1. ^*Laboratoire de Dermatochimie, Université Louis Pasteur, Clinique Dermatologique, Strasbourg Cedex, France
2. ^†Department of Dermatology, University Hospital of Leuven, Leuven, Belgium

Correspondence: Prof Jean-Pierre Lepoittevin, Laboratoire de Dermatochimie, Université Louis Pasteur, Clinique Dermatologique, CHU, F-67091 Strasbourg Cedex, France.

Received 7 April 2000; Revised 18 April 2000; Accepted 26 April 2000.

To the Editor:

Patch-testing is a well-established method of diagnosing allergic contact dermatitis (^{Wahlberg, 1995}). Patients with a history and clinical picture of contact dermatitis are re-exposed to the suspected allergens under controlled conditions to confirm the diagnosis. During the last few decades much effort has been put into standardization of allergens, vehicles, concentrations, patch test materials, and scoring of test reactions, and the method is today considered accurate and reliable. Based on patch-test results, advice is often given to the patient in order to avoid further contact with the offending chemical or product. Moreover, results obtained in large panels of patients are used to determine the prevalence of the most common contact allergens present in the standard series (^{Coenraads & Smit, 1995}). Therefore, the presence of sensitizing contaminant(s) in a patch-test material can lead to misdiagnosis and overestimation of the prevalence of a sensitization.

As part of a Research Institute for Fragrance Materials (RIFM) study,¹ we have been interested in the chemical composition of treemoss. Treemoss is a natural extract produced from a mixture of lichens, mainly Parmelia furfuracea (L.) Ach., growing on Pinus species. Due to the collection process of these lichens, small branches and pieces of bark are also present in the raw material. Chemical analysis of two ''treemoss absolute'' samples from Prodarom (Grasse, France), by fractionation on Sephadex LH20, esterification with diazomethane and Gas-Chromatography-Mass Spectrometry analysis, has shown the presence of a mixture of resin acids accounting for 11.4% (wt/wt) and 8.1% (wt/wt) of the material, respectively. As shown on the chromatogram (Figure 1), the major component is dehydroabietic acid (m/z = 314) together with abietic acid (m/z = 316) and its isomers (m/z = 316). Moreover, the presence of 7-oxo-dehydroabietic acid, an oxidation product of dehydroabietic acid, was detected at a concentration of 1.6% (wt/wt) and 1.1% (wt/wt) in the two samples, respectively. The source of resin acids in treemoss is probably the bark of the Pinus species colonized by P. furfuracea. Resin acids are the major constituents of colophony, a well-known skin sensitizer (^{Karlberg et al. 1985}). It has been shown that they were not directly sensitizers but could be transformed by air exposure into very potent sensitizers, such as 15-hydroperoxyabietic acid (^{Karlberg et al. 1988a}) or 7-oxo-dehydroabietic acid (^{Karlberg et al. 1988b}). We therefore were interested to see if ''treemoss absolute'' was able to elicit a positive patch-test reaction in colophony-sensitive subjects. Thus, 17 colophony-sensitive subjects were patch-tested with the two samples of ''treemoss absolute'' (2% and 1% in petrolatum). They were also patch-tested with two ''oakmoss absolute'' samples present in the European standard series: oakmoss (1% in petrolatum) from Trolab (Reinbek, Germany) and oakmoss (2% in petrolatum) from Chemotechnique (Malmö, Sweden). Out of 17 colophony-sensitive subjects, 12 (71%) reacted to treemoss, while nine (53%) additionally reacted to the ''oakmoss'' from Trolab, and two (12%) to the ''oakmoss'' from Chemotechnique (Table 1). It seems therefore that the content in resin acid and its oxidation products in the ''treemoss absolute'' samples was sufficient to elicit a positive reaction in most colophony-sensitive subjects. The high frequency of positive reactions to the ''oakmoss'' material present in the standard series and the discrepancy between the preparations from Trolab and Chemotechnique, was more difficult to understand. Samples of ''oakmoss absolute'' used for the manufacturing of Trolab and Chemotechnique patch-test material were thus obtained and subjected to chemical analysis. We found that the Trolab sample contained 5.6% (wt/wt) resin acids and 0.7% (wt/wt) of 7-oxo-dehydroabietic acid while the Chemotechnique sample contained less than 0.4% (wt/wt) resin acids. Interestingly, the chemical distribution of resin acids present in the Trolab sample (Figure 2) is very similar to the one found in treemoss. This suggests a contamination of ''oakmoss absolute'' by ''treemoss absolute''. Oakmoss is prepared by extraction of another lichen, Evernia prunastri (L.) Ach., mainly growing on oak trees, and has been used for decades by the perfume industry. Therefore, this material is present in the patch-test standard series, at a 1% concentration, as one of the ingredients of the fragrance mix (FM). The FM, even if it detects only 70%-80% of fragrance sensitive patients, is regarded as a marker for fragrance allergy and is used for both diagnosis and evaluation of the prevalence of allergic contact dermatitis to fragrances. The presence of resin acids and their oxidation products at a significant level in the ''oakmoss'' patch-test material, shown to be able to elicit a positive patch-test reaction in patients sensitive to colophony, could be responsible for misdiagnosis in some patients. There is therefore a need for a new patch-test material not contaminated by resin acids and for a reinvestigation of the prevalence of sensitization to oakmoss.

Figure 1.

Gas-chromatography analysis of the fraction of resin acids obtained from a treemoss extract by gel filtration on Sephadex LH-20 and esterification using diazomethane. Each peak was subjected to mass spectrometry analysis and the m/z ratio is indicated for major peaks. Chemical structures represent methyl esters of dehydroabietic acid (1) and abietic acid (2).

Full figure and legend (6K)

Figure 2.

Gas-chromatography analysis of the fraction of resin acids obtained from the ''oakmoss'' from Trolab by gel filtration on Sephadex LH-20 and esterification using diazomethane. Each peak was subjected to mass spectrometry analysis and the m/z ratio is indicated for major peaks.

Full figure and legend (5K)

Table 1 - The number of positive patch-test reactions to treemoss, oakmoss Trolab and oak moss Chemotechnique in 17 colophony-sensitive subjects.

Full table