THE SURVIVAL AND GROWTH OF MICROORGANISMS IN MASCARA DURING USE Louis A. WILSON, M.D., A. J. JULIAN, M.S., AND DONALD G. AHEARN, PH.D. In previous studies " 3 we demonstrated that all retailed eye cosmetics are typically free of significant microbial populations but are subject to microbial contamination during use. In used automatic-style mascaras, microorganisms were associated mainly with the applicator brush. In a few instances, species isolated from the mascara and from the applicator brush, a potential instrument of trauma to the eye, were the same species causing keratitis in the user. This report amplifies our previous investigations and considers the clinical significance of the use of contaminated cosmetics. MATERIAL AND METHODS Study groups To evaluate the susceptibility of eye cosmetics to contamination, new popular brand cosmetics were purchased, cultured for aerobic microorganisms, and, when observed to be free of contamination, distributed to nine study groups. Eight of nine study groups, composed of 3 student nurses and other college students, used a separate popular brand of cream-like eye mascara for nine- to -week periods. The ninth group included young women who used cake mascaras ; we examined the used cosmetics of 47 women, but only six women used a single brand of powdered mascara under study group conditions. The cosmetics were cultured biweekly and the history of product use recorded. Periodically, facial skin and fingers of the test group members were cultured for aerobic microorganisms during the study. In a few cases, the mascaras were examined after they had been From the Department of Ophthalmology, Emory University (Dr. Wilson), and the Department of Biology, Georgia State University (Mr. Julian and Dr. Ahearn), Atlanta, Georgia. Reprint requests to Louis A. Wilson, M.D., Department of Ophthalmology, School of Medicine, Emory University, Atlanta, G A 33. Atlanta, Georgia 96 in use for over six months. Used cosmetics obtained from patients of Emory University and Medical College of Georgia eye clinics were cultured for microorganisms. These cultures were compared with those obtained from the diseased outer eyes of the same patients. Testing procedure and media The mascaras were sampled by touching the applicator brushes with sterile Dacron swabs. For quantitating certain experiments, tared swabs were employed. The swabs, usually holding 3 to mg of mascara, were placed in test tubes containing ml of diluent (either sterile distilled water or isopropyl myristate) and the tubes were vigorously agitated with a mechanical mixer;. ml of a suitable dilution of this suspension was plated onto isolation agar. Mascara from within the container and powder cosmetics were sampled with an inoculating needle. When three successive samplings of a mascara yielded multiple colonies of the same organism, the mascara was withdrawn from use and cultured periodically over a storage period of at least 3 days. All other cosmetics were examined in storage at the termination of the study group. Tryptic Soy Agar (Difco) containing % sheep erythrocytes and Mycological Agar (Difco) acidified to ph 4. with lactic acid containing mg of chloramphenicol per liter were used to isolate bacteria and fungi. A more detailed description of microbiological methods was presented elsewhere. 3 RESULTS The density of aerobic microorganisms and their frequency in used automatic mascaras generally increased with increased usage (Table ). In general, the mascaras were sampled about six hours after their last use. The colony forming units per milli-
VOL. 79, NO. 4 MICROORGANISMS IN MASCARA 97 TABLE FREQUENCY OF MICROORGANISMS IN AUTOMATIC MASCARAS AFTER NINE WEEKS OF USE 3- No. Times Used 6-9- + 3 Days After Last Use Total Mascaras A, water miscible B, water miscible B, oily C, oily D, oily E, water miscible F, water miscible G, water miscible H, water miscible 9* 4 6 6 3 4 6 6 3 6 6 7 3 9 3 76 9 7 6 96 3 63 7 37 64 46 f 3 6t 7 4 7 6 6 7 * % of total mascaras yielding microorganisms, one mascara per individual. t % of mascaras yielding microorganisms 3 days after last use. t Microorganisms not isolated after 4 days. gram (cfu) in any individual mascara typically was less than cfu, but ranged to, cfu for certain brands after as few as two uses. Microbial populations fluctuated widely with highest numbers occurring in products sampled less than four hours after use and lowest numbers occurring 4 hours after use. If products were used about times (water miscible types products forming a fairly homogenous suspension in water) or about 3 times (anhydrous types), more consistent populations were noted. This was particularly true for product A. In one sample of product A, Klebsiella pneumoniae reached concentrations exceeding 3, cfu in less than three weeks of use. Products A and B (water miscible) and C (oily mascara) supported reproducing populations of bacteria, yielding microorganisms generally in increasing densities, 3 days after their last use. Product C demonstrated less than cfu at the beginning of the storage period and achieved a population of about cfu by 3 days. Portions of this anhydrous mascara after use were incubated on sterile slides positioned on a bent glass rod within a petri dish containing moistened filter paper. Within 7 hours, colonies of yeasts were observed developing within condensate on the mascara (Fig. ). The remaining products usually failed to yield microorganisms after ten days in storage. Exceptions were products E and F. A few samples of product E yielded Bacillus organisms for over a year in storage. The Bacillus organisms did not increase when introduced to new mascaras. Product F, during its nine weeks of usage, typically was sterile within four days of its last use. This was the only product that contained a combination of mercury (. to.7 parts per million), methylparaben Fig. (Wilson, Julian, and Ahearn). Development of colonies of Candida paropsilosis (arrows) on used mascara incubated 96 hours at 37 C with high humidity.
