

The claim is extraordinary, as could be the consequences, so the study should be discussed in depth.
See thread

https://academic.oup.com/humrep/article/doi/10.1093/humrep/deaa136/5860260
Some points to think about:
This is a retrospective study, single center.
Retrospective studies are good! They are the only way to get thousands of cases, but
only show association not causation.

Retrospective studies are good! They are the only way to get thousands of cases, but

Embryo storage time is short in all cases, from less than 3 months to a maximum of 24 months.
something is acting very fast? we see decrease in results already after 3 months in storage.

The storage intervals analyzed are…odd:
0-3 months, 3-6, 6-12, 12-24 (so 3, 3, 6, 12 months).
Why?
If you see an effect after just 3 months, why lump together a whole year of storage time?
0-3 months, 3-6, 6-12, 12-24 (so 3, 3, 6, 12 months).
Why?
If you see an effect after just 3 months, why lump together a whole year of storage time?
My guess: to have enough n in the last category.
As it is, there are 24.000+ cases in the first year, and just above 550 in the second one.
As it is, there are 24.000+ cases in the first year, and just above 550 in the second one.
Patients characteristics?
Women with embryos in storage for 1-2 years are 5 years older than those 0-1 year
going from 30 to 35 on average.
Maternal age has been adjusted for in the analysis, but its effect is not linear, especially going from non-AMA to AMA.
Women with embryos in storage for 1-2 years are 5 years older than those 0-1 year

Maternal age has been adjusted for in the analysis, but its effect is not linear, especially going from non-AMA to AMA.
There are ways to deal with this statistically, I do not see them used here…
I would love to see 2-3 age brackets stratified for storage time
is the storage effect maintained within the same age range?
We do not know...
I would love to see 2-3 age brackets stratified for storage time

We do not know...
What could cause this drop in reproductive results? (it is all speculative).
The authors give possible explanations: free radicals, toxicity of cryoprotectant, an effect on molecules movements, epigenetic effects…what is missing?
The authors give possible explanations: free radicals, toxicity of cryoprotectant, an effect on molecules movements, epigenetic effects…what is missing?
No mention of possible effect of opening the tank of liquid N twice every week to check level and refill manually.
We should consider that we may need to improve our operations and QC
We should consider that we may need to improve our operations and QC
Conclusion!
Interesting data, large cohort
very skewed population across groups
some unexplained analysis choices
not too strong a conclusion
So no, let's not rush to transfer stored embryos.
Let's design more, strong studies instead.




So no, let's not rush to transfer stored embryos.
Let's design more, strong studies instead.