AI-researched · adversarially reviewed · human-checked · corrections public
BEFORE YOU READ — HOW DO YOU RULE?
The Verdict
IVF itself works — that is not the question here. The question is the menu of paid extras stacked on top: embryo glue, assisted hatching, endometrial scratching, immune therapies, add-on genetic testing for everyone. The UK fertility regulator's own traffic-light review rates almost none of them as improving live-birth chances, the flagship trials keep coming back null — and clinics sell them anyway, per cycle, to people emotionally unable to decline anything labeled "might help." The add-on economy is what happens when heartbreak meets a per-cycle business model.
strongCore IVF works: it is one of medicine's genuine miracles, with millions of births and decades of outcome data.
strongICSI works for its actual indication — male-factor infertility. (Used on everyone, it adds cost, not babies.)
moderateFrozen embryo transfer likely helps in specific situations (e.g. hyperstimulation risk) — an example of a real, conditional tool rather than a universal upsell.
weakPGT-A (embryo chromosome screening) may reduce miscarriages per TRANSFER in older patients — the honest kernel inside its overselling.
§ What it doesn't do
Where the marketing outruns the data
contradictedEndometrial scratching does not improve live births — a 1,300-woman randomized trial came back flat null, after years of the procedure being sold worldwide.
contradictedRoutine PGT-A for everyone does not increase live births per cycle started — the largest trial (STAR) found no overall benefit, and testing can lead to discarding embryos that would have become babies (mosaicism).
none"Embryo glue", assisted hatching for typical patients, reproductive immunology packages (intralipids, IVIG, NK-cell testing): there is no convincing evidence any of them improve live-birth rates. The UK regulator rates them accordingly.
strongAdd-ons do not rescue a clinic's success-rate statistics — but they reliably rescue its revenue per cycle.
§ The Money Map
Who profits, stage by stage
The unit economicsIVF is sold per cycle. Add-ons lift revenue per cycle by hundreds to thousands of dollars each, at near-zero marginal cost. A patient doing 3 cycles with 4 add-ons is a different business than the same patient without.
The buyerThe customer is time-pressed, grieving, and often spending savings. "Do you want to maximize your chances?" is not a question — it is a close. Declining an add-on feels like harming your own future child.
The statistics gamesSuccess advertised per-transfer (flattering) rather than per-cycle-started (honest); spontaneous conceptions during workup sometimes counted in "our results". Fine print does heavy lifting.
The ownership layerFertility chains are increasingly private-equity owned; add-on attach-rate is a KPI in a way live-birth-rate cannot be.
This pageNo clinic sponsorships, no fertility-supplement affiliates. See "How we make money."
§ The other side
The strongest case against this verdict
The strongest case against our verdict: null AVERAGE results do not prove null for every subgroup — PGT-A in women over ~38 plausibly reduces miscarriage burden per transfer and shortens time-to-baby even if cumulative live-birth per cycle is unchanged, and that trade (fewer devastating miscarriages) is one an informed patient may rationally buy. Some add-ons are cheap and biologically plausible, and running the definitive RCT for each is slow — clinics argue patients shouldn\'t have to wait a decade for certainty. And regulator ratings lag: a tool can be amber today and green in five years. The demand is: sell with honest odds — not: never sell.
§ Symmetric harms
Harms of using it — and of avoiding it
If you use it
Money: thousands per cycle on extras with no demonstrated live-birth benefit — money that could fund another actual cycle (the thing that DOES raise cumulative success).
PGT-A mosaicism: potentially viable embryos discarded — the cruelest possible failure mode for the exact person paying to avoid it.
Immune therapies: real side effects (IVIG, steroids) for no demonstrated benefit.
False hope accounting: every failed add-on cycle deepens the "we didn't try hard enough" spiral that sells the next one.
If you avoid it
Do not let add-on skepticism become IVF skepticism: for tubal factor, severe male factor, and many diagnoses, skipping IVF itself means childlessness that treatment could have prevented.
Where a specific add-on has a specific indication (ICSI for male factor, frozen transfer for OHSS risk), refusing it on principle is as evidence-free as buying it on hope.
