j. Soc. Cosmet. (5hem., 47, 41-48 (January/February 1996) Petrolalum is not comedogenic in rabbits or humans: A critical reappraisal of the rabbit ear assay and the concept of "acne cosmetica" ALBERT M. KLIGMAN, Department of Dermatology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104. Accepted March 8, •996. Synopsis The objective was to find out whether petrolatum, previously reported to be comedogenic in the rabbit ear model, would worsen acne. Vaseline petrolatum jelly (Vaseline ©, Chesebrough-Ponds) and Vaseline pe- trolatum jelly cream (Creamy Vaseline ©, Chesebrough-Ponds) were applied to separate groups of ten acne patients each twice daily for eight weeks. No evidence of comedogenic potential was found by the follicular biopsy technique. Papulo-pustules significantly decreased in both groups. Greasiness cannot be equated with comedogenicity. The admonition for acne patients to seek "oil-free" skin care products has no scientific merit with regard to comedogenicity. Cosmetics are an unlikely cause of the relatively high prevalence of post-adolescent female acne. INTRODUCTION In a 1972 paper entitled "Acne cosmetica," Kligman and Mills reported that as many as 50% of commercial cosmetics were comedogenic in the rabbit ear model (1). Since then, this assay has been extensively used by producers of drugs and cosmetics to assess their potential for inducing acneform eruptions (2). Recently, manufacturers have added the human back test to appraise acnegenicity. Reputable manufacturers of cosmetics rou- tinely screen their products for comedogenicity in order to satisfy the consumer that their product is "non-comedogenic," a useful claim in a highly competitive market. However, neither the rabbit nor the human assay is entirely satisfactory. Serious con- troversies have sprung up regarding the reliability and relevance of the rabbit ear model. A voluminous literature reveals remarkably divergent results among laboratories and investigators. Substances deemed to be strongly comedogenic by some are declared by others to be innocuous. Reviewers have not failed to notice the high frequency of strikingly contradictory results. Long ago Frank questioned whether the rabbit ear model had any relevance for humans (3). The most extreme example of the controversies generated by the rabbit ear assay is the publications of Fulton, a prominent acneologist 41

42 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS (4). He has classified a large number of substances according to their comedogenic potential. He states categorically that manufacturers should automatically exclude in- gredients that are comedogenic at any concentration. This caveat has brought to the fore a vexing problem for manufacturers, namely, the substantial number of desirable and seemingly safe substances that have been "blacklisted" by Fulton. Nelson and Rumsfield, too, have published lists of "unacceptable ingredients" (5). They go so far as to recom- mend products that do not contain "unacceptable ingredients." Draelos, too, has sharply commented on the complexities and uncertainties of evaluating cosmetic components by the rabbit ear model (6). Especially notable in Fulton's multitudinous list of come- dogenic substances are D & C red dyes and a large variety of fatty substances that contribute to the aesthetic and functional properties of cosmetics. These have not been demonstrated to be acnegenic in humans. Because of these conflicts, I updated the rabbit ear assay in a 1990 publication, adding specifications that, hopefully, would substan- tially reduce inter-laboratory disagreements (7). REASSESSMENT OF THE COMEDOGENICITY OF OILS This wordy introduction brings us to the issue that is central to the focus of this paper. Using the original model, Kligman and Mills stated that petrolatums and mineral oils from different sources were uniformly comedogenic (1). With the updated model, it turns out that these were "false positives" (see below). This is far more than an academic controversy since it strongly impacts on the credibility of safety claims for skin care products. Seborrhoea, the excessive production of sebum, is a prerequisite for the development of acne (8). Persons with seborrhoea are also greatly discomforted and try to remove excess oil by frequent washing or by oil-absorbing papers. It is understandable that acne patients intuitively wish to avoid oily and greasy cos- metics. Indeed, it has become the universal mantra for cosmetic manufacturers to claim that their products are "oil-free." Dermatologists, too, routinely proscribe oil- containing facial products for acne sufferers. Patients are advised to read the labels and avoid medicaments and cosmetics that contain "oils." The assumption is that oily substances are intrinsically comedogenic. A justification for this belief stemmed from an earlier publication that described pomade acne in Afro-American men (9). Adult black men use a variety of greasy products daily to groom curly, kinky hair. Dense crops of open and closed comedones sometimes develop on the glabrous skin adjacent to the scalp. The case would seem to have been made for a general warning against products that contain greases and oils. As a matter of fact, the shelves are now loaded with a wide variety of products that claim that they are "oil-free." "Oils," however, comprise a great variety of chemically unrelated materials. The implication is that viscosity (greasiness) alone determines comedogenicity. Morris and Kwan have also become vexed by contradictory reports and question the usefulness of the rabbit ear test for formulating non-comedogenic cosmetics (10). The present study was undertaken to seek a resolution of these disputes. Petrolatum was selected as the centerpiece of this investigation for two reasons: (1) Petrolatum is the