Ten Years of Patch Testing to Hydroperoxides of Limonene and Linalool: Prevalence, Patterns and Clinical Insights

ABSTRACT

Background

Limonene and linalool are widely used fragrances in cosmetic and household products. Upon air exposure, they autoxidise into potent sensitisers – hydroperoxides of limonene (Lim‐OOH) and linalool (Lin‐OOH).

Objectives

The objective of this study is to investigate patch test reactions to Lim‐OOH and Lin‐OOH, and associated characteristics, reaction patterns, including doubtful reactions and co‐reactivity with other fragrance allergens.

Method

A retrospective analysis was conducted using registry data from 6719 consecutively patch‐tested patients at Gentofte Hospital, Denmark (2014–2025). Doubtful reactions were clinically subclassified as erythematous, only (E) or infiltrated (I).

Results

The prevalence of positive patch test reactions was 1.6% for Lim‐OOH and 3.1% for Lin‐OOH. Positive Lim‐OOH reactions were less likely in men and in patients with atopic, occupational or hand dermatitis, whilst positive Lin‐OOH reactions were more likely in those > 40 years and less in atopic dermatitis. Most reactions peaked on day 3/4 and were weakly positive. Doubtful I reactions more frequently progressed to positive than doubtful E. Co‐reactivity was observed between Lim‐OOH, Lin‐OOH and other fragrance allergens. Lower test concentrations reduced diagnostic sensitivity.

Conclusion

Co‐reactivity with other fragrance allergens supports true allergenicity. The currently recommended patch test concentrations (0.3% Lim‐OOH and 1.0% Lin‐OOH) seem optimal and qualify for inclusion in the European baseline series.

Limonene and linalool hydroperoxides are potent fragrance sensitisers. Amongst 6719 patch‐tested patients, 1.6% reacted to Lim‐OOH and 3.1% to Lin‐OOH. Reactions were mainly weak and peaked at day 3/4. Co‐reactivity confirmed true allergenicity. Current test concentrations (0.3% and 1.0%) appear optimal for inclusion in the European baseline series.

Abbreviations

CI

95% confidence interval

Doubtful E

doubtful reaction, type erythema only

Doubtful I

doubtful reaction, type infiltration

EBS

European baseline series

ESCD

European Society of Contact Dermatitis

Lim‐OOH

hydroperoxides of limonene

Lin‐OOH

hydroperoxides of linalool

OR

odds ratio

pet

petrolatum vehicle

1. Introduction

Fragrance compounds are widely included in cosmetic products, skincare items and household cleaning agents, with limonene and linalool amongst the most frequently used [1, 2, 3]. Upon exposure to air, limonene and linalool undergo autoxidation leading to the formation of primary oxidation products, specifically the hydroperoxides of limonene (Lim‐OOH) and linalool (Lin‐OOH), which are potent allergens [4, 5].

Despite the widespread use of limonene and linalool in consumer products [1, 2, 3] and the strong sensitising potential of their primary oxidation products [4, 5], Lim‐OOH and Lin‐OOH are at present not included in the European baseline series (EBS) but referred to as “recommended additions” [6, 7]. This is partially due to the ongoing discussion regarding the optimal patch test concentration of both compounds, but also a substantial variability in reported patch test outcomes.

Patch testing with Lim‐OOH has been reported at three different concentrations: 0.1% [8, 9, 10], 0.2% [9, 10] and 0.3% Lim‐OOH [8, 11, 12, 13], yielding positive reactions in 1.2%–9.4% [12, 14] of patients, doubtful reactions in 0.4%–17.2% [9, 12] and irritant reactions in 0.1%–9.8% [11, 15]. For Lin‐OOH, patch testing has been reported at six different concentrations: 0.25% [9, 10], 0.33% [16, 17], 0.5% [9, 10], 0.66% [16], 1.0% [8, 11, 12, 13] and 1.8% Lin‐OOH [16], eliciting positive reactions in 0.8%–20% [16, 18] of patients, doubtful reactions in 0.5%–21.9% [9, 12] and irritant reactions in 0.2%–13.6% [9, 15].

