Post-Thaw Viability — What the Number Means and How It Is Measured

Post-thaw viability is the percentage of cells in a thawed cryopreserved vial that a stated assay judges to be alive, at a stated time after thaw. It is a membrane-integrity measurement in almost all commercial usage, not a measurement of whether the cells will attach, survive or function.

A vendor stating “>80% post-thaw viability” without naming the method and the timepoint has not made a checkable claim. Those two omissions are enough to move the number by more than the difference between suppliers.

What the number is actually measuring

Every routine viability assay in this market answers the same narrow physical question: is this cell’s plasma membrane intact right now? A dye that cannot cross an intact membrane either gets in or does not. That is the whole measurement.

Membrane integrity is a reasonable proxy for necrotic death and a poor proxy for everything else. A cell can have an intact membrane and be committed to apoptosis, be unable to attach, have lost the phenotype you bought it for, or be about to die in the next four hours. All of those score as viable.

This is why the number is far less informative than its prominence on datasheets suggests, and why the more demanding measurement — plating efficiency — is the one that predicts your experiment.

Method comparison

Method	What it detects	Readout	Counts debris and non-nucleated material?	Directional bias	Typical use
Trypan blue exclusion	Membrane integrity. Trypan blue is membrane-impermeable and stains only cells with compromised membranes	Brightfield; manual haemocytometer or automated image cytometer	Yes. Brightfield cannot distinguish a cell from debris, a red blood cell, or an aggregate on morphology alone	Tends to read high on samples containing debris or non-nucleated material, because unstained non-cells are scored as live cells	The default. Cheap, fast, universal, and the method behind most published vendor numbers
Acridine orange / propidium iodide (AO/PI)	Dual fluorescence. AO is membrane-permeable and stains nucleated cells; PI is membrane-impermeable and stains only compromised cells	Two-channel fluorescence on an image cytometer	No. Only nucleated events are counted, so debris and non-nucleated material are excluded	Tends to read lower and truer on debris-laden samples, precisely because it stops counting the things trypan blue miscounts	Post-thaw counts, primary isolates, blood-derived samples, and anywhere debris is expected
Plating efficiency (attachment assay)	Whether cells attach and persist on a substrate over a defined interval	Count attached cells at a fixed timepoint against cells seeded	Not applicable — non-attaching material is washed away, which is the point	The most conservative and the most predictive. Routinely lower than either dye method	The measurement that actually predicts whether your experiment will work
Metabolic assays (MTT, resazurin and similar)	Bulk metabolic activity across a well	Absorbance or fluorescence, population-level	Not applicable	Reports population metabolic capacity, not the proportion of live cells. Metabolic rate per cell varies with state, so it is not a viability percentage	Proliferation and cytotoxicity work; a poor substitute for a post-thaw count
Flow cytometry with viability dye	Membrane integrity, per cell, with optional marker co-staining	Fluorescence per event	No, with correct gating	Method-dependent; the most information-rich and the least commonly used for a routine release spec	Purity and viability combined, typically on higher-value products

Why vendor numbers are not comparable

Six variables move the published figure, and vendors rarely disclose more than one or two.

1. The assay. As above. Trypan blue and AO/PI are not measuring the same population, because they disagree about what counts as a countable event. Trypan blue counts unstained objects as live; AO/PI counts only nucleated objects at all. On a clean sample of a homogeneous line the two can agree closely. On a post-thaw sample containing lysis debris — which is exactly the sample in question — they can diverge substantially, with trypan blue the more generous. Comparing a trypan blue figure from one vendor with an AO/PI figure from another is not a comparison.

2. The timepoint. Immediately after thaw, in the presence of residual DMSO, is the most favourable moment a vial will ever have. Membrane damage from cryopreservation continues to manifest for hours. A count at ten minutes and a count at four hours are different numbers from the same vial, and the earlier one is higher.

3. The wash. Whether cells are counted in thaw medium or after a centrifugation and resuspension changes the result, because the wash removes some fragile cells and some debris and those two removals push the number in opposite directions.

4. The instrument and its settings. Automated cytometers apply size gates, roundness filters and declustering algorithms. Two instruments running the same sample under different gate settings return different percentages. Manual haemocytometer counts add operator variance on top.

5. The denominator. This is the one that interacts with price. Some vendors quantify a vial as total cells at freeze; others as viable cells post-thaw. The catalogue that quotes total cells at freeze and separately quotes a viability percentage is asking you to do a multiplication that the vendor quoting viable post-thaw cells has already done. A vial described as one million cells with 80 percent viability contains fewer usable cells than a vial described as one million viable cells post-thaw, at whatever the two prices are.

6. Whether the figure is a specification or an observation. A release specification is a floor the vendor tests every lot against and will not ship below. A typical or representative value is an observation from some lots. These appear identically on a datasheet and mean entirely different things. Ask which one you are reading.

The measurement that actually matters

Plating efficiency — what fraction of seeded cells attach and remain attached at a defined timepoint — predicts your experiment better than any dye method, because it integrates membrane integrity, metabolic competence and adhesion capability into a single functional outcome.

It is also much less commonly specified, for the obvious reason that it is harder to pass. It depends on the coating, the medium, the seeding density and the substrate, so a vendor publishing a plating efficiency number has to publish the conditions with it, and those conditions become an implicit promise about performance in the buyer’s hands.

If a supplier will state a guaranteed attachment figure under stated conditions, that is a materially stronger commitment than a viability percentage, and it is worth paying for.

What to do on arrival

Run your own count. It takes twenty minutes and it is the only way to know what you received.

Count with the same method the vendor used, so you have a like-for-like comparison against the certificate of analysis, and record which method that was.
If the sample looks debris-heavy under brightfield, run AO/PI as well. A large gap between the two numbers is itself the finding.
Count at the vendor’s stated timepoint, then again at four hours if the result is marginal.
Seed a small attachment control at the vendor’s recommended density and count attached cells at 24 hours. This is your real answer.
Photograph the count field. If you later need to raise a claim, an image at the point of receipt is the difference between a discussion and an argument.
Raise discrepancies within the vendor’s claim window. It is short, and it is stated in the terms of sale rather than on the product page. Check it on the day the shipment arrives, not on the day you find the problem.

Boundary cases and common errors

Viability is not purity. A vial can be 95 percent viable and 60 percent the cell type you ordered. These are independent specifications against independent assays, and a strong number on one says nothing about the other.

Viability is not potency. For a functional product — neurons expected to fire, myoblasts expected to fuse, cardiomyocytes expected to beat — the viability figure is a necessary condition and nowhere near a sufficient one.

High viability with poor attachment usually indicts the protocol, not the vial. Coating chemistry, coating age, medium formulation and seeding density are the usual causes. Confirm you are running the vendor’s stated conditions before escalating.

Aggregation inflates automated counts. Clumped cells can be scored as single large events or declustered incorrectly depending on instrument settings. Post-thaw samples aggregate readily. Inspect the field rather than trusting the number.

A viability figure with no method attached is a marketing claim. Treat it as one until the method, timepoint and specification status are supplied. In our experience suppliers answer those three questions readily once they are asked directly — the information is simply not volunteered, because no one requires it to be.

Post-thaw viability: what the number means