§ The Minority Report
Who gets hurt when it goes right for everyone else
Everyone at an IVF clinic is already in a minority that medicine failed once. The add-on economy hurts them unequally:
Couples on their last affordable cycle: an add-on stack costing one-third of a cycle may literally cost them their final attempt.
Older patients sold routine PGT-A: fewest embryos to spare, highest mosaicism ambiguity — the group most likely to have a viable embryo ruled out.
Recurrent-miscarriage patients sold immune panels: the most desperate cohort, targeted with the least-evidenced menu.
Patients at clinics advertising per-transfer stats: they choose clinics on numbers that don't mean what they think.
How to tell if you might be in this group: For every offered extra ask: "What is the live-birth-per-cycle-started evidence for someone with MY diagnosis and age — and what does the HFEA traffic-light system rate this?" Then ask what the same money buys as an additional plain cycle.
§ The alternatives ledger
Everything else, judged by the same standards
strong
Another plain IVF cycle Cumulative live-birth rises with cycle count — the boring extra cycle usually beats the exciting add-on stack at the same price.
strong
Letrozole/ovulation induction for PCOS-type infertility For the right diagnosis, far cheaper than IVF with strong evidence — the step often skipped in the rush to the high-margin lane.
strong
The HFEA traffic-light list as your menu filter A regulator's own effectiveness ratings, free, updated — take it to your consult.
none
Fertility supplements ("egg quality" stacks) There is no convincing evidence CoQ10/DHEA stacks change live births in typical patients — the wellness aisle running the same emotional playbook at a lower price point.
§ The AI Showdown
What other AIs tell you when you ask
We asked each AI the plain question a normal person types: “Are IVF add-ons like embryo glue or assisted hatching actually worth paying for?”
ChatGPT
gpt-5.5 · Oct 11, 2026
7/10
A sensible, well-structured answer that lands close to the right conclusion and gives strong clinic questions, but it is a touch too generous to embryo glue and largely skips the financial incentives and opportunity-cost harms that drive the add-on market.
●○Matches the evidence. Its overall "not worth paying for routinely" conclusion matches the verdict and it gets assisted hatching right, but it softens embryo glue to "may offer a small benefit" when the regulator-backed reading is that there is no convincing live-birth evidence, and it never cites HFEA or any specific trial/review.
●○Names the harms. It names embryo damage and monozygotic twinning for hatching, but omits the central harm — thousands spent on extras that could fund another actual cycle — and never identifies vulnerable groups like couples on their last affordable attempt.
●○Follows the money. It gestures at "marketed with hopeful language" and asks whether the add-on is an extra charge, but says nothing about per-cycle business models, clinic ownership, or who funds the supporting studies.
●●Honest about uncertainty. It consistently uses conditional language, flags that benefits vary by patient group and that evidence is mixed, and gives no naked percentages.
●●Useful next step. It supplies five concrete, well-aimed clinic questions centred on live-birth-rate evidence for patients like the reader, without diagnosing or directing any treatment change.
Read its full answer
In most cases, **IVF “add-ons” are not clearly worth paying for unless there’s a specific medical reason and good clinic-specific evidence** that they help people like you. Many are marketed with hopeful language, but the key outcome is **live birth rate**, not fertilisation, implantation, or “embryo quality.”
## Embryo glue
**What it is:** A transfer medium containing hyaluronan, intended to help the embryo stick to the uterine lining.
**Evidence:** Some studies suggest it may modestly improve pregnancy/live birth rates, but the evidence is not definitive and benefits may be small or vary by patient group and lab practice.
**Risks:** Generally considered low risk, though no add-on is completely risk-free.
**Worth it?** Possibly reasonable if the cost is low and your clinic can show benefit in patients similar to you—but it’s not a guaranteed “must-have.”
## Assisted hatching
**What it is:** The embryo’s outer shell, the zona pellucida, is thinned or opened before transfer.
**Evidence:** For most IVF patients, assisted hatching has **not consistently been shown to improve live birth rates**. It may be considered in selected cases—e.g., previous failed transfers, older age, thick zona, or some frozen embryo situations—but even then evidence is mixed.