Since December 2011, Lim‐OOH (0.3%) and Lin‐OOH (1.0%) have been routinely included in the patch test panels for all dermatitis patients referred on suspicion of allergic contact dermatitis to the Department of Allergy and Dermatology, Gentofte Hospital, Denmark. Since January 2022, lower test concentrations – Lim‐OOH at 0.2% and Lin‐OOH at 0.5% – were additionally introduced.

The objective of this study was to investigate patch test reactions including doubtful reactions to Lim‐OOH and Lin‐OOH in detail, focusing on their classification, variability across readings and patient's characteristics. The aim was to add knowledge to the assessment of their suitability for baseline testing.

2. Methods

A retrospective study was conducted using registry data from consecutive patients referred for patch testing at the Department of Allergy and Dermatology, Gentofte Hospital, Denmark, between January 2014 and May 2025. ¹ All patients referred during this period were eligible for inclusion, regardless of age.

Patch testing was performed in accordance with the guidelines of the European Society of Contact Dermatitis (ESCD) [19], using Finn Chambers (SmartPractice, Canada) with Scanpor (Alpharma AS, Norway) and Lim‐OOH and Lin‐OOH allergens in petrolatum (pet.) from Chemotechnique Diagnostics (Vellinge, Sweden). Paediatric patients under the age of 12 years were tested with allergEAZE chambers (SmartPractice, Canada) due to practicality. Clinical relevance was assessed following ESCD guidelines [19]. Past relevance was mainly based on the patient's history. Reactions were routinely assessed on day 2, days 3–4 and day 5–7. The maximum reaction was used as the final patch test result.

All patients (including paediatric patients) were consecutively patch tested with Lim‐OOH and Lin‐OOH, fragrance mix I and II, and other important fragrance allergens. A complete list of the investigated allergens can be found in Table S1. A small subgroup of patients (including two paediatric patients) with known contact allergy to either Lim‐OOH or Lin‐OOH was not tested with this allergen, resulting in different study populations for the two hydroperoxides.

During the entire study period, doubtful (?+) patch test reactions were subcategorised into two subtypes – E (erythema, only) and I (infiltration) – based on their clinical appearance, as an addition to the ESCD reading scale [19], illustrated in Table 1.

TABLE 1.

Categorisation of doubtful reactions in two clinical subtypes: E and I.

Abbreviation	Description	Example of corresponding clinical reaction
?+ E	Doubtful E (erythema, only): homogeneous or nearly homogeneous erythema, occasionally with few papules. No sharp margin.
?+ I	Doubtful I (infiltration): scattered infiltration with/without erythema.

Note: Doubtful reactions (?+) type E and I and examples of corresponding clinical reactions.

Patient characteristics were extracted according to the MOAHLFA‐index [20], which includes the following variables: ‘M’ – proportion of male patients; ‘O’ – proportion of patients with occupational contact dermatitis; ‘A’ – proportion with current or past atopic eczema; ‘H’, ‘L’ and ‘F’ – proportions with dermatitis involving the hands, legs and face respectively; ‘A’ – proportion of patients with age over 40 years. Paediatric patients were stratified into three age groups: 0–5 years, 6–12 years and 13–17 years.

If a patient was patch tested more than once, each test was considered separately during analysis. Any allergen that had previously elicited a positive (+, ++, +++) response was omitted from subsequent tests for that individual.

Cases with missing values on the variables of interest were excluded from the respective analyses.

Associations between patch test reactions to Lim‐OOH and Lin‐OOH and the demographic and clinical characteristics according to the variables of the MOAHLFA‐index were investigated using a multinomial logistic regression, adjusted for sex, age (> 40/≤ 40) and atopic dermatitis. The category of negative patch test results was used as the reference group. Results are reported as crude and adjusted odds ratios (ORs) with 95% confidence intervals (CIs).

Exact 95% CIs (Clopper–Pearson method) were calculated for proportions. Differences between proportions were assessed using Fisher's exact test. For statistical comparisons, doubtful cases were excluded when comparing positive versus negative patients, and positive cases were excluded when comparing doubtful versus negative patients. Pattern and progression of patch test results following the course of patch testing were visualised using Alluvial flow diagrams, and co‐reactivity was illustrated using Venn diagrams.

All statistical data analyses were performed using R (version 4.3.2). p‐values < 0.05 were considered statistically significant.

