Barrier damaging effects of n‐propanol in occlusion‐modified tandem repeated irritation test: Modulation by exposure factors and atopic skin disease

Recent studies provide evidence for significant and previously underestimated barrier damaging effects of repeated exposure to 60% n‐propanol in healthy skin in vivo.

To increase the antimicrobial activity and consequently, efficacy, in commercial preparations n-propanol is frequently combined with isopropanol (2-propanol) or ethanol and 30% or 45% concentrations of the irritant are found in some of the most common marketed formulations recommended for hand disinfection in hospital settings. 1,2 Compared with alkaline soaps and detergents, short-chain alcohols are considered to have a relatively low skin irritation potential. 3 However, in two recent studies, we found considerable and previously unknown negative effects of 60% n-propanol on the permeability barrier function and corneocyte surface topography in the healthy human skin in vivo. 4,5 Therefore, in the present study, we aimed to extend our previous findings by investigating the effects of the applied irritant con-   6 To be eligible for participation, the AD individuals had to be in a stage of clinical remission, defined as the absence of active eczema lesions within the test area or any other body site for at least 6 weeks prior to recruitment. UV exposure within the 6 weeks preceding the study, pregnancy, or lactation were defined as exclusion criteria. The protocol (No. 14-111) was approved by the Ethics Committee of the University of Lübeck and all volunteers gave written informed consent beforehand.

| Irritants and mode of exposure
Aqueous solutions (50 μL) of n-propanol (1-propanol; Merck, Darmstadt, Germany) in concentrations of 30.0%, 45.0%, 60.0%, or 75.0%, respectively, were applied on eight previously marked test fields on the upper-mid back on 4 consecutive days (D1-D4) two times daily for 30 minutes according to a recently validated occlusion-modified tandem repeated irritation test (om-TRIT) protocol. 5 The irritants were applied to the respective test fields at the same time of the day (±1 hour) using large Finn chambers (12 mm diameter; SmartPractice, Reinbek, Germany). On four of the test fields, exposure to the respective irritant concentrations was preceded by occlusion with distilled water for 30 minutes and on four fields the same irritant concentrations were applied without preceding occlusion; two adjacent fields, respectively, exposed to distilled water only (occlusion) or left untreated (normal skin) served as controls. The volunteers were allowed to take a shower as usual, while the use of skin care products or UV exposure in the test area was not allowed for the entire duration of the study (5 days).

| Bioengineering assessment of the skin irritant response
Visual scoring, non-invasive assessment of erythema, transepidermal water loss (TEWL), and skin hydration (capacitance) were used to monitor the irritant response. The assessments were performed before the first application of the irritants (baseline) and 96 hours later (ie, at the end of the study), before tape stripping. Visual scoring was based on Kligman and Frosch, with assessment of erythema, scaling and fissuring based on a 0-4, respectively 0-3 point scale. 7 Erythema was measured with the Colorimeter CL400 (CK Electronics, Cologne, Germany) and the values were expressed in the L*a*b* system; the a*-value on the green-red axis was used for assessment of erythema.

| Natural moisturizing factor analysis
The samples for stratum corneum natural moisturizing factor (NMF) analysis were taken 24 hours after the last irritant exposure on D5 using commercially available 14-mm D-SQUAME discs (CuDerm, Dallas, Texas) and stored in sterile 1.5-mL Eppendorf tubes (Eppendorf, Hamburg, Germany) at −80 C until analysis. Six samples per test field per volunteer were collected as previously described. 12 The stratum corneum NMF components (histidine, 2-pyrrolidone 5-carboxylic acid, trans-and cis-urocanic acid) on the tapes were extracted with 400 μL of 25% (wt/wt) ammonia solution, evaporated to dryness and reconstituted in 200 μL pure water before high-performance liquid chromatography-UV analysis. 13 Extracts from two D-SQUAME discs were pooled for analysis; the NMF levels were corrected for the amount of protein and expressed as mmol NMF/g protein.

