Cholesterol

Cholesterol's comedogenic 0 is a real printed reading of the exact material, which is more than most scores on this site can claim — but it is one rabbit-ear result from 1989, with no second reading in the primary inventories we searched, and the argument usually offered alongside it, that a lipid the skin already makes cannot block a follicle, is not what the 0 rests on.

What it is

Cholesterol is a sterol: a four-ring steroid alcohol, CAS 57-88-5, molecular formula C27H46O and molecular weight 386.64, as the Cosmetic Ingredient Review's 1986 assessment records it. On a label it is simply Cholesterol. The same assessment gives its job in a formula — it is used primarily as an emulsifier and conditioner, and describes it as an emollient — and notes that it is used the same way outside cosmetics, as an emulsifying agent in pharmaceutical and veterinary products and in liposomes. It is also a lipid the body already makes. CIR's report puts the point plainly: cholesterol "is a component of skin surface lipids and sebum," skin surface lipids contain 2 to 20% sterols of which 90 to 95% is cholesterol, and sebum from the forearm "generally contains about 5% Cholesterol, half in the free form and half esterified." That fact is where most writing about this ingredient starts and finishes. It is worth holding on to, because it is the one thing on this page that is not evidence about pores.

Why we rate it Clear

We rate it Clear, at Low confidence. The confidence is low for the ordinary reason on this site: the only comedogenicity reading for cholesterol in the primary inventories we searched is a single rabbit-ear score. The number. James Fulton's 1989 survey in the Journal of the Society of Cosmetic Chemists prints a row reading "Cholesterol — 0" on a 0-to-5 scale, with no asterisk, and we read it off a 400dpi render of the journal scan rather than off a chart quoting it. Fulton's own published key: "a minimal grade of 0 to 1 is not considered significant." A 0 is the bottom of his scale. His 1984 paper, a separate study read under its own key, defines the bottom of that paper's scale as "no increase in visible follicular keratosis"; we do not carry that wording across to define the 1989 zero. What the 1989 row records is a bottom-of-scale reading, under his scoring procedure, in his animals, at his concentration. It is not a finding that nothing happened. The material identity holds, which is not something we can say about many of the numbers in circulation. Fulton names the material with the same single word the label uses. There is no near-miss neighbour here whose score has been quietly borrowed — no derivative, no fraction, no differently-processed grade standing in for the thing itself. The row is the ingredient. Six rows in that table carry a sterol name, and all six read 0: cholesterol, soya sterol, PEG-5 soya sterol, PEG-10 soya sterol, choleth-24, and a row Fulton calls "sterol esters." We report that because it is what the table says, not because we can tell you why. Fulton measured follicles; he did not explain them, and we have been caught before inventing chemistry to prop up a rating. There is one argument we are not going to make, because it is the argument everyone else makes and it is not evidence. Cholesterol is a normal constituent of sebum, so it is often described as skin-identical and therefore incapable of blocking a follicle. Origin is not a measurement. Whether a follicle treats additional cholesterol as familiar or as bulk is a question about what happens in the follicle, and nothing we read measured it. The 0 does not rest on that reasoning, and it does not need it.

What the evidence doesn’t tell you

All of the above is one assay. Fulton's method: the ingredient at roughly 10% in propylene glycol, about 1 mL to the inner ear of three New Zealand albino rabbits, five days a week for two weeks, follicles then scored for keratosis under a micrometer. Three animals, one laboratory, one reading. Fulton called the survey "not at all definitive but simply designed to stimulate research." We went looking for a second reading and did not find one. Morris and Kwan's independent 1983 rabbit-ear assay, on the same 0-to-5 scale, prints thirty-two rows and none of them is cholesterol. Fulton's own earlier 1984 table has no cholesterol row. Kanaar's 1971 ear-canal work covered six fatty acids and no sterols. The abstract of Nguyen and Maibach's 2007 assay names fourteen test materials and cholesterol is not among them. Within the legacy corpus we hold, Fulton 1989 is the only document that gives this material a score — and that corpus is not a complete census of the literature. Every time we have assumed it was closed, it has turned out not to be. We identified no human comedogenicity study of isolated cholesterol: no test that put the ingredient on its own on a human face or back and counted comedones. Its general safety is documented, and that is a different question. The CIR Expert Panel's 1986 final report concluded that cholesterol "is safe as presently used in cosmetic products," on toxicity, irritation, sensitisation and photosensitisation evidence, with clinical testing done on finished products rather than the raw ingredient. We searched both that report and CIR's 2024 re-review packet for the word "comedo" and found it in neither. Safety is not comedogenicity: a material can be non-toxic and non-irritating and still plug a follicle, and converting a safety conclusion into a comedogenic one is a substitution we are not going to make. The same caution applies to the literature cholesterol is usually marketed on. Barrier-repair research on ceramide, cholesterol and fatty-acid mixtures — Man and colleagues in the Journal of Investigative Dermatology in 1996 is the paper most often cited — has barrier function as its endpoint. It counts water loss, not comedones. It cannot tell you what cholesterol does to a follicle, in either direction. One question we could not close. Cholesterol is a constituent of sebum, and human sebum was itself the original test article in this model: Kligman and Katz published on the comedogenic properties of human sebum in the rabbit external ear canal in 1968, and Kligman, Wheatley and Mills followed in 1970 by applying sebum fractions and individual sebum components to the same model. We were not able to obtain either paper — both are paywalled to us, and the abstract servers return a captcha — so we cannot tell you whether cholesterol was among the components tested or what it did. We would rather name the gap than write as though it were not there. Finally, the model itself is contested, and the objections cut toward a reassuring 0 exactly as hard as toward an alarming 4. Draelos and DiNardo tested finished products on human backs in 2006 and found that isolated ingredient scores did not transfer to the formulas they ended up in. A 2025 review in JAAD Reviews found that no standardised comedogenicity test exists and that "non-comedogenic" is an unregulated claim. What a rabbit-ear 0 predicts about a human face has not been measured. Unproven is not disproven, and it is not proven either.