3. Results

From January 2014 to May 2025, 6856 patients were patch tested with Lim‐OOH and/or Lin‐OOH: 4635 women, 2221 men, including 226 paediatric patients (< 18 years). Amongst the 6856 patients, 6588 were tested once, 125 were re‐tested twice and 6 were re‐tested thrice during the study period.

Overall, 1.6% of patients (110/6829, CI 1.3–1.9) had a positive reaction to Lim‐OOH and 3.1% (215/6836, CI 2.7–3.6) to Lin‐OOH. In the paediatric population, 3.6% (8/225, CI 1.5–6.9) had a positive reaction to Lim‐OOH, and 1.8% (4/225, CI 0.5–4.5) to Lin‐OOH, compared with 1.5% (102/6604, CI 1.3–1.9) and 3.2% (211/6611, CI 2.8–3.6) respectively, in the adult population (Tables S2 and S3). No significant difference in the proportion of positive reactions between children and adults was observed (Lim‐OOH: OR = 2.35, CI 0.98–4.89; Lin‐OOH: OR = 0.55, CI 0.15–1.45).

3.1. Association to MOAHLFA‐Index

Associations to the MOAHLFA‐index are shown in Tables 2 and 3. A positive reaction to Lim‐OOH was significantly less likely in males and individuals with occupational dermatitis, atopic dermatitis or hand dermatitis compared to those with a negative reaction. The negative association with occupational and hand dermatitis remained significant after adjusting for sex, age and atopic dermatitis. A positive reaction to Lin‐OOH was significantly less likely in those with atopic dermatitis and significantly more likely in patients aged > 40 years.

TABLE 2.

Characteristics of patients with a positive or doubtful patch test reaction to hydroperoxides of limonene (Lim‐OOH) and associations to the MOAHLFA variables.

	Hydroperoxides of limonene (0.2% and 0.3% pet.)
	All patients tested, % (n), n = 6829	Doubtful reaction, % (n), n = 272	Positive reaction, % (n), n = 110	OR doubtful vs. negative (95% CI)	OR positive vs. negative (95% CI)	aOR a doubtful vs. negative (95% CI)	aOR a positive vs. negative (95% CI)
Male	32.4 (2216)	26.1 (71)	19.1 (21)	0.72 (0.55–0.95)	0.48 (0.30–0.77)	—
Occupational dermatitis	21.4 (1458)	18.0 (49)	11.8 (13)	0.80 (0.58–1.09)	0.48 (0.27–0.87)	0.82 (0.59–1.13)	0.49 (0.27–0.89)
Atopic dermatitis	23.4 (1599)	21.7 (59)	14.5 (16)	0.89 (0.67–1.20)	0.55 (0.32–0.94)	—
Hand dermatitis	41.5 (2833)	34.6 (94)	26.4 (29)	0.73 (0.56–0.94)	0.49 (0.32–0.76)	0.75 (0.58–0.98)	0.54 (0.35–0.83)
Leg dermatitis	1.5 (100)	1.1 (3)	2.7 (3)	—		—
Facial dermatitis	24.6 (1683)	25.0 (68)	30.9 (34)	1.03 (0.78–1.36)	1.38 (0.92–2.07)	0.98 (0.74–1.31)	1.33 (0.87–2.01)
Age > 40 years	62.5 (4269)	63.6 (173)	62.7 (69)	1.05 (0.82–1.35)	1.01 (0.68–1.49)	—
Age in years (mean ± SD)	46.7 ± 17.9	47.1 ± 18.3	47.1 ± 19.1

Note: Bold values are statistically significant (p < 0.05). Positive: +, ++ or +++ reaction.

Abbreviations: —, not applicable; aOR, adjusted OR; CI, confidence interval; n, number; OR, odds ratio; pet., in petrolatum; SD, standard deviation.

^a Multinominal logistic regression, adjusted for sex, atopic dermatitis and age (> 40/≤ 40).

TABLE 3.

Characteristics of patients with a positive or doubtful patch test reaction to hydroperoxides of linalool (Lin‐OOH) and associations to the MOAHLFA variables.