| RESULTS
The erythema (a*-), TEWL, and capacitance values at baseline and after 96-hour repeated exposure to the different concentrations of npropanol with and without previous damage to the skin barrier by occlusion in the AD and healthy control groups are shown in Table 1.
Within the study groups there were no significant differences in the visual irritation score and the barrier function parameters between the test and control fields at baseline.
3.1 | Repeated occlusion and water exposure modulate the concentration-dependent barrier damaging effects of n-propanol on healthy and atopic skin Cumulative exposure to 45% n-propanol without previous barrier damage by occlusion led to significantly increased a*-values in the AD group after 96 hours compared with baseline (P < .01). In contrast, at the same time point there were no significant changes in the a*values of test field exposed to the same irritant concentration in the healthy controls (Table 1). Previous barrier impairment by repeated exposure to water/occlusion enhanced the irritant-induced effects and led to a significant a*-value increase after exposure to the lowest applied irritant concentration (30%) in both groups (P < .05 and P < .01 after 96 hours compared with baseline in the healthy control and AD groups, respectively). For both groups, no significant differences in the a*-values following occlusion alone were found.
The relative changes of the a*-values after 96-hour cumulative exposure to the different n-propanol concentrations, with and without previous occlusion, compared with baseline (Δa*) are shown in Figure 1A. In both groups the a-value* increased in parallel to the increasing applied irritant concentration. In the healthy controls, the differences in the Δa*-values between the previously occluded and non-occluded fields were significant for all studied irritant concentrations (P < .01, P < .0001, P < .01 and P < .001 for 30%, 45%, 60%, and 75% n-propanol concentrations, respectively; Table 2).
Repeated occlusion with water alone resulted in impaired epidermal barrier function and significant TEWL increase on D5 compared with baseline in both groups (P < .01 and P < .05 in the AD and healthy control group, respectively; Table 1). In the AD group, cumulative exposure to 30% n-propanol alone was sufficient to induce a manifest damage to the epidermal barrier and the TEWL values of the respective test field on D5 were significantly higher compared with baseline (TEWL on D1 and D5 4.81 ± 0.19 and 6.66 ± 0.36 g/m 2 /h, respectively; P < .001). In contrast, in the healthy controls cumulative exposure to 30% n-propanol led to a significant TEWL increase only if the barrier function was previously compromised by occlusion/repeated water exposure. In both groups, TEWL after exposure to 45%, 60%, and 75% n-propanol was significantly increased on D5 compared with D1 and the observed effect was independent of barrier damage by previous occlusion with water (Table 1). Barrier damage prior to n-propanol exposure enhanced the irritant-induced effects and in the healthy controls the relative TEWL increase on D5, assessed as ΔTEWL (ΔTEWL = TEWL D5 -TEWL D1), was significantly higher for all irritant concentrations ( Table 2). In the AD group, the differences in ΔTEWL between the previously occluded and nonoccluded fields on D5 were significant only after exposure to 30% and 45% n-propanol, respectively. In the presence of atopic skin disease, repeated exposure to n-propanol resulted in more severe impairment of the barrier function and on D5, ΔTEWL values of the test fields exposed to the same irritant concentration under the same conditions in the AD group were significantly higher than in the healthy controls ( Figure 1B).
In both groups, cumulative exposure to all studied n-propanol concentrations with as well as without previous occlusion led to a significant decrease of capacitance of the respective fields on D5 compared with D1 ( Table 1). The decrease in capacitance was concentration dependent in both the healthy and atopic groups as shown by the increasing Δvalues parallel to the increase of the applied n-propanol concentration ( Figure 1C). In AD, the Δvalues on D5 compared with baseline were significantly greater if exposure to 45%, 60%, and 75% n-propanol was preceded by occlusion (P < .01, P < .001, and P < .05 for 45%, 60%, and 75% n-propanol, respectively). In the healthy controls the differences between the previously occluded and nonoccluded fields exposed to the same irritant concentrations were significant only after exposure to 75% n-propanol ( Table 2). The relative decrease of stratum corneum hydration on D5 was more pronounced in the healthy control group; the differences between the healthy and AD groups were significant for the test fields exposed to 45%, 60%, and 75% n-propanol with as well without previous damage by occlusion/water exposure ( Figure 1C).
3.2 | Cumulative exposure to n-propanol reduces significantly the NMF levels in healthy and atopic skin There were no significant differences in the baseline NMF levels between the groups. On D5, in both groups the NMF levels of all irritant-exposed fields, with as well as without previous damage by occlusion/water exposure, were significantly lower compared with the control fields (non-exposed/normal skin, respectively, occlusion with water); at the same time point there were no significant differences between the control fields ( Figure 2). In the AD and the healthy control group, there were no significant differences in the relative reduction of the NMF levels between the previously occluded and the corresponding, non-occluded irritant-exposed fields. In contrast to the barrier function parameters, no significant differences in the relative NMF reduction of the test fields exposed to the same concentrations of the irritant under the same conditions between the healthy and the AD individuals were found.

