Parabens: A Review of Epidemiology, Structure, Allergenicity, and Hormonal Properties

Allison L. Cashman; Erin M. Warshaw


Dermatitis. 2005;16(2):57-66. 

In This Article

Patch Testing

Several different concentrations of individual parabens, as well as paraben mixes, have been used for patch testing. Several paraben mixes are currently commercially available and are listed in Table 3 . In the United States, the individual esters were typically tested at 5% until December 1984, when the concentration was decreased to 3%.[2]

Today, patch testing is usually performed with methylparaben, ethylparaben, propylparaben, and butylparaben combined in petrolatum, each at a 4% concentration for a total of 16%.[8] There is controversy over which concentration is the most sensitive and/or specific. Higher concentrations of a paraben mixture are often more irritating than individual esters are.[9,112] Most practitioners initially patch-test patients with a paraben mix and then, if the result is positive, test with individual paraben esters. Cross-reactions between the paraben esters are fairly common. One of the largest reported studies evaluating cross-reactions was by Menne and Hjorth. They performed routine patch testing on 8,020 patients from 1971 to 1986. Positive reactions (+, ++, or +++) to the paraben mixture were found in 2% of patients. Of the 76 patients who reacted positively to the paraben mixture, 60 were tested with the individual paraben esters. Two-thirds of the patients (40 of 60) reacted to one or more of the individual esters, including butylparaben (30 of 60 patients [50%]), propylparaben (21 of 60 [35%]), and benzylparaben (12 of 60 [20%]). All patients who reacted to propylparaben reacted to ethylparaben or butylparaben or both.[2]

Paraben-containing therapeutic agents that cause allergic contact dermatitis in paraben-sensitive patients often produce false-negative patch-test reactions. This is thought to occur because the paraben concentration in the commercial product is too low (typically < 0.3%) to induce a positive reaction on normal skin.[98] Several studies have indicated that lower concentrations of parabens used for patch testing result in a lower number of positive reactions. In the study by Menne and Hjorth, individual parabens were tested at 5% until December 1984, when the concentration was reduced to 3%. They reported that after the concentration was reduced, there were no +++ reactions, unlike what had been witnessed with the 5% concentration.[2] Hjorth and Trolle-Lassen described 19 patients who reacted to a 5% paraben mixture and who were subsequently tested with four paraben esters (methyl-, ethyl-, propyl-, and benzylparaben), each at four different concentrations (5%, 1%, 0.5%, and 0.1%). All patients reacted to one or several of the individual paraben esters at 5%, but lower concentrations induced fewer reactions. Positive reactions to the 1%, 0.5%, and 0.1% concentrations were found in 13 of 19 (67%) patients, 11 of 19 (58%) patients, and 5 of 19 (27%) patients, respectively.[113]

The efficacy of topical allergy screening systems relies on the ability of the allergens to separate from the vehicle, penetrate the stratum corneum, and activate Langerhans' cells. It is known that occlusion of the patch-test site increases the penetration of topical allergens.[114,115] Although there are no studies that have evaluated the impact of vehicle preparation or occlusion specifically in patch testing with parabens, some studies have indirectly indicated an effect. Occlusion increases hydration of the stratum corneum, from 5% to 50%.[115] Parabens are lipophilic, so their penetration should increase from the change in partitioning between the stratum corneum and viable epidermis during occlusion. This theory was tested by Cross and Roberts, who evaluated (1) the in vitro human epidermal penetration of a mixture of paraben ester preservatives in ointment, ethanol, and acetone vehicles and (2) the effect of a 10-hour occlusion period on the rate of delivery in each system. The occlusion increased diffusion eightfold in ethanol and 16-fold in acetone. Surprisingly, however, decreased diffusion and flux were reported in the ointment formulation. Additionally, there was a significant difference in the epidermal flux of the parabens when applied without occlusion in ointment, ethanol, or acetone compounds, indicating a role of the actual vehicle in the penetration of the allergen.[116] These results have notable implications regarding the use of the optimal topical allergy screening system, its vehicle, and occlusion.