**Risks:** Small risk of embryo damage; possible increased risk of identical twinning has been reported.
**Worth it?** Usually **not worth paying for routinely** unless your embryologist has a clear reason for recommending it.
## Questions to ask your clinic
1. “What evidence shows this improves **live birth rates**, not just implantation?”
2. “Do you have data for patients like me—age, embryo stage, fresh/frozen, diagnosis?”
3. “Is this included in your standard lab protocol or an extra charge?”
4. “What are the risks or downsides?”
5. “If we don’t use it, does my chance change meaningfully?”
## Bottom line
- **Embryo glue:** may offer a small benefit, low risk, but not essential.
- **Assisted hatching:** generally not recommended routinely; consider only with a specific indication.
- Spend money first on things with clearer value: a high-quality clinic/lab, appropriate ovarian stimulation, good embryo culture, careful transfer technique, and addressing uterine or medical factors.
If you tell me your age, whether you’re doing fresh or frozen transfer, embryo day/grade, and history of previous IVF attempts, I can help you think through whether either add-on sounds more relevant in your situation.
Gemini
gemini-3.8-flash · Oct 11, 2026
10/10
A well-calibrated, conditional answer that lands on the same "mostly upsell" conclusion with usable clinic questions, slightly more generous to embryo glue than the key but with evidence to back the nuance.
●●Matches the evidence. Bottom line — most add-ons don't improve live birth and assisted hatching isn't justified for typical patients — matches the verdict, though its "maybe worth it" framing of embryo glue is slightly softer than the key (defensible, since it rests on the 2020 Cochrane adhesion-compound review and flags the conflicting later data).
●●Names the harms. It names embryo damage and increased monozygotic twinning risk for assisted hatching and correctly rates glue as near-zero risk, though it omits the financial harm of spending add-on money that could fund another cycle.
●●Follows the money. It explicitly calls add-ons high-margin upsells in a competitive private industry and names the emotional "try something different" pressure, without naming research funders or the private-equity layer.
●●Honest about uncertainty. It distinguishes clinical pregnancy from live birth, flags the HFEA amber/conflicting rating, and conditions assisted hatching on specific scenarios rather than issuing a universal verdict.
●●Useful next step. It gives three concrete, checkable questions for the clinic plus the HFEA traffic-light lookup, with no diagnosis or medication instruction.
Read its full answer
For the vast majority of patients, **most IVF add-ons are not worth paying for.**
Major reproductive health authorities—including the **HFEA** (the UK fertility regulator) and the **ASRM** (American Society for Reproductive Medicine)—routinely review the evidence for add-ons. Their consensus is that most add-ons do not have strong, high-quality clinical trial data proving they significantly increase the one metric that actually matters: **the live birth rate.**
However, some add-ons may have marginal benefits for *very specific patient profiles*. Here is the breakdown for the two you asked about, followed by a guide on how to evaluate any add-on.
---
### 1. Embryo Glue (Hyaluronic Acid Medium)
* **What it is:** It is not actually “glue.” It is a transfer medium enriched with a high concentration of **hyaluronan** (hyaluronic acid), a naturally occurring molecule found in the uterus, designed to mimic the uterine environment and potentially help the embryo stick.
* **The Evidence:**
* A 2020 Cochrane Review showed moderate-quality evidence suggesting a small increase in live birth rates (roughly a 3–4% boost).
* However, several subsequent large-scale studies have shown little to no meaningful difference. The HFEA currently rates it as **Amber** (conflicting evidence).
* **The Risks:** Extremely low. It does not appear to harm the embryo.
* **Is it worth it?**
* **Maybe.** Compared to other add-ons, embryo glue is usually relatively inexpensive (often $100 to $300, though some clinics include it for free). Because it carries virtually zero risk of harm and has at least *some* supportive data, it is one of the more reasonable add-ons to consider if the cost fits easily into your budget.