	Hydroperoxides of linalool (0.5% and 1% pet.)
	All patients tested, % (n), n = 6836	Doubtful reaction, % (n), n = 787	Positive reaction, % (n), n = 215	OR doubtful vs. negative (95% CI)	OR positive vs. negative (95% CI)	aOR a doubtful vs. negative (95% CI)	aOR a positive vs. negative (95% CI)
Male	32.4 (2213)	33.1 (261)	29.8 (64)	1.04 (0.89–1.21)	0.73 (0.50–1.04)	—
Occupational dermatitis	21.3 (1456)	19.6 (154)	16.7 (36)	0.88 (0.73–1.06)	0.75 (0.52–1.07)	0.91 (0.75–1.10)	0.85 (0.58–1.22)
Atopic dermatitis	23.4 (1601)	21.3 (168)	16.3 (35)	0.86 (0.72–1.03)	0.62 (0.43–0.89)	—
Hand dermatitis	41.4 (2832)	40.7 (320)	39.1 (84)	0.96 (0.83–1.12)	0.90 (0.68–1.19)	0.99 (0.85–1.16)	1.02 (0.77–1.36)
Leg dermatitis	1.5 (100)	1.7 (13)	1.9 (4)	—		—
Facial dermatitis	24.6 (1682)	23.1 (182)	27.9 (60)	0.92 (0.77–1.09)	1.18 (0.87–1.60)	0.93 (0.77–1.11)	1.18 (0.86–1.61)
Age > 40 years	62.6 (4277)	66.1 (520)	76.3 (164)	1.21 (1.04–1.42)	2.01 (1.46–2.76)	—
Age in years (mean ± SD)	46.7 ± 17.9	47.9 ± 18.1	51.2 ± 16.2

Note: Bold values are statistically significant (p < 0.05). Positive: +, ++ or +++ reaction.

Abbreviations: —, not applicable; aOR, adjusted OR; CI, confidence interval; n, number; OR, odds ratio; pet., in petrolatum; SD, standard deviation.

^a Multinominal logistic regression, adjusted for sex, atopic dermatitis and age (> 40/≤ 40).

3.2. Type and Progression of Doubtful Reactions

Overall, 4.0% (272/6829, CI 3.5–4.5) of patients had a doubtful reaction (E or I) as the maximum reaction to Lim‐OOH and 11.5% (787/6836, CI 10.8–12.3) to Lin‐OOH. Of the paediatric patients, 4.0% (9/225, CI 1.8–7.5) had a doubtful reaction as the maximum reaction to Lim‐OOH and 12.4% (28/225, CI 8.4–17.5) to Lin‐OOH.

The overall progression of patch test results of Lim‐OOH and Lin‐OOH is illustrated using Alluvial diagrams (Figure 1). Reaction patterns were similar for both hydroperoxides. Most positive reactions peaked at the second reading (day 3–4): 75.5% (83/110) for Lim‐OOH and 81.4% (175/215) for Lin‐OOH. If the late reading on days 5–7 had been omitted, 4.5% (5/110) of positive reactions to Lim‐OOH and 1.9% (4/215) to Lin‐OOH would have been missed. The majority of positive responses were of maximum weak (+) intensity (Lim‐OOH: 56.4% [62/110]; Lin‐OOH: 68.4% [147/215]), followed by strong (++) reactions (Lim‐OOH: 37.3% [41/110]; Lin‐OOH: 29.3% [63/215]). Extreme (+++) reactions were rare (Lim‐OOH: 5.5% [6/110]; Lin‐OOH: 0.9% [2/215]).

FIGURE 1.

Overall progression of patch test results to Lim‐OOH (a) and Lin‐OOH (b). Alluvial flow coloured according to strongest reaction. Patients with no reaction at all three readings are not shown in the figures. –, no reaction; +1, + positive reaction; +2, ++ positive reaction; D, day; E, doubtful E reaction; I, doubtful I reaction; IR, irritant reaction; Lim‐OOH and Lin‐OOH, hydroperoxides of limonene and linalool.

Doubtful I reactions were less frequent than doubtful E reactions but showed a progression pattern more similar to true positive reactions (Figure S1). Amongst reactions with a maximum doubtful E response, 66.8% (143/214) for Lim‐OOH and 50.9% (337/662) for Lin‐OOH peaked at the second reading (days 3–4). In contrast, 92.2% (47/51) of doubtful I reactions to Lim‐OOH and 86.8% (99/114) to Lin‐OOH peaked at the second reading. There was no manifest difference between Lim‐OOH and Lin‐OOH, between patients with atopic dermatitis and patients without atopic dermatitis, or between children and adults (Figures S2 and S3).

