Petrolatum is closer to being the ruler than the thing measured — and little has cleared it yet

A 1997 study ran a repetitive irritation test in twenty people. Irritants were applied repeatedly to the forearms, with some sites pre-treated with petrolatum, to see how much protection it gave.

There were four irritants:

• 10% sodium lauryl sulphate (SLS) — a surfactant
• 1% sodium hydroxide (NaOH) — an alkali
• 30% lactic acid — an acid
• Toluene — an organic solvent

The result: petrolatum was very effective against the first three and moderately effective against toluene.

Why toluene behaved differently is guessable from its properties. The first three are water-soluble, and petrolatum's oily film blocks the water route. Toluene is a lipid-soluble solvent and passes through that same film. It is a result that shows an occlusive is not universal.

The authors' proposal is the heart of this item: use petrolatum as the standard reference substance for these tests.

To unpack that: when a company says 'our barrier cream protects against irritation', the claim only means something against a comparator. The authors' conclusion is that petrolatum is consistent enough and effective enough to serve as that comparator.

This was twenty people, and the control was an untreated site. It was not a randomised trial, so by this site's standard it sits at 'moderate'.

A 2000 trial asked exactly this question: if you put back the lipids skin already has, is that better than a plain grease film?

The design: healthy skin was deliberately damaged two ways — repeated tape stripping of the stratum corneum and detergent perturbation of the barrier. Two treatments were then applied over 14 days.

• Skin-identical lipids in a petrolatum-rich base
• Pure petrolatum

The result, in the authors' own wording: there was no indication that skin-identical lipids are more efficient than pure petrolatum.

How to read this.

It does not mean ceramides are useless. What this trial used was a lipid mixture, not ceramide alone, and the subjects were healthy people with deliberately damaged skin. People with actual skin disease may differ.

But it does retire one claim — that 'it is a substance the skin already has, so obviously it is better.' That obviousness was not confirmed here.

The ceramide page records a trial of the same character. The two point the same way.

The 2020 trial has the cleanest design on this page: a four-arm crossover, randomised, double-blind clinical pharmacology study in fifty-one healthy volunteers with dry skin, applied to the forearms.

• The complete emollient cream (V0034CR)
• Its vehicle alone
• Vehicle + glycerol
• Vehicle + petrolatum

Transepidermal water loss (TEWL) and stratum corneum moisture (corneometry) were measured serially over 12 hours after application.

Because each person experienced all four in a crossover, differences between people do not blur the result. That is an unusually good condition for a trial in this layer.

The authors' conclusion is the whole of this item: the petrolatum component improved barrier function (reduced TEWL), and the glycerol component improved hydration.

One trial separated two different jobs. The 'moisturising' that advertising rolls into one word is at least two things, and petrolatum is on the side that stops water leaving, not the side that holds water in.

Which is why so many products contain it alongside glycerin. They are not competitors.

The limits: this was a 12-hour pharmacology study, and the authors state explicitly that it was not designed to evaluate therapeutic benefit. It is also an industry trial of a specific finished product.