Does PGT-A improve the chance of a live birth from one IVF egg retrieval? In a UK registry analysis of 111,975 treatment episodes, PGT-A was associated with lower cumulative live birth odds and lower pregnancy loss odds per pregnancy. These observational results from 2014–2018 do not establish that testing caused either difference.
- Analysis of over 110,000 UK HFEA treatment episodes: PGT-A associated with lower odds of cumulative live birth
- Lower pregnancy loss odds per pregnancy… Live birth per transfer and cumulative live birth per oocyte retrieval answer different questions
- Observational design and 2014–2018 historical data limitations… “Not a one-size-fits-all tool for every patient”
To couples undergoing In Vitro Fertilization (IVF), few options sound as intuitively reassuring as selecting a “chromosomally normal embryo” for transfer.
Preimplantation Genetic Testing for Aneuploidy (PGT-A) screens embryos for numerical chromosomal abnormalities before intrauterine transfer, allowing clinicians to prioritize euploid embryos. The working assumption seems straightforward: if you screen out aneuploid embryos and transfer only normal ones, shouldn’t the ultimate chance of delivering a baby rise?
However, a large-scale study of UK assisted reproductive technology (ART) registry data found an association between PGT-A and lower pregnancy loss odds per pregnancy, alongside lower cumulative live birth odds per egg retrieval episode compared with conventional IVF. This observational comparison does not establish causation.
PGT-A and live birth odds in the UK registry analysis
Researchers analyzed 111,975 treatment episodes recorded in the UK Human Fertilisation and Embryology Authority (HFEA) registry between 2014 and 2018. In this dataset, a single treatment episode encompasses one oocyte aspiration cycle and all subsequent fresh and frozen transfers utilizing embryos derived from that specific retrieval.
The findings showed that patients undergoing PGT-A had lower odds of delivering a baby compared to those undergoing conventional IVF:
- First-Transfer Analysis: The adjusted odds ratio (aOR) for live birth was 0.86 (95% confidence interval 0.77–0.97).
- Cumulative Analysis (All Transfers per Retrieval): The adjusted odds ratio was 0.74 (95% confidence interval 0.67–0.82).
Even after statistically adjusting for maternal age, infertility etiology, prior IVF attempts, ovarian stimulation protocols, ICSI utilization, oocyte and embryo yield, and clinic-specific effects, PGT-A was not associated with an increased live birth rate.
The study also reported the proportion of treatments with no embryo transferred in the first cycle:
- In the PGT-A cohort, 48% of treatments had no embryo transferred in the first cycle.
- In the conventional IVF cohort, 5% of cycles had no embryo transferred in the first cycle.
Embryo exclusion after screening and possible biopsy damage can reduce the number available for transfer. These are potential explanations, not mechanisms established by this registry comparison; a published response also noted that missing post-thaw survival data limited analysis of embryo loss.
Pregnancy loss odds per pregnancy
Among recorded pregnancies, PGT-A was associated with lower odds of pregnancy loss:
- First-Cycle Pregnancy Loss Odds (per pregnancy): aOR = 0.47 (95% confidence interval 0.36–0.61)
- Cumulative Pregnancy Loss Odds (per pregnancy): aOR = 0.64 (95% confidence interval 0.52–0.78)
This divergence underscores why the PGT-A debate remains intensely polarized:
| Clinical Metric | Conventional IVF | PGT-A Screened Cycles | Clinical Interpretation |
| Embryo Utilization | Higher proportion of embryos transferred | Fewer embryos may remain available after testing | Possible effect of embryo selection; mechanism not isolated here |
| No Transfer in the First Cycle | ~5% | ~48% | First-cycle observation; not a rate of never reaching transfer |
| Pregnancy Loss Odds (per pregnancy) | Reference | aOR 0.47 (first cycle); 0.64 (cumulative) | Lower observed odds; these are not risk ratios or percentage-point reductions |
| Cumulative Live Birth per Retrieval | Reference | Lower odds (aOR 0.74) | Observed association; this study does not isolate its cause |
PGT-A screens for numerical chromosome abnormalities. Selecting an embryo reported as euploid does not guarantee implantation, prevent every miscarriage, or guarantee a live birth.
However, lower pregnancy loss odds among pregnancies are not the same outcome as a higher cumulative chance of live birth from a retrieval.
As checked on 25 September 2026, the UK HFEA rates PGT-A Red for improving the chance of a baby and Green for reducing miscarriage for most fertility patients. Its separate ratings for both outcomes in older women are Grey, reflecting insufficient moderate- or high-quality evidence. These ratings do not quantify an individual patient’s benefit.
