Fragrances are widely encountered in our daily environment and are known to be a common cause of allergic contact dermatitis. We have reviewed our patch test data from 1980 to 1996 to establish whether the pattern of fragrance allergy has changed with time. During this period, 25,545 patients (10,450 male, 15,005 female) were patch tested with the European standard series. The mean annual frequency of positive reactions to the fragrance mix was 8.5% in females (range 6.1-10.9) and 6.7% in males (range 5.1-12.9). Females were 1.3 times more likely to be allergic to fragrance (P < 0.001, 95% confidence interval, CI 1.17-1.41). Males with fragrance allergy were older than females by 5.6 years (mean age 48.2 vs. 42.6 years; P < 0.001, 95% CI 3.9-7.3). The incidence of a concomitant positive patch test to balsam of Peru in fragrance-sensitive patients showed wide variation, suggesting that it is not a reliable marker of fragrance allergy. There was a positive correlation between the isomers isoeugenol and eugenol. Oak moss remained the most common overall allergen throughout the study, positive in 38.3% of females and 35.6% of males who were tested to the constituents of the fragrance mix. During the period of the study the incidence of positive tests to oak moss increased by 5% yearly (P = 0.001, 95% CI 2.2-8.7). The frequency of allergic reactions to eugenol and geraniol remained relatively constant. Isoeugenol and alpha-amyl cinnamic aldehyde sensitivity increased and hydroxycitronellal showed a slow decline. There was a striking reduction in the frequency of sensitivity to cinnamic aldehyde (by 18% yearly; P < 0.001, 95% CI 14.3-21.0) and cinnamic alcohol (by 9% yearly; P < 0.001, 95% CI 5.2-12.9); these are now uncommon fragrance allergens. These data show temporal trends which may reflect the frequency of population exposure to individual fragrances.

Skin-sensitizing chemicals exhibit dose-response relationships for the elicitation of contact dermatitis. Previously, considerable work has been carried out in which the elicitation of allergic skin reaction has been examined as a function of the applied concentration. However, the relationship between exposure time, dose and response has not been explored in any depth. The present work has extended our initial assessment of the relationship between both exposure time and concentration for para-phenylenediamine (PPD) in a group of 19 PPD-allergic volunteers. The results clearly demonstrate that a relationship exists between both exposure time and concentration. Positive responses to PPD were directly proportional to exposure time: at 5 min 16% responded; at 15 min, 38%; at 30 min, 50%; and at 120 min, 69%. A similar direct relationship was found between concentration of PPD and response: after 120 min, 22% of patients had responded to 0.01%, and 69% to 1% PPD. All exposures for 1 and 2 min were negative. Subsequent evaluation using repeated 5 min open application testing demonstrated a cumulative effect, as after 8 days 39% of the panel reacted, more than double the number that reacted to a single occluded 5-min treatment. It was noted that there was marked subject variability in exposure time and dose required to elicit an allergic response. These results are of relevance for the general interpretation of patch test data, especially with regard to risk assessment.

Mechanisms underlying susceptibility to skin irritants are not clearly understood. Cytokines play a key role in inflammation, and functional polymorphisms in cytokine genes may affect responses to irritants. We investigated the relationship between polymorphism in the tumor necrosis factor (TNF) alpha-chain gene and responses to irritants. Volunteers (n=221) tested with sodium dodecyl sulfate (SDS) and benzalkonium chloride (BKC) were divided into responders and nonresponders and high and low irritant-threshold groups. DNA was assayed for the TNF-308 polymorphism by a polymerase chain reaction-restriction fragment length polymorphism method. There was a significant increase in the A allele (P=0.030) and AA genotype (P=0.023) in both the SDS low irritant-threshold group and in SDS responders (A allele P=0.022, AA genotype P=0.048). In the BKC low irritant-threshold group, we found a significant increase in the A allele (P=0.002) and AA genotype (P=0.016). Individuals with a low threshold to both irritants demonstrated a significant increase (P=0.002) in the A allele. This is the first description of a nonatopic genetic marker for irritant susceptibility in normal individuals. Genotyping for the TNF-308 polymorphism may thus contribute to screening of individuals deemed at risk of developing irritant contact dermatitis.

To study the length of exposure time required to elicit para-phenylenediamine (PPD) allergic reactions, patients known to be allergic to PPD were recruited and patch tested. A group of 7 patients were patch tested with 1% PPD in pet. for 15 min, 30 min and for 120 min. The remaining 9 patients were patch tested with 1%, 0.3%, 0.1% and 0.01% PPD for 15 min, 30 min and for 120 min each. With exposure for 120 min, 11 of 16 subjects reacted to 1% PPD and 2 of 9 reacted to 0.01%. With exposure of 15 min, 6 of 16 reacted to 1% PPD and 0 of 9 reacted to 0.01% PPD. This study showed marked inter-individual variability in eliciting a reaction to the PPD molecule on patch testing, with regard to both the exposure time and the concentration required.

The development of clinical dermatitis in prospectively followed subjects with greater irritant reactivity has not previously been identified. The association of greater irritant reactivity with a proinflammatory cytokine polymorphism may partly explain this. Further development of the irritant threshold test could contribute to the identification of non-atopic subjects at risk of occupational skin disease.