How your barrier repairs itself

Lesson 3 of 11 · 4 min read

"Barrier repair" is printed on a lot of jars, and it sounds like one thing you either buy or you don't. It's actually a specific piece of building work happening in your outermost skin layer. Once you can picture that work, the back of the box starts telling you far more than the front does.

Bricks and mortar

Your outer layer, the stratum corneum, is built like a brick wall. The bricks are corneocytes: flat, dead skin cells packed with keratin and stacked in overlapping rows. The mortar between them is fat. Lipids fill the gaps in thin, ordered sheets, and that greasy filler is what makes the wall watertight.

So a damaged skin barrier is mostly a mortar problem. When lipids run low or get stripped out, gaps open between the bricks. Water that ought to stay in your skin evaporates off faster than usual, so skin feels tight and looks dull. Traffic runs the other way too: irritants, allergens and bacteria push further in than they normally would. That's why a worn barrier suddenly stings from products it tolerated fine last month.

Three lipids, one team

The mortar isn't a single ingredient. Three families of lipid occur naturally between the bricks: ceramides, cholesterol (the same molecule your body makes on its own) and free fatty acids.

Research on compromised skin, meaning eczema, atopic dermatitis, post-procedure skin and skin past 40, looked at what happens when you put those lipids back. One finding holds across all of it: repair runs fastest when all three go on together. A mix missing a class doesn't perform the same way, and cholesterol without ceramides is the clearest case, because on its own it can't rebuild the layered sheets and may build faulty ones instead.

Which lipid leads the mix is the part that shifts. It tracks whichever one the skin is short of, and for skin past 40, where cholesterol runs most depleted, the balance studied is cholesterol-led: roughly 3:1:1 cholesterol to ceramides to fatty acids. The same balance has been studied for skin healing from a procedure. Other situations call for a different lead, which is where the next step in this course picks up.

You can't verify a ratio at the shelf, though. Percentages aren't printed on packaging, so no ingredient list can promise you 3:1:1. The usable half of the finding is simpler anyway. A product carrying all three families beats a product built around one.

Beyond the mortar

The bricks get built too. Every corneocyte is wrapped in a hardened protein shell called the cornified envelope. The lipid mortar chemically anchors to that wrapper. When the wrapper is poorly built, as it is in atopic skin, the mortar has nothing to grip, and water leaks out no matter how good the lipids are.

Niacinamide is the one widely used cosmetic ingredient with solid evidence for supporting how that wrapper matures. From around 2 percent it pushes skin cells to make more of the proteins the shell is built from, and more ceramides alongside. That explains something that looks odd on a shelf: niacinamide keeps showing up in barrier toners and essences that contain no lipids at all. A watery formula can't deliver mortar. It can still support the bricks.

What this means at the shelf

Don't shop by the ceramide count on the front of the box: a "five ceramide complex" can be five ceramides sitting near the end of the ingredient list, where the total amount is functionally trace, while a formula with two or three of them higher up does more real work.

Turn the box over and look for the team instead. You want a ceramide (NP, AP, EOP and NS are the common ones), cholesterol, and a free fatty acid, which reads on a label as stearic, palmitic, oleic or linoleic acid. Niacinamide next to them is a good sign. If the print on the tube is too small to work with, the Skin Bliss app can scan the label and lay out the full ingredient list, so you can check for all three families yourself. A later step in this course gets into choosing an actual moisturizer, and this is the shape to have in mind by the time you do.