Where you’ll see it

On a label it appears as Cholesterol. The CIR's 1986 report recorded it in 145 cosmetic formulations registered with the FDA — mostly skin care products, eye and face makeup, hair care products and shaving preparations — at 5% or less, with the majority of products at 0.1 to 1%. Those are FDA voluntary-registration figures from the early 1980s and a use survey is not a market measurement; CIR's own report notes the concentration ranges are wide enough to introduce a two- to ten-fold error. We have not audited what the shelf looks like now. CIR released an updated re-review packet in May 2024 for its June 2024 panel meeting, containing newer use figures. Every page of it is stamped "Distributed for Comment Only — Do Not Cite or Quote," and it is a document put to the panel rather than a conclusion reached by it, so we are recording that it exists and leaving its numbers where they are. What we can record is what we read and when: the 1986 assessment as posted on CIR's site, and the May 2024 packet, both read in August 2026. We did not record a status line from the 1986 PDF, and we are not going to characterise where the assessment stands now.

Sources

Each source says what it actually is. Several widely-cited “sources” for comedogenic ratings are republishing the same 1989 assay, and counting them as independent agreement is how a thin evidence base gets made to look thick.

1. Fulton JE. "Comedogenicity and irritancy of commonly used ingredients in skin care products." J Soc Cosmet Chem 40:321-333 (1989) The primary source of nearly every comedogenic number in circulation, and the only document we hold that assigns cholesterol a comedogenicity score. Read off a 400dpi render of the journal scan rather than a list quoting it. Method: the ingredient at roughly 10% in propylene glycol, 1 mL to the inner ear of three New Zealand albino rabbits, five days a week for two weeks, follicular keratosis scored under a micrometer — three animals, a rabbit ear, not a human face. Its Table I prints "Cholesterol 0" on a 0-to-5 scale, with no asterisk, and the row is not in the paper's oils section, so it does not inherit that section's "results depend on source of raw material" caveat. Fulton's key: 0 to 1 "not considered significant," 2 to 3 "borderline," 4 to 5 "considered positive." Fulton records no supplier, purity or batch for any row, so the identity match is at the level of the printed name. He called the survey "not at all definitive but simply designed to stimulate research." acne.org, CosDNA and the consumer charts republish this row; they are the same reading quoted again, not independent corroboration.
2. Cosmetic Ingredient Review. "Final Report on the Safety Assessment of Cholesterol." J Am Coll Toxicol 5(5):491-516 (1986). doi 10.3109/10915818609141922 An expert-panel SAFETY assessment. What we read is the PDF at the URL above, filed as 115_buff3d_suppl.pdf, read in August 2026; we did not record a status line from it, and that filename does not obviously correspond to the journal-published final report, so we describe the reading rather than the document. Cited for four things and no others: the material's identity (CAS 57-88-5, C27H46O, MW 386.64); its function in a formula (emulsifier and conditioner, also described as an emollient, and the same role in pharmaceutical and veterinary products and in liposomes); its presence in skin lipid ("a component of skin surface lipids and sebum"; skin surface lipids 2 to 20% sterols, 90 to 95% of that cholesterol; forearm sebum "about 5% Cholesterol, half in the free form and half esterified"); and its 1980s use data (145 FDA- registered formulations, 5% or less, majority 0.1 to 1%). Its conclusion — "safe as presently used in cosmetic products" — rests on toxicity, irritation, sensitisation and photosensitisation data, with the clinical work done on finished products (most at 1.4% cholesterol, the highest at 6%) rather than the isolated ingredient. It contains no comedogenicity data; we searched it for the word. A safety conclusion is not evidence for or against a comedogenic rating and we do not use it as one. Its use figures are voluntary FDA registrations from the early 1980s, not a current market survey.