Per-transfer versus per-retrieval live birth: why the denominator matters
The central source of confusion surrounding PGT-A lies in the difference between two separate statistical metrics:
- Per-Transfer Success: The proportion of embryo transfers resulting in a live birth; this denominator includes transfers that actually occur.
- Per-Retrieval (Cumulative) Success: The chance of at least one live birth across transfers of embryos from one oocyte retrieval. Reporting per cycle started instead uses a different denominator.
Time-to-pregnancy (TTP) may be shorter for some patients if transfers are avoided, but HFEA notes that testing can also lengthen the time to pregnancy. A shorter treatment time is not guaranteed.
If a potentially viable embryo is excluded after an incorrect result, mosaic classification, or biopsy damage, fewer embryos remain available for transfer. With one or two blastocysts, this may reduce opportunities; it does not necessarily eliminate the entire chance from that retrieval.
ASRM’s 2024 committee opinion does not recommend routine PGT-A for all IVF patients. The named randomized trials studied selected populations and different endpoints: STAR enrolled women aged 25–40 with at least two biopsy-suitable blastocysts and found no significant overall improvement in ongoing pregnancy at 20 weeks. ESTEEM used polar-body testing in women aged 36–40 and found no substantial improvement in cumulative live birth within one year. Neither was an unrestricted all-comer study.
PGT-A study limitations and individualized treatment decisions
Drawing a blanket conclusion that PGT-A is “useless” or “harmful” would mischaracterize the study’s conclusions:
- Observational Design: This retrospective registry comparison cannot eliminate residual confounding. A published response noted that indications for PGT-A were not recorded, limiting adjustment for why particular patients received testing.
- Historical Data (2014–2018): A response by Mathieu d’Argent and colleagues in the same journal questioned applicability to contemporary practice, citing changes in testing platforms, selection of the available registry sample, and missing embryo-survival data. This is a methodological critique, not a new randomized comparison.
The question is not a binary “should everyone do PGT-A or avoid it?” Rather, it centers on identifying what the patient aims to achieve:
- Patients with Multiple Blastocysts: A 2018 decision model projected lower costs and shorter treatment for patients with more than one embryo. Those estimates depended on clinical data and modelling assumptions; they do not guarantee a saving or a live birth for an individual patient.
- Patients with Recurrent Pregnancy Loss: Avoiding another loss may be a major personal goal, but ASRM’s 2024 opinion states that definitive evidence of PGT-A benefit in recurrent pregnancy loss is lacking. Individual counselling should address this uncertainty.
- Patients with 1–2 Blastocysts: Counselling should consider the possibility of no embryo remaining for transfer, inconclusive results, and how the clinic handles mosaic findings. The balance of benefits and disadvantages cannot be determined from embryo count alone.
PGT-A must answer two distinct questions: “Which embryo should be transferred first?” and “Does testing improve this specific patient’s cumulative chance of having a baby?” As this large-scale registry analysis demonstrates, the answers to those two questions do not always align.
To explore two related but distinct questions, read about mosaic embryo transfer and classification and live birth after repeated euploid embryo transfers.
Medical Source & Study Reference
- Journal: Reproductive BioMedicine Online (RBMO), Volume 53, Issue 3 (September 2026).
- Study Title: PGT-A is associated with a reduced live birth rate in routine practice: evidence from the UK ART register
- Lead Institutions: University of Manchester; St Mary’s Hospital, Manchester University NHS Foundation Trust, UK.
- Cohort Size: 111,975 treatment episodes from the UK Human Fertilisation and Embryology Authority (HFEA) registry (2014–2018).
- Article Identifiers: Article ID: 105747 | DOI: 10.1016/j.rbmo.2026.105747 | PMID: 42275751
Additional sources checked: UK registry study abstract; HFEA PGT-A ratings; ASRM 2024 committee opinion; STAR randomized trial; ESTEEM randomized trial; 2018 cost and treatment-time model; 2026 methodological response.
※ This article provides general scientific and medical information and does not replace individualized clinical diagnosis or care. Decisions regarding PGT-A should be tailored to individual patient profiles—incorporating maternal age, reproductive history, previous miscarriages, and oocyte/blastocyst yield—through detailed counseling with an attending reproductive endocrinologist and clinical geneticist.
※ The images associated with this article were generated using generative AI (ChatGPT, OpenAI) as illustrative visual references and do not depict real individuals.
※ Copyright © 2026 THE FERTILITY NOTES. All rights reserved. Unauthorized reproduction, duplication, distribution, scraping, AI training, or data utilization of all content, including articles, photos, and images, is strictly prohibited and may result in legal liability.