3.3. Relevance and Exposures

Amongst patients with a positive reaction to Lim‐OOH, 14.5% (16/110) had both current and past relevance, 42.7% (47/110) had only current relevance and 13.6% (15/110) had only past relevance. For Lin‐OOH, 15.3% (33/215) of patients with a positive reaction had both current and past relevance, 44.7% (96/215) only had current relevance and 10.3% (22/215) only had past relevance.

Exposure data were available for 79 Lim‐OOH‐sensitised and 179 Lin‐OOH‐sensitised patients with positive or doubtful reactions of current or past relevance (Figure 2). The most common exposure was a combination of leave‐on and rinse‐off cosmetics (44.3% [35/79] and 38.5% [69/179], respectively), followed by leave‐on cosmetics only (35.4% [28/79] and 32.4% [58/179]) and rinse‐off cosmetics only (12.7% [10/79] and 21.8% [39/179]). Cleaning agents and laundry detergents were less frequent sources. Multiple product category exposures were observed in 48.1% (38/79) of patients with a relevant reaction to Lim‐OOH and in 44.1% (79/179) of patients with a relevant reaction to Lin‐OOH.

FIGURE 2.

Sources of exposure to (a) Lim‐OOH and (b) Lin‐OOH. Exposures to Lim‐OOH and Lin‐OOH amongst patients with a doubtful or positive reaction with current or past relevance. Lim‐OOH and Lin‐OOH, hydroperoxides of limonene and linalool.

3.4. Co‐Reactivity to Other Fragrance Allergens

Of the 110 patients with a positive reaction to Lim‐OOH, 40.9% (45/110) also exhibited a positive reaction to Lin‐OOH, and 22.7% (25/110) had a doubtful reaction to Lin‐OOH (Figure S4a). Of the 215 patients positive to Lin‐OOH, 20.9% (45/215) also had a positive reaction to Lim‐OOH, and 13.5% (29/215) had a doubtful reaction to Lim‐OOH (Figure S4b).

Co‐reactivity to other fragrance allergens was observed in 44.5% (49/110) of patients with a positive reaction to Lim‐OOH and 22.4% (61/272) with a doubtful reaction to Lim‐OOH (Figure 3a). For Lin‐OOH, co‐reactivity to other fragrance allergens was found in 44.7% (96/215) of patients with a positive reaction and 21.0% (165/787) with a doubtful reaction to Lin‐OOH (Figure 3b), compared to 12.1% in the overall population.

FIGURE 3.

Co‐reactivity of Lim‐OOH (a) and Lin‐OOH (b) to other fragrance allergens. Lim‐OOH and Lin‐OOH, hydroperoxides of limonene and linalool; n, number of patients. A complete list of the ‘other fragrance allergens’ is described in Table S1.

Both positive and doubtful reactions to Lim‐OOH or Lin‐OOH were associated with significantly higher co‐reactivity to other fragrance allergens compared to negative patients (Lim‐OOH positive: OR = 6.41, CI 4.27–9.57; Lim‐OOH doubtful: OR = 2.31, CI 1.69–3.12; Lin‐OOH positive: OR = 7.47, CI 5.57–10.01; Lin‐OOH doubtful: OR = 2.46, CI 2.02–2.99).

3.5. High and Low Concentrations

When evaluating patch test concentrations, the lower doses of Lim‐OOH (0.2%) and Lin‐OOH (0.5%) failed to identify 55.6% (5/9) and 59.3% (16/27) of patients respectively, who reacted positively at the higher concentrations (0.3% Lim‐OOH and 1% Lin‐OOH). Differences in reactivity between concentrations are presented in the cross‐tabulated Tables S4 and S5.

4. Discussion

The overall prevalence of positive patch test reactions to Lim‐OOH and Lin‐OOH was 1.6% and 3.1% respectively, consistent with some previous studies [8, 10, 11, 12, 13]. A potential explanation for the higher proportions of positive reactions to Lim‐OOH and Lin‐OOH reported in other centres [11, 12, 18, 21] could be the inclusion of doubtful reactions as positive.