3. Morris WE, Kwan SC. "Use of the rabbit ear model in evaluating the comedogenic potential of cosmetic ingredients." J Soc Cosmet Chem 34:215-225 (1983) An independent primary rabbit-ear assay on the same 0-to-5 scale, six years before Fulton 1989, thirty-two rows, materials applied undiluted unless stated and graded microscopically — the one legacy paper that could have given cholesterol a second reading. Cited here purely for an absence: it has no cholesterol row and no sterol row of any kind. Morris and Kwan are also on record contradicting Fulton in print on other materials, which is worth knowing when weighing a number that rests on Fulton alone.
4. Fulton JE Jr, Pay SR, Fulton JE III. "Comedogenicity of current therapeutic products, cosmetics, and ingredients in the rabbit ear." J Am Acad Dermatol 10(1):96-105 (1984). PII S0190-9622(84)80050-X Fulton's own earlier rabbit-ear paper, read from a 400dpi render of the journal scan. Method: two New Zealand albino rabbits per test, one ear dosed daily for two weeks, the other ear the control. Cited for two things: an absence — its Table I has no cholesterol or sterol row, so it neither corroborates nor contradicts the 1989 zero — and its verbatim definition of the bottom of the scale, "no increase in visible follicular keratosis." Note that its scale key is not the 1989 key: 1984 calls 1 to 2 not significant and 3 or above positive, so its numbers cannot be read against 1989's without carrying each paper's own threshold.
5. Kligman AM, Wheatley VR, Mills OH. "Comedogenicity of human sebum." Arch Dermatol 102(3):267-275 (1970). PMID 4247928. And: Kligman AM, Katz AG. "Pathogenesis of acne vulgaris. I. Comedogenic properties of human sebum in external ear canal of the rabbit." Arch Dermatol 98(1):53-57 (1968). PMID 4232034 NOT READ. Two papers we could not obtain — both paywalled to us, and the abstract servers return a captcha — cited here as a named gap rather than as evidence. Between them they applied human sebum, sebum fractions and individual sebum components to the rabbit ear model. Because cholesterol is a constituent of sebum, they are the most directly relevant unexamined literature for this page; if their full methods and results can be obtained, they might bear on whether a lipid the skin already secretes behaves any differently in this assay. We cannot say whether cholesterol was among the components tested, or what result it produced, and nothing on this page depends on either.
6. Man MQ, Feingold KR, Thornfeldt CR, Elias PM. "Optimization of physiological lipid mixtures for barrier repair." J Invest Dermatol 106(5):1096-1101 (1996) The barrier-repair literature cholesterol is usually sold on, named here so that its endpoint can be stated: it studies ceramide, cholesterol and free-fatty-acid mixtures applied to disrupted skin and measures permeability-barrier recovery. We were not able to retrieve the full text. It counts barrier function, not comedones, and it is cited on this page only to say what it therefore cannot support. Barrier data is not comedogenicity data, in either direction.
7. Cosmetic Ingredient Review. "Re-Review of the Safety Assessment of Cholesterol as Used in Cosmetics." Memorandum and data packet, released 10 May 2024 for the Panel meeting of 3-4 June 2024 A COMMENT-STAGE DOCUMENT, not a finding. Every page is stamped "Distributed for Comment Only — Do Not Cite or Quote," and its status line reads "Re-Review for Panel Consideration": it is material put in front of the expert panel so the panel can decide whether to reopen the 1986 assessment, not a conclusion the panel reached. We read it, we are disclosing that it exists, and we are not carrying its use figures onto this page as fact. It contains no comedogenicity data either. Interim packets and final reports are different document states, and we have previously cited one for the other.
8. "Comedogenicity in cosmeceuticals: a review of clinical relevance, regulatory gaps, and future directions." JAAD Reviews (2025) A recent peer-reviewed review of the whole field, covering literature from 1972 onward. It is a synthesis rather than a new measurement, and a review of this literature necessarily takes in Fulton rather than standing apart from him; what it is not is one of the consumer charts that republish his table. Cited for the field-wide caveat that no standardised comedogenicity test exists and that "non-comedogenic" is an unregulated label — a caution that governs the reassuring 0 on this page as much as it would govern a 4.

Last reviewed 2026-08-20 · How we decide

‹ Check a full ingredient list