]

• If you have unexplained infertility, ask why ICSI is being recommended despite ESHRE’s strong recommendation that “ICSI is not recommended over conventional IVF in couples with unexplained infertility.”[3]

• Ask whether any past fertilisation problem seemed sperm-related or egg-related, because HFEA says ICSI is unlikely to help if poor-quality or immature eggs were the reason fertilisation was difficult.[4]

• Ask whether the clinic expects ICSI to improve fertilisation only, or whether they believe it is likely to improve live birth in your diagnosis, because in unexplained infertility the guideline-cited trial found no significant live birth difference: “35.5% (65/183) vs. 36.7% (73/199), RR 1.03 (95% CI 0.79-1.35).”[1]

What this doesn't tell you

This evidence does not give a one-size-fits-all answer for every clinic or diagnosis.[1][3] The dossier does not provide exact HFEA treatment-by-treatment success-rate figures because the analytics tool was unavailable.[11] It also does not provide direct guideline text here for routine ICSI use in low egg numbers, PGT cycles, or donor eggs, so those situations should not be assumed either way from this article.[11]

Some of the numbers here come from unexplained infertility studies, so they may not apply to severe male-factor infertility, surgically retrieved sperm, or other specific clinical scenarios.[1][6][5] This is general information, not medical advice, and decisions about whether you should have IVF or ICSI need to be made with your fertility clinician using your semen results, egg factors, treatment history, and the lab’s experience.[3][4][7]

If you want a more personalised, evidence-based read on whether ICSI makes sense for your diagnosis, PATH can help you compare the reason you were given with the evidence and the kinds of clinics that offer ICSI for cases like yours.

Sources

1. ESHRE Management of Unexplained Infertility (2023) — Further information (p. 85): “rate between the IVF and ICSI groups (14/30 (46.7%) vs. 15/30 (50%)) (Foong et al., 2006).” “A subgroup analysis of a large RCT included 382 couples with UI, randomly assigned to IVF (n=183) and ICSI (n=199), found no significant difference in live birth rate between groups (35.5% (65/183) vs. 36.7% (73/199), RR 1.03 (95% CI 0.79-1.35)) (Dang et al., 2021).” “A subgroup analysis of an RCT included 100 couples with UI. There was no difference in pregnancy rates between IVF and ICSI (32% vs. 38%, RR 0.83, 95% CI 0.48-1.45) (Bhattacharya et al., 2001).”
2. WHO Infertility Guideline (2025) — Page 223: reference to Palermo et al., “Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet. 1992;340(8810):17-18.”
3. ESHRE Management of Unexplained Infertility (2023) — List of all recommendations (p. 10): “ICSI is not recommended over conventional IVF in couples with unexplained infertility.” Strong recommendation.
4. HFEA, “Intracytoplasmic sperm injection (ICSI)”: “Your doctor may recommend ICSI if:” and “If you’ve had treatment in the past and poor quality or immature eggs meant the sperm and the egg had trouble fertilising, then ICSI is unlikely to help.” Source returned by tool: https://www.hfea.gov.uk/treatments/explore-all-treatments/intracytoplasmic-sperm-injection-icsi
5. Jiang Liu, Zhou Jiayin, Huang Haoming et al. “Do different male infertility factors impact embryological, cumulative pregnancy and neonatal outcomes in IVF/ICSI cycles? A retrospective cohort study.” Human Reproduction Open (2025): “Some severe male infertility factors, i.e. severe oligoasthenozoospermia (OAT-S) and non-obstructive azoospermia (NOA), may be negatively associated with fertilization, embryo development, and cumulative live birth rates, but not with neonatal outcomes.”
6. Tsiartas Panagiotis, Montejo Rocio, Iliadis Stavros I et al. “ICSI with surgically retrieved sperm in azoospermia: protocol for a systematic review and meta-analysis of reproductive, perinatal, long-term, and paternal outcomes.” Systematic Reviews (2025): “Intracytoplasmic sperm injection (ICSI) with surgically retrieved sperm from the testis or epididymis enables biological fatherhood in obstructive (OA) and non-obstructive azoospermia (NOA) patients.” “comparative studies versus ejaculated sperm remain fragmented, particularly for neonatal and long-term offspring outcomes.”
7. ESHRE Management of Unexplained Infertility (2023) — IVF (p. 116): male partner’s sperm “normal based on WHO 2010 criteria (total sperm count ≥39 × 10⁶ sperm, progressive motility ≥32%).”
8. UCSF Health, “FAQ: Intracytoplasmic Sperm Injection”: “The fertilization rate in the UCSF IVF laboratory is exceptional – currently 80 to 85 percent.” Source returned by tool: https://www.ucsfhealth.org/health-articles/faq-intracytoplasmic-sperm-injection
9. “The Impact of Intracytoplasmic Sperm Injection in Non-Male Factor Infertility—A Critical Review” (PMC) returned by fertility web tool: “Studies have reported anywhere from a 5%–19% mechanical damage rate to oocytes during the ICSI procedure.” Source returned by tool: https://pmc.ncbi.nlm.nih.gov/articles/PMC8231975
10. Grounded answer from the dossier, Question 4: “ICSI is more technically dependent on laboratory skill because it requires micromanipulation of each egg.” “Not every egg is suitable for injection; ICSI is performed on mature eggs.”
11. HFEA success-rate tool output: “HFEA analytics unavailable (check SUPABASE_DB_URL and that table hfea_17_18 exists).”