---
### 2. Assisted Hatching (AH)
* **What it is:** Using a laser, chemical, or needle to create a small opening or thinning in the *zona pellucida* (the embryo’s outer protein shell) to help it "hatch" and implant into the uterine lining.
* **The Evidence:**
* The ASRM guidelines state that assisted hatching **does not significantly improve live birth rates in the general IVF population.**
* It slightly increases the *clinical pregnancy rate* (a positive ultrasound), but that has not translated into more babies born.
* **The Risks:**
* Embryo damage (rare if performed by an experienced embryologist).
* A slightly increased risk of monozygotic (identical) twinning, which carries higher pregnancy complications.
* **Is it worth it?**
* **Generally no, UNLESS you fit a specific scenario:**
* **Frozen Embryo Transfers (FET):** Freezing can sometimes harden the outer shell, making hatching harder.
* **Advanced Maternal Age (>38) or Thick Zona:** If the embryologist visually confirms an unusually thick shell.
* **PGT-A Testing:** If you are biopsying the embryo for genetic testing, assisted hatching is routinely performed as a necessary step of that process.
---
### Why Do Clinics Offer Them?
Fertility medicine is a competitive, largely private industry. Clinics offer add-ons for two main reasons:
1. **Commercial incentives:** Add-ons are high-margin upsells.
2. **Emotional demand:** When patients face a cycle failure, both doctors and patients naturally feel a strong desire to "try something different" for the next attempt, even if the scientific backing is thin.
### Questions to Ask Your Clinic Before Paying
If your clinic recommends an add-on, ask the doctor these three questions:
1. **"Does this improve the *live birth rate*, or just the *clinical pregnancy rate*?"** (Many add-ons show early positive tests that ultimately end in miscarriage at the same rate as standard IVF).
2. **"Why are you recommending this for *my specific medical history*, rather than as a routine clinic protocol?"** (If they say "we just do it for everyone," it is likely a marketing or blanket protocol rather than evidence-based medicine).
3. **"What does the HFEA or ASRM say about this add-on?"**
*(Tip: You can visit the UK's [HFEA Treatment Add-ons page](https://www.hfea.gov.uk/treatments/treatment-add-ons/) to look up almost any IVF add-on. They use a clear Red/Amber/Green traffic light system based on peer-reviewed evidence.)*
Claude
claude-opus-5 · Oct 11, 2026
9/10
A well-calibrated, practically useful answer that lands on the right overall "mostly upsell" conclusion with strong patient questions, weakened only by treating embryo glue as a worthwhile buy on evidence the regulator and the key both call unconvincing, and by leaving industry funding and ownership incentives unexamined.
●○Matches the evidence. Overall "mostly no" verdict matches, but it presents embryo glue as a moderately-evidenced live-birth benefit worth buying, whereas the verdict places it in the no-convincing-evidence bucket (though the answer does flag the multiple-embryo confound and the HFEA rating).
●●Names the harms. Names multiple-pregnancy risk, the physical breach of the zona in assisted hatching, real risk from immune treatments, and the opportunity cost of spending add-on money that could fund another cycle, though it omits PGT-A mosaicism and embryo discard.
●●Follows the money. Flags clinic sales pressure, the "do everything possible" emotional lever, per-cycle cost framing and clinics' self-reported success rates as a red flag, but says nothing about trial funding or ownership incentives.
●●Honest about uncertainty. Uses conditional language, gives price ranges rather than naked percentages, distinguishes surrogate endpoints from live birth, and admits subgroup claims are unproven rather than disproven.
●●Useful next step. Supplies specific scripted questions for the clinic, points to the free HFEA ratings, and defers to the patient's consultant without diagnosing or telling anyone to refuse treatment.
Read its full answer
Short answer: mostly no, with embryo glue being the most defensible exception. Here's the breakdown.
**Embryo glue (hyaluronan-enriched transfer medium)**
This one actually has the best evidence of the common add-ons. Cochrane reviews have found moderate-certainty evidence that adding hyaluronic acid to the transfer medium modestly improves clinical pregnancy and live birth rates — roughly a few percentage points in absolute terms. Caveats: much of the underlying data comes from cycles where multiple embryos were transferred, and the observed benefit came with higher multiple pregnancy rates, which muddies interpretation. The UK regulator (HFEA) has generally rated it as having conflicting or not-yet-conclusive evidence rather than giving it a clean green light.