9 AMERICAN JOURNAL OF OPHTHALMOLOGY APRIL, 97 TABLE COMMON MICROORGANISMS OCCURRING IN MASCARAS USED FOR NINE WEEKS Species A B c D Brand E F G H Staphylococcus epidermidis Bacillus Micrococcus Corynehacterium Streptococci Nongroup A Group A Gram-negative rods Candida Molds Total samplings 64* 6 3 3 9 9 6 4 7 7 3 3 3 3 74 73 4 6 7 6 4 73 Incidence (%) per total number of samplings. (.4%), propylparaben (.7%), and propylene glycol (3%). The two products (A and C) supporting reproducing populations were purchased during 97 and did not contain preservatives on chemical analyses. In contrast, the other products contained concentrations of methyl- and propylparabens ranging from. to.9%, combined in some instances with up to.7 ppm mercury. During 973, new mascaras representing products A and C were redistributed to a few select individuals and samples were challenged with bacteria. None of these new mascaras supported reproducing populations of microorganisms after weeks of use. Preliminary analyses indicated that both products now contained methyl- and propylparabens plus mercury. Only a few cosmetics were observed more than six months. Of these, a single mascara of product E, used approximately times, yielded populations of Pseudomonas aeruginosa, Staphylococcus epidermidis, diphtheroids, and Candida paropsilosis, each in excess of S X IO 4 cfu after 3 days in storage. Other samples of this brand, used for only nine weeks with as high as 4 cfu at the beginning of a storage period, failed to yield microorganisms on culture within a ten-day storage period. A sample of product B that was used about months yielded established populations of 9 cfu of S. epidermidis. The most common microorganisms encountered in all brands used for nine weeks were ". epidermidis and Corynehacterium (Table ). These species, part of the normal flora of the outer eye, often occurred in the eyes and on the fingers of the test group members. Established pathogens such as S. aureus and C. albicans were isolated from the mascaras but they always occurred as a few isolated colonies and, with a single exception, were not obtained from the same mascara on two successive samplings. A budding cell of C. albicans was found on the hair of an applicator brush of a cosmetic stored for six months (Fig. ). Only this mascara supported a reproducing population of C. albicans, Forty-seven used, powder-based cosmetics representing 6 brands were examined for microorganisms. In general, these cosmetics were used longer than the automatic creamy mascaras, yet only 4 yielded microorganisms. Cephalosporium, Fusarium, Aspergillus, Pénicillium, Trichoderma, and Bacillus organisms were the most common contaminants encountered. Populations were usually slight but, in four instances, extremely high concentrations of fungal propagules repre-
VOL. 79, NO. 4 MICROORGANISMS IN MASCARA 99 senting the genera Fusarium and Pénicillium were cultured both from the eyelid margins and the cosmetics. Scanning electron microscopic studies of one of these cosmetics showed the typical conidia and conidiophore of Pénicillium organisms (Fig. 3). Growth of the fungi on powdered mascaras was generally found at the periphery and upper surface of the cosmetic. In a few instances the plastic containers holding the cosmetic appeared to be etched by fungal growth. Fungal contaminants in powdered mascaras remained viable and evidenced growth in storage, some for over a five-year period. Six women used the same brand of powdered mascara for nine weeks. Of these, only a single mascara yielded sparse bacteria Fig. (Wilson, Julian, and Ahearn). Budding cell of Candida albicans (arrows) on bristle of mascara applicator brush (X,). Fig. 3 (Wilson, Julian, and Ahearn). Conidiophores and conidia of Pénicillium species developing on powdered makeup (X 3,).
