Time-Lapse Imaging Doesn't Improve IVF Live Birth Rates: The TILT Trial
Last updated: February 2026
Journal Club #5. We read the latest fertility research so you don’t have to.
The Paper
“Clinical effectiveness and safety of time-lapse imaging systems for embryo incubation and selection in in-vitro fertilisation treatment (TILT): a multicentre, three-parallel-group, double-blind, randomised controlled trial”
Bhide P, Chan DYL, Lanz D, et al. Lancet, 2024;404(10449):256–265.
Read the full paper on PubMed →
DOI: 10.1016/S0140-6736(24)00816-X | Trial registration: ISRCTN17792989
Why This Matters
Your clinic may offer time-lapse incubation, often branded as “EmbryoScope” or “24/7 embryo monitoring.” The pitch: a camera inside the incubator photographs your embryos every few minutes, the embryo stays undisturbed (no opening the incubator to check on it), and an algorithm analyzes the footage to pick the best one. It sounds like better science. Clinics charge $500 to $1,000 per cycle for it.
This trial tested both parts of that pitch separately. Does undisturbed culture help? Does the selection algorithm help? 1,575 patients. Seven centres across the UK and Hong Kong. The answer to both questions: no.
The Three-Arm Design
Most trials compare A to B. This one compared A to B to C, and that’s what makes it valuable.
Arm 1: Time-lapse everything. Embryos cultured in a time-lapse incubator (undisturbed) and selected using the time-lapse algorithm. This is the full add-on package clinics sell.
Arm 2: Undisturbed culture only. Embryos cultured in the same time-lapse incubator but selected by standard morphology (an embryologist looking at them, not an algorithm). This isolates the “undisturbed culture” claim.
Arm 3: Standard care. Conventional incubator. No time-lapse. No algorithm. The way IVF has been done for decades.
525 patients per group. Double-blind (patients and trial staff). Embryologists were unblinded by necessity: you can’t hide which incubator you’re using from the person operating it.
Key Findings
Live birth rates: no significant differences
| Arm | Live births | Rate |
|---|---|---|
| Time-lapse (culture + algorithm) | 175/520 | 33.7% |
| Undisturbed culture only | 189/516 | 36.6% |
| Standard care | 172/522 | 33.0% |
The full time-lapse package (33.7%) performed within a percentage point of a conventional incubator (33.0%). Paying $500 to $1,000 bought a 0.7 percentage point difference that was not statistically significant.
The statistics confirm it
Time-lapse vs standard care: adjusted OR 1.04 (97.5% CI: 0.73–1.47). Absolute difference: +0.7 percentage points (97.5% CI: -5.85 to 7.25).
Undisturbed culture vs standard care: adjusted OR 1.20 (97.5% CI: 0.85–1.70). Absolute difference: +3.6 percentage points (97.5% CI: -3.02 to 10.22).
The 97.5% confidence intervals (Bonferroni-corrected for two primary comparisons) include zero in both cases. Neither comparison reached significance.
Undisturbed culture alone did slightly better than the algorithm
Look at the numbers again. The middle group, embryos in the time-lapse incubator but selected by a human embryologist, had the highest live birth rate at 36.6%. Adding the selection algorithm dropped it to 33.7%. That difference also wasn’t significant, but the direction matters: the algorithm didn’t help, and the point estimate suggests it may have made things marginally worse.
Safety was equivalent
79 serious adverse events across all three groups (28, 27, and 24 respectively). All were judged unrelated to the trial intervention.
The Lancet Editorial
The accompanying editorial by Bergh and Lundin chose a title that leaves little room for interpretation: “No improvement in livebirth rates by time-lapse technology.” Published alongside the trial in Lancet, 2024.
The Post-Publication Debate
Campbell et al. published a critical review in Reproductive Biomedicine Online (2025) questioning aspects of the experimental design. The trial authors (Bhide et al.) responded that the concerns were “largely addressed in the paper.” The exchange is worth noting for completeness, but the trial’s result, a three-arm randomized design with 1,575 patients showing no benefit, speaks for itself.
What This Means for Patients
“24/7 embryo monitoring” is a marketing phrase, not a clinical benefit. The camera watches. The live birth rate doesn’t change. If your clinic lists time-lapse as a premium add-on, you’re paying for a feature that makes embryology more convenient for the lab, not more effective for you. Similar to AI embryo selection: the technology benefits the workflow, not the pregnancy test.
This trial had no manufacturer funding. The TILT trial was funded by Barts Charity and academic grants. No time-lapse incubator company paid for it. Compare that to the iDAScore AI trial (Journal Club #4), where Vitrolife funded its own product’s trial, had employees as co-authors, and still couldn’t demonstrate noninferiority. When even industry-funded trials can’t show benefit, and independently funded trials confirm it, the evidence base is clear.
The UK’s HFEA already rates time-lapse as “amber” (insufficient evidence of benefit). This trial should push it further.
The Add-Ons Trilogy
This is the third in a series of large randomized trials that all point the same direction.
- Journal Club #3: ICSI vs IVF. 824 women. ICSI without severe male factor: no benefit, worse fertilization rates, potential harm in women under 32. Add-on cost: $1,000–$2,500.
- Journal Club #4: AI embryo selection. 1,066 patients. iDAScore AI failed noninferiority versus a trained embryologist. Funded by the manufacturer. Add-on cost: $500–$1,500.
- This trial. 1,575 patients. Time-lapse incubation and selection: no improvement in live births. Add-on cost: $500–$1,000.
Three procedures. Three large, well-designed trials. Zero evidence of benefit. Combined add-on cost if a clinic bundles all three: $2,000 to $5,000 per cycle, on top of the base IVF price. That money buys a lot of things. Evidence-based improvement in your live birth rate is not one of them.
Limitations
- Seven centres (5 UK, 2 Hong Kong). Results may differ in clinics with different lab environments or patient populations.
- Embryologists were necessarily unblinded to the incubator type, which could introduce bias in embryo handling or culture decisions.
- The undisturbed culture arm showed a non-significant 3.6 percentage point improvement. A larger trial might detect a real but small benefit from undisturbed culture specifically, separate from the algorithm.
- Women using donor gametes were excluded. Results apply to patients using their own eggs.
- The trial enrolled women across their first, second, or third IVF/ICSI cycles. Subgroup effects by cycle number were not reported.
Practical Takeaway
When your clinic offers time-lapse incubation, ask one question: does this improve my chance of a live birth? The TILT trial, the largest double-blind RCT on the topic, says no. A conventional incubator and a trained embryologist picking your blastocyst by morphology produced the same live birth rate as the full time-lapse package.
Your money is better spent on a clinic with strong lab fundamentals and transparent outcomes data than on add-ons that look impressive on a brochure.
This is part of EuroFertile’s Journal Club. Summaries of recent fertility research, written for patients, not doctors. Browse all research summaries →
Choosing a clinic for your cycle? Get matched to clinics in your budget → | Estimate your costs →