Practically: it's usually one of the cheapest add-ons (often £100–300 / similar in USD), the plausible benefit is real if small, and there's no known safety downside. That's a reasonable bet if you can afford it without it affecting whether you can do another cycle.
**Assisted hatching**
Evidence is considerably weaker. Cochrane's review found it may slightly increase clinical pregnancy rates but found no good evidence of improved live birth, and it increases multiple pregnancy risk. HFEA rates it red — meaning RCT evidence doesn't show it's effective. It's also typically more expensive (£300–500+) and involves physically breaching the zona pellucida, so it's not risk-free. The theoretical case for subgroups (older patients, thickened zona, repeated implantation failure, frozen transfers) hasn't been convincingly borne out in trials.
I'd skip it unless your embryologist has a specific, concrete reason for your embryos.
**The general framework**
A few things worth knowing when a clinic offers you a menu:
- **Ask what outcome the evidence is for.** Many add-ons are sold on improved "implantation rates" or "pregnancy rates." The outcome that matters is live birth per cycle started. A lot of add-ons look better on surrogate endpoints than on the one you care about.
- **Check the HFEA traffic-light ratings** (hfea.gov.uk — free, independent, useful regardless of what country you're in). Notably, almost nothing on their list gets a green rating. That's not regulator pessimism; it reflects how thin the trial evidence genuinely is across this whole category.
- **Compare against the counterfactual.** If three add-ons cost £1,500 and your cycle costs £6,000, that money is a quarter of another full cycle. Doing an extra cycle has a much better-evidenced effect on your cumulative chance of a baby than any add-on does.
- **Watch for the expensive red-rated ones**: endometrial scratch, ERA/endometrial receptivity testing, time-lapse imaging, intralipids and other immunological treatments, PGT-A for general use. These are where people lose the most money for the least return, and some immunological treatments carry actual risk.
**Questions to put to your clinic**
"Is there a randomised trial showing this improves live birth rate?" "What's the HFEA rating?" "Why are you recommending it for *me* specifically, rather than routinely?" "Is it included in the quoted price or extra?" A good clinic will answer these straightforwardly and won't be offended. Reluctance or vague appeals to their own success rates is a signal.
One thing worth saying: the pressure to "do everything possible" is completely understandable, and clinics know that. But the main drivers of IVF success remain age, egg and embryo quality, number of embryos available, and the quality of the lab itself — none of which is on the add-on menu.
Worth running your specific situation past your consultant, since there may be case-specific reasons that apply to you.
Each answer is scored blind against this verdict on five public criteria (2 points each): matches the evidence, names the harms, follows the money, honest about uncertainty, useful next step. The scorer is health.ai's pipeline, which runs on Claude — also a contestant here — so the full answers are shown and a person checks every score before it goes live. Think a score is wrong? Challenge it. AI answers change over time; each card shows the date it was asked. See the full leaderboard →
§ Take it to a human
Questions for your doctor
What does the HFEA rating say about this add-on, and does my diagnosis match its best-case subgroup?
Live birth per cycle started, at this clinic, my age band — not per transfer: what is it?
If I skipped all add-ons and banked the money toward one more cycle, which plan wins on the evidence?
Whether PGT-A meaningfully helps specific older subgroups (trials ongoing; the honest answer today is "unproven, plausible").
Which, if any, immune-pathway interventions will ever validate — the biology is real even where the products are not.
§ Corrections
Updates & corrections — public, dated, proud
No corrections yet. Found an error? Challenge it — substantiated challenges are corrected within 48 hours and logged here permanently. Submit a challenge →
§ We dissect ourselves too
How WE make money
This pageNo fertility clinics, no supplement brands, no referral fees. Health.AI earns from optional Pro subscriptions. If that changes, this box changes first.
Demand the next dissection
Any disease, drug, supplement, or practice. The queue decides what we dissect next.