| DISCUSSION
The aim of this study following the om-TRIT protocol was to investigate the relationship between the applied n-propanol concentration and the relative contribution of workplace-relevant co-exposures on T A B L E 1 Erythema (a*-value), transepidermal water loss (TEWL), and capacitance at baseline (D1) and after 96-hour (D5) repeated exposure to different concentrations of n-propanol (n-PrOH) with and without previous damage by repeated occlusion/water exposure in healthy controls (healthy; N = 20) and individuals with atopic dermatitis (AD; N = 20) Abbreviations: AU, Arbitrary units; ctrl, non-exposed (normal skin) site; Occl. Ctrl, occlusion with distilled water.
Note: Data are presented as mean and SEM. Level of significance P < 0.05. *P < .05. **P < .01. ***P < .001. shown to reduce the activity of phospholipase A2 (PLA 2 ), one of the key enzymes involved in the maintenance of barrier homeostasis and lipid processing in the skin. 19 In an earlier publication, we provided first evidence for significant reduction in the stratum corneum NMF levels after om-TRIT with n-propanol and/or SLS and demonstrated that short-chain alcohols may cause skin dryness through interaction with both the skin lipids and reduction of NMF. 5 The results of the present study confirm and extend these observations by showing a significant decrease in the NMF levels after exposure to all tested n-propanol concentrations in both healthy and atopic skin. The T A B L E 2 Differences in the Δvalues of the test fields exposed to the same irritant concentrations with and without previous damage by occlusion/water exposure in the healthy and atopic dermatitis (AD) groups  showed that atopic skin is more susceptible to damage by even low concentrations of weak workplace irritants, such as 2% acetic acid, that would not exert a significant negative impact on healthy skin. 33 In addition, in another study based on the same model, we observed significant differences in the severity of the barrier function impairment and inflammatory response induced by repeated single and concurrent exposure to detergents and alkaline agents in AD. 12 In the present study we found that cumulative exposure to 30% n-propanol, applied as a single irritant, was sufficient to induce damage to the epidermal barrier in AD, whereas the same exposure had no significant effect on healthy skin, unless the barrier function had been previously impaired. The pattern of skin reactivity to cumulative exposure to n-propanol was similar to the one observed in our previous studies and there were significant differences in the severity of barrier impairment of the test fields exposed to the same irritant concentration under the same conditions in the AD compared with the healthy control group. To the best of our knowledge, this is the first study on cumulative exposure to short-chain alcohols and at the same time first om-TRIT study in atopic skin, and its findings provide further experimental evidence for the increased susceptibility to irritant damage in AD.
The For correct interpretation of our findings, it is important to mention that the present study did not intend to analyze the cumulative effects of repeated exposure to short-chain alcohols in comparison with soaps or detergents, as these classes of irritants differ in terms of physicochemical properties, interact with different components of the epidermal barrier, and consequently exert distinct effects on the noninvasive parameters for assessment of the skin barrier function.
Although soaps and detergents maybe in general more irritating than alcohols, the multi-parametric approach with instrumental and biochemical parameters used in the present study provides solid evidence for significant barrier damaging effects of even low concentrations of n-propanol in vivo. The consistent effects found in the present and in our earlier investigations confirm the validity and reproducibility of the om-TRIT model for experimental studies on cumulative skin irritation in healthy and at-risk individuals. As in addition to n-propanol and mentioned, most of the marketed alcoholbased hand disinfectants contain also isopropanol or ethanol, it would be important to assess the cumulative effects of tandem exposure to different combinations of short-chain alcohols in future studies.