The numbers were reversed in the paediatric group, with 3.6% positive to Lim‐OOH and 1.8% to Lin‐OOH; however, this difference was not statistically significant, most likely attributed to the limited sample size of the paediatric cohort. The reversed results may reflect differing exposures; Bennike et al. [1] identified linalool as the most labelled fragrance allergen in 5588 cosmetic products in Denmark, whereas Botvid et al. [2] found limonene to be the most frequently labelled fragrance allergen in 1179 cosmetic products specifically for children. Our results are further supported by the findings of Pesqué et al. [22], who reported the highest prevalence of Lim‐OOH sensitisation in young children, with rates decreasing with increasing age [22], whilst Sukakul et al. [11] found that sensitisation to both hydroperoxides was more prevalent in patients under 40 years, although this difference was statistically significant only for Lim‐OOH.

Analysis of the progression of patch test results identified D3/4 as the optimal timepoint for detecting positive patch test responses. Evaluation of the two types of doubtful reactions indicated that the type I reactions possessed a pattern more similar to a positive reaction than the type E. We find this plausible, as infiltration is one of the most important characteristic features of true contact allergy [23, 24, 25]. Interpretation of patch test results is difficult – especially interpretation of weak positive/doubtful patch test reactions [26, 27, 28]. Meticulous interpretation of patch test results, by use of current international guidelines (including timepoints for readings) is essential. We suggest this sub‐classification of doubtful reactions (E/I) to aid diagnostic decision‐making and avoid false positive or negative outcomes.

Doubtful I reactions appeared more homogeneous and resembled the pattern of positive reactions, suggesting that they may represent weak allergic responses. In contrast, doubtful E reactions were heterogeneous and less easily interpreted. Our findings indicate that doubtful I reactions should not be disregarded; if consistent with the patient's history, symptoms and exposure, additional testing such as a repeated open application tests (ROAT) should be considered. Christensson et al. [16] tested four concentrations of linalool hydroperoxides (0.25%–1.8%, corresponding to 2%–11% oxidised linalool). The 1.0% Lin‐OOH (6% oxidised linalool) elicited the highest number of weak positive reactions with few doubtful or irritant responses, whereas 1.8% increased the number of strong positive, doubtful and irritant reactions. These findings further suggest that some doubtful reactions may represent low‐grade allergic responses. Nevertheless, an even higher concentration might not only increase the risk of active sensitisation but also reduce pH, thereby enhancing skin irritation. The 1.0% Lin‐OOH and 0.3% Lim‐OOH concentrations were found to be the most suitable in this study and remain the most extensively explored [29] patch test concentrations to date.

Our findings also indicate that testing with lower concentrations (0.1% Lim‐OOH and 0.5% Lin‐OOH) leads to a substantial proportion of overlooked cases – 55.6% for Lim‐OOH and 59.3% for Lin‐OOH respectively. Therefore, testing at higher concentrations for both substances is recommended to improve detection of fragrance contact allergy and reduce false negatives. This should also apply to testing children according to our data. In a recent multicentre European study, the recommendation to test children with lower concentrations was made solely based on high rates of positive reactions in the youngest age group [30]. However, the study did not present centre‐specific data, nor were doubtful reactions or clinical relevance addressed. Ogueta et al. [29] argued that, as patch testing at higher concentrations (0.3% Lim‐OOH and 1.0% Lin‐OOH) frequently results in doubtful or irritant responses that are difficult to interpret, these hydroperoxides should be reserved as recommended additions for experienced clinicians.

Cosmetic products were the main source of exposure to both Lim‐OOH and Lin‐OOH. This aligns with a recent study of contact dermatitis amongst adolescents in Denmark, in which 33.8% reported a history of contact dermatitis linked to use of a cosmetic product [31]. In the current study, most sensitised patients had been exposed to a combination of stay‐on and rinse‐off products, representing cumulative exposure. Exposure to a mixture of fragrance allergens is known to cause both additive and synergistic effects during sensitisation and elicitation of allergic contact dermatitis [32, 33, 34].