6 AMERICAN JOURNAL OF OPHTHALMOLOGY APRIL, 97 on each sampling. The average sampling time since their last use, six hours, was similar to that for the creamy mascaras. Populations of bacteria included S. epidermidis, Bacillus organisms, and diphtheroids in concentrations of less than cfu. Analysis of the powdered mascara gave concentrations of approximately.% of methyl- and propylparaben. Clinical observations No correlation of eye irritation with contamination of the mascaras used by the study groups was possible since the mascaras were removed from use when they exhibited significant populations of the same species on three successive samplings. During clinical examinations, females, ranging in age from 6 to 4 years, who were heavy daily users of eye area cosmetics had symptoms and clinical signs suggesting longterm infection of the outer eye. Cultures of the eyelid margins of a woman with a recent eye irritation and of one of her mascaras used months yielded confluent growth of S. epidermidi*. After discontinuing the use of the cosmetic and initiating normal daily facial washings including eyelid margin hygiene, the patient improved symptomatically. Cultures of the eyelid margin showed insignificant populations two weeks later. In storage the mascara demonstrated a reproducing population of S. epidermidis of over cfu. Seven patients with symptoms and clinical findings of chronic blepharitis were examined. Specimens of eye area cosmetics and the eyelid margins of four women (cosmetics of three were unavailable) yielded many S. epidermidis organisms. The regimen described above was instituted for the four women whose cosmetics were cultured and within one month three of the four showed marked improvement. The remaining patient required therapy with a topical antibiotic. The three patients whose cosmetics were not cultured were advised to discontinue use of their cosmetics and to apply topical antibiotics to the eyelid margins. All responded to treatment. Cultures of the eyelid margins of women with chronic blepharitis yielded confluent growth of S. epidermidis. The cosmetics of those women were contaminated with the same bacterium. Follow-up of these patients was not possible. Aside from a previously reported case of fusarial keratitis, no corneal ulcers could be directly related to the use of contaminated cosmetics. However, according to personal communications from A. J. Gay (June 97), F. M. Polack (August 974), and D. B. Jones (August 974), corneal ulcération by P. aeruginosa occurred in female patients who abraded a cornea while applying mascara. Four such cases were documented and in each instance P. aeruginosa organisms were cultured from the infected cornea of the patient and from her contaminated mascara. DISCUSSION Retail eye mascaras appear to be typically free of microbial contamination. The cosmetics are subject to contamination by the consumer but the efficacy of the cosmetic preservative seems to control the ultimate fate of the microorganisms. Water-based mascaras were most susceptible to contamination, but even anhydrous oily mascaras and pressed powders harbored microbes that readily developed with the addition of moisture. During normal use, changes in humidity and temperature provide moisture by condensation sufficient to support microbial growth. This was particularly true if preservatives were lacking, as in two of the three brands readily susceptible to contamination in the early phases of this study. The manufacturers have since added microbial inhibitors to these specific products. Most brands tended to harbor bacteria for longer time periods after they had been used repeatedly but during the nine weeks of observed use, they still retained the capacity to inhibit microbial growth. The diversity of species and density of populations typically were fewer than those noted previously in a more random sampling of used cosmetics.
VOL. 79, NO. 4 MICROORGANISMS IN MASCARA 6 The cosmetics examined in this earlier study were generally in use much longer than the nine weeks of this study. Observations of the microbial populations in a few mascaras over a six-month period, and preliminary results of the repeated challenge of mascaras with bacteria, indicate that the inhibitory capacities of various brands differ. Twenty-two women had symptoms and clinical signs of bacterial blepharitis and heavy densities of ^. epidermidis were isolated from their eyelid margins and eye cosmetics. Often considered nonpathogenic outer eye flora, a recent study has incriminated this species in staphylococcal blepharoconjunctivitis. 4 Our clinical findings support this observation and indicate that used mascaras should not be applied to hide the signs of this disorder. Indeed, such a practice may prolong the course of the disease and result in its recalcitrance to antimicrobial therapy. We previously advised avoiding eye area makeup prior to and following ocular surgery. Considering the frequency of S. epidermidis as a contaminant in used eye cosmetics and the recent report of two cases of postoperative endophthalmitis due to this bacteria, warning. it seems prudent to repeat this SUMMARY Over ISO mascaras representing eight popular brands were examined for their susceptibility to microbial contamination during their use by study group members. Additional mascaras from patients with symptoms and clinical findings of long-term blepharitis also were investigated. Early in the study, two brands without preservatives supported reproducing populations of microorganisms, including potential eye pathogens. These products, as currently manufactured, were recalcitrant to microbial attack. Microbes associated with the facial skin and fingers of the study group users were typically isolated from mascaras after use. Initial microorganisms isolated from mascaras were usually transients. Establishment of reproducing populations within the cosmetics appeared related to the number of uses, personal habits of the user, and the formulation of the product. Four patients with staphylococcal blepharitis and cosmetics heavily laden with Staphylococcus epidermidis showed marked clinical improvement when they stopped using the contaminated cosmetics. The application of used eye area makeup prior to and following ocular surgery should be avoided. ACKNOWLEDGMENTS This study was supported in part by funds awarded by the FDA (contract 3-74-6) to Emory University. We thank R. Gerdes, M.D., of ScanAtlanta for preparing the electron photomicrographs, and the FDA Laboratories, Washington, D.C., for chemical analyses of cosmetics. REFERENCES. Kuehne, J. W., Ahearn, D. G., and Wilson, L. A. : Incidence and characterization of fungi in eye cosmetics. Dev. Ind. Microbiol. :73, 97.. Wilson, L. A., Kuehne, J. W., Hall, S. W., and Ahearn, D. G. : Microbial contamination in ocular cosmetics. Am. J. Ophthalmol. 7:9, 97. 3. Ahearn, D. G., Wilson, L. A., Reinhardt, D. J., and Julian, A. J. : Microbial contamination in eye cosmetics. Contamination during use. Dev. Ind. Microbiol. :,974. 4. Valenton, M. J., and Okumoto, M. : Toxin producing strains of Staphylococcus epidermidis (albus). Isolates from patients with staphylococci blepharoconjunctivitis. Arch. Ophthalmol. 9:6, 973.. Valenton, M. J., Brubaker, R. F., and Allen, H. F.: Staphylococcus epidermidis (albus) endophthalmitis. Report of two cases after cataract extraction. Arch. Ophthalmol. 9:94,973.