PGT-A: screening embryos for chromosome number
Fewer miscarriages, and in the best trials, fewer babies.
This is the most heavily sold add-on in fertility care, and it has the most evidence. The evidence is not what the marketing suggests.
The trial in the previous topic found ongoing pregnancy per patient randomized of 41.8% with screening against 43.5% without.
The larger trial, and the one that measured what patients actually want. 1,212 women aged 20 to 37 with three or more good blastocysts, followed through up to three successive single transfers within twelve months:
| Outcome | With screening | Conventional IVF | Difference (95% CI) |
|---|---|---|---|
| Cumulative live birth | 77.2% (468 of 606) | 81.8% (496 of 606) | −4.6 points (−9.2 to −0.0) |
| Cumulative clinical pregnancy loss | 8.7% | 12.6% | −3.9 points (−7.5 to −0.2) |
Read plainly: in good-prognosis patients, adding the test was associated with about 28 fewer babies per 1,000 women treated, and about 39 fewer miscarriages per 1,000. Both effects are real, and the live birth difference sits at the very edge of statistical significance.
One design point matters for how much weight to put on this. The trial was set up to test whether conventional IVF was no worse than screening, allowing a margin of 7 percentage points. Conventional IVF passed that test comfortably.
A woman for whom the experience of miscarriage is the dominant fear may reasonably choose differently from a woman optimising for the highest chance of a baby. That is not a question the evidence answers. It is a question about what you would rather risk, and nobody should resolve it for you.
The part patients are least often told. Some embryos come back "mosaic", neither clearly normal nor clearly abnormal, and historically many were discarded. A blinded trial transferred embryos without using the result to select them, then unblinded: across 484 normal, 282 low-grade mosaic and 131 medium-grade mosaic embryos, live birth and miscarriage rates were equivalent. No mosaicism or uniparental disomy was found in the resulting pregnancies or newborns. In a companion mapping study, when mosaicism affected fewer than half the cells in one biopsy, only 1% of those abnormalities were present elsewhere in the embryo.
A five-to-ten-cell biopsy is being used to classify an embryo of roughly 200 cells. For low and medium grade mosaic results, that classification does not predict the outcome.
the value of PGT-A as a routine screening test for all patients undergoing in vitro fertilization has not been demonstrated, and that
the value of PGT-A to lower the risk of clinical miscarriage is also unclear.
On cost, an analysis of 158,665 US cycles found that in patients under 35 the test produced worse clinical outcomes at higher cost, and that from the patient's perspective the incremental cost per live birth favoured not testing from under 35 through age 38.