Accurate measurement of hydroperoxide isomers in cosmetics is limited by low thermal stability, lack of chromophore and interfering secondary oxidation products [35]. Detection requires sensitive techniques such as two‐dimensional HPLC coupled with electrospray ionisation tandem mass spectrometry (HPLC/ESI‐MS/MS) [35]. Clinically, Lin‐OOHs as low as 0.2 μg/g in shampoo have been associated with eyelid dermatitis in a child [36], whilst 8.4 μg/g Lin‐6‐OOH and 5.6 μg/g Lin‐7‐OOH in deodorant have been linked to adult dermatitis using the 2D HPLC‐MS method [37].

Our data indicate a notable co‐reactivity between Lim‐OOH and Lin‐OOH, further indicating a similar pattern of sensitisation by coupled exposure via cosmetic and household products [1, 2, 3, 38, 39, 40, 41]. The substantial co‐reactivity with other fragrances – 44.5% of patients positive to Lim‐OOH and 44.7% positive to Lin‐OOH – supports the classification of these two hydroperoxides as true fragrance allergens. This co‐reactivity was also reported by Sukakul et al. in a study of Swedish patients [42].

Both hydroperoxides fulfil the main criteria of inclusion of a substance in the EBS [43] – the substances (limonene and linalool) are common in the environment [1, 2, 3]; the results of testing are reliable (few severe (+++) reactions) [13], and have been assessed by ROATs with different vehicles [4, 44, 45]; and relevant in the patient population ≥ 0.5% – confirmed in this study. Hence, we recommend inclusion of 0.3% Lim‐OOH and 1.0% Lin‐OOH to the EBS, alongside meticulous interpretation of test results – preferably by using our sub‐classification of doubtful reactions – to avoid misclassification.

The strength of this study includes a large dataset with what we consider high quality data. Patch testing procedures, readings and registration of data were performed by specialised staff following uniform criteria and supervision. Additionally, the used allergens and patch testing equipment were purchased from the same distributors during the entire study period. However, fewer patients were patch tested during the COVID pandemic of 2019–2020. Data of the two lower concentrations of Lim‐OOH and Lin‐OOH were only available from 2022.

5. Conclusion

Positive patch tests to Lim‐OOH and Lin‐OOH are frequent; they have the characteristics of allergic reactions. Co‐reactivity with other fragrance allergens supports their role as true contact allergens and inclusion in the EBS for screening of fragrance contact allergy. Based on our finding, we recommend patch testing with Lim‐OOH at 0.3% and Lin‐OOH at 1.0%, including for paediatric patients, as lower concentrations reduce diagnostic sensitivity. Careful interpretation of test results is essential to avoid misclassification.

Author Contributions

Rebekka Søgaard: conceptualization, data curation, formal analysis, investigation, methodology, validation, visualization, writing – original draft, writing – review and editing. Sofia Botvid: conceptualization, investigation, methodology, project administration, validation, visualization, writing – original draft, writing – review and editing. Jeanne Duus Johansen: conceptualization, funding acquisition, investigation, methodology, writing – original draft, supervision, writing – review and editing. Jakob Ferløv Baselius Schwensen: methodology, supervision, writing – review and editing.

Funding

This work was supported by The Danish Environmental Protection Agency under the Danish Ministry of Environment and Gender Equality.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Consecutively patch‐tested fragrances at Gentofte Hospital from 2014 to 2025.

Table S2: Paediatric patch test results for Lim‐OOH (0.3% and 0.2%), stratified into three age groups: 0–5, 6–12 and 13–17 years of age.

Table S3: Paediatric patch test results for Lin‐OOH (1% and 0.5%), stratified into three age groups: 0–5, 6–12 and 13–17 years of age.

Figure S1: Progression of doubtful patch test results of Lim‐OOH (a and b) and Lin‐OOH (c and d).

Figure S2: Progression of patch test results of Lim‐OOH in patients with atopic dermatitis (a) and in children (b).

Figure S3: Progression of patch test results of Lin‐OOH in patients with atopic dermatitis (a) and in children (b).

Figure S4: Co‐reactivity of Lim‐OOH (a) to Lin‐OOH (b) and vice versa.

Table S4: Comparison of patch test results for hydroperoxides of limonene at 0.3% and 0.2%.

Table S5: Comparison of patch test results for hydroperoxides of linalool at 1% and 0.5%.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.